A Redditor thanked his father for gifting him something that he plucked out of the trash. Alternative-Run363 claims that his dad enjoyed a big ‘W’ by fishing out a PC packing an Intel Core i5-9400F, an Nvidia GeForce GTX 1660, and more. That’s a ‘W’ indeed, since it all seems to work, and Windows 11 is installed and activated. It's certainly better than the kipper bones, tin cans, or old boots that Thomas Jasper Cat Sr. usually found in the trash.

In this latest case of finding computer treasure in the trash, we don’t get much of a backstory regarding the find. However, it certainly looks like the PC system in the picture has seen better days. You can clearly see dust and grime has built up in this machine over the years. From the single picture, it is difficult to determine if there are dust filters missing or if they were just unmaintained. According to the social media posting, there's a glass side panel that was removed for the photo.

So, let’s take a closer look at what Alternative-Run363 managed to snag for the grand total of zero dollars and zero cents:

• CPU: Intel Core i5 9400F
• CPU Cooler: Asetek Shark Gaming AiO
• GPU: Asus Phoenix Nvidia GeForce GTX 1660*
• Motherboard: Asus TUF B360-Plus Gaming motherboard*
• RAM: 16GB dual-channel
• Storage: 500GB Kingston A2000 M.2 NVMe SSD
• Case: InWin chassis with at least two system fans
• PSU: Unknown, unknown condition
• OS: Windows 11 was installed by Alternative-Run363, and they found out it was activated
• (*We guessed these precise component models after some pixel peeping)

Nvidia has reportedly decided to phase out some of the last remnants of the Turing graphics card era, according to rumors shared on China's Bobantang tech forums. This means it has decided to "completely discontinue" the GeForce RTX 2060, the 2060 Super, as well as the GTX 1660 and 1660 Super. Removing these popular options will leave quite a void in Nvidia's budget to midrange offerings, a high volume segment.

This comes as we're seeing a few final gasps of life in the way of Turing-based graphics card deals. Amazon has the RTX 2060 for $179 right now, an all-time low. You can also find the Zotac GTX 1660 for $179 and the Zotac GTX 1660 Super for $199... but why would you buy those slower cards for the same or higher prices?

Earlier this month we reported about whispers from supply chain sources, stating that the GeForce RTX 2060 and RTX 2060 Super were about to be discontinued. Now that appears to have happened, if the latest rumors are to be believed.

All the above mentioned Nvidia Turing graphics cards remain available at US-based online retailers, but that could change quickly once the situation filters through to retailers looking to replenish stocks.

On the topic of stocks and inventories, the source also indicates that there remains a "small amount of inventory," of the GTX 1660/Super TU116 GPUs. However, Nvidia hasn't wielded the knife on the GTX1650 and 1630 TU117 GPUs, notes the forum poster. These continue to be supplied to graphics card makers.

If you check out our GPU hierarchy feature, you will see that all Nvidia has left to address a significant and popular consumer segment, in the wake of the alleged Turing cull, is the Ampere generation GeForce RTX 3050. Sadly for the RTX 3050, it lived in the value and performance shadow of the RTX 2060 cards and, even with this removed, it is roundly beaten by AMD's Radeon RX 6600 / XT cards.

Recently we wrote about how AMD's graphics cards offer the best value across the full gamut of performance segments, except for the very top halo graphics cards. Some Cyber Monday price probing reveals that the best RTX 3050 is still priced on a par with the RX 6650 XT (both in the $250–$270 range), but is completely crushed by the RX 6600 which can be found for $209.

There are signs of Nvidia shuffling around some GeForce RTX 30-series SKUs with the new RTX 3060 8GB and the new RTX 3060 Ti with GDDR6X. However, with the phasing out of so many midrange Turning cards, Nvidia needs to widen the appeal of the RTX 3050. Perhaps it can make a desktop RTX 3050 Ti and reduce the price of the current RTX 3050 to better suit its performance level.

Nvidia may be priming the supply chain ahead of some new Ampere choices for the rather neglected entry-level to midrange market segment. This appears to be logical, but please add salt to the Bobantang sourced news about it happening right now. We look forward to some compelling new options from Nvidia, in this supposedly vacated space, to fight back against AMD and the new competition from Intel.

