Palit Group and Galax have announced that the latter's GPU brand will continue to operate and release hardware, despite reports to the contrary following an announcement of an internal restructuring at Palit Group. It follows reports of Galax ceasing operations. The initial announcement came from Ronaldo Buassali from TecLab, part of Galax Brazil, whose now-deleted X post suggested Galax had exited the market amidst the change. Both Palit and Galax have come forward to clarify that, apart from the internal restructuring, nothing is happening to the Galax brand.

Palit Group has owned Galax since 2007, but the manufacturer has operated independently. Now, it's moving under the direct control of Palit Group, consolidating support, logistics, and R&D to streamline the business and make it more efficient.

Galax's management will be centralized at Palit Group headquarters, where it will run alongside Palit. Its iconic GPUs, such as the Hall of Fame (HoF) lineup, don't seem to be going anywhere.

In its official statement, Palit said, "GALAX is not ceasing operations. We remain fully committed to the development, production, and support of our high-performance hardware. [...] The recent internal transitions are part of a pre-planned global initiative to integrate brand management for enhanced operational efficiency and cross-departmental synergy."

Galax has been headquartered in Hong Kong since its inception, but with the new management, it will now be run directly out of Palit in Taiwan. "Building upon the successful integration of internal operations between GALAX and Palit, the management of the “GALAX,” “KFA2,” and the flagship “HOF” (Hall of Fame) brands is now being centralized at the Palit Group headquarters," confirmed Palit.

To further back all this up, Galax came forward with its own statement clarifying that it is "continuing to develop, produce, and support [our] award-winning high-performance hardware."

If you go to Galax Brazil's website right now, an official communication notice is still up informing customers that Palit will now handle all RMA and warranty concerns; the global site hosts no such notice. "This move is designed to strengthen the brand’s global presence, not diminish it," according to Galax. It is unclear if any employees have been laid off in the process; however, the aforementioned note from Galax Brazil noted that employees had lost access to systems.

For the first time since its launch in January, Nvidia's GeForce RTX 5090 has fallen below its MSRP thanks to this discount at OverClockers in the UK, with broader savings also available in Europe, a sign that the great GPU drought might be coming to an end.

The RTX 5090 is the best Graphics card on the market if you want extreme performance and a price tag to match. With its 32GB of memory, enormous bandwidth, and excellent performance, all of its flaws mostly revolve around the fact that, up until now, it has been extremely hard to get hold of and, as a result, very expensive.

Now, you can buy one at OverClockers UK for £1,879.99. Less than the MSRP of the Founders Edition and £280 less than the list price of this particular Palit version.

Beyond the UK, there are also signs of a general softening of the RTX 5090 market across Europe. A similar dip below MSRP has recently been reported in Finland by VideoCardz. As with this UK discount, the margins of discount below MSRP are very slim. Rather, it's the landmark moment of the card dropping below MSRP for the first time that has caught our attention.

Here at Tom's Hardware, we're observing multiple UK stockists with decent amounts of 5090s, and OC's website claims it has shifted more than 30 of these discounted cards today alone.

5090 prices in the U.S. remain exorbitantly high, unfortunately. The cheapest card in stock at Newegg, for instance, is this $2,919 Gigabyte version, fully $900 more than the Founders Edition MSRP.

With the gorgeous weather in the UK today, staying inside and playing computer games is probably not the best plan, plus, with the heat these massive graphics cards kick out, you do not want to be trapped in a room with them. If, however, you've been waiting with bated breath to get your hands on one of Nvidia's RTX 5090 behemoths, now's your chance.

It's finally back in stock and not on sale for a horrendously over-inflated MSRP price (the original MSRP is egregious enough), you can pick up this GPU with a slight discount and also grab a free premium edition of DOOM: The Dark Ages game up until May 21st.

This deal is available at Overclockers, where the Palit GeForce RTX 5090 GameRock graphics card is on sale for £2,099. This is a slight discount from the list price of £2,159 and the cheapest price I can find for this giant GPU in the UK.

The RTX 5090 comes with an uncompromising 32GB of GDDR7 VRAM on a 512-bit memory interface, which means you get a whopping 1792GB/s bandwidth capability. The core clock of this card is 2017MHz, with a boost clock of up to 2407MHz, using its 21760 CUDA cores to power your games.

The Palit GameRock RTX 5090 is contained in a large 3.5-slot chassis with triple-fan cooling and large heatsinks. It's brightly coloured, and will stand out in any build. You will need a large PC case to house this beast and a decent power supply to feed its 600W needs and cover transient power spikes.

If you're considering an RTX 5090 in your system, please check out our review of the Nvidia RTX 5090 and see how it performs across our gaming benchmarks, compare it against the competition, and previous generations of GPUs in our GPU Hierarchy.

Gainward and Palit have released their respective Small Form Factor Ready RTX 40 series, adhering to Nvidia's definition of an SFF-friendly form factor for GPUs and PC cases. Palit classifies its SFF variation within the 'Infinity 3' series and Gainward introduces it under 'Python-III' branding.

The cards on offer are analogous, which shouldn't be surprising as the respective brands belong to the same parent company- Palit Group. While having multiple brands for the same AIC partner isn't exclusive only to this group, it's rare to see both brands use the same design and specifications, including clock speed. 

With a uniform design consisting of three cooling fans and no RGB. Ideal for those who prefer a more subdued aesthetic. There is a total of twelve SKUs involving GeForce RTX 4080, RTX 4070 Ti Super and RTX 4070 Super as a non-overclocked or overclocked variation. The overclocked RTX 4080 Super 16GB is clocked up to 2580 MHz, with RTX 4070 Ti and RTX 4070 Super up to 2640 MHz. The only difference is the packaging. Neither of the companies mentioned the warranty period of these variations either on the site or in its online manual.

Palit claims on its website that the Super Infinity 3 series is 40% smaller than its standard-size RTX 40 series cards- which also applies to its Gainward SSF-friendly counterpart. The variations from both brands measure 294 x 116 x 49.5 mm- which is within Nvidia's 304 X 151 X 50mm spec. That said, this only means it qualified to be compatible with those small form factor cases that are designed to accommodate Nvidia's standard which must have a minimum clearance of 312 x 154.5 x 50mm.

While having a smaller profile than a conventional variant is appreciated, users question the obfuscation of what 'SFF friendly' should be and if Nvidia explored this category before setting the dimension standard and passing it on to respective AIC partners and PC case makers. While having a set standard makes sense for a certain form factor, having a smaller profile doesn't necessarily mean it's compatible with a mini-ITX case. Some would also agree that contrary to what Nvidia has set, it doesn't truly help SFF builders and enthusiasts. That said, the SFF category itself is broad as it involves users with different needs- HTPC, general purpose, gaming- or all of the above.  

Whether Nvidia would revisit this decision for the upcoming RTX 5000 series remains to be seen. Some AIC partners might venture to have a proper mini-ITX-friendly graphics card, such as how Zephyr did with its RTX 4070 ITX Sakura Blizzard GPU. But would need respective makers to experiment and implement. Naturally, it depends on whether respective manufacturers could- and for which model. 

A few months ago, Asus introduced its variants of Nvidia's SSF-friendly graphics card, the Asus Prime which cover the 4000 range from the 4060 Ti to the 4080 Super. We may see more companies bringing its Nvidia SFF-ready graphics cards, or simply confirm if its existing SKUs are within Nvidia's preferred SFF-friendly dimensions like how MSI did.

Palit Group, a specialized GPU manufacturer and one of Nvidia's biggest AIB partners, just previewed a hybrid GPU cooler on social media. The company shared a 12-second clip of a Game Rock Nvidia GeForce RTX graphics card that has triple cooling fans and a water block attached. The clip thus seemingly shows a video card that uses a hybrid water and air cooling solution to keep its temperatures down.

While we don't have much information about the GPU yet, Palit's Game Rock GPU line often consists of performance cards, with the 4070 Ti and more powerful cards in the family. GPUs from the higher end of the spectrum usually draw more power and often run hotter, so we would expect RTX 4080 (Super) and RTX 4090 cards from Palit to have this dual cooling solution as an option.

If graphics cards don't have adequate cooling, they can soon hit thermal limits, which in turn limits their performance - not nice. However, if you keep a GPU's temperature low enough, you can maintain high boost clock speeds, or even achieve overclocking records. Palit could also be developing this hybrid cooling solution in preparation for the upcoming RTX 5000 series GPUs we expect to arrive this year.

Palit isn't the first company to produce a hybrid cooling solution for a GPU. Asus launched a 3080 Ti that used a similar system in 2021, and even EVGA offered a hybrid-cooled RTX 2080 Super back in 2020. You could even get a third-party hybrid GPU cooler from Arctic way back in 2015 if you wanted to replace the stock cooler on your video card.

Nevertheless, this teaser is the first time that Palit will try this hybrid setup for its GPUs. But while using both air and water cooling will likely increase the performance of any graphics card, the question remains whether the larger space that the video card will take up will allow it to fit in your case, plus the additional cost you will have to pay, will make it worthwhile.

We will know more about this new hybrid GPU at Computex 2024. However, we do not recommend buying the card right at launch — it's better to wait for reviews and real-world performance numbers before purchasing. That way, you're sure you're getting one of the best graphics cards and the hybrid cooling solution is worth paying a premium for. 

Palit has built a smaller profile, fanless RTX 3050 KalmX 6GB GDDR6 with reduced VRAM and CUDA cores. Interestingly, this GPU has a DVI port, which is likely for those with older systems and monitors looking for a replacement graphics card. Although the GeForce RTX 3050 6GB can't compete with the best graphics cards, it retains 2nd gen RT and 3rd gen Tensor cores with DLSS, RTX, AV1 decoding, and NVENC encoding/decoding.

Along with the reduced VRAM capacity, the GPU's memory bus is 96-bits. CUDA cores are reduced to 2,048 from the 2,560 typically found in the standard RTX 3050 8GB configuration introduced two years ago. Because of these step-down specs, Palit could get this card's TDP to 70W, which otherwise would be 130W. The only PSU requirement is 300W, with no auxiliary power connectors. Due to the lowered configuration, it shouldn't be surprising that this card will be cool enough for a passive cooler, valid to its sub-branding 'KalmX.'

This GPU is a dual-slot card with HDMI 2.1, DP 1.4a, and dual-link DVI for video output. Palit's RTX 3050 6GB form factor would make it ideal for HTPCs in SFF cases, as it measures 16.3 x 13.7 cm. This probably would be excellent for those who wish to replace aging GPUs like the GTX 950 and GTX 1050 Ti.

MSI also has two new variants, where ComputerBase received the RTX 3050 Ventus 2x 6G OC for a review. The other version is the RTX 3050 LP 6G OC. Both have different coolers and PCB sides, but with 20 MHz clock speed added over the default clock speed while maintaining 70-watt TDP. Based on its initial testing, the RTX 3050 6G loses more than 20% performance compared to the RTX 3050 8G.

This isn't bad since the GPU has significantly reduced TDP and power consumption, which is suitable for the intended audience base that is unlikely to play new (or dated) PC games. German-based Mindfactory also shows it is selling the RTX 3050 6G made by both MSI and Palit from €189 onwards, which is about US$ 204.24, and the cheapest RTX 3050 8G costs € 239.93, viz US$ 259.28. It's safe to say this should be available in other retail channels from respective countries and AICs.

Nvidia has quietly launched its newest entry-level GPU, the GeForce RTX 3050 6GB (h/t TechPowerUp). This out-of-step release (we are still waiting for an RTX 4050 for desktops) cripples the original RTX 3050 8GB by reducing the VRAM quota and providing a narrower memory bus. Moreover, the new RTX 3050 6GB models are based on a smaller GPU, Nvidia’s GA107, with fewer CUDA than the original 8GB model. However, two positive things come from these cuts: a cheaper MSRP and a significantly lower TBP.

Before further discussion, let's compare the two desktop GeForce RTX 3050 graphics cards as straightforward as possible via a table. Below is a side-by-side comparison of the key specs of the new RTX 3050 6GB, which is starting to become available in retail, and the model we reviewed almost exactly two years ago.

*A GeForce RTX 3050 8GB OEM model was also released based on the smaller GA107 but was far closer in specs to the original GA106 model.

Looking at the big picture, the new desktop graphics card seems to be drastically cut down from the GeForce RTX 3050 8GB that launched two years ago. We question whether this 6GB card should be named an RTX 3040 6GB instead. It has cuts to so many performance-impacting specs like memory capacity, bandwidth, GPU cores, and GPU clock speeds that it's expected to have a distinctly different character. We will find out exactly how different it is when we get one of these new models for a detailed examination and extensive real-world testing.

In addition to the drawbacks, there are some positives to the new GeForce RTX 3050 6GB model. Probably most important for Nvidia’s product planning is that the new model is supposed to be about $60 cheaper than the GeForce RTX 3050 8GB models we spied on Amazon today. The efficient GA107 with lower clocks and cut-down memory subsystem means that these new RTX 3050 6GB cards can run at spec in under 75W, meaning they can be used in systems without a spare power connector.

While TPU doesn't mention any new RTX 3050 6GB models being sold by retailers, tracking down some Nvidia partner models online isn't difficult. Austria's e-Tec is selling MSI's GeForce Ventus 2X 6G OC model, Shopee in Taiwan has a listing for an Asus Dual GeForce RTX 3050 V2 OC 6GB model, and Mindfactory has an MSI GeForce RTX 3050 LP 6G OC (low profile), an MSI GeForce RTX 3050 Ventus 2X 6G OC, a Palit GeForce RTX 3050 KalmX 6GB (passive, and spotted previously) and a Palit GeForce RTX 3050 StormX 6GB listed. We should be getting low-profile, fanless, and other compact models of this GPU, thanks to its low TBP.

That isn’t the end of the early listings, another interesting thing was spotted by Twitter / X’s Harukaze5719: a Chinese listing for an RTX 3050 6GB that shows a bare PCB, with no external power connector present (see image embedded above).

Still, what you really need to know about the RTX 3050 6GB is that performance is likely to be quite poor. Remember when the RTX 2060 was derided for not having enough memory back in 2019? Now, five years later, the RTX 3050 6GB has the same amount of memory but half the bandwidth thanks to its 96-bit interface, plus fewer cores and overall worse performance.

The RTX 3050 8GB was already generally slower than the 2060, incidentally, as you can see in our GPU benchmarks hierarchy, so this will be even worse. If you're still thinking the 3050 6GB can't be all bad, we suggest looking around for an RTX 2060 — those were selling for $200 last year, before inventory dried up.

Russian-based retailer Extremecomp listed Palit's RTX 40 Super series graphics cards with specifications. The website did not have the GPU's images uploaded, but the manufacturer's code listed on the website matches the ones from Momomo's recent post.

The source listed the upcoming Palit RTX 40 Super series under Gaming Pro/ Pro OC, Jetstream OC, and Dual variants. Based on the placeholder information, Palit has readied the RTX 4080 Super, RTX 4070 Ti Super, and RTX 4070 Super with 12GB and 16GB GDDR6X variants.

Based on how Palit's RTX 40 non-super versions are configured, its GPU's base clock is the same as used on Founder's Edition. The OC suffix has varying boost clock speeds, which the website didn't list.

The RTX 4080 Super is on the 256-bit memory bus, with other variants on 192-bit. The Asus RTX 4070 Super OC Edition uses the 12VHPWR connector rather than the 8-pin connector used in its non-super OC card. Some PSU brands have already started providing the much-awaited 12V-2x6 cables for existing or newer power supplies. Since these are hoped to be much safer standards, new RTX 4070 / 4070 Ti Super users can get a 12V-2x6 cable from respective vendors for that piece of mind, assuming their existing power supplies qualify for the new cable.

The RTX 4070 Ti Super now has 16GB GDDR6X, while the non-super version is limited to 12 GB. The RTX 4070 Super appears to have two memory variants - 12GB GDDR6X and 16GB GDDR6X.

The difference between Palit's Dual and Dual OC is the boost clock speed, while its base clock is the same as Founder's Edition, according to its RTX 40 non-super specifications. Assuming it is the same here, The RTX 4070 Super 12GB has a 40 MHz increase in base clock over the RTX 4070 Super 16GB. The previous leaks have more details about the RTX 40 Super's specifications, but this is the first we've seen a leak specifying RTX 4070 16GB GDDR6X from a brand.

Earlier leaks showed Nvidia's schedule to announce and allow the reviews of the new RTX 40 Super Series throughout January, ready for purchase from respective dates. If those leaked schedule details are accurate and unchanged, we may see both memory variants of the RTX 4070 Super on Jan. 8, followed by its review and sales embargo lifted on Jan. 17.

A tweet from @fpsojisan_yt reveals that Palit is allegedly cutting the physical size of its PCIe connector in half for its RTX 4060 graphics cards. This change makes some sense as the RTX 4060 is only wired to use eight lanes of PCIe connectivity. At the same time, while the RTX 4060 ranks among the best graphics cards, it's mostly by virtue of being the least expensive Ada Lovelace architecture GPU from Nvidia.

All the RTX 4060 cards we're aware of have so far come with an x16 physical PCIe connector, even though the AD107 GPU (and AD106 on the RTX 4060 Ti) only uses eight lanes of connectivity. The remaining eight lanes are not electrically connected and are there more for stability. These cards from Palit are the first time we've seen an RTX 4060 with a cut down PCIe connector, and in fact it's a rare bird of a graphics card that has anything other than an x16 physical connector — even the RX 6500 XT and RX 6400, which only use four electrical lanes, usually have x16 connectors.

The neutered PCIe x8 interface is generally used to reduce manufacturing costs and reduce the physical size of the actual chip. External interfaces generally don't scale at all with newer process nodes, so dropping from 16 to eight or even four lanes can save on die size. This can be useful on lower-end GPUs like the 4060, since most graphics cards don't really take full advantage of an x16 interface, especially at Gen 4 speeds.

The thing is, while changing the physical connector to an x8 connector might make the cards more flexible in that they can now fit in x8 motherboard slots, in practice there are almost no consumer motherboards that provide x8 slots. Even when a slot only has four or eight lanes of connectivity, nearly all motherboards still opt to provide a physical x16 slot. There are more workstation and server boards with x8 slots, but those likely wouldn't be using RTX 4060 cards in the first place.

In (very) rare cases, the x8 connector of these cards could be beneficial for users installing the RTX 4060 as a secondary graphics card inside their machine. There may also be some older PCs that only had an x8 physical slot... maybe. But even Micro-ATX and often mini-ITX motherboards still provide at least one physical x16 slot.

For consumer use, the shorter x8 connector could compromise stability as well. Not that the RTX 4060 needs the massive coolers that we've seen on GPUs like the RTX 4090 and 4080, but the cards are still long enough that providing an x16 connection would provide more support near the end of the card. So, over time, these Palit cards might be slightly more prone to sag.

The benefits of an x8 connector are very niche for most consumers and gamers. Nearly all modern desktop motherboards feature full-sized x16 slots, with the primary slot linked directly to the CPU to improve latency and speed. Putting an RTX 4060 into any slot not connected directly to the CPU could reduce performance, and PCIe 3.0 or even 2.0 slots on older PCs would likewise impact performance. But if you happen to have a system that only has a PCIe x8 slot (let us know in the comments what board you're using!), Palit's RTX 4060 cards might be useful.

Even though most users will never benefit from a physical 8-lane connector, Palit's unique design will give niche users a rare opportunity to install the RTX 4060 in an x8 slot should they need it.

Nvidia's upcoming GeForce RTX 4060 Ti aims to be one of the best graphics cards. However, if the latest listing from Russian distributor Marvel (via momomo_us) and rumors are accurate, the GeForce RTX 4060 Ti specifications look underwhelming on paper compared to the GeForce RTX 3060 Ti. However, the GeForce RTX 4060 Ti does wield Nvidia's latest Ada Lovelace architecture, so there's definitely a performance uplift, but the question is just how much?

The GeForce RTX 4060 Ti rumoredly features 4,352 CUDA cores, 11% less than the GeForce RTX 3060 Ti. This is because so much of the increased performance on the GeForce RTX 4060 Ti comes from the new architecture. Accordingly, the Geforce RTX 4060 Ti would also have fewer Tensor and RT cores. But, again, the Ada-based graphics card uses the latest 4th-generation Tensor and 3rd-generation RT cores, which improve AI and ray tracing performance over the last-generation GeForce RTX 3060 Ti.

According to early speculation, the GeForce RTX 4060 Ti might debut with a 2,310 MHz base clock and 2,535 MHz boost clock. Some premium, heavily-overclocked models will allegedly boost up to 2,685 MHz. Unfortunately, FP32 performance is a terrible metric for measuring gaming performance. But for discussion's sake, the GeForce RTX 4060 Ti would deliver 22.06 TFLOPs, 36% higher FP32 performance than the GeForce RTX 3060 Ti. The figure looks impressive but remember that FP32 performance doesn't translate to real-world gaming performance.

GeForce RTX 4060 Ti Specifications

*Specifications are unconfirmed.

Russian distributor Marvel listed four custom Palit GeForce RTX 4060 Ti graphics cards with 8GB of GDDR6 memory that operates through a 128-bit memory interface. The specifications align with other retailers' GeForce RTX 4060 Ti listings and the original leak. However, the memory configuration certainly has us wondering what Nvidia is doing with the GeForce RTX 4060 Ti.

The GeForce RTX 3060 Ti has a 256-bit memory interface. Even the GeForce RTX 3060 has a 192-bit bus. A 128-bit bus with 8GB of GDDR6 memory is the kind of setup that's on a GeForce RTX 3060 8GB or a GeForce RTX 3050. It's not a combination that you typically find on a Ti-tier SKU.

The GeForce RTX 4060 Ti reportedly wields 18 Gb/s memory, but even the faster memory won't help the Ada graphics card punch out more memory bandwidth. Limited by a 128-bit memory interface, the GeForce RTX 4060 Ti supplies 288 GB/s, 36% below the GeForce RTX 3060 Ti with its 14 Gb/s chips and 256-bit memory interface. Nonetheless, the GeForce RTX 4060 Ti's more enormous L2 cache (32MB) will help mitigate the lower memory bandwidth, similar to AMD's Infinity Cache.

Palit GeForce RTX 4060 Ti

The GeForce RTX 4060 Ti may already be around the corner. The latest rumors claim that Nvidia will start shipping GeForce RTX 4060 Ti graphics cards to the channel on May 5. That explains why we're seeing more listings from retailers worldwide.

There's no firm launch date yet, but Nvidia allegedly wants to launch the GeForce RTX 4060 Ti before the end of May. Despite all the leaks and retailer listings, the pricing for the GeForce RTX 4060 Ti remains unknown. So far, GeForce RTX 40-series graphics cards have launched with premium pricing. If we look at previous launches like the GeForce RTX 4090 ($1,599), GeForce RTX 4080 ($1,199), GeForce RTX 4070 Ti ($799), and GeForce RTX 4070 ($599), the upward trend in pricing is evident. So it would be foolish to think that the GeForce RTX 4060 Ti will debut at $399, the MSRP for the GeForce RTX 3060 Ti. We're not saying it can't happen, but it's unlikely.

The official launch of Nvidia's GeForce RTX 4070 graphics card is allegedly a few days away, so we are not surprised by the surplus of images of yet-to-be-announced add-in-boards featuring the cut-down AD104 'Ada Lovelace' GPU shown by momomo_us, alexwifi42 and Italian vendor Drako.it  (and found by our colleagues at VideoCardz). 

On the specifications side of matters, images of GeForce RTX 4070 graphics cards by Gigabyte, Palit, MSI, and Zotac show us nothing new, as we already know from the leaks (which should be taken with a grain of salt, yet at this point they are more than likely to be accurate) that these AIBs carry a cut-down AD104 graphics processor with 5,888 CUDA cores and paired with 12GB of GDDR6X memory.

Gigabyte is prepping four GeForce RTX 4070 graphics card designs, with only the WindForce and Eagle models shipping with standard 8-pin auxiliary PCIe power connectors, while the Gaming and Aero models will have 12VHPWR 16-pin connectors. All four designs will have three fans, based on the leaked images. 

One of the things that strikes the eye with the pictures of the upcoming GeForce RTX 4070 graphics cards from Gigabyte, Palit, MSI, and Zotac is that most of them are triple-wide with a massive cooling system, just as my colleague Mark reported yesterday about the hoard of MSI's GeForce RTX 4070 AIBs.

While it is not particularly surprising that many of GeForce RTX 4070-based graphics cards will be huge, considering the overclocking potential of the GPU that powers them, it is reasonable to expect more compact AIBs too. This is perhaps where MSI's GeForce RTX 4070 Ventus 2X comes into play. The board features a dual-fan two-wide cooling system and will likely fit into most of PCs used by gamers. Of course, different GeForce RTX boards feature different power connections, some keep using well-proven eight-pin plugs, while others employ the all-new 12VHPWR.

Apparently, budget-friendly options from Zotac and Palit will also have dual-fan designs and 12GB of GDDR6X memory, yet take the information with a grain of salt as this comes from unofficial sources.

It isn't uncommon for hardware makers and retailers to list yet-to-be-announced products weeks before they should officially start selling. It looks like this trend is repeating with Nvidia's GeForce RTX 4070. At least four vendors are already listing their RTX 4070-badged products, perhaps as a way to generate buzz and interest among consumers ahead of their official release, as noticed by @momomo_us and @harukaze5719.

MSI briefly listed its GeForce RTX 4070 Ventus 2X 12G graphics board on its website. The listing indicated that the product would carry 12GB of memory onboard and would feature a cooling system with two fans suggesting that we are dealing with a fairly compact graphics card, one of the first for the Ada Lovelace microarchitecture. Obviously, this is a preliminary listing, but since it is an official website, we can be more or less sure that it accurately reflects what MSI plans to offer. That said, take the news with a pinch of salt until the official release date arrives. 

Based on listings from a retailer, it seems that MSI also intends to release GeForce RTX 4070 Ventus 3X (with a bigger cooler), two Suprim-branded models, and two Gaming-badged boards aimed at enthusiasts and which are more than likely to enter the ranks of some of the best graphics cards available. So far, MSI and its partners seem to be offering at least eight GeForce RTX 4070 boards, which is a rather broad line-up. Prices of Geforce RTX 4070 offerings from MSI are likely to be significantly higher than the $599 allegedly recommended by Nvidia, though preliminary listed products tend to be overpriced in general.

Meanwhile, Gainward, Palit, and PNY also appear to prep multiple GeForce RTX 4070 versions, though these companies (or rather their partners) have yet to reveal more details about their offerings.

Listings of multiple GeForce RTX 4070 graphics cards weeks before launch could suggest that the product will be widely available right after the formal launch. Nvidia's GeForce RTX 4070 reportedly uses a cut-down version of the AD103 graphics processing unit with 5888 active CUDA cores, a chip that has been in mass production since late summer 2022. By now, the GPU developer should have plenty of silicon qualified for GeForce RTX 4070, so makers of graphics cards will have plenty of graphics chips to build products on.

It's nice to see more graphics cards coming into stock, and even nicer to see them come down in price. Today you can pick up a Palit GeForce RTX 3070 Ti GamingPro 8GB GPU from CCL Computers for £800. 

This RTX 3070 Ti SKU comes with 6144 cores, with a base clock of 1575 MHz and a boost clock of 1770 MHz, 8GB GDDR6X VRAM, PCIe 4.0 x16, and with four outputs via a single HDMI 2.1 and three DisplayPort 1.4. See our review of the RTX 3070 Ti and how it compares to the GPU hierarchy. 

If you've been looking to build your own PC recently, you've more than likely already experienced how hard it's been to get your hands on a 30-series GPU, this could be a great opportunity to snap one up whilst stocks last. 

The Palit GeForce RTX 3070 Ti GamingPro is in the premium mid-tier GPU range and has more than enough power to run all the latest games and applications. We are seeing the price of graphics cards drop slightly recently, and it is amazing to see a card even go on sale and drop another £40 off of the asking price. 

More GeForce RTX 3070 Ti Deals

UPDATE 7/29 03:27 PT: 

This graphics card was developed by Galax. Galax even lists a similar non-LHR graphics card on its website. The original story incorrectly named Palit as the developer of Galakuro’s GeForce RTX 3070 LHR graphics card. While Galakuro, Galax, Gainward, KFA2, and Palit brands belong to one owner and share engineering and manufacturing resources, they are different business entities. The focus of the story still stands: there is a short version of Nvidia’s GeForce RTX 3070 LHR that is available in Japan and will likely hit other markets under different brands.

Original Article:

Gone are the days when add-in-board (AIBs) makers would introduce Mini-ITX graphics cards based on top-of-the-range GPUs. Today, virtually all premium graphics AIBs are large and can barely fit into many typical desktops. To address such systems, Galax's partner in Japan this week introduced a GeForce RTX 3070 LHR graphics card that is shorter than other RTX 3070 boards on the market, including Nvidia's own RTX 3070 Founders Edition. 

Galakuro's GeForce RTX 3070 LHR (GG-RTX3070-E8GBOC/DF/LHR) carries Nvidia's GA104 GPU with 5888 CUDA cores clocked at up to 1725 MHz as well as 8GB of 14 GT/s GDDR6 memory that connect to the processor using a 256-bit interface. The card has four display outputs (three DisplayPort 1.4a, one HDMI 2.1) as well as one eight-pin auxiliary PCIe power connector. Like other products designed by Galax, this one is compatible with Galax's XtremeTuner software. 

The AIB does not feature an extreme GPU clock or an enhanced VRM to ensure superior overclocking potential of the card. Its main selling point is a 238 mm length: this is 4 mm shorter when compared to Nvidia's own GeForce RTX 3070 Founders Edition and is about 120mm shorter when compared to typical GeForce RTX 3070 boards from AIB makers. Lower length of the card will allow its installation into PCs that cannot house a 400-mm graphics card, such as smaller ATX, Micro-ATX, or even larger Mini-ITX systems.  

There is a catch though. Galax's GeForce RTX 3070 LHR has a rather massive cooling system with two 102mm Wings fans that makes it 54 mm thick and 128mm high, which will likely make it incompatible with certain PCs. 

The 238mm GeForce RTX 3070 LHR made by Galax is currently available only in Japan from its partner Galakuro Gaming. Hermitage Akihabara reports that the board has been announced, but it does not say anything about its price.  

What remains to be seen is when (and if) the Galax-made 238mm GeForce RTX 3070 LHR board is set to arrive to the U.S. and Europe. There are loads of cases that cannot comfortably accommodate a full-size 400mm graphics card and many of their owners are looking forward to upgrading their graphics cards, so chances that Galax will bring the board to these markets are fairly high.

As the launch dates for Nvidia's GeForce RTX 3070 Ti and GeForce RTX 3080 Ti edge closer, more information about their general specifications are revealed. As reported by VideoCardz, Leadtek and Palit Microsystems have added general specifications of their upcoming GeForce RTX 3080 Ti graphics boards to Korean National Radio Research Agency (RRA) database. As this is not an official announcement, a modicum of scepticism is required.

In the RRA listing there is an indication of 12GB of GDDR6X memory for the Titanium variant of the RTX 3080, as discovered by @harukaze5719. Last week an alleged leaked MSI presentation that covered the company's upcoming Suprim-series graphics cards confirmed memory configurations of Nvidia's long-rumored GeForce RTX 3070 Ti and GeForce RTX 3080 Ti graphics cards: 8GB of memory for the former and 12GB of GDDR6X memory for the latter.

We have previously reported, citing unofficial sources, that Nvidia's GeForce RTX 3080 Ti is based on Nvidia's GA102 GPU with 10,240 CUDA cores, has a 384-bit memory interface, and carries 12GB of GDDR6X memory. Performance of the unit should be close to that of Nvidia's range-topping GeForce RTX 3090 in cases where the latter's 24GB of onboard memory is not a factor. Meanwhile, for applications that require significant amounts of DRAM, the RTX 3090 will be unbeatable. 

The GeForce RTX 3070 Ti is based on Nvidia's GA104 graphics processor with 6,144 CUDA cores and a 256-bit interface that will be used for 8GB of 19 Gbps GDDR6X memory, according to unofficial information.  

It is expected that Nvidia will introduce at least the GeForce RTX 3080 Ti on May 31 and start its sales in June. Still, we advise to take everything unofficial with a grain of salt as plans tend to change. 

Palit Microsystems, which owns Galax, Gainward and KFA2, is reportedly working on a range of GeForce RTX 30-series graphics cards that feature cryptocurrency mining limiters which severely degrade the financial viability of using the cards for mining. Last week, we learned that Palit was prepping GeForce RTX 3070 and 3080 graphics cards with limited hash rates (LHR), and this week VideoCardz reported that the GeForce RTX 3060/3060 Ti LHR boards are also on the way from Galax. 

The new Galax GeForce RTX 3060/3060 Ti For Gamers boards will be based on Nvidia's LHR GA106 and GA104 GPUs featuring new IDs and paired with appropriate firmware, according to VideoCardz, which cited Galax as its source; however, the cards aren’t currently listed on the vendor’s site. 

Palit's FG-series GeForce RTX 3060 will reportedly carry 12GB of GDDR6 memory with a 192-bit interface; whereas, the GeForce RTX 3060 Ti will come with 8GB of GDDR6 memory with a 256-bit interface. Both will be equipped with custom triple-fan cooling systems, but their clock rates will be in line with Nvidia's recommendations: up to 1,777 MHz for the RTX 3060 and up to 1,665 MHz for the RTX 3060 Ti.

Unofficial reports claim Nvidia is quietly rolling out its existing graphics processors with cryptomining limiters enabled by a combination of a new GPU ID, firmware and driver. In particular, it is expected that Nvidia's lineup of crypto-limited graphics cards will include both existing and new models, such as the RTX 3090 (GA102-302), RTX 3080 Ti (GA102-225), RTX 3080 (GA102-202), RTX 3070 Ti (GA104-400), RTX 3070 (GA104-302), RTX 3060 Ti (GA104-200), RTX 3060 (GA106-302), and GeForce RTX 3050/3050 Ti.

There is a good reason why Nvidia and its graphics card partners reportedly want to deny usage of GeForce GPUs, some of the best graphics cards for gaming, for cryptomining. Mining obviously uses a GPU at around 100% load 24/7, something that a chip for a client PC is not meant for. As a result, failure rates because of mining are almost guaranteed to be growing. 

Keeping in mind that Nvidia controls over 80% of the market of discrete desktop graphics cards selling around 9 million GPUs per quarter, increased failure rates would clearly hit the company and partners badly. Just 1% of 9,000,000 is 90,000 RMA cases, and makers of graphics cards (and possibly Nvidia itself) may not be ready to process an overwhelming number of RMA cases in a timely manner. Still, very few GeForce RTX LHR graphics cards have been officially announced so far.

Because it's hard to get a modern graphics card, some suppliers are now trying to meet demand for entry-level options by re-launching older products. Earlier this month we reported that Japanese vendor Kuroutoshikou decided to re-start sales of its GeForce GTX 1050 Ti boards. This past weekend the card made it into retail, but its price is far from what we usually consider an entry-level. The unit costs ¥22,800 ($210 with taxes), according to Akiba PC Hotline. Meanwhile, such boards can cost even more in the U.S.

The Kuroutoshikou GF-GTX1050Ti-E4GB/SF/P2 is produced by Palit Microsystems and is based on Nvidia's GP107-400 graphics processor, with 768 CUDA cores (featuring the Pascal architecture and introduced in October 26, 2016) paired with 4GB of 128-bit GDDR5 memory with a 7 GT/s data transfer speed. The card has three display outputs: DisplayPort 1.4, HDMI 2.0b and a DVI-D for legacy monitors (fitting for a legacy architecture). As for compatibility, the board does not require an auxiliary PCIe power connector and will fit into most desktops, including Mini-ITX systems, but not low-profile machines.

Nvidia's 4.5-years old GPU will hardly deliver sufficient performance in modern games (unless maybe you're happy to dial down most settings). Its performance should be comparable to or higher than that of Intel's Iris Xe G7 (96 EUs) integrated GPU, so it is still faster than the vast majority of integrated graphics solutions that are used today.

Given that the GeForce GTX 1050 Ti is outdated in terms of performance and architecture, demand for such a product would be close to zero under normal conditions. But today the GPU market does not act normally, which is why the price of the product exceeds $200. In fact, Akiba PC Hotline claims that a GeForce GTX 1050 Ti was priced at ¥11,000 with taxes in October, 2019. So, in less than two years this outdated product has become twice as expensive in, at least in Japan.

But if you thought that was bad, it's worse in other countries. Based on data from CamelCamelCamel, new GTX 1050 Ti boards were recently sold for betwee $399 – $599 at Amazon.com. So maybe it's time to dig through your closet and see if your old graphics card could finance a new CPU or SSD upgrade. 

One of the first Nvidia CMP 30HX models (via momomo_us) has emerged at Microless, a retailer in Dubai, for $723.84. The graphics card is part of Nvidia's latest Cryptocurrency Mining Processor (CMP) lineup that's suppose to convince cryptocurrency miners to lay off GeForce gaming graphics cards.

The Palit Nvidia GeForce CMP 30HX (NE630HX017J9-1160X) arrives with a 1,530 MHz base clock and a 1,785 MHz boost clock. The clock speeds are identical to Palit's GeForce GTX 1660 Super GP gaming graphics card. Although Microless didn't reveal the core count, the CMP 30HX is expected to wield 1,408 CUDA cores since it's derived from TU116 (Turing) silicon, like the GeForce GTX 1660 Super.

Similar to its gaming counterpart, the Palit Nvidia GeForce CMP 30HX rocks 6GB of GDDR6 memory that clocks in at 14 Gbps. Leveraging a 192-bit memory interface, the graphics card supplies a memory bandwidth of up to 336 GBps. Built for mining Ethereum, the Palit Nvidia GeForce CMP 30HX doesn't come with any display outputs. 

Rated for 125W, the Palit Nvidia GeForce CMP 30HX only depends on a single 8-pin PCIe power connector. The graphics card delivers Ethereum mining performance up to 26 MH/s before optimizations.

The value added tax (VAT) rate in the United Arab Emirates (UAE) is 5%, so the Palit Nvidia GeForce CMP 30HX should cost around $689.37. Keep in mind that computer hardware is frequently more expensive outside of the U.S., though. Besides, we don't know how much of the price is due to the retailer's markup on the Palit Nvidia GeForce CMP 30HX.

Due to the graphics card shortage and cryptocurrency mining craze, graphics cards are obnoxiously expensive right now. Not so long ago, we saw custom GeForce GTX 1660 Super models selling between $599 and $899. A week later, the same graphics cards are starting from $699. It looks like the CMP 30HX isn't completely dead on arrival, but there are certainly cheaper and more effective options out there, such as the GeForce GTX 1060 6GB or Radeon RX 590.

You know the graphics card market is in a bad place when vendors resort to rereleasing five-year old graphics cards. Kuroutoshikou, a Japanese vendor, has announced that its GeForce GTX 1050 Ti (GF-GTX1050Ti-E4GB/SF/P2) will hit the domestic market in mid-March.

In reality, the GF-GTX1050Ti-E4GB/SF/P2 is a rebranded version of Palit's GeForce GTX 1050 Ti StormX. Based on the GP107 (Pascal) silicon, the graphics card is equipped with 768 CUDA cores with a 1,392 MHz boost clock and 4GB of 7 Gbps GDDR5 memory. The GeForce GTX 1050 Ti is rated for 75W so it doesn't require any external PCIe power connectors, making it a good plug-n-play option for entry-level gamers, even though it is no longer among the best graphics cards.

The GeForce GTX 1050 Ti's revival isn't a coincidence though. It was Nvidia itself who decided to replenish its partners with Pascal GPUs in the middle of the ongoing graphics card crysis. Nvidia's actions also paved the way for other vendors to get rid of their old Pascal stock, including Palit who might launch new specialized GeForce GTX 1060 models for cryptocurrency mining.

We've already started seeing more GeForce GTX 1050 Ti availability here in the U.S. Sadly, the pricing leaves much to be desired. While Kuroutoshikou's GeForce GTX 1050 Ti will arrive in Japan with a price tag of ￥20,727 (~$190.97), custom models in the U.S. market currently retail between $330 and $600. That's pretty insane since the GeForce GTX 1050 Ti has five years under its belt now and had launched for $139.

With how ridiculous pricing is right now and the graphics card shortage, picking up a pre-built PC, especially one of the best gaming PCs, suddenly doesn't sound like a bad idea anymore.

The cryptocurrency mining business is booming again and, according to a PC Gamer report, Palit may be getting ready to ride the wave. The company has registered multiple P106 graphics cards with the Eurasian Economics Commission (EEC). As always, these registered models may or may not make it to the market.

Palit, like many other brands, launched the P106-100 back in 2017. The graphics card was based on the GP106 Pascal silicon, which is the same die that powers the highly successful GeForce GTX 1060. In effect, the P106-100 and GeForce GTX 1060 were literally the same graphics card, aside from the removal of the video outputs on the first.

As PC Gamer noted, the new part numbers sport a different nomenclature in comparison to the original part numbers. The registered part numbers fall in line with Palit's last-generation GTX 16-series graphics cards. However, it isn't likely that the manufacturer will update graphics card to the TU116 Turing silicon. Given the recent revitalization of the cryptocurrency mining business, Palit may just be dumping leftover inventory that it may have in its warehouse.

Nvidia has recycled Turing to launch its latest Cryptocurrency Mining Processor (CMP) lineup of graphics cards. The 30HX is Nvidia's entry-level crypto-mining SKU. And as we've seen, it's likely based around the TU116 silicon, the die inside the GeForce GTX 1660 Super. Nvidia rates its 30HX with an Ethereum hash rate up to 26 MH/s. According to Minerstat, the P106-100 is still good for a hash rate of 21 MH/s, which isn't bad for a four-year-old graphics card.

The 30HX and 40HX are scheduled to launch in the first quarter of the year.  But Nvidia hasn't revealed the pricing for any of its CMP devices yet. However, we've already seen mentions of the 30HX and 40HX in the latest GeForce Game Ready 461.72 WHQL driver, so they should be right around the corner. If Palit prices the new P106-100 just right, the manufacturer could definitely steal some of the cryptocurrency profit pie from Nvidia's CMP.

Extravagant DIY builds have long become mainstream, so many people are looking for hardware that matches design of their existing rigs. Those with a snow-white PC want to use white components, which are sometimes hard to find. Of the many GeForce RTX 3060-based graphics boards announced thus far, there is only one white card, spotted by Hermitage Akihabara and made by Galax. This card is currently not even listed on Galax's website and appears to be only available in one Japanese store.

The graphics board is called the Galax Galakuro GeForce RTX 3060 12GB (GK-RTX3060-E12GB/OC/White) and it carries Nvidia's GA106-A300 GPU with 3584 CUDA cores clocked at up to 1807 MHz (slightly up from 1780 recommended by Nvidia) in boost mode as well as 12GB of 15 Gbps GDDR6 memory with a 192-bit interface. 

The GPU is cooled down by a rather massive cooling system incorporating an aluminum heatsink, heat pipes, and two fans which occupy 2.5 slots. Keeping in mind that the board is rated for a 170W TDP, it is not surprising that it comes with a rather advanced cooler. On the one hand, the massive cooler will improve the card's overclocking potential. On the other hand, not every PC can actually accommodate the white Galax Galakuro GeForce RTX 3060 12GB.

The board has one 8-pin PCIe power connector as well as four display outputs (three DisplayPort 1.4, one HDMI), which is now the standard configuration for performance mainstream graphics boards.

Given that white builds are getting traction, a GeForce RTX 3060 graphics card in white would gain quite some popularity among modders and DIY enthusiasts. Unfortunately, Galax currently does not list the board on its website and for now it will be available only from one store in Japan for ¥57,980 without tax ($546).

As tweeted by @Harukaze5719, Palit has added one more indication there may be additional (future) RTX 3060 variants. Palit filed listings for new RTX 3060 model names with the NRRA, and one model in particular showcases the RTX 3060 with 6GB of VRAM.

Even before the official RTX 3060 announcement from Nvidia, there were a number of rumors floating around. Considering the history of the -60 line of GPUs, many have assumed Nvidia planned to make a 6GB variant of the RTX 3060. We had GTX 1060 6GB and 3GB (and even 5GB in Asia), so this isn't without precedent.

Of course, this is just a listing for a product name. We've seen listings like this numerous times in the past, where a graphics card manufacturer will pump out a model name, and it never gets used. It may come to fruition, or it may not. We'll have to wait and see where the fate of the 3060 6GB will lie.

It's worth noting that the mobile RTX 3060 already exists in a 6GB variant, so this isn't exactly a difficult switch. Instead of 16Gb chips on each channel, all Nvidia needs to do is put in 8Gb chips. Six channels, 32-bits each, and you end up with either a 6GB or 12GB card. But how would a 6GB variant of the 3060 play out?

From our own experiences testing 6GB graphics cards like the RTX 2060, we've found that 6GB is about as low as you want to go with a mid-range card. Especially at 4K resolutions, you might have to turn down a couple of settings (like 4K texture packs) in very graphically demanding games to prevent VRAM bottlenecking.

There are other incentives to buying an RTX card, of course, like DLSS support. Ray tracing often gets more hype, but the Tensor cores are potentially more useful for lower-spec PCs. There are already a few dozen games with DLSS (about 24 with DLSS 2.0), and the list continues to grow. Nvidia just announced that Nioh 2: Complete Edition and Mount & Blade II: Bannerlord now have DLSS 2.0, for example, and Unreal Engine also has a new free marketplace plugin for DLSS that should make it even easier to implement. So, even though a 6GB card might not make a lot of sense for ray tracing, it's definitely not out of the question.

The card would have to be priced well to make it attractive over the more favorable 12GB offering. Right now, unfortunately, in the world of virtually impossible to buy graphics cards, it's doubtful good prices will even be a thing. Everything from GTX 1050 Ti through RTX 3090 is basically sold out, with extreme scalper prices on eBay. However, from Nvidia's standpoint, a 6GB model does make sense — especially if they are suffering from the video memory shortages that have reportedly plagued GDDR6 (and GDDR6X) production.

As usual, Nvidia won't comment on the existence or potential for a desktop 3060 6GB. Our take is that, given where games are heading and current pricing, it will be a tough sell unless the price is really good (and actually something you can find). The GTX 1060 3GB was in a similar situation a few years back, and we never recommended it without concerns about the lack of VRAM. A 3060 6GB would be the modern equivalent, and at the right price, consumers probably would be okay with the reduced memory capacity. Or, you know, just buy a 3060 laptop.

Nvidia's GeForce RTX 30-series consumer graphics cards and its Ampere architecture are generally not the company's most economical solutions in terms of power consumption, at least compared to its launches over the last several years. Yet, with a proper power management and cooling system, you can get an Ampere into a Mini-ITX system. Palit Microsystems is the first graphics card maker to bring Ampere to the small form factor.  

Palit Microsystems, which happens to be one of the world's largest graphics card makers with its Palit, Gainward, Galax, and KFA2 brands, is among the first to announce Mini-ITX add-in-boards running Nvidia's GA106-300 graphics processing units with 3584 CUDA cores as well as 12GB of GDDR6 memory. 

So far, the company has introduced four Mini-ITX GeForce RTX 3060 boards: the Palit GeForce RTX 3060 StormX, the Palit GeForce RTX 3060 StormX OC, the Gainward GeForce RTX 3060 Pegasus, and the Gainward GeForce RTX 3060 Pegasus OC. For some reason, Palit has yet to add Mini-ITX GeForce RTX 3060 cards into its Galax lineup and actually unveil any KFA2-branded GeForce RTX 30-series products. 

All four Mini-ITX GeForce RTX 3060 graphics cards use a dual-slot cooling system with one fan and have four display outputs – four DisplayPort 1.4a connectors as well as one HDMI 2.1 port. The cooling systems have multiple heat pipes and can stop the fan under light loads, yet they have different shrouds, and it is unclear whether the coolers used by the Gainward and Palit brands are actually different. Furthermore, it is unclear whether the boards use the same PCB design. 

The cards are said to consume up to 170W of power and feature one eight-pin PCIe auxiliary power connector. Measuring 170mm x 125 mm, the boards can physically be installed into a Mini-ITX chassis, yet not all SFX power supply units have an eight-pin power connector. 

From a performance point of view, the Palit and Gainward-branded Mini-ITX GeForce RTX 3060 graphics cards feature a default GPU clock of 1320MHz and a boost clock of 1777MHz. Meanwhile, Palit has yet to disclose boost frequencies of OC versions of these cards, which probably means that the boards will be available later. 

Palit does not list MSRPs of its products on its websites, so we do not know whether Mini-ITX versions of its GeForce RTX 3060 will cost more than regular models with large cooling systems.   

Graphics card manufacturer Palit has unveiled via its Twitter account an all-new RTX 3070 SKU for the JetStream graphics card lineup. As discovered by VideoCardz, the card appears to favor a low key aesthetic and features a fully blacked-out shroud with very little or no RGB visible in the images. 

The Palit RTX 3070 JetStream is equipped with a dual-fan cooler design, paired with an all-black shroud and PCB. The PCB seems to be rather short and may extend just half-way into the cooler, as the 8-pin power connectors are installed near the dead-center of the card. Supporting this is the rear backplate, with a massive cutout at the end, which can allow air to pass thru the heatsink freely. It appears to be slightly bigger than most cutouts found in other Ampere graphics cards, indicating the extra short PCB was designed that way for cooling reasons.

The card is pretty average in size for an RTX 3070 AIB partner card, the thickness is just above 2 PCIe slots, and the edge of the card extends to what appears to be approximately over an inch above the PCIe bracket.

For now, this was just a sneak-peek. Palit has yet to release a product page for the RTX 3070 JetStream. But I'm sure they will soon as the RTX 3070 launch is just four days away.

As the launch of the RTX 3090 starts, Palit introduces the RTX GameRock lineup for the 3070, 3080, and 3090 series of graphics cards.And with a diamond aesthetic these cards shine with full RGB.

Speaking of power, the RTX 3090 GameRock OC variant is speced to a whopping 420W of power, the highest we've seen on any RTX 3090 right now. I would expect very high overclocks to be achieved on this unit as the RTX 3080 and RTX 3090 are known to be very power limited when it comes to increasing performance.

The entire front of the shroud is coated in transparent diamond-looking material which allows RGB lights to shine thru, flanked by two brackets on the top and bottom. The RTX 3090 model specifically is a triple-slot card with a triple-fan cooler design, the heatsink is clearly visible and not obstructed much from the shroud itself. The backplate is like other Ampere partner cards we've seen, with cutouts at the backplates end allowing air to travel directly thru the graphics card from the rear fan.

The card's triple 8-pin PCI-E power plugs are centrally located on the card. Palit decided to not use any extensions which would allow the connectors to stay at the end of the card, so managing the PSU cables will be somewhat challenging, especially when you've got three 8-pin connectors instead of the usual two.

Pailt did a great job with the card's design, the implementation of RGB is beautiful, the diamond aesthetic should please DIY builders who want to assemble a system around that theme (especially a good pairing with Trident Z Royal's from G.Skill). The RTX 3090 variant especially is gorgeous, showing off a full-fat heatsink that will really fill out an ATX chassis. But, if you want something a bit smaller, Palit offers RTX 3080 and RTX 3070 models with smaller footprints.

There is no word on when these specific SKUs will get released into the wild. With the launch of the RTX 3090 today I would expect RTX 3080 and 3090 models to be coming soon. Expect to wait until October 15th before you can buy the RTX 3070 GameRock, as that's the official launch date of the RTX 3070 from Nvidia.

If it's silence that you're after above all else, you may be interested in Palit's latest GT 1650 KalmX graphics card. This GPU comes with a huge passive heatsink, and while that means it won't be quite as happy to stay in boost as much as actively cooled cards, it does mean it'll be whisper quiet. No, actually, it'll be even quieter than whisper-quiet -- so quiet that you'll probably be able to hear the coil whine if you're in a quiet room.

Thankfully, the card doesn't have a very high TDP and is able to draw its maximum of 75W straight from the motherboard slot without the need for a PCI-Express power cable. Clock speeds sit at a base clock of 1485 MHz with a boost clock of 1665 MHz. That's the standard boost clock for reference Nvidia GTX 1650 cards, though how much the card will be able to hit that will depend strongly on the kind of case it's running in, how much airflow is provided from external sources, and the card's orientation. 

The cooler uses two nickel-plated heatpipes which draw heat from a copper base, leading to the large fin stack. The cooler also covers the VRM and memory, so there is no need to worry about those overheating. Despite being quite big though, the cooler does fit within the two-slot profile, though we strongly recommend you keep the third adjacent slot empty for airflow purposes.

Memory sits at the standard 4 GB of GDDR5, which runs over a 128-bit memory interface at 8 GB/s to generate a memory bandwidth of 128 GB/s.

There's no word on pricing or availability just yet.

Hot on the heels of recent GeForce GTX 1660 Ti leaks, famous leaker VideoCardz has released images that claim to be a couple of custom GeForce GTX 1660 Ti models from big names EVGA and Palit.

The GeForce GTX 1660 Ti purportedly comes equipped with the Turing TU116 die, 1,536 CUDA cores and 6GB of GDDR6 memory operating clocked at 1,500MHz (12,000MHz effective) across a 192-bit memory bus. It reportedly has a 1,500MHz base clock and a boost clock that climbs to 1,770MHz. Although Nvidia has not many official announcement about the card being released, VideoCardz's report today claims the GeForce GTX 1660 Ti will come out on February 22.

EVGA GeForce GTX 1660 Ti XC Gaming

According to VideoCardz, EVGA initially plans to release single-fan models of the GeForce GTX 1660 Ti while saving the dual-fan models for a later date. Its two planned graphics cards are said to be the GeForce GTX 1660 Ti XC and GeForce GTX 1660 Ti XC Black Gaming, with the latter expected to come with a lower out-of-the-box boost clock. The publication did not list clock speeds for the EVGA models.

Both graphics cards reportedly are Mini-ITX form factor and employ EVGA's redesigned 2.75-slot cooler. The cooling solution utilizes a single Hydro Dynamic Bearing (HDB) cooling fan, which is supposed to increase cooling performance and reduce fan noise by up to 15 percent. Another advantage of using a HDB fan instead of the conventional sleeve-bearing fan is the prolonged lifespan. EVGA also claims that the upraised "E" pattern on the fan contributes to noise reduction as well.

Palit GeForce GTX 1660 Ti StormX

It appears there is a Mini-ITX trend going on with the GeForce GTX 1660 Ti. Much like EVGA's offerings, Palit's GeForce GTX 1660 Ti StormX and GeForce GTX 1660 Ti StormX OC also appear to fit the Mini-ITX form factor but with a dual-slot, single-fan cooler.

Russian retailer DNS Shop briefly listed the specifications for both Palit models, pointing to the Palit GeForce GTX 1660 Ti StormX following Nvidia's reference specifications, which means a 1,500MHz base clock and 1,770MHz boost clock. As its name suggests, the OC version is expected to come with a higher boost clock in the range of 1,815MHz.

Graphics card manufacturer Palit recently registered various new products with the Eurasian Economic Commission (ECC), including graphics cards using Nvidia's rumored GeForce GTX 1160, GTX 1660 and GTX 1660 Ti. This new listing disputes rumors that the GeForce GTX 1160 is fake; although, Lenovo has previously listed the graphics card with its Legion laptops.

These past months have been an interesting time for Nvidia rumors. But one thing is certain, Nvidia's GTX branding will continue to coexist alongside the current RTX 20-series, whether it be via the GTX 11-series or GTX 16-series. In either case, they are not expected to have real-time ray tracing.

Here's a table comparing the potential specifications of the GeForce GTX 1660 Ti, GTX 1660, and GTX 1160.

*=unconfirmed

The GeForce GTX 1660 Ti, GTX 1660 and GTX 1160 are said to feature TU116 silicon. The GTX 1660 Ti allegedly comes with 1,536 CUDA cores and 6GB of GDDR6 memory. According to a recent HardOCP article, the GeForce GTX 1660 Ti is reportedly priced at $279 and will be announced on February 15. The GeForce GTX 1660 supposedly sports 1,280 CUDA cores and GDDR5 memory. It purportedly will be available in March for $229.

As for the GeForce GTX 1160, the only specification we've heard so far is around onboard memory. The graphics card seemingly comes equipped with 3GB or 6GB of memory. It's uncertain whether it'll be GDDR5, GDDR5X or GDDR6.

Several graphics cards OEMs have released lower-performing versions of the Nvidia GT 1030 that swap the GDDR5 memory of the reference spec for slower DDR4.

Nvidia’s GT 1030 reference spec has a memory bandwidth of 48GB/s. This is calculated from the GDDR5 memory’s 3GHz clock, which is effectively 6GHz due to the nature of double-data-rate RAM, and the 64bit-wide memory bus. Recently, however, lower-spec’d GT 1030 graphics cards have shown up from MSI, Gigabyte, and Palit. Instead of having the standard 3GHz GDDR5 memory, these cards have DDR4 memory that runs at one-third of the reference card’s memory clock.

For example, MSI’s GT 1030 2G LP OC, which is based on Nvidia’s reference spec, is listed with GDDR5 that has an effective memory memory clock of 6008MHz (~3GHz x 2), whereas the GT 1030 2GD4 LP OC is listed with DDR4 memory that has an effective memory clock of 2100MHz. MSI doesn’t explicitly list memory bandwidth as a part of its spec sheets, but a two-thirds drop in effective memory clock implies a two-thirds drop in memory bandwidth. The GT 1030 2GD4 LP OC should have only 16GB/s of memory bandwidth compared to the reference card’s 48GB/s. Both MSI and Gigabyte differentiate their DDR4-based GT 1030 cards by name with a “D4” denotation, whereas Palit does not.

According to MSI’s spec sheets, the DDR4-based GT 1030 cards, at least, have 10W lower power consumption compared to the reference card, which has 30W total power consumption.

This isn’t the first time we’ve seen lower-performing implementations being released in the middle of a GPU’s life cycle. Asrock, which only recently entered the GPU market, also released lower-clocked versions of the AMD RX 560 and RX 550. Although they do not dictate what graphics card OEMs can produce, AMD and Nvidia do rely on reference specs to segment their product stacks. It’s harmful and confusing to consumers when graphics card OEMs, or maybe even the GPU manufacturers themselves, blur their product segmentation after launch with revised products.

Introduction & Overview

Nvidia's GeForce GTX 1080 is the company's gaming flagship right now (check out our Nvidia GeForce GTX 1080 Graphics Card Roundup for in-depth analysis of the top models). But its GeForce GTX 1070 is undoubtedly a better value play. After all, the GP104-powered card outperforms the old GeForce GTX 980 Ti for hundreds of dollars less.

There's a lot of variation out there, though. You'll find 1070s selling anywhere from under $400 (£300) to over $500 (£400), and it's amazing to see what Nvidia's add-in board partners have done with this platform. To help you size up the current field, our team continues testing GeForce GTX 1070 cards. For now, we have four different models to compare. And we go deep. Really, each page of this piece could be its own complete review. We focus on the manufacturing quality and technical features of each card, along with power consumption, clock rates, cooling, and acoustics.

The gaming performance of every factory-overclocked board within a given chip class is usually pretty similar to begin with. But that's more true now than ever. This is a result of features like GPU Boost 3.0, which allows manufacturers to safely extract as much headroom as possible from a processor. Very little is left on the table, even if you have access to extreme overclocking hardware.

We will continue to update this roundup as new test samples become available. Some are already in the lab; others are en route (and there's at least one model we need to bring over from Tom's Hardware DE still).

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

The cards are presented in alphabetical order, and their order has nothing to do with performance in our roundup or a preference for individual products.