In a shocking move, MSI Russia has has reduced the warranty period for 18 graphics cards, as reported by PC Gamer. MSI typically offers a 36-month warranty on graphics cards; however, the brand will only back the affected models for six months now.

MSI Russia shared the news on the company's warranty webpage but didn't share its exact reasoning. The list includes AMD Polaris and Navi cards, as well as Nvidia's Turing and latest Ampere products representing some of the best graphics cards for gaming. 

*Data from Minerstat

MSI Russia didn't point to cryptocurrency mining, but it still comes to mind as a potential driver for the drop in warranty, especially considering the first six models are obviously tailored towards cryptocurrency miners, such as the famed Radeon RX 580 Miner. 

It's a bit surprising to see Ampere cards on MSI's list, since the Ampere SKUs are GeForce gaming graphics cards. But cryptocurrency miners have repurposed Ampere, such as the GeForce RTX 3060 for mining activities. 

Mining cryptocurrency takes a heavy toll on graphics cards, since mining farms operate 24/7. Graphics cards used for mining are more likely to suffer a premature death, which is particularly problematic given today's graphics card shortage. Graphics card vendors aren't eager to replace these cards, and, in fact, many vendors explicitly state that cryptocurrency mining isn't covered under their warranty.

Nvidia's answer to crytocurrency mining was to introduce the Cryptocurrency Mining Processor (CMP) product line. However, CMP options aren't very appealing to casual miners, considering their price tag and warranty. Although Nvidia never acknowledged it, many CMP devices are using rewarmed Turing silicon. For example, the CMP 30HX performs similarly to the GeForce GTX 1660 Super in mining Ethereum. We've seen the CMP 30HX on the market with a price tag as high as $723 with warranty periods up to three months. It's easy to see why miners would avoid the CMP lineup since the GeForce counterparts are often cheaper and feature longer warranties.

If the rumor about Nvidia giving its Ampere silicon an overhaul is accurate, graphics card vendors won't have to resort to shortening warranty periods to protect their GeForce products from mining. The chipmaker is reportedly reinforcing its anti-mining algorithm on the new Ampere dies, so that mining performance is gimped straight out of the box.

This shouldn't affect shoppers outside of Russia,  but we've reached out to MSI to see if the company can shed any more light and if we could see similar action in other countries.

Nvidia announced its Cryptocurrency Mining Processor (CMP) just three weeks ago. Today, VideoCardz shared what appear to be renders of the first custom CMP 30HX, which originates from Gigabyte's camp.

Although Nvidia hasn't admitted it, there is enough evidence to suggest that most of its CMP graphics cards are based on the Turing arechitecture. The chipmaker advertises the 30HX with an Ethereum hash rate of up to 26 MH/s, and the graphics card comes with 6GB of memory, a 125W TDP, and requires only one 8-pin PCIe power connector. The 30HX's performance and partial specifications basically describe Nvidia's GeForce GTX 1660 Super, and the latest Gigabyte CMP 30HX helps confirm the early suspicions.

It's not hard to see that the Gigabyte CMP 30HX is a close copy of the brand's GeForce GTX 1660 Super OC 6G. The only apparent difference is that Gigabyte removed the display outputs. The graphics card uses the exact WindForce 2X cooler with the same pair of 90mm semi-passive cooling fans as its gaming counterpart. It also retains the single 8-pin PCIe power connector.

The CMP 30HX is in all likelihood based on the TU116 silicon. The jury is still out on whether the 30HX will come with the same 1,408 CUDA cores as the GeForce GTX 1660 Super. Mining Ethereum requires a certain level of compute performance after which memory bandwidth becomes the main factor.

According to VideoCardz's report, Nvidia's AIBs have already started mass-producing the CMP 30HX and 40HX. The 30HX and 40HX are scheduled to debut in the first quarter of this year, so these cryptomining devices are right around the corner. However, Nvidia left something very important out of its CMP announcement: the pricing.