MORE: Best Deals

Benchmark Results

Gaming

We test every card after a suitable warm-up period to avoid unfair differences in GPU Boost frequencies. All benchmarks are run six times; the first one is used to get the GPU hot again.

These cards are all press samples operating at the same settings as retail models in our best effort to ensure one vendor doesn't get a leg up on another using non-representative clock rates.

The following galleries each contain four images, covering two tested resolutions. We put our focus on QHD (2560x1440) and UHD (3840x2160), plotting out average and minimum frame rates for each resolution in separate graphs.

Ashes of the Singularity

Battlefield 4

The Division

Grand Theft Auto V

Hitman

Project CARS

Rise of the Tomb Raider

The Witcher 3

All of the factory-overclocked cards offer similar performance, more or less. That's why our primary focus centers on evaluating the more technical aspects of each board design, along with their coolers. This is where differences in noise, power, and temperatures are most likely to come from.

Power Consumption

We begin by comparing the power consumption of each card in our gaming loop, stress test, and at idle. Depending on the power targets specified by each manufacturer, we sometimes measured substantial differences. We're ignoring the decimal places in our bar graphs, since those values are too small and would be within measurement tolerances.

We also found that some cards with lower power targets started to throttle during our stress test, resulting in lower power consumption numbers. This did not, however, have a negative impact on general gaming performance for any of the tested cards, as the stress test merely represents a worst-case scenario.

We confirmed that MSI's retail cards will ship with a slightly lower power target (max. 240 to 250 watts) after an internal discussion and evaluation of our measurements.

This also applies to the BIOS versions with OC mode enabled by default, which employ a roughly 20 MHz-higher base and GPU Boost frequency. In the interest of fairness, we tested both MSI cards using normal mode, without the overclocked base and GPU Boost rates. This doesn't affect our power consumption measurements, though.

Noise

For the following comparison, we divide all of the gallery's bar graphs into gaming loop and idle, even if "noise" in practice spans a wide spectrum and the "character" of the sound varies a lot, with each card having highly individual results. Therefore, it is important not to compare just the absolute numbers, but also the frequency spectrum we're presenting.

Many of the cards implement a semi-passive mode, where their fans remain off when the card is idle. Thus, we refrained from taking measurements in that state. Even in our anechoic chamber, levels of 22 dB(A) and below merely represent ambient noise.

Temperatures

For this comparison, we divide the gallery's bar graphs into gaming loop, stress test, and peak temperatures measured on the MOSFETs.

We occasionally compared the temperatures on our benchmark table with those measured inside a closed case and found them to be no more than two or three Kelvin higher.

Since temperatures in a closed case also depend heavily on the enclosure's cooling performance, the only representative and reproducible values are those measured on our benchmark table. Those are the ones we compare.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Nvidia GTX 1070 Founders Edition

The GeForce GTX 1070 Founders Edition is based on the same GPU as Nvidia's GeForce GTX 1080, only with 25% of its CUDA cores and texture units disabled. As we learned in our Nvidia GeForce GTX 1070 8GB Pascal Review, though, the processor's back-end still serves up 64 ROPs and 2MB of L2 cache. Also, GDDR5X memory is replaced with standard GDDR5, while the 1070 sports a less sophisticated cooling solution and a lower power target of 150W.

The Founders Edition board is actually what Nvidia used to call its reference design. This time around, however, the company decided it wanted a piece of the premium card sales as well. Renaming its own offering is supposed to help sidestep the negative connotation that sometimes accompanies reference implementations.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The injection-molded aluminum cooler cover is metallic silver and black in color. Its design exudes quality, but the metal construction also contributes to the card's relatively high weight of more than 35 ounces (one kilogram).

The back of the board is covered by a two-piece plate. It's purely aesthetic and does nothing for cooling. If you have a couple of cards next to each other in SLI or if you'd simply prefer more airflow, the backplate can be unscrewed without a problem.

There's an illuminated GeForce GTX label on top of the card, and an eight-pin auxiliary power connector is positioned toward the end.

Nvidia's angular and rugged design is certainly a matter of taste. But it certainly stands out. In a windowed case, you can count on turning heads.

Peeking down through the card's end reveals the heat sink's fins. The mounting frame offers up to three screws for attaching additional brackets. You'd use those for stabilization in a case.

The slot plate features five outputs, four of which can be used simultaneously in a multi-monitor setup. In addition to one dual-link DVI-D interface, the rear bracket also exposes an HDMI 2.0 output and three DisplayPort 1.4-compatible connectors. The rest of the plate is peppered with openings to encourage better airflow.

Board & Components

In addition to the one phase for memory, only four out of six available power phases for the GPU are implemented.

Nvidia leans on Samsung for its GDDR5 memory. Eight K4G80325FB-HC25 modules are each able to store up to 8Gb (32x 256Mb). Voltages are specified between 1.305V and 1.597V, depending on the clock rate.

The 4+1-phase system relies on the sparsely documented µP9511P PWM controller. Since this controller can't communicate directly with the VRM's phases, Nvidia utilizes 53603A chips for solid PWM drivers (gate drivers) and controlling the power MOSFETs (primarily of type 4C85N).

Two capacitors are installed right below the GPU to absorb and equalize peaks in voltage. The board design looks tidy and well thought-out.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. This also shows that both curves drop as the GPU's temperature rises.

After warm-up, the GPU Boost frequency flirts with 1.7 GHz under load. This is mirrored by our voltage readings. While we measured up to 1.031V in the beginning, that value later dropped as low as 0.812V, if only temporarily.

Combining the measured voltages and currents allows us to derive a total power consumption we can easily confirm with our instrumentation by taking readings at the card's power connectors. In fact, let's start with the measured power consumption values in the following table:

These charts go into more detail on power consumption at idle, during 4K gaming, and under the effects of our stress test. The graphs show how load is distributed between each voltage and supply rail, providing a bird's eye view of load variations and peaks.

Temperature Results

As you can see from the 1070's cooler, Nvidia went with a simple copper heat sink instead of the GeForce GTX 1080's vapor chamber-based solution. Given the price of this card, that's a rather incomprehensible cost-cutting measure.

The beefy frame helps with rigidity, but it was also designed to host thick thermal pads, which dissipate waste heat from the MOSFETs and memory modules.

Unfortunately, Nvidia's cooling compromises have an effect on the temperatures we measured:

Not only does GP104 touch 176°F (80°C), it actually exceeds that value in a closed case. During the gaming loop in an open case, we measured 172°F (78°C) at the GPU package.

We observed similar power consumption during the torture loop, which is due to the fact that this card hits its power target. Not surprisingly, we also see corresponding temperatures.

Sound Results

As a consequence of the higher temperatures, Nvidia's fan has to work harder. During our stress test, the fan speed rose slowly in the beginning due to a lower clock rate. However, after about 30 minutes, it eventually caught up with what we saw during our gaming loop.

The obvious consequence of a spinning fan is noise. Yet, our acoustic measurements are still relatively moderate when the card is idle, despite the sound's slightly snarly character.

Under prolonged load, the noise levels reach almost 42 dB(A). That's still acceptable for a radial fan. At this point, the frequency spectrum becomes rather wide, so the noise doesn't feel too intrusive:

Overall, the radial fan solution is workable. But it's miles away from making this a quiet, gentle card.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

NEW: Asus ROG Strix GTX 1070

Asus uses its ROG Strix brand to denote the high-end solutions in its portfolio. But where does the ROG Strix GeForce GTX 1070 land in comparison to other companies with premium models?

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The cooler cover is made of anthracite-colored matte plastic. This does look a bit boring, and it doesn't feel particularly inspiring in-hand, either. Perhaps the card's strengths lie elsewhere.

Weighing in at 1048 grams, Asus' board lands in the heavyweight category. You shouldn't have any trouble securing it into your case, though. A little more critical is its 30cm length. The card's height of 12.5cm is rather average, and its 3.5cm width is typical of all dual-slot designs. A total of three 90mm fans ensure the right amount of airflow and pressure cool the heat sink underneath.

The back of the board is covered by a single-piece plate made of what looks like anodized aluminum. Asus calls this its Aura RGB Lighting Backplate, and it's adorned with a back-lit ROG logo. This backplate makes it necessary to plan for an extra 5mm of clearance behind the card, which may be relevant in multi-GPU configurations.

The card's top sports a Republic of Gamers label with LED background lighting and an eight-pin power connector.

A peek at the end and bottom of the ROG Strix GeForce GTX 1070 reveals that its fins are oriented vertically. They won't allow any waste heat to exhaust out the back. Instead, hot air is pushed from the top and bottom, warming up other components in your case, along with your motherboard underneath. As such, this design decision is rather counterproductive.

The slot plate features five display outputs, of which a maximum of four can be used simultaneously in a multi-monitor setup. In addition to one dual-link DVI-D connector, the bracket also hosts two HDMI 2.0b ports and two DisplayPort 1.4-capable interfaces. No doubt, this is a nod to Rift and Vive owners who need at least one HDMI output for their HMD. The rest of the slot plate is dotted with openings for airflow, though they're not functional due to Asus' fin design.

Board & Components

Asus uses its own circuit board for the ROG Strix GeForce GTX 1070. As far as we can tell, the company doesn't make any concessions in its layout or component choice. However, we also don't see any exclusive traits that'd set this configuration apart.

This card uses eight Samsung K4G80325FB-HC25 modules with a capacity of 8Gb (32x 256Mb). Each chip operates at voltages between 1.305 and 1.597V, depending on the selected clock frequency.

But let's get back to the PCB and power supply. Asus employs an almost oversized 6+1-phase design, wherein the six GPU phases are supplied by uPI Group's uP9511 eight-phase buck controller. The GPU's voltage regulation is implemented using one IR3555 PowIRstage per phase. This chip includes a gate driver, control and synchronous MOSFETs, and a Schottky diode, saving a lot of board space.

The memory's one phase is provided by a uP1666 2/1-phase synchronous buck controller. This phase takes quite a bit of load, so to help it out, Asus' card utilizes two UBIQ Semiconductor QM3054M6 N-channel MOSFETs in parallel on the high side and two QM3056M6 N-channel MOSFETs on the low side.

Asus relies on its homegrown "Super Alloy Power" chokes, which the company says guarantee a particularly stable power supply. What Asus refers to as its SAP technology, MSI calls Military Class. In the end, both are meant to ensure increased durability and longer life expectancy for capacitors. Of course, there is no way for us to verify that claim in a review, but it certainly sounds good.

Asus' ROG Strix GeForce GTX 1070 uses an ITE 8915FN to monitor and control the current. Two familiar capacitors are installed right below the GPU to absorb and equalize voltage peaks.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. This also shows that both curves drop as the GPU's temperature rises. Asus imposes a relatively low power target of approximately 170W, which in turn causes a relatively frenetic GPU Boost frequency that's primarily limited by the power consumption cap.

After a warm-up run through our variable gaming load, the card's GPU Boost clock rate settles at an average 1946 MHz, down from a starting point of 2015 MHz. Under a more constant load, it falls to an average of 1734 MHz.

The voltage measurements look similar. Readings around 1.062V drop to 1.025V as the board's frequency slides.

Combining the measured voltages and currents allows us to derive a total power consumption we can easily confirm with our instrumentation by taking readings at the card's power connectors.

As a result of restrictions imposed by Nvidia, whereby the lowest attainable frequencies are sacrificed to hit higher GPU Boost clock rates, the power consumption of many factory-overclocked cards is disproportionately high when they're idle. This one can only go as low as 291 MHz. The following table shows what impact that has on our measurements:

These charts go into more detail on power consumption at idle, during 4K gaming, and under the effects of our stress test. The graphs show how load is distributed between each voltage and supply rail, providing a bird's eye view of load variations and peaks.

The 5.1A we measure provides a comfortable margin below the PCI-SIG's 5.5A maximum for a PCIe slot, especially if you're using this card on an older motherboard. Asus only feeds the memory and one GPU phase through the PCIe slot; the other five phases are powered through the auxiliary eight-pin connector.

Temperature Results

Asus' backplate doesn't play an active role in cooling the ROG Strix GeForce GTX 1070. It's mostly aesthetic, though the plate does contribute to the card's structural rigidity.

The ROG logo's back-lighting is achieved using a simple scattering film with an embedded LED. This diode is then plugged into a socket on the card's PCB.

A basic frame on the front cools most of the memory modules, though its shape could be better. While three of the ICs receive some extra cooling thanks to a thermal pad connecting them to the cooler's heat sink, there is one module that gets almost no cooling at all. 

The sink employs a classic heat pipe direct touch design using flattened and sanded pipes. There are four 8mm heat pipes and one 6mm pipe. To dissipate up to 170W, this configuration is more than sufficient.

Reaching temperatures of up to 144°F (62°C), Asus' ROG Strix GeForce GTX 1070 is still in the green. It's running cool enough to ensure the highest possible GPU Boost frequencies allowed by the low power target.

One look at an infrared image of the board reveals the impact of a well-designed cooler on voltage regulation circuitry. As a result of Asus' work, the PCB endures very little thermal stress.

The same holds true when we run our stress test and observe the temperature rising only marginally. This cooler is absolutely sufficient for what it's being asked to do.

Sound Results

Hysteresis is perfectly implemented, allowing the fan curve to leave a positive impression throughout its range. Even after an hour, the three fans spin at less than 1600 RPM.

Registering 37 dB(A) under full load, Asus' card lands in the middle of our pack for noise. But a closer analysis of the frequency spectrum sheds more light on where that reading's peaks appear. Although the three fans do generate some bearing noise, the sound is balanced well. At a purely subjective level, this is less annoying to listen to than the deeper tones produced by other cards with nominally lower dB(A) values.

There's also some electrical noise caused by the voltage converters. But it's in the high-frequency range that most enthusiasts won't be able to perceive.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1060 Roundup

MORE: All Graphics Content

NEW: Gigabyte GTX 1070 G1 Gaming 8G

If there's one thing we've come to expect from Gigabyte's G1 Gaming cards, it's the highest possible performance at a reasonable price. The company's GeForce GTX 1070 G1 Gaming 8G follows that familiar mission with three fans and a fairly bold-looking fan shroud.

Technical Specifications

Exterior & Interfaces

The fan shroud is made of anthracite-colored matte plastic with decorative orange highlights. Weighing in at a modest 797 grams, the G1 Gaming 8G is a true flyweight. A length of 28.2cm should fit easily in most cases, and a height of 11.5cm is about average. The 3.5cm width is typical of dual-slot designs.

A total of three 80mm fans are meant to ensure the right amount of airflow and pressure. The fact that you can't tell these fans are fairly small is due to the card's proportions, which make it look bigger than it actually is.

The back of the board is covered by a single-piece plate made of what looks like anodized aluminum and adorned with a white Gigabyte logo (there is no back-lighting). This backplate makes it necessary to plan for an extra 5mm of clearance behind the card, which may be relevant in multi-GPU configurations.

Up top, the card is branded with a Gigabyte logo back-lit by an RGB LED. There's also a fan-stop indicator and an eight-pin auxiliary power connector.

A peek at the end and bottom of the GeForce GTX 1070 G1 Gaming 8G reveals that its fins are oriented vertically. They won't allow any waste heat to exhaust out the back. Instead, hot air is pushed from the top and bottom, warming up other components in your case, along with your motherboard underneath. As such, this design decision is rather counterproductive.

The slot plate features five display outputs, of which a maximum of four can be used simultaneously in a multi-monitor setup. In addition to one dual-link DVI-D connector, the bracket also hosts one HDMI 2.0b port and three DisplayPort 1.4-capable interfaces. The rest of the slot plate has openings cut into it, which look like they're meant to help with airflow. In this configuration, however, they're not functional due to Gigabyte's fin design.

Board & Components

Gigabyte's PCB is proprietary with some conservative (yet interesting) component choices.

This card uses eight Samsung K4G80325FB-HC25 modules with a capacity of 8Gb (32x 256Mb). Each chip operates at voltages between 1.305 and 1.597V, depending on the selected clock frequency.

One unique addition is the Holtek HT32F52241 32-bit ARM Cortex-M0+ MCU, which Gigabyte uses for controlling the card's RGB effects.

Gigabyte employs an almost oversized 6+2-phase design, wherein the six GPU phases are supplied by uPI Group's uP9511 eight-phase buck controller.

The GPU's voltage regulation is implemented using one Alpha & Omega AON6508 N-channel AlphaMOS on the low side of each phase and one AON6414A on the high side. What's interesting about these two MOSFETs is the low internal resistance across their performance range, as well as the low control values for the gates. It is thus safe to assume that this choice is an attempt to minimize voltage converter losses.

Gigabyte uses Foxconn's Magic coils, which are fully encapsulated and easy to install during automated assembly.

The memory's two phases are supplied by an unmarked PWM controller, which should roughly correspond to a uP1666 2/1-phase synchronous buck controller. It already comes with an integrated bootstrap Schottky diode and gate driver. The high- and low-side MOSFETs are similar to what we found on the GPU's power phases. Just the coils are slightly smaller.

Current monitoring is handled by a triple-channel Texas Instruments INA3221. Two familiar capacitors are installed right below the GPU to absorb and equalize voltage peaks.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. This also shows that both curves drop as the GPU's temperature rises. Gigabyte uses a power target of ~185W, which in turn causes a relatively frenetic GPU Boost frequency. Short-term drops in clock rate hint that we, at least temporarily, hit the card's power limit.

After a warm-up run through our variable gaming load, the card's GPU Boost clock rate settles at an average 1936 MHz, down from a starting point of 1974 MHz. Under a more constant load, it falls to an average of 1759 MHz.

The voltage measurements look similar. Readings around 1.062V drop to 1.043V as the board's frequency slides.

Combining the measured voltages and currents allows us to derive a total power consumption we can easily confirm with our instrumentation by taking readings at the card's power connectors.

As a result of restrictions imposed by Nvidia, whereby the lowest attainable frequencies are sacrificed to hit higher GPU Boost clock rates, the power consumption of many factory-overclocked cards is disproportionately high when they're idle. This one can only go as low as 240 MHz. The following table shows what impact that has on our measurements:

These charts go into more detail on power consumption at idle, during 4K gaming, and under the effects of our stress test. The graphs show how load is distributed between each voltage and supply rail, providing a bird's eye view of load variations and peaks.

The 3.6A we measure provides lots of margin below the PCI-SIG's 5.5A maximum for a PCIe slot, especially if you're using this card on an older motherboard. Gigabyte only feeds the memory through the PCIe slot; the other six phases are powered through the auxiliary eight-pin connector.

Temperature Results

Gigabyte's backplate doesn't play an active role in cooling the GeForce GTX 1070 G1 Gaming 8G. It's mostly aesthetic, though the plate does contribute to the card's structural rigidity.

The built-in, two-stage heat sink for VRM and coils deserves a positive mention. We also like the fact that the memory modules are cooled directly by this simple design's stabilizing frame.

A direct consequence of this simplicity, unfortunately, is an uninspired and cheap-looking thermal solution. It consists of a finned sink with a thick aluminum base plate that also serves as a mount for two 8mm heat pipes in a flattened direct touch configuration. These pipes don't even cover the GPU entirely. They're hardly sufficient for moving excess heat. It would have been more appropriate for Gigabyte to use three heat pipes instead.

Of course, a maximum GPU temperature of 147°F (64°C) keeps this card in the green. But a safe thermal ceiling isn't attributable to the cooler, but rather Gigabyte's very low temperature target. After hitting that ceiling, the fans are audibly hard at work.

Looking at our infrared images, we see that the GPU package (and thus the PCB) is slightly warmer than the cooled GPU itself. A reading of 172°F (78°C) on the voltage converters is also bearable. But that's not surprising given dedicated cooling.

The same holds true when we run our stress test and observe the temperature rising only marginally. This cooler is merely acceptable, since it's already operating at its limit.

Sound Results

Hysteresis is perfectly implemented; there is no annoying on/off/on loop to worry about. However, permanent fan speeds above 2000 RPM are certainly nothing to celebrate, since they're clearly audible.

Although we've heard worse than 38.3 dB(A) under full load, Gigabyte's fans are still significantly louder than many other GeForce GTX 1070s, and unnecessarily so.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1060 Roundup

MORE: All Graphics Content

Gigabyte GTX 1070 Mini ITX OC

Even the longest can fall short when it comes to mini-ITX enclosures. Suddenly there are other attributes that matter more than just performance, like length, power, and where all of that hot air goes after it's blown off the GPU.

With its GeForce GTX 1070 Mini ITX OC, Gigabyte adapts well to the limitations of small cases, creating a card that's short and still wields Nvidia's powerful GP104 processor. That leaves us with just two questions: how well can the chip be cooled, and how well does Gigabyte manage the trade-off between less surface area and noise output?

This board's price makes it an interesting offering, since Gigabyte deliberately avoids gimmicks and concentrates on the qualities that matter.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The cooler cover is made of black plastic with orange lacquered highlights. Weighing just 605g, this card is a featherweight among the other 1070s. A length of just 6 3/4 in (17.2 cm) also makes the GeForce GTX 1070 Mini ITX OC conveniently short. The height (12.5cm from the upper edge of the motherboard) and depth (3.5cm, the same as most dual-slot cards) are perfectly acceptable.

Gigabyte deliberately doesn't use a continuous single-piece backplate; in ITX projects, thicker structures attached to the back of a card often cause collisions with CPU coolers or memory. The smaller plate at the end of the card shouldn't pose any problems, and it does serve a real purpose.

The top of the card bears an unlit Gigabyte logo, and the eight-pin auxiliary power connector sits at the top's back edge. We can appreciate the simple, yet functional design. Gigabyte even held off on eye-catching lighting, since few mini-ITX enclosures are windowed.

Horizontally-oriented fins direct heated air to the back and front of the card. Accordingly, the slot plate has some openings for ventilation.

You get a total of four display outputs, all of which can be used simultaneously in a multi-monitor setup. In addition to the two dual-link DVI-D connectors, there is also one full-sized HDMI 2.0b port and a DisplayPort 1.4-capable interface.

Board & Components

Gigabyte designed an extra-short PCB for this card. The true art is finding a configuration in which the VRMs are arranged in such a way that potential hot-spots are elegantly avoided, despite the lack of space. In practice, Gigabyte's implementation works well.

Like Nvidia's GeForce GTX 1070 Founders Edition, this card uses eight Samsung K4G80325FB-HC25 modules, each able to store up to 8Gb (32x 256Mb).

Such a short card presents issue with placing the memory. On one hand, distance to the GPU cannot be reduced. On the other, the modules must be kept as far away as possible from the voltage regulation circuitry, else they overheat.

The 4+1-phase system relies on the same uP9511P for PWM control as Nvidia's reference card. Gigabyte places this chip on the back of its PCB, though. The memory's one power phase is controlled by a uP1728 on the top side of the board. The converter rails are doubled, a trick that allows each of them to be advertised as two phases, rather than one. Of course, Gigabyte uses an obligatory INA3221 for monitoring current, too.

Gigabyte implemented a voltage regulation solution that allows it to keep the card short without risking thermal issues. Foxconn coils and 6414 high-side MOSFETs are placed on the front of the PCB, while the five gate drivers responsible for controlling the individual phases are banished to the back.

This is also where the 6508 low-side MOSFETs are found. Two MOSFETs are used for each converter circuit, halving the already low internal resistance of the circuits a second time.

The advantage of this setup is that the low-side MOSFETs can be cooled from above by the cooler, while the low-side chips benefit from increased surface area due to the doubling of MOSFETs. This allows them to be cooled by a small screwed-on plate, preventing problematic hot-spots.

Furthermore, two capacitors are installed right below the GPU to absorb and equalize peaks in voltage.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. This also shows that both curves drop as the GPU's temperature rises.

After warming-up in our gaming workload, the GPU Boost frequency fluctuates between 1778 and 1860 MHz. Under a more constant and taxing load, the clock rates drop significantly to a range between 1584 and 1607 MHz.

The voltage measurements respond similarly. While we observe ~0.975V in the beginning, this value later drops as low as 0.893V. It is easy to tell that Gigabyte had to put clock rates on a diet in order to manage power consumption and thermal output.

Combining the measured voltages and currents allows us to derive a total power consumption we can easily confirm with our instrumentation by taking readings at the card's power connectors.

As a result of restrictions imposed by Nvidia, whereby the lowest attainable frequencies are sacrificed to hit higher GPU Boost clock rates, the power consumption of many factory-overclocked cards is disproportionately high when they're idle. Gigabyte achieves a good compromise, though. Its lowest frequency is 164 MHz, only slightly above Nvidia's reference.

These charts go into more detail on power consumption at idle, during 4K gaming, and under the effects of our stress test. The graphs show how load is distributed between each voltage and supply rail, providing a bird's eye view of load variations and peaks.

Temperature Results

Naturally, power consumption directly affects temperatures. The question of how well Gigabyte's compact GeForce GTX 1070 Mini ITX OC copes with thermal energy can only be answered by looking closely at its cooling solution. The company employs two 1/3-inch (8mm) and one 1/4-inch (6mm) heat pipes made of copper-composite material. They make direct contact with the GPU, accelerating dissipation out through the heat sink's fins.

The small plate for cooling the memory modules is seated on an aluminum heat sink. Three heat pipes are embedded in it, and they're directly connected to the finned cooler.

Gigabyte actively cools the six high-side MOSFETs using the cooler's integrated heat sink. It also draws heat from the coils with thermal pads, and applies slight pressure through them to combat vibration, which we come to know as coil noise. This solution kills two birds with one stone; it eliminates hot-spots and noise simultaneously.

The smaller cooling plate on the back covers all 12 low-side MOSFETs and the gate drivers, dissipating excess heat through the back of the card. As you might imagine, this helps with the small board's thermal challenges quite a bit. Registering 151°F (66°C) on an open test bench, and 154°F (68°C) to 156°F (69°C) in a closed case, the GPU's temperatures are completely acceptable.

A relatively high idle temperature between 118°F (48°C) and 122°F (50°C) is attributable to the card's passive mode. Gigabyte's small cooler doesn't really seem suitable for that purpose. In a compact enclosure, we'd suggest keeping the fan spinning at all times, even slowly. Fortunately, this behavior can be adjusted manually through software.

The infrared images reveal no troublesome hot-spots. Nvidia's GPU, Samsung's memory, and the voltage regulators are cooled well, particularly given a dense PCB.

Temperatures rise a bit during our stress test, but they're not high enough to cause concern. The bottom line is that Gigabyte's concept works.

Sound Results

Since the temperatures during our gaming and torture workloads are similar, the observed fan speeds are about the same as well.

There is no start-up peak in the graph, which would normally indicate the jump from passive to active mode. The installed 90mm fan tops out at 3300 RPM and is stable at speeds as low as 400 RPM. As a result, no tricks are necessary to get it up and running. Gigabyte keeps a good grip on hysteresis as well, and that might be helped by the fact that the fan is activated early, just below 132°F (56°C).

Although the 41.8 dB(A) we measure after a prolonged workload isn't particularly quiet, it's also also not loud enough to bother us. Even Nvidia's GeForce GTX Founders Edition is significantly louder. The high-frequency parts of the VRM still generate quantifiable noise, but they're barely audible. Low-frequency buzzing is rather moderate, and the part that actually is measurable is due to airflow.

Gigabyte does a good job with this small card's cooling; it'd be hard for us to suggest significant improvements.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

MSI GTX 1070 Gaming X 8G

MSI managed to build a reputation for itself with the Gaming series, and its GeForce GTX 1070 Gaming X 8G strives to live up to high expectations set by those predecessors. The coming employs a conservative default clock rate and a good-enough, low-noise cooler.

There's a higher-end family called Lightning in MSI's portfolio, so it remains to be seen how much the manufacturer held back to create the more mid-range Gaming X version. At the same time, this also isn't MSI's entry-level GeForce GTX 1070. That distinction goes to the simpler Armor 8G and Aero 8G.

We're testing a press sample of the Gaming X with a non-standard firmware, as it turns out. In this version, the OC mode is active by default, resulting in a ~20 MHz-higher core and GPU Boost clock rate. In practice, the differences between this card and the retail one are negligible, so we didn't re-flash the BIOS for our tests. Instead, we switched back to the standard mode via vendor-supplied software. Thus, the GPU's frequency in our benchmarks is still equivalent to the retail model.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

MSI's fan shroud is made of relatively thin plastic, though it's mounted on a large frame for improved stability. Weighing in at 1080g, this card isn't particularly light. It's 28.4cm long, 13.5cm tall, and 3.5cm wide, similar to most dual-slot cards. The rotor blades of the double ball bearing fans have a diameter of 95mm.

The back of the board is covered by a single-piece plate with some holes for ventilation, but it doesn't contact the PCB through thermal pads and consequently does nothing for cooling. Unfortunately, this backplate makes it necessary to plan for an extra 5mm of clearance behind the card, which may be relevant in multi-GPU configurations.

Using the card without its backplate proved difficult due to integrated spacers. Furthermore, the screws attaching the backplate have threaded portions that are too short to secure the frame and VRM cooler once the plate is gone. That's why we had to abstain from running the card without its backplate, limiting our view during the IR temperature testing.

The top of the card is dominated by an illuminated MSI logo and three visible heat pipes (two 6mm and one 8mm). Six- and eight-pin auxiliary power connectors are positioned at the end of the card and rotated by 180°. MSI's black and red color scheme should be familiar by now, similar to the card's shape.

An open end reveals horizontally-oriented fins, which channel airflow towards the card's front and back. This is an advantage since a lot of heated air exhausts out the slot cover. What remains isn't directed out the top, close to your CPU. Although we like centrifugal coolers able to blow everything out the back, this is a next-best solution.

The slot plate features five display outputs, of which a maximum of four can be used simultaneously in a multi-monitor setup. In addition to one dual-link DVI-D connector, the bracket also hosts one HDMI 2.0b port and three DisplayPort 1.4-capable interfaces. The rest of the slot plate is dotted with openings for airflow.