With the existing graphics card shortage, pricing is crazy right now. For reference, the GeForce GTX 1660 Super launched two years ago for $229. Presently, custom GeForce GTX 1660 Super models start at $599 and scale up to $899. For the CMP 30HX to succeed, Nvidia will have to price the graphics card very attractively. If not, cryptocurrency miners would just stick to the GeForce GTX 1660 Super since it offers the same level of Ethereum as the CMP 30HX. The GeForce RTX 3060 was also a very convincing graphics for mining Ethereum too before Nvidia's nerf.

MSI and other card partners typically have several different SKUs of the same card. These range from reference models with branded stickers to high-end models with improved power delivery and custom cooling. The premium models typically offer some kind of improvement over the reference spec, be it performance, cooling, acoustics or all of the above. In that light, MSI’s GTX 1660 Super Gaming X card sits at the top of the stack for MSI's GTX 1660 Super options, but how does it rank among the best graphics cards in its class?  

The card comes with a factory overclock, more robust power delivery, as well as the Twin Frozr 7 thermal design to keep things cool. Pricing on this card is $269.99 at Newegg and is close to the high end of GTX 1660 Supers, which currently start at at $230. But with AMD’s new Radeon RX 5600 XT now delivering 2060-class performance while occupying the $279-plus price band, the $250-$300 market becomes quite a busy, complicated space.

As with most mid-range and lower-end cards, Nvidia offered reference specifications but not an actual reference/Founder’s Edition card for the 1660 Super. So it's up to the board partners to do as they wish (within reason of course) with clocks speeds, power delivery and cooling solutions. With this Gaming X, power is delivered via a single 8-pin connector, and is full size (it's 9.7 inches long) but should fit into most builds.

Features

The GTX 1660 Super is based on the Turing architecture and uses the TU116 GPU found in the GTX 1660 and 1660 Ti. The GTX naming, as opposed to RTX, tells us there is no support for ray tracing. The TU116 die is manufactured using the 12nm FFN (FinFET Nvidia) process and made by TSMC. The GPU consists of 6.6 billion transistors with a die size of 284 mm squared. The GTX 1660 Super has 1,408 shaders, 88 TMUs, and 48 ROPs, which is trimmed down from the full version (1,536/96/48) found in the GTX 1660 Ti.

The GTX 1660 Super uses 6GB of GDDR6, as does the 1660 Ti (the 1660 uses GDDR5). The memory uses a 192-bit bus with six 32-bit memory controllers. The 6GB capacity is a sweet spot for 1080p gaming. While 4GB capacity is capable, there are many titles out already (this number grows monthly) that when using 1080p ‘ultra’ settings, eclipses that mark, resulting in negative results as the buffer gets dumped to slower off-card memory. We’ve seen what this can do to the RX 5500 XT in other reviews (even when it is using the faster PCIe 4.0 x16 slot).

The MSI GTX 1660 Super Gaming X we're looking at here comes with a 1,530 MHz base clock and 1,750 MHz (14 Gbps GDDR6 effective) memory clock. The boost clock is listed as 1,830 MHz, 45 MHz above the reference specification. TDP on the reference card is listed as 126W, so this model may use a bit more. MSI suggests a 450W or larger power supply.

The TU116 GPU core includes the Turing NVENC encoder, rather than the Volta encoder included in the original GTX 1650 and the TU117 chip. NVENC encoding on the Turing-based cards decreases CPU utilization and improves performance, especially with higher-resolution encoding and decoding.

Below is a complete list of specifications for the budget range of Nvidia GPUs.

Design

The MSI GTX 1660 Super Gaming X measures in at 9.7 x 5 x 1.6-inches, which makes this a dual-slot card. Since it is a true two-slot solution, you will be able to use the slot below so long as whatever is plugged in there doesn’t protrude above the slot line, as space in between will be tight. The card is by all accounts full-size, running almost the full width of an ATX motherboard with its height just barely sticking up above PCIe expansion slot height. But it's compact enough that it should fit in most non-SFF systems. As always though, be sure to check your chassis specifications before buying.

The card’s Twin Frozr 7 cooler sports a black shroud around the two large fans. The plastic cover uses sharp angles, along with a chevron-like pattern between them. Also located on the front are four tastefully implemented RGB LEDs that sit above and below the two fans. In addition, the top of the card also uses RGB lighting, which illuminates the MSI Twin Frozr 7 writing along with the gaming dragon. Protecting the back of the card is a simple backplate.