Shedding that often-unused DVI connector would have made room for even more ventilation, which may have helped the cooling solution's effectiveness. Ah well.

Board & Components

A glance at the board reveals a well-arranged layout. It uses eight Samsung K4G80325FB-HC25 modules, each able to store up to 8Gb (32x 256Mb). Each chip operates at voltages between 1.305 and 1.597V, depending on the selected clock frequency.

The 8+2-phase power subsystem relies on the sparsely documented µP9511P for PWM control. However, unlike Nvidia's Founder Edition card, the controller is placed on the front rather than the back. Since this controller can't communicate directly with the VRM's phases, MSI utilizes gate drivers to talk to the dual-channel MOSFETs. And because MSI is using dual- rather than single-channel MOSFETs, it can use space on the PCB more efficiently. The company also goes with Super Ferrite Chokes, which are a bit classier than what you typically find on reference designs.

Two capacitors are installed right below the GeForce GTX 1070 Gaming X's GPU to absorb and equalize voltage peaks.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. This also shows that both curves drop as the GPU's temperature rises.

After warm-up, GPU Boost frequencies fall as low as 1962 MHz under load. This is mirrored by our voltage measurements. While we observed up to 1.062V in the beginning, just like Nvidia's Founders Edition, that value drops to 1.05V later on.

Combining the measured voltages and currents allows us to derive a total power consumption we can easily confirm with our instrumentation by taking readings at the card's power connectors.

As a result of restrictions imposed by Nvidia, whereby the lowest attainable frequencies are sacrificed to hit higher GPU Boost clock rates, the power consumption of many factory-overclocked cards is disproportionately high when they're idle. This one can only go as low as 253 MHz. The following table shows what impact that has on our measurements:

These charts go into more detail on power consumption at idle, during 4K gaming, and under the effects of our stress test. The graphs show how load is distributed between each voltage and supply rail, providing a bird's eye view of load variations and peaks.

Temperature Results

As it pertains to cooling, MSI has a lot to offer. Its sturdy frame covers almost all of the PCB not already covered by heat sinks for the VRMs.

Thermal pads between the memory modules and frame, along with the voltage regulators and cooling plate, provide the interfaces needed to transfer heat away from those areas. However, we would have liked thinner and more efficient design, since fan speeds (and thus airflow) are rather low.

The massive cooler has horizontally-oriented fins as well as four nickel-plated 1/4-inch (6mm) copper heat pipes, plus an 8mm one. The direction they face is more or less irrelevant since they consist of sintered composite material, relagating unreliable axial grooves or meshes to ancient history. MSI furthermore relies on a massive nickel-plated heat sink.

A rather restrictive 149°F (65°C) temperature target does result in the fans reacting faster with a shorter start-up time. But this is unfortunately foiled by their throughput and maximum required speed.

Since we couldn't test the GeForce GTX 1070 Gaming X 8G without its backplate, we had to take our measurements with that cover in place. On the bright side, one of the plate's ventilation openings is right above a memory module positioned between the VRMs and GPU. The measured 176°F (80°C) is acceptable.

Temperatures actually dropped during our stress test, since the card throttled back its clock rate, lowering power consumption.

Sound Results

Since the temperatures during our gaming and torture workloads are similar, the observed fan speeds are about the same as well. The start-up behavior and well-implemented hysteresis, which prevents multiple on/off cycles once the fans start spinning, is clearly visible.

In addition, the start-up speed is chosen in such way that the fans will certainly continue to start reliably, even as they age. The same goes for when the fans turn off.

When the card is idle, there is no noise thanks to its semi-passive mode. Naturally, there's no need to take acoustic measurements.

Running at full load barely raises the noise level to 33 dB(A). This value doesn't go any higher, even during our torture test. The card is so quiet that we were able to measure >5 kHz frequencies coming from the voltage regulators, which most people would barely be able to hear. It's only when the GeForce GTX 1070 Gaming X 8G cranks out frame rates in the hundreds that you hear the VRMs chirp a little louder.

It seems like the cooler is almost overkill for its job. On the other hand, we prefer being safe to sorry. MSI obviously does a good job.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

Palit GTX 1070 GameRock Premium Edition

JetStream, Super JetStream, GameRock, and GameRock Premium Edition: Palit sure does provide a lot of GeForce GTX 1070-based options. The card we're testing is just as bulky as its name thanks to the oversized cooler.

Despite all of the cooling headroom you'd seemingly get, this version does have issues with hysteresis, causing the fans to start and stop during warm-up. Unfortunately, the situation isn't any better, even after downloading an available firmware update. Consequently, we're putting further updates to this piece on hold until we're offered a real solution.

Just like the GTX 1080-based version, this card looks massive at first glance. But we'll have to test how much of that bold appearance translates into real-world benefits.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The fan shroud is made of relatively thick, white plastic. The top and front are decorated with metallic blue and  brushed metal highlights.

A weight of 1053g makes this card about 180g lighter than the 1080-based version. However, with its length of 28.7cm, height of five inches (12.8cm), and two inches (5.2cm) of width, it boasts the same dimensions as its higher-end counterpart and also spans three slots. The two massive 100mm fans with rotor diameters of 96mm further emphasize the card's size.

Flip the card over and it's covered by a one-piece backplate without any openings for ventilation. It is decorated with a highly visible GameRock logo and requires an additional 5mm for clearance. Since there are no thermal pads between the plate and PCB, the backplate serves decorative purposes only. While it is perfectly possible to use the card without this plate, removing it requires taking off other components, likely voiding Palit's warranty.

The top of the card is dominated by a centered, brightly lit Palit logo. An eight-pin auxiliary power connector is positioned at the board's end and rotated 180°. This isn't a humble product by any means; it stands out and wears its heft proudly.

At its end, the card is completely closed off, which makes sense since the fins are positioned vertically and won't allow any airflow in that direction anyway.

The slot bracket features five display outputs, of which four can be used simultaneously in multi-monitor setups. In addition to one dual-link DVI-D connector, which doesn't loop through any analog signal, you also get one HDMI 2.0b port and three DisplayPort 1.4-capable interfaces. The rest of the slot plate is peppered with openings that look like they're meant to facilitate airflow. However, they're more decorative than functional.

Board & Components

The board looks clean enough, similar to the 1080-based version. It uses eight Samsung K4G80325FB-HC25 modules, each able to store up to 8Gb (32x 256Mb). Each chip operates at voltages between 1.305 and 1.597V, depending on the selected clock frequency. However, caution should be advised! Some manufacturers have switched to Micron memory modules, which is recognizable in the BIOS naming scheme (86.04.26.xx.xx versus Samsung's 86.04.1E.xx.xx).

The 8+1-phase system, like Nvidia's reference cards, relies on the sparsely documented uP9511P for PWM control. Also like Nvidia's own implementation, the controller finds a home on the back of the PCB. All eight of the GPU's phases are realized using this component, which is actually designed as a 6+2-phase jack-of-all-trades.

The DC/DC voltage converters' dual-channel MOSFETs are controlled directly, as these eight SiC632s are so-called driver MOSes. They combine the actual power MOSFETs for high-side and low-side, as well as the gate driver and Schottky diode on one chip. This is certainly more cost-efficient and a plus for compact designs, especially when a large number of voltage converter circuits are in play.

In contrast to Palit's GTX 1080 model, this board's memory gets power from one phase (instead of two). It's controlled by the same undocumented chip used on Nvidia's reference board, which should be almost identical to the well-known 1728. A dual N-Channel model is used for the MOSFETs, which combines both high-side and low-side.

The Foxconn coils are middle-class. Depending on the layout, they operate more or less quietly. And as is often the case, a well-known INA3221 handles current monitoring.

Two capacitors are installed right below the GPU to absorb and equalize peaks in voltage.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. This also shows that both curves drop as the GPU's temperature rises.

The graphs clearly show that the GPU Boost frequency after warm-up and under load falls from an excellent 2076 MHz to a still-good 1975 MHz (and sporadically a little lower). It is also apparent that voltage follows the sinking clock rates. While we measured up to 1.062V in the beginning, this value later drops as low as 0.901V.

Combining the measured voltages and currents allows us to derive a total power consumption we can easily confirm with our instrumentation by taking readings at the card's power connectors.

Since manufacturers sacrifice the lowest possible frequencies to gain an extra GPU Boost bin due to Nvidia's restrictions, the GTX 1070 GameRock Premium Edition's power consumption is slightly higher at idle. Palit sets the first GPU Boost step at 316 MHz.

These charts go into more detail on power consumption at idle, during 4K gaming, and under the effects of our stress test. The graphs show how load is distributed between each voltage and supply rail, providing a bird's eye view of load variations and peaks.

Temperature Results

Of course, power dissipated as waste heat needs to be dealt with as efficiently as possible. So, we start by looking at the backplate, which doesn't do any real cooling and instead leaves that job to Palit's 2.5-slot thermal solution. The entire structure matches what you get on the GTX 1080 card. It appears almost decadently oversized given how much less power a 1070 consumes. On the other hand, this isn't a bad problem to have. We'd rather have too much cooling than not enough.

A copper sink moves heat away from the GPU and spreads it through a total of five pipes (three 8mm and two 6mm). Palit chose to orient the sink's fins vertically, which results in short, straight 8mm pipes that work more efficiently. The two smaller pipes don't do much except provide additional area to support the transport of heat away from the sink and towards the cooler's edges.

The performance of this truly monstrous cooler leaves little to be desired. Since the temperatures only get to 151°F (66°C) during our gaming loop (154°F/68°C in a closed case) and 151°F (66°C) as well during our stress test (158°F/70°C in a closed case), the fans only need to run at low power, which should have a positive impact on the measured noise level.

The transfer of heat away from the VRMs works perfectly, despite low fan speeds and very little airflow. That massive cooler and its endless fins works wonders.

During our stress test the temperatures do rise a bit at the hottest spot, despite low average power consumption. All other areas remain at perfectly safe values, though.

Sound Results

Let's talk a bit about the one single quality that dominated our impression of this graphics card: its noise level. While some folks might enjoy the deep roar of a well-oiled machine, we the sound of silence from our graphics cards. When it's disturbed, the culprit is usually a fan or sometimes the VRM's coils. But a maximum of 1000 RPM for the two fans should be no reason to declare a state of acoustic emergency.

To examine this behavior in more detail, we need to take a closer look at the fan curve, which unfortunately reveals an unpleasant surprise. Since the fans generally start late and keep quiet, it takes diligent measurement of tachometer signals and PWM values to detect poor hysteresis. Palit confirmed this for us, even.

When the card is idle, no noise is measurable due to its semi-passive mode. We thus abstained from trying to take any readings.

The values we measured under load are blissful, and the machine is a purring kitten. A result of 35.1 dB(A) is good considering the temperatures involved. Only the mid-range coils tend to stick out a bit. If it wasn't for their audible chirp disturbing the calm of night, you might even question whether the card was running at all.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

Passive Cooling Shootout: Which One Is Better?

Our German team has a bit of an obsession with passive cooling. In case you missed it, check out their Nvidia GeForce GTX 1050 Ti Passive Cooling Mod. As we saw in that piece, at least some airflow was necessary to keep Nvidia's 75W card working properly. Nowadays you can buy a couple of different retail products that sport passive cooling, without the need for our do-it-yourself effort. One is XFX's Radeon RX 460 Heatsink Edition, and the other is Palit's GeForce GTX 1050 Ti KalmX. Both sport 4GB of GDDR5 memory.

Does either card work the way it's advertised? A specially designed cooler with optimized fin spacing is obviously on the right track, but we're professionally skeptical and obligated to test in every way, shape, and form. You'll see us benchmarking on an open-air chassis and in a closed case with single- (front/positive pressure) and dual-fan (front and back/negative pressure) configurations.

XFX and Palit apparently came up with different approaches for utilizing the airflow in your case, based on the way they have their fins positioned. At a glance, internal convection may be adequate to cool Palit's card, while the XFX solution appears to need some additional help from a fan.

Based on their respective GPUs, Palit's GTX 1050 Ti KalmX should obviously be the faster card. But that's not our primary concern: the boards first have to work the way each company's marketing department advertises before we even start thinking about gaming. Right now, cooling is everything!

Our test system and methodology should already be familiar to you from How We Test Graphics Card. But if that story is new to you, we encourage you to check it out before digging in to this piece.

The following table provides a quick overview:

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

XFX Radeon RX 460 Heatsink Edition

XFX's Radeon RX 460 Heatsink Edition looks a lot like the older HIS Radeon HD 7750 iCooler card. That company, now under the XFX (Pine) umbrella, no longer offers passively-cooled AMD cards, but instead shifted to exotic single-slot solutions. The cooler size and fin arrangement we're testing today are clearly based on older HIS models, though.

MORE: Best Deals

MORE: Best PC Builds

Exterior

Our first impression of the Radeon RX 460 Heatsink Edition is that it's small and cute, weighing in at just 337g. The heat sink extends a bit beyond the PCB, resulting in an overall length of 20.2cm (from the display output bracket to the end of the cooler). Its 12.5cm height (motherboard slot to the top of the heat pipe) is workable in most cases, and 3.5cm width requires two slots.

Flipping the card over, we can see its PCB measures just 16.5cm long. XFX fortunately dispenses with a backplate and wrap-around cooler, helping to avoid potential conflicts in compact mini-ITX cases.

Both nickel-plated heat pipes run from top to bottom. This adds 3cm to the card's height, but is necessary since the cooling fins are oriented horizontally.

The end of the card is completely open, which is appropriate for a passive solution. Up front, the output bracket sports a number of holes for air to escape, while exposing one DVI-D port, a DisplayPort 1.4-capable interface, and an HDMI 2.0 connector.

Circuit Board Layout And Power Supply

The voltage regulation circuitry is placed near the display output bracket, which makes perfect sense when you consider the card's power comes from a motherboard slot. The +12V pins are right there at the end.

XFX follows AMD’s requirements and uses Samsung K4G80325FB-HC28 memory modules, each with a capacity of 8Gb (32x256Mb). They reach a maximum of 1750 MHz at voltages between 1.305V and 1.597V. This card enjoys a thermal advantage compared to Radeon RX 470, which uses faster modules.

XFX gets creative with the Richtek RT8880C, a true dual-output PWM controller with three integrated drivers for the GPU's power phases. At the same time, it's able to control the memory's one phase. This saves space and simplifies layout significantly.

The high side of each GPU phase is equipped with one QM3092M6 N-channel MOSFET; the low side sports two QM3098M6 N-channel MOSFETs. Somewhat crudely-cast ferrite shells for the coils are adequate, but not what we'd call stellar. Since the GPU's maximum of 45W are divided up between these phases, the currents flowing through them shouldn't be an issue.

The memory uses one QM3098M6 N-channel MOSFET for the high and low sides, as well as a smaller Foxconn-style coil in the usual molded cup.

Power Consumption

This card's power consumption is manageable, which is a basic requirement for passive cooling. It hits a searing 62W in our gaming loop, and drops to 58W in our stress test. Given the cooling available to it, we simply cannot expect much more.

Our peak measurement represents a brief moment in time; it can't be used to characterize the board's power consumption over longer periods.

The following graphs show two-minute runs in our gaming loop and stress test; they provide the basis for calculating average power consumption.

Now we get to look at our current measurements, which fall below the PCI-SIG's 5.5A limit for the 12V motherboard slot.

Cooler Assembly

XFX's cooler is easily removed by loosening four screws surrounding the GPU package. Two 6mm nickel-plated copper alloy heat pipes are sandwiched between a small, yet beefy, copper sink and an aluminum block above it. The aluminum block supports the actual cooler assembly, while the pipes dissipate waste heat through the fin array.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Palit GeForce GTX 1050 Ti KalmX

Palit introduced its first passively-cooled board in the GeForce GTX 750 Ti KalmX and followed up with the GeForce GTX 1050 Ti. Nothing much has changed visually, other than the new substructure. But how effective is Palit’s passive implementation and what are its limits? After all, the 750 Ti was a 60W board, while 1050 Ti is rated at 75W.

A known limitation of this card is its relatively constrained clock, which starts at 1291 MHz and is rated for a GPU Boost frequency of 1392 MHz.

MORE: Best Deals

MORE: Best PC Builds

Exterior

The card weighs 466g, only a bit more than XFX's board. Its 18.3cm length, 13.7cm height, and 3.6cm width are all manageable in a dual-slot form factor.

Palit dispenses with a backplate and overhanging rear fins, in contrast to its older passive coolers, preventing possible conflicts with CPU coolers in mini-ITX enclosures.

Palit again employs two nickel-plated heat pipes within a copper slug to dissipate thermal energy as effectively as possible. The sink, with its embedded aluminum fins, juts out the end and top edge of the PCB to capture circulating air. The fins are oriented vertically, similar to Palit's previous-gen design. Although this is better for natural convection, it benefits less from air flowing from the front of the case.

Vertically-oriented fins also benefit when the card is installed standing up, as you'd find in a cube-style case.

The display output bracket is limited to one DVI-D port, one HDMI 2.0 interface, and DisplayPort 1.4-capable connector. This leaves a lot of room for the many honeycomb-shaped air openings (though they seem fairly useless, given Palit's fin orientation).

Circuit Board Layout And Power Supply

Palit’s power target for this card is modest, as you might guess from the missing six-pin auxiliary connector. Similar to XFX, Palit also situates its voltage regulation circuitry near the output bracket, closer to the PCIe slot's relevant pins.

Although the PCB can accommodate three power phases for the GPU, only two are implemented.

In contrast to XFX, Palit uses memory modules from Micron, each with a capacity of 8Gb (32x256Mb). These operate at 1752 MHz and are not actively cooled, which creates a bit of an issue since Micron's memory runs somewhat hotter than Samsung's. Fortunately, we have the right tools to quantify the difference.

Back to the GPU's voltage regulation, which is controlled by a uPI Semiconductor uP9509. Each of the high-side GPU phases employs a 4C019 N-channel MOSFET, while the low side has two 4C024 N-channel MOSFETs. Standard ferrite coils, poured into cups, are manually fastened to the board.

The memory's power comes from a Richtek RT8128 synchronous buck PWM controller, as well as an On Semiconductor NTMFS4C024N single N-channel MOSFET for the high and low side.

Power Consumption

This card's power consumption is manageable, as it was for XFX's passively-cooled RX 460. The GeForce GTX 1050 Ti KalmX barely reaches 58W in our gaming loop, and slightly exceeds that figure during the stress test. Given a power target of 60W, this card simply can’t deliver much more.

Our peak measurement represents a brief moment in time; it can't be used to characterize the board's power consumption over longer periods.

So, both of the cards we're testing consume roughly the same amount of power, even though the GeForce's performance is expected to be appreciably higher.

The following graphs show two-minute runs in our gaming loop and stress test; they provide the basis for calculating average power consumption.

Now we get to look at our current measurements, which fall below the PCI-SIG's 5.5A limit for the 12V motherboard slot.

Cooler Assembly

Two 6mm nickel-plated copper alloy heat pipes are sandwiched between a small, yet beefy, copper sink and an aluminum block above it. The aluminum block supports the actual cooler assembly, while the pipes dissipate waste heat through the fin array.

Both vendors’ coolers differ in their fin orientations and resulting heat pipe structures. Palit at least also tries to establish thermal contact between the voltage regulation circuitry and cooler using thicker thermal pads. Do they help? Again, we have the tools to answer that.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Temperatures, Clock Rates, And Performance

Temperature Behavior For Different Setups

Unfortunately, we have to dash your hopes for a completely passively-cooled card right out of the gate. Both cards do work when they're installed in an open test bench, albeit just barely. But as soon as you build them into a case (even a large one full of holes), both cards eventually hit their limits. And that's despite no other heat sources in there to complicate things; our CPU is water-cooled, after all. We even left the enclosure's side and back walls as open as possible. While Palit's GeForce GTX 1050 Ti KalmX simmered away with virtually no GPU Boost headroom, the XFX Radeon RX 460 Heatsink Edition frequently displayed a black and white checkerboard pattern on the screen. The expected protection mechanisms didn’t even kick in once, and at 93°C you're in trouble.

Each measurement requires a full hour to generate, so we're limiting ourselves to three scenarios for the thermography analysis: the open setup, a closed case with a single fan (positive pressure), and two fans (negative pressure). For the temperature comparisons, however, we're comparing all four results.

We clearly see that XFX's much smaller cooler falls behind and requires at least one case fan to help the GPU.

The voltage regulators tell a similar story in the chart below. The MOSFETs on Palit's card sometimes have to carry more load due to the missing third phase, but they also benefit from thermal pads on the heat sink.

We already suggested that the Micron memory on Palit’s card runs hotter than Samsung memory, despite a nominally better cooler. But because XFX's Radeon RX 460 Heatsink Edition runs into trouble before it can benefit from cooler memory, the thermal story is actually reversed.

Unless you use two fans, the memory on Palit's board runs consistently hotter than Micron's specified 85°C maximum, so that the card’s otherwise good measurements melt away. These readings just aren't recommended for long-term use.

IR Thermographic Images

To complement our charts, we also used the high-resolution thermal camera to document the three most important test cases, including readings from the heat pipe.

Open Bench Table

Closed Case (Positive Pressure, One Fan)

Closed Case (Negative Pressure, Two Fans)

The measurements show very clearly that components not measured by on-board sensors climb to temperatures you wouldn't want to see for extended periods, based on their manufacturers' specifications. Palit would be well-advised to lower its memory voltage and frequency, or switch to cooler-running Samsung modules. Nvidia, which sells its GPU and memory bundles, could also play a role in this.

Achieved Clock Rates

Here's where it gets exciting, since high temperatures always have a negative impact on sustainable clock rates! We let each card run in a test loop for an hour, after which we determined the average frequency over several more minutes. You really need to make these measurements over an extended period because the real performance drop becomes evident after at least 30 minutes or so.

While the passively-cooled Radeon RX 460 only ends up showing small clock differences, it always remains below its advertised base frequency. Even two fans don't allow it to maintain a constant 1090 MHz. Obviously, XFX doesn't provide enough cooling surface.

Palit's GeForce puts in a better showing for two reasons. First, its cooler is in a completely different class due to more than 100g of extra metal. Also, Nvidia's GPU Boost technology works more precisely. GP107 limps along in a fanless case, but perks up a bit on the open bench, finally reaching full strength with the help of some air flow.

As for winners and losers, it it appears that Palit's GeForce GTX 1050 Ti only loses a little, while the slower XFX Radeon RX 460 Heatsink Edition doesn't do well at all. A better cooler would have improved its standing, since the Baffin GPU isn't as bad as this card makes it look.

Manufacturers: you have to test your products under real-world conditions. It's not enough for your R&D departments to simply mount hardware on exposed motherboards and test in an air-conditioned room. Installing these cards in closed bench tables would have shown why the concept doesn't work as-implemented.

Gaming Performance

We're using a couple of recent games as examples of what happens when these cards get hot. Sniper Elite 4 is playable using the Medium setting, even after heating both boards. Surprisingly, the GeForce GTX 1050 Ti's advantage isn't as large as its $40 premium might suggest it should be. Apparently, the DirectX 12-enabled engine favors AMD's GPU in this title.

DirectX 11-based titles are known to go the other way, and Watch Dogs 2 shows Palit's GeForce GTX 1050 Ti to be a fair bit faster, even with the same power specification and waste heat. Larger clock rate variations again become more significant, since GPU Boost reacts much more sensitively to heat.

The takeaway is that better cooling always pulls better performance along with it.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Summary And Conclusion

Light and shadows always go hand in hand. In the end, Palit and XFX both have skeletons in their closets here. Here's what we found:

XFX Radeon RX 460 Heatsink Edition

This card is well thought-out, but suffers because of its smallish cooler. Those horizontally-oriented fins virtually rule out self-convection.

We strongly advise XFX to use a larger cooler, which would take advantage of the full depth allowed by a wide slot bracket. A clock rate reduction to 1000 MHz might not make marketing happy, but it'd facilitate a lower voltage and allow power consumption to slide below the 50W mark.

The Radeon RX 460 Heatsink Edition isn't much more expensive than actively-cooled products based on the same GPU at least. With a bit of skill, you could even drop an ultra-quiet 92mm or 120mm fan on top of it to improve the card's behavior. That'd kind of defeat the purpose though, right?

Palit GeForce GTX 1050 Ti KalmX

Palit’s adaptation is significantly faster in our benchmarks, but it suffers from memory temperatures that climb too high. You can get actively-cooled cards that are barely audible for around $30 less, which kind of blows the point of this card away. Complicating matters for gamers in the U.S., Palit's hardware isn't readily available.

The GeForce GTX 1050 Ti KalmX would only be worth pursuing if you helped it out with a case fan. That is, if it didn't also have issues with memory temperatures, which might not even benefit from more airflow. If you like to tinker, we do know of thermal pads that'd fit. But you're better off spending less money to optimize a more readily-available card with active cooling that can be dialed down.

Conclusion

If you have at least one case fan able to push air through your enclosure, a passive card sounds like it could be a good idea. Palit's GeForce GTX 1050 Ti comes closest to making that dream a reality due to its finer-grained GPU Boost control and better heat sink.

However, both cards fail to live up to their marketing. The value of these cards is lost on us until they can do what they promise.

Again, we encourage manufacturers to test and optimize their products under real-world conditions. Only then will they preemptively spot issues like insufficient cooling area or overheated memory modules.

It's only fair to recommend not purchasing either of these cards. Any decent, actively-cooled alternative easily knocks the stuffing out of either passive model, regardless of whether you prefer AMD or Nvidia. This passive experiment was a nice try, but next time use better coolers and more conservative power targets.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Introduction & Overview

Nvidia launched its GeForce GTX 1060 6GB in July of 2016 to inevitable comparisons with AMD's Radeon RX 480 8GB. Although the 1060 was faster in DX11 games, it also commanded a premium that was harder to justify than the uncontested GeForce GTX 1070 and 1080.

A month later, Nvidia quietly rolled out a 3GB version of the 1060 to battle the 4GB RX 480. Its GPU took quite a haircut in the process, though, dropping from 1280 to 1152 CUDA cores and affecting performance far more than model name suggests. Presumably, Nvidia couldn't risk the 3GB and 6GB models appearing too similar at 1920x1080.

After wrapping up our initial GeForce GTX 1080 and GeForce GTX 1070 round-ups, Tom's Hardware DE set to work on a collection of 1060s, 3GB and 6GB alike. This first incarnation includes eight different boards from a field that spans anywhere from under $200 (£150) to over $300 (£250). Each individual review goes incredibly deep, covering manufacturing quality, technical features, power consumption, clock rates, cooling, and acoustics.

The gaming performance of every factory-overclocked board within a given chip class is usually pretty similar to begin with. But that's more true now than ever. This is a result of features like GPU Boost 3.0, which allows manufacturers to safely extract as much headroom as possible from a processor. Very little is left on the table, even if you have access to extreme overclocking hardware.

We will continue to update this roundup as new test samples become available.

The final analysis of each card is listed below for quick and easy comparison.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

Benchmark Results

We test every card after a suitable warm-up period to avoid unfair differences in GPU Boost frequencies. All benchmarks are run six times; the first one is used to get the GPU hot again.

These cards are all press samples operating at the same settings as retail models in our best effort to ensure one vendor doesn't get a leg up on another using non-representative clock rates.

The following galleries each contain four images, covering two tested resolutions. We put our focus on QHD (2560x1440) and FHD (1920x1080), plotting out average and minimum frame rates for each resolution in separate graphs.

Ashes of the Singularity

Battlefield 4

The Division

Grand Theft Auto V

Hitman (2016)

Project CARS

Rise of the Tomb Raider

The Witcher 3

All of the factory-overclocked cards offer similar performance, more or less. That's why our primary focus centers on evaluating the more technical aspects of each board design, along with their coolers. This is where differences in noise, power, and temperatures are most likely to come from.

Power Consumption

We begin by comparing the power consumption of each card in our gaming loop, stress test, and at idle. Depending on the power targets specified by each manufacturer, we sometimes measured substantial differences. We're ignoring the decimal places in our bar graphs, since those values are too small and would be within measurement tolerances.

We also found that some cards with lower power targets started to throttle during our stress test, resulting in lower power consumption numbers. This did not, however, have a negative impact on general gaming performance for any of the tested cards, as the stress test merely represents a worst-case scenario.

Noise

For the following comparison, we divide all of the gallery's bar graphs into gaming loop and idle, even if "noise" in practice spans a wide spectrum and the "character" of the sound varies a lot, with each card having highly individual results. Therefore, it is important not to compare just the absolute numbers, but also the frequency spectrum we're presenting.

Many of the cards implement a semi-passive mode, where their fans remain off when the card is idle. Thus, we refrained from taking measurements in that state. Even in our anechoic chamber, levels of 22 dB(A) and below merely represent ambient noise.

Temperature

For this comparison, we divide the gallery's bar graphs into gaming loop, stress test, and peak temperatures measured on the MOSFETs.

We occasionally compared the temperatures on our benchmark table with those measured inside a closed case and found them to be no more than two or three Kelvin higher.

Since temperatures in a closed case also depend heavily on the enclosure's cooling performance, the only representative and reproducible values are those measured on our benchmark table. Those are the ones we compare.

Nvidia GeForce GTX 1060 Founders Edition

For a reference design - sorry, Founders Edition - this card's appearance is quite neat. But are there any good reasons to pick Nvidia's implementation over competing models? After all, it's heavy and employs a completely different cooling solution than the competition.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

Despite using a much shorter PCB, Nvidia's Founders Edition is still 10 inches (25.4cm) long. It is 4¼ inches (10.7cm) tall and 1½ inches (3.8cm) wide, too. The slot bracket is the widest part. Behind it, the card and cooler only measure about 3.5cm together.

For the shroud, Nvidia again uses a mix of aluminum and plastic. This time the company favors economy, though. The cover can be detached in one piece, including the fan.

There's still an illuminated GeForce GTX logo up top, and the cooler's internals are completely hidden.

In contrast to the coolers you often seen on cards with short PCBs, the housing's overhang doesn't have a hole to feed the fan with air. Instead, the back side is covered. There is no backplate either, further reducing cost.

The back hosts Nvidia's well-known, gill-shaped air intakes. These are more fragile, so you'll want to be careful positioning this card in a case. 

The output bracket carries over from Nvidia's GeForce GTX 1070 and 1080 Founders Editions. It's dominated by three DisplayPort 1.4-ready connectors, one HDMI 2.0 output, and one dual-link DVI-D interface.

Board & Components

The center of the card is naturally dominated by its graphics processor, even if the GP106-400-A1 is significantly smaller than the GP104 on GeForce GTX 1070 and 1080. There are other noticeable differences as well, like the on-board memory. Nvidia's GeForce GTX 1060 only uses six of the available eight module emplacements, each populated with Samsung K4G80325FB-HC25 ICs. The modules have a capacity of 8Gb (32x 256Mb), and operate between 1.305 and 1.597V, depending on clock rate. In total, they add up to 6GB of GDDR5.

Unfortunately, the PWM controller isn't documented. It’s made by uPI Semiconductor and bears the model number uP9509, which means that it’s probably the uP9511P’s smaller sibling (the latter controller is what we found paired to the GP104 processor).

The memory modules and one of the GPU phases get their power through the motherboard’s PCIe slot. The two remaining GPU phases and the card’s accessories draw power from the six-pin power connector. We'll take a closer look at what this means in terms of load distribution across the rails on the next page.

When it comes to voltage regulation, Nvidia uses only one Dual N-Channel MOSFET, the E6930, per phase for both the high and low side; separate gate drivers aren’t needed. This highly integrated component explains the empty spaces on the board.

The GPU’s three phases are completely sufficient, and their distribution makes more sense here than on AMD's Radeon RX 480.

Apart from the six-pin power connector, which appears to have taken a wrong turn somewhere, Nvidia's reference GeForce GTX 1060 actually looks pretty good. And given a relatively low amount of waste heat, its axial fan isn't a bad choice either.

Power Results

The GeForce GTX 1060 FE does hit its power and temperature limits.

In order to test overclocking, then, we not only increased the GPU Boost frequency, but also the power target to 116 percent while setting the fan to 100 percent. The results certainly speak for themselves, too. Our sample held a stable 2050 MHz through the most taxing passages of our test sequence. Without this forced bit of headroom and the additional fan speed, though, the graphs look shakier:

For comparison, let’s take a look at the voltages during normal operation as well. It’s plain to see that the GPU Boost clock and voltages drop once the limit is reached.

The lowest GPU frequency measured at idle was 139 MHz.

We took our measurements using a variable low-pass filter, so we mention short load peaks only as a side note (see the grayed-out bar in the chart below), since they're rarely relevant in practice. 

Power Connector Load

Now we're getting into the nitty gritty. Measurements across the different rails during our taxing gaming and stress tests show that Nvidia manages to distribute the load well with a bias to the six-pin power connector. The 3.3V rail isn't used anymore, which is why you won't find it in our charts.

Here are the detailed graphs. Clicking on them brings up the high-resolution versions.

The PCI-SIG’s technical specifications cover current exclusively, so that’s what’s shown in the chart below (power consumption only tells half of the story, after all). With less than 5A for the motherboard slot, Nvidia plays it safe with its GeForce GTX 1060. The maximum, according to the PCI-SIG, is 5.5A.

We have detailed graphs for the current measurements as well. Once again, clicking on them brings up the high-resolution versions.

Temperature Results

Up top, we immediately notice the power connector's strange position. It’s situated in a part of the cooler that protrudes beyond the actual PCA. This necessitates a number of cables to attach to the board.

The implementation is anything but elegant, and it prevents Nvidia's partners from building shorter 1060s. Although the card is only 17.5cm long, it doesn't have any space to accommodate a power connector.

Remove the four screws securing the cooler's body and it comes right off. There’s a massive copper heat sink and metal frame underneath. The closed cooling fin design reminds us of the GeForce GTX 1070, and it should provide ample performance given the 1060's 120W TDP.

The massive frame serves double duty by keeping everything in place and cooling the voltage regulation circuitry/memory modules.

The GeForce GTX 1060 generates similar numbers during the gaming loop (119W) and stress test (122W). In light of this, it’s hardly surprising that the two temperature curves look a lot alike. GPU Boost 3.0 is very restrictive (or already well-optimized, if you prefer). It adjusts the clock rate and voltage to keep the card exactly where it’s supposed to be.

The infrared picture tells us that the cooling solution does its job well. All four of the voltage converter areas are in great shape as well.

During the stress test, the GPU’s three voltage converters reach 97 degrees Celsius, which is just within the acceptable range. Then again, nobody runs stress tests for hours, so this result is more theoretical, and not so practical, in nature.

Sound Results

Next, we want to quantify just how loud the GeForce GTX 1060 gets under these conditions. The largest source of noise is generally the fan, so we look at that first. Its rotational speed takes a while to plateau, finally stabilizing after about 15 minutes. The two curves are almost on top of each other, which is again hardly surprising seeing that the temperatures the cooler deals with are almost identical as well.

At idle, we measure 31.4 dB(A). This is a good result that’s barely above the noise level encountered in a living room. The sound produced by Nvidia’s radial fan is noticeably more pleasant than that of AMD's reference Radeon RX 480. Nvidia's fan is a bit louder, though.

The good news continues with our gaming loop results. After reaching its maximum temperature, the GeForce GTX 1060 emits just 35.4 dB(A). Motor and bearing noises can’t really be heard above the relatively pleasant swooshing noise produced by the card. The sound isn't as low and growling as what we hear from AMD's competing offering. For more on that card, check out AMD Radeon RX 480 8GB Review.

Even a casual glance at this graph reveals that the majority of the GeForce GTX 1060’s operating noise comes from the fan blades and airflow. The voltage regulation circuitry doesn't contribute at all. This means that the DHE cooling solution performs well, doing its job quietly. Then again, it's not that big of a challenge to dissipate 120W.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

EVGA GeForce GTX 1060 SC Gaming

With the 6GB GeForce GTX 1060 SC Gaming, EVGA added a model that roughly follows Nvidia's reference design, but uses a cooler that eliminates the Founders Edition's overhangs. This gives us a much shorter card seemingly ideal for HTPCs.

This model has some minor weaknesses, though. We've already discussed them with EVGA, and the company apparently plans to update its BIOS after checking it with Nvidia. Improved VRM cooling is on the table as well. However, since EVGA allows customers to change its cooler, you can apply our suggested modifications in a few manual steps without voiding the card's warranty.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The cooler cover is a low-cost affair made of anthracite-gray plastic and devoid of gaudy extras. Perhaps that's a plus. In compact cases, the card probably won't be visible anyways. Functional and affordable are interesting attributes in their own right.

Measuring just 7¼ inches (18.4cm) long from the slot cover, 4 1/5 inches (10.5cm) tall from the top edge of the motherboard slot, 1 3/8 inches deep, and weighing just 20 ounces (576g), the GeForce GTX 1060 SC Gaming comes across as a light-weight dwarf. And if space if your chassis is still an issue, it's easy to cut another third of an inch off the shroud if you have the right tools.

EVGA also abstained from attaching a backplate for reasons of cost efficiency, which frees up roughly one-fifth of an inch (5mm) for the CPU cooler in a mini-ITX enclosure. Sometimes that's the difference between success and failure in a compact build.

There's an EVGA logo up top, but it isn't back-lit. Moreover, a simple six-pin power connector sits at the end of the card and is not rotated.

A closer look at the bottom shows that, in spite of the card's compact design, EVGA employed a heat pipe-based cooling solution anyway. Our upcoming measurements speak to the wisdom of that decision.

Horizontally-aligned fins help some of the GPU's waste heat escape out the slot cover, which is fairly free-flowing thanks to a number of holes cut into it. This principle is also apparent at the end of the card, where hot air dumps out into the middle of your case. Good circulation is critical as a result.

The slot cover exposes five outputs, four of which can be used simultaneously to drive multi-monitor configurations. In addition to one dual-link DVI output (which lacks an analog signal), you also get one HDMI 2.0 interface and three DisplayPort 1.4-capable connectors.

Board & Components

EVGA's PCB is similar to Nvidia's reference design, but still unique in detail. For instance, while the GeForce GTX 1060 Founders Edition is manufactured in China, EVGA relies on a Taiwanese supplier.

Right off the bat we see four power phases, just like Nvidia's card. Three are meant for the GPU and one corresponds to the memory. Two of the three GPU phases are fed by the auxiliary power connector. The third GPU phase and the one for the memory get power from the motherboard slot.

In theory, the PCB could have enabled three phases and a maximum of six regulator circuits via doubling for the GPU, plus one phase for the memory. However, this minimalist configuration with just one regulator circuit per phase, similar to Nvidia's reference design, seems to be more popular given its lower cost.

The three GPU phases are controlled using one uPI Semiconductor uP9505P, with each regulator circuit realized as a highly integrated E6930 Dual N-Channel MOSFET, which combines the gate driver, high- and low-side MOSFET, and Schottky-diode in one convenient package.

Again, this is meant to cut costs and conserve space. But concentrating all of the components in one package also comes with a significant disadvantage: the creation of thermal hot-spots. This is made worse by the fact that two of these MOSFETS are positioned fairly close to each other. Since the PCB was designed to facilitate doubling, one of the two MOSFETS could have been moved into the second regulator circuit's vacant space.

Two capacitors are installed right below the GPU to absorb and equalize voltage peaks.

Only six of the eight available memory emplacements are populated with Samsung K4G80325FB-HC25 modules. Each one adds 8Gb (32x256Mb) and operates at voltages between 1.305 and 1.597V, depending on clock rate. In total, they add up to the GeForce GTX 1060 SC Gaming's 6GB of GDDR5.

Power Results

A compact graphics card typically enjoys less thermal headroom due to its smaller cooler. Naturally, we want to track the correlation between clock rate and voltage.

The data shows us that frequency drops slightly as the card heats up during a gaming session.

The 2012 MHz GPU Boost frequency we initially measure falls to 1949-1962 MHz under load. Voltage behaves similarly: after first measuring 1.062V, that value dropped to 0.975V on average after warming up (the actual range was 1.025 to 0.9V). All of this points back to GPU Boost 3.0 and a relatively low power target that limits consumption to just over 120W.

The lowest GPU clock rate measured with an idle card was about 253 MHz, and at that frequency, power consumption was similar to Nvidia's reference design.

We took our measurements using a variable low-pass filter, so we mention short load peaks only as a side note (see the grayed-out bar in the chart below), since those peaks are rarely relevant in practice. Nevertheless, that value is quite high, hinting at less than optimal primary-side smoothing prior to the voltage regulators.

A low power target causes GPU Boost to cut the voltages dramatically during our stress test, resulting in power consumption numbers that are lower than what we measure during a worst-case gaming workload.

Power Connector Load

Load distribution from the four voltage regulators to their respective power phases isn't problematic at all. However, there's little headroom for overclocking, even if it wasn't already limited by the firmware's power target and voltage settings.

Here's the data for the gaming and stress tests:

Power consumption on its own only tells us part of the story. The PCI-SIG specifies a maximum of 5.5A at the motherboard slot, and we're showing a reading under 5A, which puts EVGA's GeForce GTX 1060 SC Gaming in the green.

Of course, we have the data for measured currents as well:

The Cooler, Its Performance & A Simple Mod

Fortunately, the headaches caused by EVGA's cooler can be improved in just a few steps. Since the company does permit removal of its thermal solution without voiding the card's warranty, we can describe our modification in detail and encourage you to try it yourself. As long as EVGA doesn't implement changes of its own to augment performance, this procedure is worthwhile.

Once the four screws holding the cooler above the GPU package are removed and the cooler lifted off, two flat 8mm heat pipes become visible. They're meant to dissipate heat from the copper sink to the fins on both ends of the cooler. Their composite material should be able to deal with curvature and orientation in a reasonably efficient manner.

However, it's apparent that the VRM's four dual-channel MOSFETs and the memory packages receive no active cooling. This is exactly where we will apply our minor modification to affect big change with little effort.

The built-in 9cm fan relies on conventional rotor blades, which are quite steep to generate more pressure. Unfortunately, the heat pipes cover several important areas, such as the VRMs, meaning little of that extra airflow actually reaches the PCB's surface.

As far as the GPU is concerned, we measure 163 to 165°F (73 to 74°C) during our gaming loop and a maximum of 172°F (78°C) in a closed case. When it comes to the processor itself, EVGA's cooler does a respectable job. The stress test reveals a more gradual ramp up due to a lower frequency and less power consumption. But the final temperatures are nearly identical.

As our infrared readings demonstrate, though, a lack of active cooling on the VRMs has negative consequences. Even on an open test bench, temperatures during our gaming loop reach 234°F (112°C). Our stress test gets them up to 246°F (119°C). Although that sounds bad, it's not out of spec for the built-in MOSFETs, which are rated at up to 302°F (150°C). However, the excess heat travels through the multi-layer circuit board's copper traces and spreads over a large area.

The PWM controller mounted on the back of the board (visible just right of the VRM label as a faint rectangle) measures approximately 221°F (105°C). Again, this is within an acceptable range, since only temperatures as high as 302°F (150°C) would end in an OTP-related shut-down. We're more worried about the capacitors. Their through-hole leads work like little antennas, effectively transporting heat inside. With well over 212°F (100°C), and up to 234°F (112°C) during the stress test, this could lead to problems over time.

After all, these measurements were taken in an air-conditioned room at 72°F (22°C). In the height of summer, smaller PC cases heat up considerably. The same readings collected under less forgiving circumstances would likely turn out to be even higher.

Furthermore, the fact that the circuit board just below the GPU package is hotter than what the processor's own diode measures means the GPU is cooling the PCB, which is counterproductive. We do hope EVGA considers implementing something similar to our remedy.

A Safety Modification

The fact that the heat pipes cover the MOSFETs can easily be used as leverage. For the sake of simplicity, this example focuses exclusively on reducing the larger hot-spot of the two upper GPU VRMs by applying a stack of two thermal pads to bridge the tenth of an inch (2.5mm) difference between them and the heat pipe. Of course, you'd want to set up all four MOSFETs like this, yielding an even better result.

This superimposed image clearly shows the position of the heat pipes in relation to the components underneath, along with the position of our small cooling aid.

Retaking the measurement with an infrared camera left us amazed. Despite our meager investment and basic materials, the impact on cooling was astonishing. A reading seven to eight Kelvin lower for only two of the four MOSFETs puts us in a temperature range that allows some confidence in long-term reliability, even in small cases.

Sound Results

Hysteresis is not an issue for most manufacturers, and EVGA typically fares well here. After applying the most recent BIOS update, the GeForce GTX 1060 SC Gaming's semi-passive mode worked well without any annoying start/stop cycles. The threshold at which the fan spins up is set to ~149°F (65°C), but the GPU's current load isn't taken into account.

It's possible to lower the GPU's temperature four or five Kelvin more by configuring a custom fan curve to give the card a bit of extra air (important for mini-ITX projects). That helps the GPU Boost clock rates, too.

When the card is idle, noise is not measurable due to a semi-passive mode. A reading of 34.3 dB(A) under load is commendable, and that's primarily due to a conservative fan curve. However, even at 1300 RPM the noise level remains below the still-acceptable 36 dB(A) mark.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

Gainward GeForce GTX 1060 6GB Phoenix GS

With its GeForce GTX 1060 6GB Phoenix "GS", Gainward aims for a slightly more demanding enthusiast without giving in to excessive overclocking madness. You can think of the company as a sort of luxury brand belonging to Palit, so it benefits from the organization's substantial resources. It's only a shame that our U.S. readers don't have access to the board. According to Gainward's webpage, the GeForce GTX 1060 6GB Phoenix "GS" is only available in Europe and Asia.

Gainward labels its top-performing cards in a given product family with the Golden Sample (GS) suffix, and that's exactly what our German lab received for testing.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The cooler cover is made of black plastic with various metal highlights. It is designed to be used as an attachment, just like the fan modules. Thus, a Palit card could easily be transformed into a Gainward one and vice versa.

Although the card weighs a moderate 31oz (875 grams), it is fairly compact, measuring 9½ inches (24cm) long and 4 3/5 inches (11.7cm) tall. However, it's also 4.7cm wide, nearly requiring three expansion slots and making installation in mini-ITX motherboards difficult.

The back of the board is covered by a single-piece plate that lacks openings for ventilation and does nothing to cool the PCB. Because of the backplate, though, you must plan for an additional one-fifth of an inch (5mm) of clearance to avoid collisions with your CPU's cooler, particularly in small form factor cases.

The top of the card features a Gainward logo back-lit with customizable RGB LEDs. A single six-pin power connector is rotated by 180° and positioned at the end of the card.

A closer look at the bottom reveals that the heat pipes aren't embedded in a heat sink, but rather lay flat behind the base of the board. You can bet we'll explore how well this cost-effective implementation holds up to taxing graphics workloads.

Horizontally-oriented fins allow some waste heat to escape from an externally-facing slot bracket loaded with ventilation holes. This also applies to the card's other end, though, where hot air dumps out into the middle of your case. Good circulation is critical as a result.

The slot cover exposes five outputs, four of which can be used simultaneously to drive multi-monitor configurations. In addition to one dual-link DVI output (which lacks an analog signal), you also get one HDMI 2.0 interface and three DisplayPort 1.4-capable connectors.

Board & Components

The PCB that Palit and Gainward use features some prominent differences compared to Nvidia's reference design.

Right off the bat we see four power phases, just like Nvidia's card. Three are meant for the GPU and one corresponds to the memory. Two of the three GPU phases are fed by the auxiliary power connector. The third GPU phase and the one for the memory get power from the motherboard slot.

While, in theory, Gainward's PCB offers the option to use one more power phase, the company went with a rarely-used and rather inexpensive NCP81174 for its PWM controller, which can only control up to four phases. Thus, the vacant emplacement you see in the image above could only have been populated by another parallel regulator circuit (via doubling).

The three GPU phases utilize a highly integrated Vishay SiC632, which combines the gate driver, high- and low-side MOSFET, and Schottky-diode in one convenient package. Meanwhile, the memory is driven by a pair of N-Channel MOSFETs, including one SM4503NHKP (30A, high-side) and one SM4377NSKP (50A, low-side), both manufactured by Sinopower. The MOSFETs are also fairly cost-effective choices.

Overall, the board reflects that Gainward didn't spend big on pricey components. Even the Foxconn Magic chokes are middle-of-the-road. They get their job done in a rather unremarkable way.

Two capacitors are installed right below the GPU to absorb and equalize voltage peaks.

Only six of the eight available memory emplacements are populated with Samsung K4G80325FB-HC25 modules. Each one adds 8Gb (32x256Mb) and operates at voltages between 1.305 and 1.597V, depending on clock rate. In total, they add up to this card's 6GB of GDDR5.

Power Results

The GPU Boost frequency and core voltage curves are particularly interesting to us. Our data shows that frequency drops slightly as the card heats up during a gaming session.

The 2025 MHz GPU Boost frequency we initially measure falls to 1949 MHz under load. Voltage behaves similarly: we measured 0.975V on average, with readings varying between 1.025 and 0.9V. All of this points back to GPU Boost 3.0 and a relatively low power target that limits consumption to a maximum of 130W.

The lowest GPU clock rate measured with an idle card is about 253 MHz.

We took our measurements using a variable low-pass filter, so we mention short load peaks only as a side note (see the grayed-out bar in the chart below), since those peaks are rarely relevant in practice.

A low power target causes GPU Boost to cut the voltages dramatically during our stress test, resulting in power consumption numbers that are lower than what we measure during a worst-case gaming workload.

Power Connector Load

Load distribution from the four voltage regulators to their respective power phases isn't problematic at all. However, there's little headroom for overclocking, even if it wasn't already limited by the firmware's power target and voltage settings.

Here's the data for the gaming and stress tests:

Power consumption on its own only tells us part of the story. The PCI-SIG specifies a maximum of 5.5A at the motherboard slot, and we're showing a reading under 5A, which puts Gainward's GeForce GTX 1060 6GB Phoenix "GS" in the green.

Of course, we have the data for measured currents as well:

The Cooler And Its Performance

As mentioned, Gainward's backplate is purely aesthetic; it doesn't help cool the PCB at all. Four screws secure it to the top of the board, shielding the card from curious customers.

Though the Gainward and Palit shrouds look a little different, everything else is mostly identical (at least up to the geometry of their respective 9cm fans; Gainward went with a conventional straight edge).

Again, the delicate-looking cooler sports horizontally-oriented fins, as well as a quartet of 4mm heat pipes made from sintered composite material. Those pipes are not integrated into the bottom of the cooler. Instead, they merely touch the back of the thin base plate. Expensive copper is nowhere to be found.

We're glad to see that the base plate has a specific area to help cool the MOSFETs. However, the coils have to live without the benefit of active cooling.

Registering between 156 and 158°F (69 to 70°C) in our gaming loop, and a maximum of 167°F (75°C) in a closed case, the cooler does a fair job. It would almost assuredly perform better if the heat pipes were attached more securely, though.

Measurements collected with the backplate removed also reveal acceptable values. A peak of approximately 181°F (83°C) from the VRMs is reasonable, in our opinion.

While the readings generated during our stress test aren't exactly worrisome at ~192°F (89°C), they also reveal the cooling solution's weaknesses.

Gainward's cooling performance is thus regarded as merely acceptable. We get the impression that one member of the finance team was assigned to each engineer, keeping costs in check more than enthusiasts like to see.

Sound Results

Hysteresis is not an issue for most manufacturers. However, Gainward implemented a fan curve with too-low of an initial rotational speed for the transition between passive and active mode. As a result, the fans simply stop, and need to be restarted repeatedly. Even the best hysteresis won't help when a manufacturer overestimates the quality of its components this grossly.

When the card is idle, noise is not measurable due to a semi-passive mode. A reading of 35.6 dB(A) under load is acceptable, but not great. So much air has to be moved through the thermal solution because Gainward's cooler isn't as efficient as it should be. Spending a few dollars more would have gone a long way. On the bright side, we measure very little low-frequency bearing noise. What remains is the sort of acoustic energy that's easily absorbed by a case with sound insulation.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

Gigabyte GeForce GTX 1060 G1 Gaming 6G

Gigabyte's GeForce GTX 1060 G1 Gaming 6G is not the manufacturer's least-expensive version. Rather, it targets price-conscious gamers looking for a good deal, but still interested in feature-rich hardware. Thus, this factory-overclocked card represents the mass-market model of Gigabyte's GeForce GTX 1060s. The company reserves shorter ITX-oriented models for its system integrator business and doesn't aggressively advertise them directly to enthusiasts. Instead, the GeForce GTX 1060 G1 Gaming 6G employs a full-size PCB covered by a relatively large cooler.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The fan shroud is made of black plastic and seems fairly inconspicuous, despite some orange detail work. The two 3½-inch (9cm) fans are well known from previous Windforce-equipped models, and (from an aesthetic perspective) seem a bit lost on this otherwise long card.

Weighing in at just under 29 ounces (813g), this card is light for its size, particularly given measurements of more than 11 inches (28cm) long, 4 1/3 inches (11cm) tall, and 1 1/3 inches (3.5cm) wide. It occupies two full expansion slots, and is neither particularly small nor really oversized.

We're glad that Gigabyte decided against the almost obligatory third fan present on many of its coolers. Frankly, another one would have been completely superfluous for this card. A little more distance between the fans might have worked even better, though.

The card is covered by a single-piece backplate and lacks openings for ventilation. But it is connected to the board in two spots via thermal tape to aid active cooling. Because of the backplate, though, you must plan for an additional one-fifth of an inch (5mm) of clearance to avoid collisions with your CPU's cooler, particularly in small form factor cases.

Up top, there's a back-lit Gigabyte logo and one indicator for the semi-passive mode. Both are enabled as customizable RGB LEDs. An eight-pin auxiliary power connector is rotated 180° and positioned at the end of the card.

Vertically-oriented cooling fins mean that hardly any air escapes through the slot bracket, even though narrow slits in the metal suggest otherwise. The vertical fins are cheaper to implement, since they make it easier to arrange the heat pipes. You just have to be diligent about exhausting all of that waste heat out of your case.

The slot cover exposes five outputs, four of which can be used simultaneously to drive multi-monitor configurations. In addition to one dual-link DVI output (which lacks an analog signal), you also get one HDMI 2.0 interface and three DisplayPort 1.4-capable connectors.

Board & Components

A look at the board reveals that it was inspired by Gigabyte's GeForce GTX 1080 G1 Gaming and GeForce GTX 1070 G1 Gaming, but with far fewer components since the 1060 is electrically less complex.

Gigabyte relies on a rather unusual 6+1-phase design, with a uPI Semiconductor uP9511P PWM controller for the GPU's six phases, leaving two emplacements vacant. The memory's power comes from a seventh phase controlled by a uP1666; it consists of a single regulator circuit.

Since the PWM controller can't communicate directly with each phase's MOSFETs, Gigabyte utilizes separate PWM drivers (gate drivers) to control the two single-channel MOSFETs per phase for the range of the DC/DC voltage converter (low- and high-side). The second low-side MOSFET, which the higher-end cards do have, is missing in this case.

Two capacitors are installed right below the GPU to absorb and equalize voltage peaks.

Only six of the eight available memory emplacements are populated with Samsung K4G80325FB-HC25 modules. Each one adds 8Gb (32x256Mb) and operates at voltages between 1.305 and 1.597V, depending on clock rate. In total, they add up to this card's 6GB of GDDR5.

Power Results

Before addressing power consumption directly, let's take a look at the relationship between clock rate and voltage. As we've seen already, the GPU's frequency drops significantly once the card is fully exercised with a gaming workload.

After warm-up, the initial GPU Boost clock rate of 2000 MHz falls to a still-respectable 1974 MHz under load. The slope of this curve is mirrored by our voltage measurements, which aren't much worse either (on average, we measured 1.05V). This value drops slightly, but only sporadically.

The lowest GPU clock rate measured at idle is about 240 MHz.

We took our measurements using a variable low-pass filter, so we mention short load peaks only as a side note (see the grayed-out bar in the chart below), since those peaks are rarely relevant in practice.

Power Connector Load

Since Gigabyte's GeForce GTX 1060 G1 Gaming doesn't use the motherboard slot's 3.3V rail, we left it out of our chart.

The power distribution we observe shows a heavy bias toward the auxiliary power connector, which supplies about 100W.

Here's the data for the gaming and stress tests:

Power consumption on its own only tells us part of the story. The PCI-SIG specifies a maximum of 5.5A at the motherboard slot, and we're showing a reading under 3A, which puts Gigabyte's GeForce GTX 1060 G1 Gaming 6G in the green.

Of course, we have the data for measured currents as well:

The Cooler & Its Performance

When Gigabyte first launched this card, it didn't have thermal pads on the backplate to help draw heat from the board. This changed during mass production, however, after we passed along our feedback.

The revised version wasn't available at the time this review was originally written, so we had to implement the changes Gigabyte made ourselves. Similar to the GeForce GTX 1080 G1 Gaming, there are now two pads on the plate. They're in slightly different places, though, since a 1060's hot-spots aren't quite the same. One corresponds to the VRMs, while the other sits opposite Nvidia's GPU.

We started by taking readings with the card as it shipped from Gigabyte, and then introduced our modifications, which we can judge based on their effect on thermal performance.

The cooler employs vertically-oriented fins and two angled 6mm heat pipes made of a sintered composite material. They're flattened to sit flush with the heat sink, and directly touch the GPU. For cost reasons, the sink is all-aluminum, while the supporting base plate is fashioned from steel to improve stability.

It deserves a positive mention that the heat sink for the voltage regulation circuitry is built into the cooler and does an excellent job.

A measurement between 152 and 154°F (67 to 68°C) during our gaming loop, peaking at 158°F (70°C) in a closed case, proves that the cooler does an acceptable job. Based on our do-it-yourself attempt at replicating Gigabyte's modified version, we'd expect to see temperatures about two Kelvin lower.

Measurements taken with the backplate removed show a maximum of ~160°F (71°C) around the VRMs. That's an excellent value for what often appears as a troubling hot-spot on competing cards. Thermal pads should help even more, too.

In both our gaming workload and stress test, Gigabyte's cooler works the way it should, yielding satisfactory infrared images.

Although we were initially skeptical of how well the two flattened heat pipes would fare, the thermal solution performs as well as we could ask. Two 8mm or three 6mm heat pipes would have made more sense to us since they would have covered the GPU completely. But it's good to see Gigabyte cooling the front and back of its card in such a way that hot-spots are minimized.

Sound Results

Implementing exemplary hysteresis is a matter of honor for Gigabyte, and the company doesn't disappoint. With that said, a fan speed just above 1400 RPM shows where this solution's performance comes from. Of course, you can always play with the fan curve yourself. Before we do that, let's see what the stock noise levels sounds like.

When the card is idle, its noise level is imperceptible due to a semi-passive mode. There's no point taking readings in that state.

Under load, a 34 dB(A) reading is commendable, even if it isn't completely silent. In a closed case, the GeForce GTX 1060 G1 Gaming would only be audible as a soft whisper. The bearing noises we measure are acceptable, and there is very little unwanted structure-borne noise.

Relatively aggressive factory overclocking turns this fairly quiet card into a capable performer. It doesn't slip up in any of the disciplines we consider important, and lands at the top of our results in each individual test. A high GPU Boost frequency, which remains stable even after we warm the card up, make this an interesting choice for additional overclocking.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

Inno3D GeForce GTX 1060 Gaming OC

Inno3D's GeForce GTX 1060 Gaming OC isn't the company's flagship, but rather a bread-and-butter workhorse in its 11-card portfolio of GP106-based products. We consider this a good thing. As one of the more affordable GeForce GTX 1060s (at least in Europe), the Gaming OC board appeals to a broader audience. You get a similar level of performance without spending an extra $40 or $50 (£20-30) on bells and whistles you might not need. It'll be interesting to see where Inno3D cut costs though, and what features it dropped to achieve a more attractive price.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The fan shroud is made of black plastic and features red detailing that looks quite nice. We won't say the same for those piano-black bolt-ons that serve little purpose other than to collect fingerprints. After installing the card, you'll feel compelled to clean it off again.

Weighing in at 29 ounces (820g), this card is fairly light. It measures 11 inches (28cm) long, five inches (12.5cm) tall, and 1 1/3 inches (3.5cm) wide. It monopolizes two expansion slots and is neither particularly small nor oversized—it's just average. The blades of both fans have a diameter of 9cm, which is pretty common as well.

The back of the board is covered by a single-piece plate with some ventilation holes. Unfortunately, there are no thermal pads connecting it to the circuit board, so the plate does nothing to help with cooling. Because of the backplate, though, you must plan for an additional one-fifth of an inch (5mm) of clearance to avoid collisions with your CPU's cooler, particularly in small form factor cases. A lack of SLI support means you don't have to worry about fitting two GeForce GTX 1060s back to back in a case.

The top of the card features an unlit Inno3D label, and the shroud up there is narrow enough that you see the cooling fins behind it.

An eight-pin auxiliary power connector is located at the end of the card facing out, rather than back. As a result, the heat sink is cut out right above the connector. Interestingly, this can be explained by the fact that Inno3D uses Nvidia's reference PCB for the GeForce GTX 1070 to build its GTX 1060 Gaming OC.

Vertically-oriented cooling fins mean that hardly any air escapes through the slot bracket, even though narrow slits in the metal suggest otherwise. The vertical fins are cheaper to implement, since they make it easier to arrange the heat pipes. You just have to be diligent about exhausting all of that waste heat out of your case.

The slot cover exposes five outputs, four of which can be used simultaneously to drive multi-monitor configurations. In addition to one dual-link DVI output (which lacks an analog signal), you also get one HDMI 2.0 interface and three DisplayPort 1.4-capable connectors.

Board & Components

A glance at the board gives us a sense of déjà vu; as mentioned, Inno3D relies on the longer GeForce GTX 1070 reference board. Of course, the components it uses to build the 1060 are simpler, but because Nvidia's GP106 processor is pin-compatible, it's not a problem to use the 1070's PCB.

The 5+1-phase design, similar to Nvidia's GeForce GTX 1070 Founders Edition, relies on the uPI Semiconductor uP9511P as its PWM controller. All five of the GPU's power phases are fed by the eight-pin auxiliary connector; only the memory's phase is driven through the motherboard's PCIe slot.

The voltage of the five GPU phases and one memory phase is controlled by no-name dual-N-channel MOSFETs that require external gate drivers. Inno3D's coils of choice are clones of Foxconn's Magic series, which we'd expect to find in the lower price segment.

Two capacitors are installed right below the GPU to absorb and equalize voltage peaks.

It's worth noting that only six of the eight available memory emplacements are populated with Samsung K4G80325FB-HC25 modules. Each one adds 8Gb (32x256Mb) and operates at voltages between 1.305 and 1.597V, depending on clock rate. In total, they add up to this card's 6GB of GDDR5.

Power Results

Before addressing power consumption directly, let's take a look at the relationship between clock rate and voltage. As we've seen already, the GPU's frequency drops significantly once the card is fully exercised with a gaming workload.

After warm-up, the initial GPU Boost frequency of 1974 MHz falls to an acceptable average of 1949 MHz under load. The slope of this curve is mirrored by our voltage measurements: what started at 1.043V in the beginning falls to an average of 1V. The drops we observe correspond closely with dips in GPU clock rates.

The lowest GPU clock rate measured at idle is about 203 MHz.

We took our measurements using a variable low-pass filter, so we mention short load peaks only as a side note (see the grayed-out bar in the chart below), since those peaks are rarely relevant in practice.

Power Connector Load

Since Inno3D's GeForce GTX 1060 Gaming OC doesn't use the motherboard slot's 3.3V rail, we left it out of our chart.

The power distribution we observe shows a heavy bias toward the auxiliary power connector, which supplies about 95W.

Here's the data for the gaming and stress tests:

Power consumption on its own only tells us part of the story. The PCI-SIG specifies a maximum of 5.5A at the motherboard slot, and we're showing a reading under 3A, which puts Inno3D's GeForce GTX 1060 Gaming OC in the green.

Of course, we have the data for measured currents as well:

The Cooler & Its Performance

The question remains: what kind of performance can we expect out of this budget card's cooler? According to the label on its backplate, we should be enjoying the benefit of HerculeZ Design!

Inno3D's backplate is screwed into the cooler from behind and into the PCB from the front. Spacers measuring almost 3mm tall allow some airflow between the board and backplate.

This isn't one of the beefiest coolers we've tested. It employs horizontally-oriented fins, two nickel-plated 6mm heat pipes above the base plate, and four nickel-plated 6mm pipes towards the card's rear. For cost reasons, the heat sink and base plate are both made from aluminum.

It deserves a positive mention that the heat sink for the MOSFETs is built into the cooler and does an excellent job. Unfortunately, the coils do not benefit from this implementation.

A measurement between 140 to 142°F (60 to 61°C) during our gaming loop, peaking at 145°F (63°C) in a closed case, shows that Inno3D's cooler works amazingly well.

Measurements taken with the backplate removed and the cooler mounted directly to the PCB show that Inno3D's thermal solution still has some performance in reserve. Results from our gaming loop and stress test reveal no abnormalities, except for extremely low temperatures around the VRMs thanks to an integrated sink responsible for drawing heat away from that area.

In both our gaming workload and stress test, Inno3D's thermal solution works the way it should, leaving us with one of the coolest boards in today's round-up.

Although the pressure to cut costs is apparent in several place, this cooler does a great job. Performance was clearly one area Inno3D refused to compromise.

Sound Results

A picture-perfect implementation of proper hysteresis is critically important to Inno3D. However, fan speed readings as high as 1400 RPM point to where this card's great cooling performance comes from. We think there's room to dial back a bit, since even at 1000 RPM there'd be enough cooling headroom to keep Nvidia's GP106 operating at under 167°F (75°C), while at the same time reducing noise quite a bit.

When the card is idle, its noise level is imperceptible due to a semi-passive mode. There's no point taking readings in that state.

Under load, a 35.7 dB(A) reading is neither loud nor really quiet, but instead average. If you're looking for a much quieter result, take advantage of the cooler's headroom and use a third-party tool like MSI Afterburner to dial in a lower fan speed.

A spectrum analysis turns up significant vibrations in the low-frequency range. Given this card's price tag, though, it's hard to expect perfection.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

MSI GeForce GTX 1060 Gaming X 3G

The introduction of Nvidia's GeForce GTX 1060 3GB means we all have to pay more attention to naming. Not only does this model come equipped with half as much memory, but its GPU also sports 1152 active CUDA cores instead of the 6GB version's 1280. That means GP106-300 offers exactly 10 percent less theoretical compute power than the original GP106-400. Then there's the issue of on-board RAM. A measly 3GB could prove to limit current titles, even at 1920x1080.

These cards were first introduced in Asia, which doesn't come as a surprise since the Asian market has a different focus. Cards like this GeForce GTX 1060 "Light" will find plenty of satisfied customers among the millions of MMORPG and MOBA players.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The cover above the GeForce GTX 1060 Ti Gaming X 3G’s cooler is made of relatively thin plastic. It’s similar to the shrouds used on the larger cards in MSI's Nvidia-based line-up. A massive frame under the cover keeps the card stable.

At 1001 grams, this board isn't exactly light. Incidentally, it weighs in at just two grams less than the 6GB version. You can probably attribute the difference to the memory modules used on this model, which we'll discuss shortly.

The card’s other physical attributes mirror the 6GB version. Its dimensions are 28.4 x 13.5 x 3.5cm (L x H x D), spanning two expansion slots and standing relatively tall. The circumferences of the two dual-ball-bearing fans are also the same at 9.5cm.

The back of the board is hidden under a plate that consists of a single large piece. It has holes to allow air circulation, but doesn’t cool the card in any meaningful way since it’s not connected via thermal pads. The plate also adds 5mm to the back of the card, which can be a deal-breaker if you're butting up against a large cooler or memory modules.

Three nickel-plated heat pipes made from composite material are observable on the bottom of the GeForce GTX 1060 Ti Gaming X 3G. The two 6mm pipes travel all the way through, transporting waste heat to both parts of the cooler.

MSI's logo on the top of the card lights up. Again, the two 6mm heat pipes make a prominent appearance, as does an 8mm pipe. The eight-pin power connector sits on the end of the card, turned by 180°.

The rear of the card is completely open. The cooler’s fins are positioned horizontally, directing airflow towards the slot cover and the rear. This design has the advantage that a lot of its heated air gets blown out the back of your chassis.

The slot cover sports five connectors, four of which can be used concurrently in a multi-monitor setup. There’s one dual-link DVI-D connector without an analog signal, joined by two HDMI 2.0 connectors and three DisplayPort 1.4 outputs.

The rest of the slot cover is mostly taken up by vents. Airflow could have been improved by omitting the largely pointless DVI connector and including an adapter instead. This would have allowed for better cooling performance.

Board & Components

A quick look at the GeForce GTX 1060 Gaming X 3G’s board proves it to be pretty much identical to the 6GB version. All five phases are powered through the auxiliary connector. The memory modules have their own phase supplied through the motherboard’s PCIe slot.

The GPU power supply’s five phases are controlled by a uP9511P made by uPI Semiconductor Corp. Dual N-channel MOSFETs in the M381 line are used for the high and low side voltage converters. This setup eliminates the need for separate gate drivers, which saves both space and money.

The memory’s single phase has one small uP1641P also supplied by uPI Semiconductor Corp. It’s joined by a different voltage converter, the SM7320, which is also a dual N-channel MOSFET.

MSI uses the same coils we've seen for years. Their quality is just what you’d expect from a mass-market product in this category with decent encapsulation. That's to say they're better than Foxconn’s Magic coils.

An INA1221 is in charge of the card’s power, and we'll soon see that setting a high power target through the firmware doesn’t really benefit real-world performance. This is especially true if the power target is pushed as far as MSI takes it. Nvidia implements its own internal limiter that won't let overclockers get very far.

Only six of the eight available memory emplacements are populated with Samsung K4G41325FC-HC25 modules (32x 128 Mb). Their voltages range between 1.305 and 1.597V, depending on clock rate. Altogether, this amounts to just 3GB of graphics memory.

Two capacitors are added right below the GPU to smooth out voltage peaks. This design is similar to Nvidia's reference implementation.

Power Results

First, we'll look at GPU Boost clock rates. The factory setting initially yields an aggressive 1974 MHz, though that measurement drops slightly to 1949 MHz once the card is fully warmed up.

Of course, the voltage curve dips similarly, starting at 1.062V and ending at 1.043V, corresponding perfectly to the slightly lower GPU Boost step.

While we use several games with different rendering paths and quality settings, we're only testing at one resolution: 1920x1080. Benchmarking at higher resolutions just doesn't make sense on a card with 3GB of memory.

The GeForce GTX 1060 Gaming X 3G’s power consumption at idle and low loads turns out to be very similar to the GP106-400-based card. However, differences between the two boards increase as the load goes up. Doom doesn’t use the most power this time around; that honor goes to Metro: Last Light instead. Frame rates really suffer, though. The Witcher 3 is another example of a game that doesn't run smoothly unless the quality settings are significantly relaxed.

At playable settings, gaming power consumption is much more reasonable at around 102 to 103W. We take this as an indication that 3GB of graphics memory holds the card back more than we’d like. From here on in, we’re sticking with Doom to illustrate real-world performance.

The gray bar represents power consumption based on those load peaks that made it through our filters to the smoother curve. That bar doesn't have any practical significance since the peaks we measured are too brief for them to matter (even if the shortest-duration ones were already filtered out by this point).

Power Connector Load

Next, we drill down a bit more to look at how the loads are distributed between the two different 12V rails (motherboard slot and external power connector) during the realistic gaming load and stress test. Readings of 21W during gaming and 30W during the stress test show that the motherboard slot doesn’t see a lot of usage.

This is due to the fact that the memory and board components have the motherboard slot to themselves. The GPU draws all of its power through the auxiliary power connector. Now it's easier to explain why MSI uses an eight-pin connector, even though the GeForce GTX 1060 Gaming X 3G’s power consumption remains lower than 150W during all of our usage scenarios.

Here are the corresponding graphs for gaming and our stress test. Click on them for a larger version.

The PCI-SIG’s specifications only apply to current, meaning power consumption results on their own aren't comprehensive enough. Our readings put the motherboard slot just over 2.5A. Given a ceiling of 5.5A, this is most certainly on the safe side with lots of room to spare.

Of course, there are larger graphs for the current measurements as well.

The Cooler & Its Performance

The GeForce GTX 1060 Ti Gaming X 3G’s board is sandwiched between a heavy frame on top, and the aforementioned back plate on the bottom. Since these two components are screwed together, they basically create one massive and very solid part.

The frame is also screwed to the slot cover, hiding most of the board's top. Thermal pads connect it to not just the memory modules and MOSFETs, but also the two PWM controllers and one voltage converter responsible for the card’s other components.

The cooler is identical to what we found on MSI's GeForce GTX 1060 Gaming X 6G. It’s dominated by a nickel-plated heat sink that's screwed on in a way that presses its flattened fins onto the top of the cooler’s base. MSI doesn't use a massive plate with the heat pipes pressed into it for the GTX 1060, wheres the company did for its GeForce GTX 1070 and 1080.

Our new measurement and analysis software package enables us to examine benchmark results in new ways. The graph below shows the relationship between temperature and gaming performance. The actual frame rates during our strategically-chosen test run stay consistently at 79 to 80 FPS. The only aberrations are some small spikes due to the game.

This means that the GPU Boost clock rate isn't affected by a thermal limit. We know from our other benchmarks what really holds the GeForce GTX 1060 Gaming X 3G back. We’ll see the small valley that was produced by the game in our other graphs as well. This is how synchronized the measurements have become with our new software.

The thermal reading from below the GPU's package is consistent with the sensor output from within the GPU for our gaming loop and stress test. Meanwhile, the voltage converters’ and memory modules’ temperatures are well within the green range. Right underneath the memory modules, we measure temperatures of significantly less than 70 °C, even though those components are rated for up to 85 °C. The voltage converters come in at a cool 73 °C.

During the stress test, the area around the five voltage converters reaches 75 °C, and the hottest parts of the memory modules stay below 75 °C. MSI's cooling solution does its job, that's for sure.

Sound Results

The two large fans barely spin faster than their minimum RPM. Even during the stress test, they remain below 900 RPM. From 50 cm away, this isn't really audible.

The fans don’t even spin at idle, so noise results aren't necessary. During gaming, the GeForce GTX 1060 Gaming X 3G posts 29.3 dB(A). Its profile is dominated by deeper bearing noises, whereas the actual fan noise can’t really be heard above the card’s overall soft hum.

This result, combined with the temperature results, show that the card’s cooler is larger than it needs to be. Although the cooling performance is appreciated, everything comes at a price. The question whether that price is too high segues perfectly to our conclusions.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

MSI GeForce GTX 1060 Gaming X 6G

After reviewing its GeForce GTX 1080 and 1070, MSI is presenting us with a third graphics card armed with the same cooler. Naturally, on a GeForce GTX 1060, it feels a little oversized. The big thermal solution's advantage is quite obvious though; it's probably going to be even quieter than either of the two higher-end models. Otherwise, this card looks pretty similar. The most apparent difference is its lack of an SLI connector.

Even if there's not a lot about the card's appearance to introduce, our performance results are still worth going into depth on. And there's an important caveat we need to present upfront:

We test all samples as they arrive, out of the box, exactly as most gamers use them. Bundled control software or third-party tools are deliberately avoided. Once again, MSI tried to exploit this policy by delivering a so-called golden sample with a modified firmware, just as it did in the case of our GeForce GTX 1080 Gaming X 8G. The sample employs higher clock rates (and perhaps a higher power target) than the BIOS you'd receive from an off-the-shelf board.Therefore, we cannot grant this card any award, no matter how good the technical implementation may be. In addition, we also have to disqualify this card from our comparative summary; it just wouldn't be fair to the other contenders.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

Similar to its higher-end relatives, this card's fan shroud is made of relatively thin plastic. Underneath, you'll find a massive frame that improves the structure's stability.

Weighing 35.4 oz (1003 grams), the GeForce GTX 1060 Gaming X 6G isn't light (it's only about 100g lighter than the GTX 1080 version). The card measures 28.4cm long, 13.5cm tall, and 3.5cm wide, monopolizing two full expansion slots and plenty of area vertically, coming off the motherboard. This is due to the large fan blades, which stretch out to a diameter of 9.5cm.

The back of the board is covered by a single-piece plate with some ventilation holes. Unfortunately, there are no thermal pads connecting it to the circuit board, so the plate does nothing to help with cooling. Because of the backplate, though, you must plan for an additional one-fifth of an inch (5mm) of clearance to avoid collisions with your CPU's cooler, particularly in small form factor cases. A lack of SLI support means you don't have to worry about fitting two GeForce GTX 1060s back to back in a case.

MSI's logo on the top of the card lights up. Again, the two 6mm heat pipes make a prominent appearance, as does an 8mm pipe. The eight-pin power connector sits on the end of the card, turned by 180°.

The black and bright red colors follow MSI's corporate design, while, again, the card's shape resembles GeForce GTX 1070 and 1080 boards in the Gaming X line-up.

The rear of the card is completely open. The cooler’s fins are positioned horizontally, directing airflow towards the slot cover and the rear. This design has the advantage that a lot of its heated air gets blown out the back of your chassis.

The slot cover sports five connectors, four of which can be used concurrently in a multi-monitor setup. There’s one dual-link DVI-D connector without an analog signal, joined by two HDMI 2.0 connectors and three DisplayPort 1.4 outputs.

Where you don't find display outputs, the rest of the bracket features openings to facilitate airflow. If MSI chose to leave off the seldom-used DVI connector, there'd be even more space for ventilation. As you may have noticed, Nvidia went that route with its recent GeForce GTX 1080 Ti announcement, bundling a DP-to-DVI dongle for those who still need the interface.

Board & Components

All five phases are powered through the auxiliary connector. The memory modules have their own phase supplied through the motherboard’s PCIe slot.

The GPU power supply’s five phases are controlled by a uP9511P made by uPI Semiconductor Corp. Dual N-channel MOSFETs in the M381 line are used for the high and low side voltage converters. This setup eliminates the need for separate gate drivers, which saves both space and money.

The memory’s single phase has one small uP1641P also supplied by uPI Semiconductor Corp. It’s joined by a different voltage converter, the SM7320, which is also a dual N-channel MOSFET.

MSI uses the same coils we've seen for years. Their quality is just what you’d expect from a mass-market product in this category with decent encapsulation. That's to say they're better than Foxconn’s Magic coils.

Two capacitors are added right below the GPU to smooth out voltage peaks. This design is similar to Nvidia's reference implementation.

Only six of the eight available memory emplacements are populated with Samsung K4G41325FC-HC25 modules (32x 128 Mb). Their voltages range between 1.305 and 1.597V, depending on clock rate. Altogether, this amounts to just 3GB of graphics memory.

Power Results

Before addressing power consumption directly, let's take a look at the relationship between clock rate and voltage. In the case of MSI's GeForce GTX 1060 Gaming X 6G, this illustrates that frequency doesn't drop significantly during gaming, even after an imposed warm-up period.

Our observed GPU Boost rate falls from 2000 MHz to a still-respectable 1974 MHz over time and under load. The curve's slope is mirrored in our voltage readings. A 1.05V starting point pulls back ever-so-slightly to 1.043V by the end of our run.

The lowest GPU clock rate measured at idle is about 240 MHz.

We took our measurements using a variable low-pass filter, so we mention short load peaks only as a side note (see the grayed-out bar in the chart below), since those peaks are rarely relevant in practice.

Power Connector Load

Since MSI's GeForce GTX 1060 Gaming X 6G doesn't use the motherboard slot's 3.3V rail, we left it out of our chart.

The power distribution we observe shows a heavy bias toward the auxiliary power connector, which supplies just under 100W.

Here's the data for the gaming and stress tests:

Power consumption on its own only tells us part of the story. The PCI-SIG specifies a maximum of 5.5A at the motherboard slot, and we're showing a reading under 3A, which puts MSI's GeForce GTX 1060 Gaming X 6G in the green.

Of course, we have the data for measured currents as well:

The Cooler & Its Performance

Behind every successful high-performance graphics card is a well-designed cooler. In this case, MSI's implementation is pretty elaborate given the GP106 GPU's thermal ceiling. In contrast to MSI's GeForce GTX 1070 and 1080 Gaming X boards, the 1060 Gaming X relies on a closed frame that not only keeps the board nice and stable, but also helps cool the memory modules and VRMs.

Thermal pads provide the bridge necessary to transfer heat from the memory and power circuitry, though we would have preferred a thinner and more efficient design. After all, MSI uses low fan speeds that don't move much air.

The frame is screwed to the backplate, and ~3mm-tall spacers allow some air to flow between the plate and PCB.

As mentioned, this cooler is almost identical to the one used on the GTX 1070 and 1080 Gaming cards. It uses the same horizontally-oriented fins and heat pipe configuration. The direction they face is more or less irrelevant, since they're made of sintered composite material.

A massive nickel-plated sink takes the heat from those pipes and dissipates it out through aluminum fins.

The cooler does its job almost inaudibly. And yet it performs incredibly well, enabling readings around 145°F (63°C) during our gaming loop, and topping out at 149°F (65°C) in a closed case.

Our measurements with the backplate and stabilizing frame removed show that MSI won't let this card go up in flames. The gaming loop and stress test fail to turn up any apparent issues.

It becomes clear during our gaming test sequence that the lower five voltage regulator phases handle most of the load, while the memory's VRM remains fairly cool thanks to its less demanding task.

In the end, MSI's cooling solution does a great job (as it should, since it was designed for loads almost 100 percent higher).

Sound Results

Thanks to an exemplary implementation of hysteresis and fan speeds around 800 RPM, this card is almost inaudible.

When the GeForce GTX 1060 Gaming X 6G is idle, its noise is not measurable due to a semi-passive mode. Thus, we abstained from taking those readings.

Since this card's fans are the same ones used to cool MSI's GeForce GTX 1070/1080 Gaming boards (meaning they're almost overkill), our test chamber remains frighteningly quiet. An excellent 29.3 dB(A) result drops below the threshold of minimal background noise for a room inside a quiet apartment.

All we found were minimal vibrations in the low-frequency range caused by the slow-spinning fan motors. That's basically nit-picking though, since these vibrations aren't audible.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

Palit GeForce GTX 1060 Super JetStream

Palit's GeForce GTX 1060 Super JetStream is virtually identical to Gainward's GeForce GTX 1060 Phoenix GS, with only a few small variations. One thing they share in common, unfortunately, is a lack of availability in the U.S.

Maybe this shouldn't come as a surprise. After all, the two cards come from the same factory. Where does this leave the Super JetStream in our tests, though, and could a few differences set Palit's card apart from Gainward's?

For a long time, the Super JetStream represented the top of Palit's factory-overclocking efforts. Only the higher-end GeForce GTX 1070 and 1080 have more aggressively-tuned GameRock Premium Edition models. In the GTX 1060 market, Palit decided against such a halo offering (and with good reason, as unnecessary fragmentation costs money).

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The Super JetStream's fan shroud is made of black plastic with metal highlights. This cover is designed to be modular, as are the fans. Thus, a Palit card can easily be transformed into a Gainward model and vice versa by simply removing four screws.

Weighing in at 31 ounces (878 grams), the Super JetStream is moderately hefty, yet quite compact. Similar to the Gainward model, it measures only 24cm long, 11.7cm tall, and a substantial 4.7cm wide, practically requiring three expansion slots. Needless to say, you'd have a hard time wedging it into a mini-ITX-based system.

The back of the board is covered by a single-piece plate that lacks ventilation holes and does nothing to help with cooling. Because of the backplate, though, you must plan for an additional one-fifth of an inch (5mm) of clearance to avoid collisions with your CPU's cooler.

The top of the card features a backlit Palit logo with customizable RGB LEDs. A six-pin power connector is rotated 180° and positioned at the end of the PCB. This makes sense from a technical point of view, since the cooling fins don't need to be made any shorter.

A closer look at the heat pipes on the bottom reveals that they aren't embedded in a cooler block or heat spreader. Instead, they lay flat behind the base of the board (we already saw the same concept on Gainward's implementation). Whether this budget-oriented approach holds up any better in our tests remains to be seen.

Horizontally-oriented fins allow some waste heat to escape from an externally-facing slot bracket loaded with ventilation holes. This also applies to the card's other end, though, where hot air dumps out into the middle of your case. Good circulation is critical as a result.

The slot cover exposes five outputs, four of which can be used simultaneously to drive multi-monitor configurations. In addition to one dual-link DVI output (which lacks an analog signal), you also get one HDMI 2.0 interface and three DisplayPort 1.4-capable connectors.

Board & Components

The PCB that Palit and Gainward use features some prominent differences compared to Nvidia's reference design.

Right off the bat we see four power phases, just like Nvidia's card. Three are meant for the GPU and one corresponds to the memory. Two of the three GPU phases are fed by the auxiliary power connector. The third GPU phase and the one for the memory get power from the motherboard slot.

While, in theory, Palit's PCB offers the option to use one more power phase, the company went with a rarely-used and rather inexpensive NCP81174 for its PWM controller, which can only control up to four phases. Thus, the vacant emplacement you see in the image above could only have been populated by another parallel regulator circuit (via doubling).

The three GPU phases utilize a highly integrated Vishay SiC632, which combines the gate driver, high- and low-side MOSFET, and Schottky-diode in one convenient package. Meanwhile, the memory is driven by a pair of N-Channel MOSFETs, including one SM4503NHKP (30A, high-side) and one SM4377NSKP (50A, low-side), both manufactured by Sinopower. The MOSFETs are also fairly cost-effective choices.

Overall, the board reflects that Gainward didn't spend big on pricey components. Even the Foxconn Magic chokes are middle-of-the-road. They get their job done in a rather unremarkable way.

Two capacitors are installed right below the GPU to absorb and equalize voltage peaks.

Only six of the eight available memory emplacements are populated with Samsung K4G80325FB-HC25 modules. Each one adds 8Gb (32x256Mb) and operates at voltages between 1.305 and 1.597V, depending on clock rate. In total, they add up to this card's 6GB of GDDR5.

Power Results

The GPU Boost frequency and core voltage curves are particularly interesting to us. Our data shows that frequency drops significantly as the card heats up during a gaming session. Whether that's due to a higher-quality GPU or the clock rate's higher starting point compared to Gainward's version is a question we can't easily answer.

Otherwise, we observe a lot of similarities. The 2025 MHz we initially measured falls to an acceptable 1949 MHz under load. Voltage behaves similarly: we measured 0.975V on average, with readings varying between 1.025 and 0.9V. All of this points back to GPU Boost 3.0 and a relatively low power target that limits consumption to a maximum of 130W.

The lowest GPU clock rate measured with an idle card is about 253 MHz, just like Gainward's take on the same implementation.

We took our measurements using a variable low-pass filter, so we mention short load peaks only as a side note (see the grayed-out bar in the chart below), since those peaks are rarely relevant in practice.

One set of readings that does stand out is the noticeably higher gaming power consumption, especially at lower resolutions. Meanwhile, up top, Palit's GeForce GTX 1060 Super JetStream fares the same as Gainward's GeForce GTX 1060 6GB Phoenix "GS". Since Palit is slightly faster at 1080p, though, the results remain proportionate.

A low power target causes GPU Boost to cut the voltages dramatically during our stress test, resulting in power consumption numbers that are lower than what we measure during a worst-case gaming workload.

Power Connector Load

Load distribution from the four voltage regulators to their respective power phases isn't problematic at all. However, there's little headroom for overclocking, even if it wasn't already limited by the firmware's power target and voltage settings.

Here's the data for the gaming and stress tests:

Power consumption on its own only tells us part of the story. The PCI-SIG specifies a maximum of 5.5A at the motherboard slot, and we're showing a reading of about 5A, which puts Palit's GeForce GTX 1060 Super JetStream in the green (though a bit higher than Gainward's card).

Of course, we have the data for measured currents as well:

The Cooler & Its Performance

Although we already covered Gainward's cooler, it's worth discussing Palit's as well since there are a couple of details that differ. Like the backplate Gainward uses, this one is purely aesthetic; it doesn't help cool the PCB at all. Four screws secure it to the top of the board, shielding the card from curious customers.

The fan shrouds that Gainward and Palit use look a little different. Gainward's card is equipped with slightly more powerful fans (6.6W, 2700 RPM max.), while Palit's Super JetStream only includes 4.8W models, which top out at 2500 RPM. In addition, the 9cm blades employ a different geometry, which in Palit's case is slightly wavy. Their surfaces is smooth, without grooves.

It's also worth mentioning that the fans rotate in opposite directions. This should help improve efficiency due to more chaotic air flow.

Again, the delicate-looking cooler sports horizontally-oriented fins, as well as a quartet of 4mm heat pipes made from sintered composite material. Those pipes are not integrated into the bottom of the cooler. Instead, they merely touch the back of the thin base plate. Expensive copper is nowhere to be found.

We're glad to see that the base plate has a specific area to help cool the MOSFETs. However, the coils have to live without the benefit of active cooling.

Registering between 162 to 163°F (72 to 73°C) during our gaming loop, and a maximum of 169°F (76°C) in a closed case, the cooler does a fair job. It would almost assuredly perform better if the heat pipes were attached more securely, though. The stress test shows a similar picture.

The temperatures we measure on Palit's GeForce GTX 1060 Super JetStream are one to two Kelvin above Gainward's, which could be due to the higher power numbers we observed and its weaker fans.

Measurements collected with the backplate removed also reveal acceptable values. A peak of approximately 180°F (82°C) from the VRMs is reasonable, in our opinion. Interestingly, the different fan geometry clearly helps cool the PCB more efficiently.

With readings around 189°F (87°C), our stress test doesn't reveal any serious problems.

Palit's cooling performance is basically acceptable. We get the impression that one member of the finance team was assigned to each engineer, keeping costs in check more than enthusiasts like to see.

Sound Results

Hysteresis is not an issue for most manufacturers. However, Palit implemented a fan curve with too-low of an initial rotational speed for the transition between passive and active mode. As a result, the fans simply stop, and need to be restarted repeatedly. Even the best hysteresis won't help when a manufacturer overestimates the quality of its components this grossly.

A minimum value of 400 to 450 RPM (~600 RPM as a start-up pulse) would have been necessary, corresponding to about 18 percent as a PWM controller signal. Knowing this helps us manually modify the fan curve accordingly, thus getting rid of the annoying start-stop cycles.

When the card is idle, noise is not measurable due to a semi-passive mode. A reading of 35.8 dB(A) under load is acceptable, but not great. So much air has to be moved through the thermal solution because Palit's cooler isn't as efficient as it should be. Spending a few dollars more would have gone a long way. On the bright side, we measure very little low-frequency bearing noise. What remains is the sort of acoustic energy that's easily absorbed by a case with sound insulation.

MORE: Nvidia GeForce GTX 1080 Roundup

MORE: Nvidia GeForce GTX 1070 Roundup

MORE: All Graphics Content

The GTX 1050 Ti KalmX builds on Palit's first passively-cooled card, the GeForce GTX 750 Ti. That was a 60W board, though; the 1050 Ti is rated for 75W. Can it still be cooled passively? Based on what we learned during our own Nvidia GeForce GTX 1050 Ti Passive Cooling Mod, we have some doubts. Without extra airflow, there's just too much heat to maintain full performance.

Palit's card works around this with a more conservative power budget, as you might guess from the missing six-pin connector. Because the PCB is borrowed from a standard GTX 1050 Ti, the pads for that connector are there--just unused. Still, Nvidia's GP107 processor operates at the 1050 Ti's reference 1,291MHz with a rated GPU Boost frequency of 1,392MHz.

The memory is configured at 1,752MHz, allowing Palit to leave the modules uncooled except by normal air circulation.

Weighing just over 16 ounces (466g), the card is fairly light, and with measurements of 7.2 x 5.4 x 1.4 inches wide (LxHxW), the dual-slot form factor should fit almost anywhere.

For a number of reasons, Palit didn't slap a plate on the back of its card. Chief among them is avoiding compatibility issues with CPU coolers in mini-ITX machines.

Two nickel-plated 6mm heat pipes in a copper sink dissipate thermal energy as evenly as possible through aluminum fins. The sink protrudes from the end and top of the PCB to facilitate better air circulation, and the fins themselves are oriented vertically, similar to past passive solutions from Palit. Although this promotes natural convection, it benefits less from front-to-back airflow in the chassis.

The fin alignment is most helpful when the card is installed vertically, as you'd see in many mini-ITX enclosures. In that case, the horizontal fins used on HIS' and XFX's passively-cooled Radeon RX 460 cards are at a disadvantage.

There's quite a bit of space between the fins, which is typical for passively-cooled thermal solutions relying on natural convection.

Palit even directly cools its card's VRMs. As we mentioned, though, the memory modules don't receive this benefit. Instead, they rely on airflow in the case to remove heat.

Outputs are limited to one DVI-D port, one HDMI connector, and a full-sized DisplayPort interface. This leaves room for plenty of openings for ventilation. They're almost pointless, though, because the vertical fin orientation encourages air to move perpendicular to the output bracket.

Expect the GTX 1050 Ti KalmX 4GB to sell for around $150, although prices might be higher at first.

We're looking a face-off between Palit's card and passively-cooled Radeon RX 460s from HIS and XFX. On paper, the Nvidia-based board has the advantage based on clock rates and architecture, but we really need to determine if GPU Boost is still an effective performance enhancement when this card is installed in a case and under load.

Introduction & Overview

Since we published our GeForce GTX 1080 review, Nvidia overcame availability issues, and prices on the top-end gaming card settled into the range we were told to expect at launch. Today, you'll find GeForce GTX 1080 cards selling online between $600 (£750) and $770 (£900).

When Nvidia introduced GeForce GTX 1080, we only had the Founders Edition board (the company's reference design) in our possession. While its rear exhaust and a high-quality thermal solution turned heads, we knew boards from Nvidia's partners could bring lower prices, factory overclocking, and more cooling designs to the table.

All of these cards employ the same GP104 processor, so gaming performance takes a back seat to acoustic, electrical, and thermal readings. Each board is reviewed on its own page, where we dissect build quality, differentiating features, power consumption, clock rate analysis, cooling, and acoustic measurements in depth.

Including Nvidia's GeForce GTX 1080 Founders Edition, we have 10 cards compared in this roundup so far. We'll continue adding GeForce GTX 1080 cards to this roundup as we review them.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

The final analysis of each card is listed below for quick and easy comparison. The cards appear in alphabetical order, and do not reflect any valuation or preference for individual products.

MORE: Best Deals

Benchmark Results

Gaming

We test every card after a suitable warm-up period to avoid unfair differences in GPU Boost frequencies. All benchmarks are run six times; the first one is used to get the GPU hot again.

These cards are all press samples operating at the same settings as retail models in our best effort to ensure one vendor doesn't get a leg up on another using non-representative clock rates.

The following galleries each contain four images, covering two tested resolutions. We put our focus on QHD (2560x1440) and UHD (3840x2160), plotting out average and minimum frame rates for each resolution in separate graphs.

All of the factory-overclocked cards offer similar performance, more or less. That's why our primary focus centers on evaluating the more technical aspects of each board design, along with their coolers. This is where differences in noise, power, and temperatures are most likely to come from.

Power Consumption

We begin by comparing the power consumption of each card in our gaming loop, stress test, and at idle. Depending on the power targets specified by each manufacturer, we sometimes measured substantial differences. We're ignoring the decimal places in our bar graphs, since those values are too small and would be within measurement tolerances.

We also found that some cards with lower power targets started to throttle during our stress test, resulting in lower power consumption numbers. This did not, however, have a negative impact on general gaming performance for any of the tested cards, as the stress test merely represents a worst-case scenario.

We confirmed that MSI's retail cards will ship with a slightly lower power target (max. 240 to 250 watts) after an internal discussion and evaluation of our measurements.

This also applies to the BIOS versions with OC mode enabled by default, which employ a roughly 20 MHz-higher base and GPU Boost frequency. In the interest of fairness, we tested both MSI cards using normal mode, without the overclocked base and GPU Boost rates. This doesn't affect our power consumption measurements, though.

Noise

For the following comparison, we divide all of the gallery's bar graphs into gaming loop and idle, even if "noise" in practice spans a wide spectrum and the "character" of the sound varies a lot, with each card having highly individual results. Therefore, it is important not to compare just the absolute numbers, but also the frequency spectrum we're presenting.

Many of the cards implement a semi-passive mode, where their fans remain off when the card is idle. Thus, we refrained from taking measurements in that state. Even in our anechoic chamber, levels of 22 dB(A) and below merely represent ambient noise.

Temperature

For this comparison, we divide the gallery's bar graphs into gaming loop, stress test, and peak temperatures measured on the MOSFETs.

We occasionally compared the temperatures on our benchmark table with those measured inside a closed case and found them to be no more than two or three Kelvin higher.

Since temperatures in a closed case also depend heavily on the enclosure's cooling performance, the only representative and reproducible values are those measured on our benchmark table. Those are the ones we compare.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Nvidia GeForce GTX 1080 Founders Edition

The GeForce GTX 1080 Founders Edition is Nvidia's current desktop flagship by its own design. Despite an emphasis on craftsmanship in the company's marketing materials and generally improved efficiency, the reference design will have a hard time trying to justify its high price compared to compelling solutions from add-in board partners. This is doubly applicable since the restrictive power target of just 180W and a 172°F (80°C) temperature target are real limitations.

As with the GeForce GTX 1070 Founders Edition, Nvidia chose a mechanical-looking design and a radial fan. The card's true two-slot form factor also makes it a good option in multi-GPU machines.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The GeForce GTX 1080 Founders Edition's shroud is made of injection molded aluminum, and colored both silver and black. The metal cover coveys plenty of quality, but also results in a weight of more than 35 ounces (one kilogram). Both this 1080 and Nvidia's GeForce GTX 1070 Founders Edition have an almost identical weight (the 1080 is just one-third of an ounce, or 10 grams heavier).

Around back, the board is covered by a two-piece plate that serves no practical purpose other than to facilitate a more finished appearance. If you want to increase airflow in a multi-card configuration with 1080s back to back, the plate can be unscrewed without causing a problem.

The top of the card is dominated by a glowing, green GeForce GTX logo. The eight-pin auxiliary power connector is positioned at the end of the card. The sharp-edged and mechanical design may be a matter of taste, but it certainly is distinctive.

A peek at the card's back reveals fins and a mounting frame. Three screw holes are provided for attaching brackets to stabilize the card in a case.

Five outputs populate the rear bracket, four of which can be used simultaneously in a multi-monitor setup. You get one dual-link DVI-D connector, one HDMI 2.0b port, and three DisplayPort 1.4 outputs. The rest of the plate is ventilated to maximize exhaust flow.

Board & Components

A glance at the PCB reveals that it offers significantly more space than is actually used. In addition to one power phase for the memory, five of the six possible phases for the GPU are implemented. There's even space for an extra power connector, if it's needed. 

As you no doubt know, Nvidia taps GDDR5X memory from Micron for this board. Eight of these memory chips operating at 1251 MHz are connected to the GPU through an aggregate 256-bit interface, enabling a theoretical bandwidth of 320 GB/s.

The 5+1-phase system relies on the sparsely documented µP9511P PWM controller. Since this controller can't communicate directly with the VRM's phases, Nvidia utilizes 53603A chips for solid PWM drivers (gate drivers) and controlling the power MOSFETs (primarily of type 4C85N).

Two capacitors are installed right below the GPU to absorb and equalize peaks in voltage. The board design looks tidy and well thought-out.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. This also shows that both curves drop as the GPU's temperature rises.

After warm-up and under load, the GPU Boost clock rate at times drops to GP104's base frequency of 1.605 GHz. This is mirrored in our voltage measurements. While we measured up to 1.062V at first, that number temporarily drops as low as 0.881V.

Combining the measured voltages and currents allows us to derive a total power consumption we can easily confirm with our instrumentation by taking readings at the card's power connectors. In fact, let's start with the measured power consumption values in the following table:

These charts go into more detail on power consumption at idle, during 4K gaming, and under the effects of our stress test. The graphs show how load is distributed between each voltage and supply rail, providing a bird's eye view of load variations and peaks.

Temperature Results

To dissipate the GP104 GPU's waste heat, Nvidia's GeForce GTX 1080 Founders Edition employs a real vapor chamber solution. The 1070 Founders Edition does not benefit from this same technology. The vapor chamber is a compact module that's attached to the PCB via four screws and positioned over the GPU package.

A radial fan pulls air from inside the chassis and blows it across heat sink fins on the vapor chamber, exhausting that air through the output bracket. The mounting frame is not only used to stabilize the card, but it also helps cool the voltage regulators and memory ICs.

The vapor chamber's performance is slightly better than the 1070 Founders Edition's copper heat sink, as you might expect. Still, temperatures rise to almost 185°F (85°C) during our stress test and 180°F (82°C) during the gaming loop. Then again, that's hardly shocking since the 1080 is rated 30W higher.

Sound Results

Initially, both fan speed curves stay relatively close to one another. But at a certain point in our benchmark, FurMark forces the RPM measurement to burst much higher. During the more real-world gaming loop, however, the fan cruises along just above 2100 RPM.

Noise levels are relatively low when the card is idle, even if the fan's sound has a slightly snarly character. Nvidia doesn't gift the 1080 Founders Edition with a semi-passive mode. But then again, it wouldn't be particularly useful on a card with a radial fan anyways.

Even under longer gaming loads, noise levels stay below the 42 dB(A) mark. Not bad for a radial fan. However, during our purposely taxing stress test, the fans ramp up beyond 46 dB(A). The frequency spectrum is rather wide though, so the white noise doesn't feel too intrusive.

Overall, Nvidia's thermal solution is workable. The radial fan is great for exhausting hot air from the card's back, but it's miles away from making the 1080 Founders Edition quiet.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

EVGA GeForce GTX 1080 FTW Gaming ACX 3.0

FTW is the shortened version of For The Win, and with that abbreviation EVGA sets the bar pretty high for itself. We do like that the card can be disassembled without voiding its warranty. The terms of EVGA's coverage are thus very favorable to water-cooling and modding enthusiasts. This is downright uncommon, which is why we want to point it out upfront. But we still don't know anything about the 1080 FTW Gaming ACX 3.0's technical attributes...yet, at least.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The cooler cover is made of light metal. Together with the underlying acrylic plate, the extra tacked on are primarily designed to provide some eye candy. As a result, the whole top of the card looks like it's bathed in color; built-in RGB LEDs light up the shroud thanks to numerous recesses. Right out of the gate, EVGA's FTW is the brightest card in our test field.

Weighing 38oz (1077g), this card isn't particularly heavy. But it's not a lightweight either. Measuring 11 inches (or 27.7cm), it isn't excessively long. It's five inches (or 12.5cm) tall, though, and 1 3/8 inch (3.5cm) wide, matching many dual-slot graphics cards.

The back of the board is covered by a single-piece plate with several openings for ventilation. You'll have to plan for an additional one-fifth of an inch (5mm) beyond the backplate, which could negatively affect configurations with multiple cards right next to each other. It is perfectly possible to use the card without that cover, of course. However, removing it also necessitates pulling off the cooler.

The top of the card features EVGA's logo. Two eight-pin power connectors are positioned at the end of the card, right where we'd expect to find them. As with most designs that strive to be unique, this one is a matter of personal preference. We're sure it'll find its fans, though. While there are undoubtedly fancier cards available, being fancy isn't always a compliment either.

Fins visible from the end and bottom of the 1080 FTW Gaming ACX 3.0 show that they're positioned vertically, and won't allow air to flow out of the back. Instead, exhaust is pushed from the top and down, toward the motherboard.

The rear bracket exposes five connectors, four of which can be used simultaneously in a multi-monitor setup. In addition to one dual link DVI-D connector, you also get one HDMI 2.0b port and three full-sized DisplayPort 1.4 outputs. Ventilation holes dot the rest of the bracket. They don't serve any purpose though, given EVGA's fin orientation.

Board & Components

EVGA leverages its own circuit board design. At first glance, the card seems to employ a very clean layout that sparks our curiosity due to its many on-board components.

EVGA naturally uses the Micron GDDR5X memory modules that Nvidia sells along with its GPU. Eight of them operating at 1251 MHz are connected to GP104 through an aggregate 256-bit interface, enabling up to 320 GB/s of bandwidth.

Unlike Nvidia's reference design, EVGA's 5+2-phase system relies on an NPC81274 from ON Semiconductor as its PWM controller. Even if the card is advertised as having 10 power phases, that's technically a little deceiving. In reality, there are five phases, each of which is split into two separate converter circuits. This isn't a new trick by any means. It does help improve the distribution of current to create a larger cooling area. Furthermore, the shunt connection reduces the circuit's internal resistance. This is achieved with a NCP81162 current balancing phase doubler, which also contains the gate and power drivers.

For voltage regulation, one highly-integrated NCP81382 is used per converter circuit, which combines the high-side and low-side MOSFETs, as well as the Schottky diode, in a single chip. Thanks to the doubling of converter circuits, the coils are significantly smaller. This can be quite an advantage since the current per circuit is smaller as well. As a result, conductors can be reduced in diameter while retaining the same inductance.

Yet, compared to the KFA²/Galax GeForce GTX 1080 HoF, which uses this effect for a total of 15 converter circuits, EVGA achieves rather modest results as far as cooling is concerned. As we get into the 1080 FTW's benchmark results, we'll present some data backing that claim.

Current monitoring is enabled through a three-channel INA3221.

The memory modules are powered by two separate phases controlled with a NPC81278, which integrates the gate driver and PWM VID interface. A NTMFD4C85N by ON Semiconductor combines both high-side and low-side MOSFETs in one chip.

Two familiar capacitors are installed right below the GPU to absorb and equalize voltage peaks.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. EVGA uses a very high power target that, in turn, allows for a relatively constant GPU Boost clock rate. It only drops slightly as temperature increases, and the observed voltage behaves similarly.

After warming up in our gaming workload, the GPU Boost frequency, which initially started at 2 GHz, settles at a stable value of 1936 MHz. This falls to 1848 MHz under constant load.

Our voltage measurements look similar: while we observe up to 1.062V in the beginning, that number dips to an average 1.031V.

Combining the measured voltages and currents allows us to derive a total power consumption we can easily confirm with our instrumentation by taking readings at the card's power connectors.

Since Nvidia forces its partners to sacrifice the lowest possible clock rate in order to gain an extra GPU Boost bin, this card's power consumption is disproportionately high as it idles at 253 MHz. EVGA handles this setback fairly well, though. The consequences are listed below:

These charts go into more detail on power consumption at idle, during 4K gaming, and under the effects of our stress test. The graphs show how load is distributed between each voltage and supply rail, providing a bird's eye view of load variations and peaks.

Temperature Results

Update, 11/26/16: In the original version of this round-up, we exposed an issue with EVGA's ACX 3.0 thermal solution. The company used our feedback to improve its design, in the process giving customers a couple of options for modifying their existing cards. What follows is a review of those options and their impact on cooling/noise. We're replacing the previous data with results from our updated testing, since that most accurately reflects production hardware.

Naturally, heat output is directly related to power consumption, and the 1080 FTW's ability to dissipate that thermal energy can only be understood by looking at its cooling solution.

As with the 1080 Founders Edition, the backplate is mostly aesthetic; it doesn't serve much practical purpose. At best, it helps with the card's structural stability.

Unfortunately, the way the voltage regulation circuitry was originally cooled exposed a design flaw. This problem was evident in the mounting plate, which does draw heat away from the VRMs and memory. The plate has a cut-out for the coils, meant to keep the VRMs and memory from heating each other up. Unfortunately, it also reduced the cooling surface for an area right below the center of the fan rotor, where airflow is weak to begin with.

Under normal circumstances, the company's decision might have still turned out alright. But the GeForce GTX 1080 FTW Gaming ACX 3.0 has an exorbitantly high power target, which chases the card up to and beyond 230W under load. This puts a lot of stress on the VRMs. Given those conditions, the limited surface area just couldn't keep up.

The cooler uses a nickel-plated heat sink, along with four quarter-inch (6mm) and two one-third-inch (8mm) flattened heat pipes. Its actual capacity is adequate. But a look at the GPU temperatures shows that the upper limit of this short, dual-slot solution is in sight. We originally measured up to 167°F (75°C) during our gaming loop and up to 172°F (78°C) in a closed case. Under maximum load, the reading rose to 171°F (77°C), and 176 to 178°F (80 to 81°C) in a closed case. That was just too high, especially when you consider the data was collected in an air-conditioned room set to 72°F (22°C).

Re-Testing After EVGA's Thermal Modification

Clearly, something needed to be done. Back in November, we published EVGA Addresses GeForce GTX 1080 FTW PWM Temperature Problems, linking to new BIOS versions for five different cards. EVGA also offers optional thermal pads to any affected customer who requests them. Customers who do not feel comfortable updating the BIOS or who damage their card installing the thermal pad will receive EVGA’s full support, the company says.

For our part, we committed to a re-test once EVGA implemented its planned changes, and that’s what you see going live alongside our GeForce GTX 1080 round-up.

In order to address some of the challenges originally posed by EVGA’s card, we drilled two holes into the backplate, right above the two hottest points identified in our previous tests, and cut a circular part off the thermal pads to give our infrared camera a clearer view of the board. Since we are using the same card as before, the results are directly comparable.

Meanwhile, several images have surfaced in forums allegedly showing that the thermal pads used in manufacturing don't completely fill the gap between the front plate and memory modules. Working with enthusiasts online, we analyzed several EVGA cards and couldn’t find evidence of this gap issue, as the picture below shows:

In every case, the pads were placed correctly, even if they weren’t a tight fit. Furthermore, the pictures on the Internet all show the pads sticking to the front cover, and the gap is between the memory module and thermal pad. To better understand what this means, it’s necessary to know a bit about how graphics cards are manufactured.

Thermal pads are almost exclusively glued onto the memory modules first. The front plate is attached later, together with the backplate. Thus, if a thermal pad is stuck to a plate now, at the very least it had to have firmer contact at some earlier point in time.

Our observation is that if the backplate is removed, and then the additional screws holding the front panel are removed as well, that plate is bent in such a way that it leaves a slightly angled gap, which looks very similar to these images. This is also the case if the backplate has been reattached, but the additional screws holding the front panel have not been screwed back in during reassembly.

There could, however, be gaps in the case of a GeForce GTX 1070 with this same type of cooler if the GDDR5 memory modules are a different height than the GDDR5X modules on 1080 cards. And since the thermal pads on the memory modules are kept as thin as possible, that could lead to the reported behavior.

Soon it won’t matter, though. As of mid-November, EVGA will make its thermal pads 0.2 mm thicker, just to be 100% sure there aren’t any issues. This change applies to the pads included in the thermal mod we requested, and those in mass production as well. Furthermore, the newest retail cards will come with the new BIOS versions already installed.

Because we are curious to see how much the new pads affect EVGA’s cards, we’ll test them in three steps. First, we’ll test the pad between the back of the board and the backplate. Second, we’ll test the previously installed pad and another pad between the front plate and heat sink’s cooling fins. Third, we’ll combine all of the aforementioned modifications together with the new BIOS.

Doing so should tell us if it’s really necessary to flash EVGA’s firmware and accept the disadvantage of higher noise levels. If the thermal pads themselves do the trick, perhaps it’s possible to ignore the software side altogether.

EVGA supplies a small plastic bag containing one large thermal pad (for the back) and a smaller one (for the front). They’re complemented by a supply of original EVGA thermal paste, which we’ll use later when it comes time to prepare the heat sink for modifications to the front plate. The backplate, however, is very easy to remove.

According to EVGA, you should install the larger of the two thermal pads as shown below. However, this positioning means that, in some places, the thermal pad will have direct contact with a protective foil (rather than the backplate itself). EVGA uses this foil to seal most of the holes in the backplate. We decided to leave the backplate in its original condition, but recommend removing these thermally unfavorable coverings to create more contact surface and improve airflow.

The darker area on the PCB shows where the thermal pad was positioned during installation. Since the visible area is quite large, it can be assumed that the adhesive thermal pads create good contact. This also speaks to their quality.

The voltage regulators are completely covered, and the RAM is covered at least in part. This result could be optimized further if the thermal pad was positioned about four-fifths of an inch (two centimeters) further to the right, as seen from the back of the PCB. In that case, two rectangular cut-outs need to be made in the top and bottom corners on the right-hand side so that the thermal pad doesn't cover two holes needed for screws.

Applying the thermal pads to the front panel is a bit trickier, since EVGA explicitly recommends covering the long-hole cut-out for the coils, creating a closed surface. Furthermore, this closed surface must also be tall enough to provide sufficient contact with the cooling fins of the main heat sink.

The image taken after re-disassembly shows visible stripe-shaped impressions where this contact happens. We made sure to include the area around the coils, which was one of the hottest parts of the board during our previous measurements. This should significantly reduce the heat propagation inside the board, especially towards memory and GPU.

Measurements During The Gaming Loop

Our metrics are demanding, so the card’s power consumption increases to values that very few games actually reach. It’s reasonably safe to assume that the temperatures we measure are indeed representative of a worst-case scenario.

When the card is idle, the memory and GPU both remain below 104°F (40°C).

The first measurements we’re presenting come from the GPU diode after all three stages of the thermal modification. We’ll compare those readings to our results prior to EVGA’s fix.

The temperature measurements without flashing the BIOS look very similar, which of course is due to the old fan curve. This curve changes significantly in the modified BIOS.

Since those temperatures also have a direct influence on GPU Boost clock rates, we add frequency results to the chart below:

With the original fan curve, GPU Boost frequencies follow the temperatures. Physically modifying the card with thermal pads doesn’t significantly change the GPU temperature or GPU Boost values. Only the new BIOS, with its corresponding higher fan speed and acoustic output, cool the GPU noticeably, enabling more aggressive frequencies.

This wasn’t the focus of our original results, though. The focus there was excessive temperatures measured on completely different parts of the board. To follow up, we need our IR camera.

Original Measurements Without the Modification

To recap, our original round-up results showed the memory modules at their officially specified thermal limits, even during our Metro: Last Light gaming loop.

We measured the card in its original state a second time, with the backplate attached, just to be really sure. The noise levels remained more or less the same.

Measurements After Installing the Thermal Pad on the Back

First, we want to see how much the thermal pad between the circuit board and backplate improves cooling on its own. This may serve as an example in the future, predicting what another poorly-cooled card might gain from similar modifications. The upside here is that we don’t need to remove the cooler; we can perform the mod without handling thermal paste or worrying about parts we took off.

As we can see, memory modules and MOSFETs are about nine degrees Kelvin cooler! However, the memory temperatures are still too high during our stress test. We also note that the GPU package (not the GPU itself) is hotter due to a significantly warmer backplate. Now we know why the temperature of the GPU diode was a little higher with thermal pad in place compared to the original readings without the pad.

It's also noteworthy that the card is a little bit quieter, which could be due to the lower overall GPU diode temperatures, even if the peaks are a little higher at times. On the other hand, 0.5 dB(A) doesn’t make an audible difference:

Measurements with Thermal Pads on Front and Back

We are going to throw in another thermal pad and connect the upper side of the front plate to the large heat sink using thermal paste. This solution clearly improves heat dissipation, especially for the coils and area around the VRMs. A closer look at the measurements shows that the VRMs are already 16 degrees Kelvin cooler than the first test without thermal pads. That’s incredibly significant.  

The memory enjoys plenty of relief as well: almost 15 degrees Kelvin in the gaming loop and at least seven degrees during the stress test. These modifications improve the card's cooling enough to allow some reserves for warmer days. With them, EVGA has no reason to shy away from the competition.

The thermal mod with both pads in place leaves us with similar results; we observe similar noise levels as the original, too.

Measurements of all Modifications, Including the BIOS

Even if it doesn't seem necessary at this point, we still have one more option for improving thermal performance. As such, we flashed the BIOS on our card to EVGA’s latest, available on the company’s website.

The fans spring into action and jump to high RPM, yielding impressive results. The GPU Boost frequency increases, but so does the noise level. Originally, the fan spun at 1300 RPM; now it’s over 1500 RPM. That ~200 RPM jump exacts a notable price that we’ll quantify shortly.

A look at the temperatures makes us wonder why the original BIOS didn’t have the fan spinning 100 RPM faster already. We think the new firmware’s fan curve is a little too aggressive. Even half of the increase would have sufficed.

An incredible 25-degree Kelvin drop on the VRMs in our gaming loop and stress test is a testament to EVGA’s effort. But was it really necessary to go to such extremes?

Now we want to see what these measures do to acoustics. A reading of almost 41 dB(A) the card is distinctly audible and what most would consider loud. The only upside is a more uniform fan noise that can be filtered relatively easily with an insulated case.

A more moderate curve would have left the card between 37 and 38 dB(A), and sufficiently cooled. In the end, the BIOS update gets a small question mark, though we can at least understand it’s easier to implement than a hardware modification.

Conclusion

Two months passed after we first contacted EVGA at the beginning of September and the publication of this update on Tom’s Hardware DE. The company should have probably reacted faster to counter the forum posts from unhappy customers discussing poor thermal performance.

In the end, though, it’s the final result that matters, and EVGA did successfully solve its issue. If the ACX 3.0-equipped cards are used in cases with at least moderate airflow, the BIOS update is superfluous. EVGA would be well-advised to simply install the two thermal pads in mass production and leave the new firmware on its site as an optional download. Otherwise, you’re going to be subjected to 41 dB(A) of noise for a slightly higher GPU Boost frequency. Adding the thermal pads is enough to achieve what gamers want: a cool, but still pleasantly quiet card. Enthusiasts willing to go with the noisier option for a bit of extra performance can flash the BIOS or, even easier, set up a custom fan curve with EVGA’s Those who absolutely want to go with the noisier option to gain a smidgen of extra performance could either just flash the BIOS, or (even easier) set up a custom fan curve with EVGA's own Precision XOC software.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

Galax/KFA² GTX 1080 Hall of Fame

This card, which is sold in Germany under the name KFA² GeForce GTX 1080 Hall of Fame, is marketed internationally using the Galax brand. Both are exactly the same, though. They come from Galaxy Microsystems, which also refers to itself as Galaxy and Galaxytech. You won't find the company's cards on Newegg or Amazon. Rather, the only way we've found to buy the 1080 HoF is through galax.com.

Why might you feel compelled to poke the company's site? To begin, the 1080 Hall of Fame has a very different look. It sports a white shroud, white body panels, a white backplate, and even a white PCA. Anyone with an MSI Titanium-series motherboard (Z170A or X99) has a match made in heaven.

Beyond the premium paint scheme, we're also impressed by Galax's bundled extras. Among the extraneous gimmicks, you also get a structural support to keep the heavy card from flexing in its slot.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces 

The cooler shroud is made of blindingly white plastic with aluminum highlights for additional eye candy. Overall, Galax went for a mechanical appearance with lots of bold edges and corners.

Weighing 46oz (1315g), this card is very heavy, which explains the support you get in the box. With a length of 12.5in (31.7cm), smaller cases may have a hard time accommodating the 1080 Hall of Fame. A height of five and one-third inches (13.5cm) from the motherboard slot's top edge isn't too subtle either. Although this is a dual-slot card on paper, some smaller parts are an extra 2mm wider, resulting in a total width of 1.46in (3.7cm).

The back of the board is covered by a single-piece aluminum plate with ventilation holes cut into it. Plan for an additional one-fifth of an inch (5mm) in depth beyond the plate, which may become relevant in multi-GPU configurations, particularly if your motherboard's PCIe slots are two spaces apart.

We heavily discourage using this heavy card without its backplate. Despite the front-side mounting and bundled cooling frame, structural stability would suffer significantly without that support.

The top of the card features a back-lit "Hall of Fame" logo and two white auxiliary power connectors.

Vertically-oriented fins prevent air from exiting the front or back of the card. A button on the output bracket lets you set the fans to run a maximum speed; it serves no other purpose and definitely isn't a BIOS switch.

The rear bracket features five outputs, of which a maximum of four can be used simultaneously in a multi-monitor setup. In addition to one dual-link DVI-D connector (be aware that there is no analog signal), the bracket also exposes one HDMI 2.0b and three DisplayPort 1.4-ready outputs. The rest of the plate is mostly solid, with several openings cut into it that look like they're supposed to improve airflow, but don't actually do anything.

Board & Components

As far as memory is concerned, the 1080 HoF is the same as its competition. It uses GDDR5X memory modules from Micron, which are sold along with Nvidia's GPU to board partners. Eight memory chips (MT58K256M32JA-100) transferring at 10 MT/s are attached to a 256-bit interface, allowing for a theoretical bandwidth of 320 GB/s.

The rest of the PCB is considerably more interesting, though. KFA²/Galax proudly advertises a 12+3-phase layout. That's not technically true, so how does the company arrive at the number on its marketing material?

Let's start with a look at the GPU's voltage regulation. Unlike other manufacturers, KFA²/Galax went with International Rectifier's good old IR3595A, which we used to find on a lot of AMD graphics cards. This chip can be used as either a single-loop or dual-loop PWM controller. All six phases provided in loop one are used for the GPU. The rest are regulated by other means.

The reasons why KFA² went with the IR3595A, even though it still uses VRM11 instead of OVR4, are likely found in the fact that it offers firmware programmers more options for customization and complete control over voltage regulation. This also provides some distinct advantages for extreme overclocking, since the chip allows the use of additional extensions. In short, there are benefits. On the other hand, it may conflict with the sensor loops of third-party tools like GPU-Z if monitoring is activated.

But where does the 12-phase marketing claim come from? The 1080 HoF has a small IR3599, which implements a so-called doubler that splits each phase in two individual converter circuits. Thus, there are six true phases and 12 converter circuits.

This isn't a new trick. It does improve the distribution of current, creating a larger cooling area for the MOSFETs. Furthermore, the shunt connection also reduces the circuit's internal resistance as a whole.

Since such a high number of converter rails also leads to a shortage of space, this card uses a highly integrated IR3555 with high- and low-side MOSFETs, a gate driver, and Schottky diode on a single chip, freeing up a little bit of real estate. Spoiler alert: compared to EVGA's GTX 1080 FTW, this concept works very well when it comes to cooling the card.

In order to fit all of the requisite components on its circuit board (that's 12 coils for the GPU and three for the memory/memory controller), Galax relies on vertically-stacked coils encapsulated in a ferrite frame. They're more susceptible to vibration though, and tend to buzz. We'll explore their impact on acoustics shortly.

Three phases are reserved for the GDDR5X and memory controller, employing conventional N-channel MOSFETS for the high- (MDU 1514) and low-sides (MDU 1511), along with an external gate driver. Additionally, a uP9509 is used as the PWM controller. The coils are the same ones used for the GPU.

Current monitoring is handled by an INA3221. Two familiar capacitors are installed right below the GPU to absorb and equalize peaks in voltage.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which show such striking similarity that we deliberately put their graphs one after the other. KFA²/Galax sets a very high power target that facilitates a relatively constant GPU Boost frequency. It only drops slightly as the temperature increases, and our voltage readings mirror that behavior.

After warming up in a variable-load gaming scenario, the GPU Boost clock rate that started at 2025 MHz stabilizes at 1987 MHz, and then slides to 1923 MHz under constant load. While we measure up to 1.062V at first, voltage later drops to an average of 1.043V.

Summing up measured voltages and currents, we arrive at a total consumption figure we can easily confirm with our test equipment by monitoring the card's power connectors.

As a result of Nvidia's restrictions, manufacturers sacrifice the lowest possible frequency bin in order to gain an extra GPU Boost step. So, the GTX 1080 HoF's power consumption is disproportionately high as it idles at 278 MHz.

Now let's take a more detailed look at power consumption when the card is idle, when it's gaming at 4K, and during our stress test. The graphs show the distribution of load between each voltage and supply rail, providing a bird's eye view of variations and peaks:

Temperature Results

Naturally, heat output is directly related to power consumption, and the 1080 HoF's ability to dissipate that thermal energy can only be understood by looking at its cooling solution.

As with the 1080 Founders Edition, the backplate is mostly aesthetic; it doesn't serve much practical purpose. At best, it helps with the card's structural stability. But the thermal solution is completely sufficient anyway, as our benchmarks show.

KFA² applies a black coating to the back of its plate, improving the absorption of heat. The mounting and cooling frame is made from injection-molded aluminum and placed on top of the PCB. Then it's attached to the backplate with several screws. This structure, which sandwiches the board, also cools the voltage regulators and memory modules through thermal pads.

Why does this concept work so well on the GTX 1080 HoF and not on EVGA's GTX 1080 FTW? Looking at the top of the frame, we see a number of fins that not only increase the cooler's surface area but also take better advantage of available airflow.

The heat sink itself is one of the largest and heaviest you'll find, aside from the solutions that Palit and Gainward use. But KFA²/Galax doesn't stop there.

The cooler relies on a copper block topped with one quarter-inch (6mm) pipe and three more measuring one-third of an inch (8mm) in diameter. During our gaming test, we observed temperatures of only 151°F (66°C) on an open bench, and 154°F (68°C) to 156°F (69°C) in a closed case. Clearly this card's thermal performance is good.

But the high power target does take its toll during our stress test. Power consumption as high as 240W heats the GPU up to 162°F (72°C) on an open bench and 165°F (74°C) in a closed case. Then again, none of us actually play FurMark.

One look at the infrared images reveals the effectiveness of well-planned VRM and memory cooling, even without the need for additional built-in heat sinks. On the other hand, KFA²/Galax's massive mounting/cooling frame wouldn't fit on smaller cards.

A reading of 154°F (68°C) below the armada of VRMs sets a new record, pleasantly surprising us in the process. The GPU enjoys competent cooling as well. In the end, the package is hotter than its graphics processor.

During our stress test, the cooler has to handle an additional 46W, driving temperatures higher. Yet even a reading of 167°F (75°C) is still mild compared to some of the other cards we're benchmarking.

Sound Results

We started looking at Galax's 1080 HoF back in July, and the initial results identified some acoustic issues. Our German team contacted the manufacturer directly and, together with its R&D department, solved the problems we wrote about. The newest cards have been modified according to our guidance, and as of August 2016, they're in mass production. If you buy today, you'll get one of the improved boards.

Let's start with the fan curve, which shows that KFA2/Galax made a conscious decision to eschew semi-passive operation, even at idle. The complaint we originally had was a relatively high 33 percent minimum duty cycle. This is what we saw:

A 33.6 dB(A) result is hardly what you'd call quiet. Our first attempt to tweak the fan controls down to 28 percent duty cycle still left us at 32.6 dB(A). Low-frequency peaks were particularly obtrusive, and the sandwiched PCB caused those sounds to resonate in our chassis.

Under full load, the fans wound up to maximum performance and became even more noticeable. During our stress test they peaked near 38 dB(A). We could have lived with 36.1 dB(A) during our gaming loop, but those low frequencies just dominated the otherwise tolerable sounds.

After the testing we published in July, KFA²/Galax passed our measurements on to its fan supplier with a request for comment and a demand for improvement. When the dust cleared, the manufacturer provided a history of the issue.

In almost every card, regardless of vendor, the fan modules are screwed directly to cover. Depending on the resonance frequency of this structure, there is a chance it'll match the fan noise and form an acoustically disastrous alliance. In order to prevent that, Galax installs those modules without any direct contact.

The fan vendor's measurements not only confirmed our objections, but also proved the effectiveness of its solution. This included a blind test carried out with employees and third-parties.

As the saying goes, trust but always verify. That's why we want to get our hands on current versions of KFA²/Galax GeForce GTX 1080 Hall of Fame and GeForce GTX 1070 Hall of Fame, which would allow us to retake our measurements. We'll publish the results in an update when this happens.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

Gigabyte GTX 1080 G1 Gaming

Although Gigabyte sells six different versions of the 1080, we're starting with the company's GeForce GTX 1080 G1 Gaming. On the next page we add its 1080 Xtreme Gaming model. Once we make the rounds and cover more of the competition's hardware, we'll circle back around to see what else Gigabyte has to offer.

Let's start with a summary of the 1080 G1 Gaming's technical specifications:

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

Gigabyte's shroud is made of thin plastic and doesn't exude the same high quality as the metallic covers over some of its previous Windforce-equipped models. The trade-off is that this card weighs just 31oz (871g), despite measuring 11 inches (28.4cm) long, 4 1/3 inches (11cm) high, and 1 3/8 inches (3.5cm) wide.

The card is covered by a single-piece backplate. It lacks openings for ventilation, but is connected to the hot spots via thermal tape. Plan to accommodate an additional one-fifth of an inch (5mm) in depth beyond the plate, which may become relevant in multi-GPU configurations.

While it is perfectly possible to use this card without its backplate, removing it requires disassembling the whole cooler and likely voiding Gigabyte's warranty.

The top of the card is dominated by a centered Gigabyte label, glowing in bright colors, and an LED indicator for the card's silent mode. The eight-pin power connector is rotated 180° and positioned at the end of the card. The design may be a matter of taste, but we don't think it's anything to write home about.

At its end, the card is completely closed off. This makes sense, as the fins are positioned vertically and won't allow any airflow down there anyway.

The rear bracket features five outputs, of which a maximum of four can be used simultaneously in a multi-monitor setup. In addition to one dual-link DVI-D connector (be aware that there is no analog signal), the bracket also exposes one HDMI 2.0b and three DisplayPort 1.4-ready outputs. The rest of the plate is mostly solid, with several openings cut into it that look like they're supposed to improve airflow, but don't actually do anything.

Board & Components

Gigabyte's Xtreme Gaming card, dissected on the next page, employs a different layout than the G1 Gaming, which has just enough space for one eight-pin power connector.

The card uses GDDR5X memory modules from Micron, which are sold along with Nvidia's GPU to board partners. Eight memory chips (MT58K256M32JA-100) transferring at 10 MT/s are attached to a 256-bit interface, allowing for a theoretical bandwidth of 320 GB/s.

The 8+2-phase power supply relies on the sparsely documented µP9511P as a PWM controller, just like Nvidia's reference cards. Since this component can't communicate directly with the MOSFETs on the VRM phases, Gigabyte utilizes separate PWM drivers (gate drivers) to talk to a total of three single-channel MOSFETs per phase. We were pleasantly surprised by this rather elaborate solution.

Not using more cost-effective dual-channel MOSFETs (and thereby sacrificing space on the card) may have been a deliberate choice to improve efficiency and dissipate heat more effectively, especially since the third line of MOSFETs is positioned on the back of the board and may benefit from cooling through the board's backplate. This is a unique feature in the field of 1080s we're reviewing.

As seen previously, two capacitors are installed right below the GPU to absorb and equalize voltage peaks. The board layout seems pretty dense, but it's organized well.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which show such striking similarity that we deliberately put their graphs one after the other. Notice how the two curves don't drop as much in the face of rising temperatures compared to Nvidia's reference design.

Where the GPU Boost frequency falls to 1873 MHz under load (and even further during our stress test), this movement is mirrored by our voltage measurements. We recorded up to 1.062V in the beginning (similar to the Founders Edition model), and the value drops below 0.962V later on (still above the readings collected for our Founders Edition sample).

Summing up measured voltages and currents, we then arrive at a total consumption figure we can easily confirm with our test equipment by monitoring the card's power connectors.

As a result of Nvidia's restrictions, manufacturers sacrifice the lowest possible frequency bin in order to gain an extra GPU Boost step. So, Gigabyte's power consumption is disproportionately high when idle. In all fairness, the company manages this behavior relatively well compared to some of its competition.

Our complete power measurements are as follows:

Now let's take a more detailed look at power consumption when the card is idle, when it's gaming at 4K, and during our stress test. The graphs show the distribution of load between each voltage and supply rail, providing a bird's eye view of variations and peaks:

Temperature Results

Naturally, heat output is directly related to power consumption, and the 1080 G1 Gaming's ability to dissipate that thermal energy can only be understood by looking at its cooling solution.

Gigabyte relies on its aging Windforce design. In this case, the copper heat sink gives way to a simpler (and more affordable) direct heat touch solution that facilitates contact between the cooler's three partially flattened pipes made of sintered composite material and the GPU.

The rest of the plate is meant to cool the memory and (commendably) the voltage regulators. Added thermal pads ensure good contact.

The slightly better performance of this cooler compared to Nvidia's Founders Edition card is also reflected by the temperature graph, as a limit of 158°F (70°C) is not, or hardly (162°F/72°C in a closed case) exceeded.

Unfortunately, upon a closer look at the infrared images, it becomes clear that this card does have troublesome areas. While the voltage regulation circuitry stays relatively cool at 180°F (82°C), two smaller SMD capacitors glow at an unpleasant 189°F (87°C).

During our torture loop, those capacitors (R990 and R991) even reach temperatures of 212°F (100°C), despite a marginal increase in heat measured at the voltage regulators.

To test the two shunts more accurately, we change the distance to just 2.75in (7cm) and swap the IR camera's lens, since the areas we need to measure are just 1mm². A look at the specs and an email to Gigabyte quickly confirmed that our findings shouldn't cause any problems.

Even after a FurMark run, the measured temperature proves acceptable if the specifications are to be trusted. Of course, that's not taking into account the PWM controller residing right above the two capacitors on the board's top side. It generates a fair amount of heat as well, resulting in two hot spots right on top of one another.

Our initial testing showed that the severity of this problem could easily be mitigated with an additional thermal pad, so Gigabyte implemented our suggestion on its retail cards.

With the backplate mounted (sporting thermal pads between the RAM modules and VRM), the "outside" temperatures appear as follows:

Sound Results

Since power consumption during our gaming and stress tests is similar, it comes as little surprise that the fan curves are also a close match. The behavior during start-up and an overall clean profile reliably prevents multiple on/off cycles once the fans spin up. In addition, the start-up speed is chosen in such a way that the fans should continue to start reliably even as they age. The same goes for turning the fans off once the card cools down.

When the card is idle, its noise is not measurable thanks to a semi-passive mode that keeps the fans from spinning. However, running at full load for a while easily raises the noise level to 40 dB(A). It gets even louder during our torture test.

To be frank, we're not impressed with the card's acoustics. They can be improved by manually adjusting the fan curve and increasing the temperature target to 167°F (75°C). Doing so reduces the maximum GPU Boost frequency though, and you may start to hear an audible chirp coming from the coils.

Considering what other manufacturers achieve using aggressive coolers, the G1 Gaming's solution almost feels unambitious. It's missing the tender loving care competing vendors put into their own 1080-based cards. Though the G1 Gaming isn't bad in this regard, the end result isn't particularly good either.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

Gigabyte GTX 1080 Xtreme Gaming

It took a while for us to get our hands on Gigabyte's GeForce GTX 1080 Xtreme Gaming, but without giving too much away, the wait was worthwhile.

Gigabyte advertises this card's extended feature set, so you know we have to pay particular attention to its value-adds. Upon opening the over-sized box, we discover loads of useful extras.

While wrist protection and a new mouse pad are nice to have (even if they're more Gigabyte swag than essentials), the VR bracket does deserve special mention. That's because the GeForce GTX 1080 Xtreme Gaming comes with a front panel bracket that exposes USB 3.0 I/O and HDMI outputs for easy-to-reach connectivity.

Of course we'll cover the stacked fans and heat pipes, but let's start with a word on the 1080 Xtreme Gaming's overall performance and how we tested it.

We try to benchmark every card as it ships, right out of the box. That means we typically avoid installing vendor-specific software and intentionally use default firmware settings. Many utilities are only available for Windows, and direct comparisons are complicated by the sheer number of options available for each card. Nevertheless, in this case we're including measurements for Gigabyte's optional OC mode.

The company's four-year warranty is exemplary, even if it requires registering the 1080 Xtreme Gaming online.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

The cooler cover is made out of a light alloy, with silver highlights on dark anthracite. Diagonal LED-lit struts cross above the center fan. You can control the lighting using Gigabyte's Xtreme Engine software. Because the three fans are stacked above each other in two planes, the 47oz (1330g) 1080 Xtreme Gaming is still significantly shorter than many of its competitors.

Measuring approximately 11 inches (28cm) long, five inches (13cm) tall, and almost two inches (5cm) wide, excluding the backplate, this card is anything but compact. But it's still small enough to squeeze into cases only able to accommodate 12-inch expansion cards.

Gigabyte covers the 1080 Xtreme Gaming's rear with a single-piece backplate that does nothing for cooling performance, but also doesn't hurt thanks to openings for ventilation and sufficient spacing.

Plan to accommodate an additional one-fifth of an inch (5mm) in depth beyond the plate, which may become relevant in multi-GPU configurations, particularly if your motherboard's PCIe slots are two spaces apart. Of course, it's possible to use the card without its backplate, though removing it also requires disassembling the cooler. That'll likely void Gigabyte's warranty.

The top of the card is dominated by a centered Gigabyte logo and LED indicator for the card's silent mode. Again, the color and lighting effect is controlled via software. Two eight-pin power connectors are turned 180° and positioned at the end of the card. As with so many board partner designs, Gigabyte's style is a matter of taste.

At its end the card is completely closed off, which makes sense, since the fins are positioned vertically and don't allow any airflow through the front or back anyway. The output bracket sports the same five display connectors as Nvidia's reference design, and four of them can be used simultaneously in a multi-monitor configuration.

In addition to one dual-link DVI-D connector (be aware that there is no analog signal), the bracket also exposes one HDMI 2.0b and three DisplayPort 1.4-ready outputs. The rest of the plate is mostly solid, with several openings cut into it that look like they're supposed to improve airflow, but don't actually do anything.

There's a real surprise on the other end, though. Gigabyte installs two additional HDMI outputs back there, which it means for you to connect to the front-panel accessory mentioned earlier. The company even gives you cables to make the hook-up. Alternatively, you can unscrew the front-panel jacks and install them into an included expansion bracket for extra HDMI connectivity on the back of your PC.

This type of connection sacrifices two of the rear DisplayPort outputs to enable what Gigabyte calls its Xtreme VR Link. As an option for owners of VR HMDs, we think the accessory is ingeniously simple. It addresses the hassle of hooking up HDMI and USB ports to the back of your PC with much more accessible I/O.

Board & Components

The card uses GDDR5X memory modules from Micron, which are sold along with Nvidia's GPU to board partners. Eight memory chips (MT58K256M32JA-100) transferring at 10 MT/s are attached to a 256-bit interface, allowing for a theoretical bandwidth of 320 GB/s.

Even if Gigabyte advertises 12+2 power phases, the µP9511P PWM controller, manufactured by uPI Semiconductor, is nothing but a purebred 6+2-phase model. The six GPU phases are doubled by dividing each one via separate converter circuits. First, this draws hot-spots further apart, and second, it halves each converter's current.

Gigabyte uses a total of 12 Fairchild FDMF6823Cs, which are highly integrated DrMOS modules that combine a driver IC, two power MOSFETs, and a bootstrap Schottky diode into a thermally enhanced, 6x6mm package. The memory is controlled separately through a uP1666 and utilizes two separate phases, each of which connected to a Fairchild FDMD3604AS with PowerTrench technology. The coils are well-known magic chokes produced by Lenovo; they're more or less middle-of-the-road, quality-wise.

Two capacitors are installed right below the GPU to absorb and equalize peaks in voltage. The board layout seems pretty loaded, but it's well-organized and reflects careful component placement.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage. We also need to revisit the subject of software configuration since Gigabyte's default settings are quite conservative and really don't do the card justice. With a little bit of manual work in the company's bundled utility, it's possible to keep the 1080 Xtreme Gaming above a 2 GHz GPU Boost frequency (or 2.1 GHz+, in our case).

Full disclosure: to achieve the settings shown in the screen shot below, we enlisted the help of a hose piping cold air from an air conditioning unit. As you might imagine, this card might be a good candidate for liquid cooling, too.

In Gaming mode, the GPU frequency remains relatively stable, even through the challenging Metro Last Light loop at 3840x2160. Setting the software to OC mode should effortlessly push it past 2 GHz; after all, it's already almost there in Gaming mode.

After warm-up, the GPU Boost frequency falls to 1987 MHz at times under load. During our stress test, it drops to 1800 MHz on average. The corresponding voltage readings fall to 1.05 and 0.9V.

Summing up measured voltages and currents, we then arrive at a total consumption figure we can easily confirm with our test equipment by monitoring the card's power connectors.

As a result of Nvidia's restrictions, manufacturers sacrifice the lowest possible frequency bin in order to gain an extra GPU Boost step. So, Gigabyte's power consumption is disproportionately high when idle. In all fairness, the company manages this behavior relatively well compared to some of its competition. Its lowest clock rate hovers at 291 MHz.

During our manual overclock, we increased the voltage by a maximum of 100mV and set the power target to 150 percent. We hit the ceiling at exactly 230W, and the card couldn't be provoked into going any higher.

Our complete power measurements are as follows:

Now let's take a more detailed look at power consumption when the card is idle, when it's gaming at 4K, and during our stress test. The graphs show the distribution of load between each voltage and supply rail, providing a bird's eye view of variations and peaks:

Temperature Results

Building a thermal solution with three 10cm fans onto a card that measures less than 30cm long required some clever work on Gigabyte's part. The company calls this a stack setup, and it requires overlapping the fans somewhat; the center fan is actually mounted lower than the other two and given unique blade geometry.

While fans mounted side by side may leave dead zones without ample airflow, a stacked setup purportedly mixes air more thoroughly. We're told this should be quantifiable in our temperature measurements, particularly when it comes time to scan for hot-spots.

We strive to avoid parroting marketing material, but the following diagram from Gigabyte does do a good job explaining what's going on.

The massive copper plate is meant to cool both the GPU and memory. A VRM heat sink is placed a little farther back, integrated directly into the cooler. This makes a lot of sense; the hot MOSFETs don't heat the memory modules through the copper plate thanks to a bit of strategic separation.

The backplate is attached with four screws and it's completely aesthetic. At best, it may help with the card's structural integrity.

The thermal solution and its fan curve deliver downright chilly temperatures using Gigabyte's default settings. We measured 65°C (67°C inside a chassis) under the influence of fast-spinning fans.

A peak value of 81°C on some spots below the MOSFETs is quite commendable. To be sure, Gigabyte's card stays cooler than much of its competition, even after long periods of heavy use.

Even during our stress test, the results don't change much. If any part of this card's design deserves special recognition, it's the broadly distributed DrMOS and efficient VRM sink, which dissipates waste heat right into the pipes running over it.

Sound Results

Gigabyte knows how to nail exemplary hysteresis. The activation of its fans is short and crisp, and they continue to run at an almost inaudible 670 RPM after that.

In that light, the high default maximum fan speed is difficult to understand. While it does ensure relatively low temperatures, it also has a negative (though not necessarily dramatic) impact on the card's noise levels.

When the card is idle, its semi-passive mode ensures silence. Naturally, there's no reason to take measurements in that state.

Running under full load raises the noise level to almost 36 dB(A), which is an acceptable value. It just doesn't exploit the card's full potential. Install Gigabyte's software, though, and you can complement the various overclocking modes with predefined fan profiles.

Setting the card to "silent mode" drops its noise output to just over 34 dB(A), which you might call excellent given the sound's unobtrusive character, even at higher speeds.

Unlike the fans used by Galax, Zotac, and Palit, Gigabyte's show no narrowband emissions in the low-frequency range that might be transferred to the housing and cause resonance.

It'll probably remain Gigabyte's secret why the 1080 Xtreme Gaming employs such an aggressive profile by default, only to hit record-low temperatures. We can say that configuration isn't ideal, except for those who want to take full advantage of the company's maximum power target and higher voltge. Otherwise you almost need to use the Xtreme Engine software to get this card's behavior tuned optimally.

MORE: Best Deals

MORE: Hot Bargains @PurchDeals

MSI GTX 1080 Gaming X 8G

The GeForce GTX 1080 Gaming X 8G is a somewhat conservative offering from MSI. This isn't a disadvantage per se, since the company combines reliable factory overclocking with a solid list of specs that doesn't break the bank. Besides, MSI has its own premium Lightning model for enthusiasts who demand a more sophisticated graphics card. Then, at the other end of the spectrum, there are simpler/cheaper Armor 8G and Aero 8G cards. MSI's GTX 1080 Gaming X fits somewhere in the middle.

We're testing a press sample with an MSI-supplied firmware. In this version, the OC mode is active by default, resulting in approximately 20 MHz higher base and GPU Boost frequencies. Since differences between the retail model and our press version are negligible in practice, we decided not to re-flash the BIOS and instead switched the card back to standard (rather than OC) mode using MSI's bundled software. In the end, our card's clock rates match the retail version.

Technical Specifications

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Exterior & Interfaces

As with most third-party GeForce GTX 1080s, the Gaming X's shroud is made of relatively thin plastic. Under the hood, though, a massive frame provides better stability than we've seen elsewhere.

Weighing in at 38.8oz (1101g), the Gaming X 8G isn't exactly light. It measures 11.2in (28.4cm) long, five and one-third inches (13.5cm) high, one and three-eighths inches (3.5cm) wide, and occupies two full slots. The rotor blades of its double ball-bearing fans have a diameter of three and three-fourths inches (9.5cm), which explains the card's towering height.

The back of the board is covered by a single-piece plate sporting holes for ventilation. But at no point is it connected to the PCB via thermal pads, and thus it does nothing to aid cooling. Plan for an additional one-fifth of an inch (5mm) in depth beyond the plate, which may become relevant in multi-GPU configurations.

Using the card without its backplate is difficult due to well-integrated spacers. Furthermore, the plate is attached with screws that are too short to secure the frame and VRM cooler once the backplate is gone. As a result, we had to do our testing with this piece of metal in place, notably limiting our view during the IR testing.

The top of the Gaming X is dominated by an illuminated MSI logo and three visible heat pipes (two 6mm pipes and one that measures 8mm in diameter). One eight- and one six-pin power connector are positioned at the end of the card and rotated by 180°. The choice of colors adheres to MSI's black and bright red corporate design, while the Gaming X's shape matches previous-gen models.

The end of the card is completely open, which is a happy coincidence since the cooler's fins run horizontally and thus channel air toward the back and output bracket. We consider this an advantage for a couple of reasons. First, a significant amount of hot air exhausts out of the case before it has a chance to heat anything up. Second, the air that remains can more easily be directed away from the CPU and its cooler.

The rear bracket features five outputs, of which a maximum of four can be used simultaneously in a multi-monitor setup. In addition to one dual-link DVI-D connector (be aware that there is no analog signal), the bracket also exposes one HDMI 2.0b and three DisplayPort 1.4-ready outputs. Openings for airflow dot the rest of the bracket. If the often-unused DVI port had been left off and replaced by a bundled adapter, the card would have had more space for ventilation to improve cooling efficiency.

Board & Components

The 1080 Gaming X uses GDDR5X memory modules from Micron, which are sold along with Nvidia's GPU to board partners. Eight memory chips (MT58K256M32JA-100) transferring at 10 MT/s are attached to a 256-bit interface, allowing for a theoretical bandwidth of 320 GB/s.

Similar to Nvidia's reference design, the 8+2-phase power system relies on a µP9511P as its PWM controller. Unlike the Founders Edition board, however, this controller resides on the front of MSI's card rather than its back.

The PWM controller can't communicate with the VRM MOSFETs directly, so MSI utilizes PWM drivers (gate drivers) to talk to them. By using dual-, rather than single-channel MOSFETs, MSI can lay out its PCB more efficiently. And for coils, MSI goes with SFCs (Super Ferrite Chokes), which are a bit classier than the usual suspects we're accustomed to finding.

In addition, two capacitors are installed right below the GPU to absorb and equalize voltage peaks, just like Nvidia's reference implementation.

Power Results

Before we look at power consumption, we should talk about the correlation between GPU Boost frequency and core voltage, which are so similar that we decided to put their graphs one on top of the other. We'll also see that clock rate doesn't drop significantly during gaming, even after the Gaming X heats up.

After warm-up, the GPU Boost frequency falls to 1936 MHz at times. This behavior is mirrored by our voltage measurements. Although we observe up to 1.062V in the beginning, voltage later drops to 1.05V. The graph itself remains mostly flat though, just as we saw when we recorded frequency over time.

Summing up measured voltages and currents, we arrive at a total consumption figure we can easily confirm with our test equipment by monitoring the card's power connectors.

As a result of Nvidia's restrictions, manufacturers sacrifice the lowest possible frequency bin in order to gain an extra GPU Boost step. So, the GTX 1080 Gaming X's power consumption is disproportionately high as it idles at 253 MHz. The consequences of this are apparent in the following table:

Now let's take a more detailed look at power consumption when the card is idle, when it's gaming at 4K, and during our stress test. The graphs show the distribution of load between each voltage and supply rail, providing a bird's eye view of variations and peaks:

Temperature Results

MSI put its focus on the GeForce GTX 1080 Gaming X 8G's cooling system, and it shines as a result. A solid frame covers almost the entire board (or at least the part not already covered by the VRM cooler, which is implemented as a big plate with fins).