The card has a fair amount of style and RGB glow, but it won’t stick out like a spotlight. Its black-and-grey design should blend in with most build themes.

The Twin Frozr 7 cooler was able to keep this card cooler than EVGA’s solution and its beefier 2+ slot cooler, all while being whisper quiet throughout our testing.

MSI’s Twin Frozr 7 sports dual 90mm fans to get air across the heatsink. The Torx fan 3.0 uses a dispersion type fan blade that is specially curved,  which MSI says accelerates airflow and increases its effectiveness. MSI implements its Zero Frozr technology here, so that the fans remain off on idle and low loads. This allows for complete silence on the desktop or light activity.

The heatsink below the fans is large, with a relatively dense fin array and runs the entire length -- and most of the height -- of the card. It uses a nickel-plated copper base plate to mate to the die to the heatsink. Just above the base plate are three s-shaped heat pipes, each making two passes through the fin stack. The Memory and VRMs are cooled by a simple black baseplate.

Taking a closer look at the power delivery, the Gaming X uses a 4+2 configuration, with four phases for the GPU and two for the memory. The GPU voltage is controlled by an OnSemi NCP81610 controller with a total of eight channels -- no doublers are in use. The MOSFETs are also OnSemi brand NCP302045 DrMOS. Delivering power to the capable (but not overbuilt) VRMs is a single 8-pin PCIe connector. Combined with the PCIe slot, a total of 225W of in-spec power will be delivered.

Connectivity on this card consists of four total ports: three v1.4a DisplayPorts, which supports lossless DSC and 8k@30 Hz using a single cable or 8k@60 Hz with DSC enabled. In addition to that is a single HDMI (2.0b) port.

How We Tested MSI’s GTX 1660 Super Gaming X

Recently, we’ve updated our GPU test system to a new platform and swapped from a six-core i7-8086K to an eight-core Core i9-9900K. The CPU sits in an MSI Z390 MEG Ace motherboard, along with 2x16GB Corsair DDR4 3200 MHz CL16 RAM (CMK32GX4M2B3200C16). Keeping the CPU cool is a Corsair H150i Pro RGB AIO along with a 120mm Sharkoon fan for general airflow across the test system. Storing our OS and gaming suite is a single 2TB Kingston KC2000 NVMe PCIe 3.0 x4 drive.

The motherboard was updated to the latest (at this time) BIOS (version 7B12v16) from August 2019. Optimized defaults were used to set up the system. We then enabled the memory’s XMP profile to get the memory running at the rated 3200 MHz CL16 specification. No other changes or performance enhancements were enabled. The latest version of Windows 10 (1909) was used and is fully updated as of December 2019.

As time goes on we continue to build our database of results back up based on this test system. For now, we will include GPUs that are close in performance to the card that is being reviewed. Here we have a Zotac GTX 1660 Amp, EVGA GTX 1660 Super SC Ultra, and the EVGA GTX 1660 Ti XC. Representing AMD, in this case, is an XFX RX 590 Fat Boy, and Gigabyte RX 5500 XT Gaming OC 8G.

Our list of test games is currently Tom Clancy’s The Division 2, Ghost Recon: Breakpoint, Borderlands 3, Gears of War 5, Strange Brigade, Shadow of The Tomb Raider, Far Cry 5, Metro: Exodus, Final Fantasy XIV: Shadowbringers, Forza Horizon 4 and Battlefield V. These titles represent a broad spectrum of genres and APIs which gives us a good idea of the relative performance difference between the cards. We’re using driver build 441.20 for the Nvidia cards and Adrenalin 2020 Edition 20.1.1 for AMD cards.

We capture our frames per second (fps) and frame time information by running OCAT during our benchmarks. In order to capture clock and fan speed, temperature, and power, GPUz's logging capabilities are used. Soon we’ll resume using the Powenetics-based system used in previous reviews as soon as the equipment is ready.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

While we do not have the AMD RX 5600 XT results listed below, we know from our review it wipes the floor with 1660 Ti while priced at $279 or $289 (SEP/SRP). This poses a problem for such high priced 1660 Supers, like this MSI card, which is just $10-$20 cheaper but notably slower.

For this class of cards, testing at 1080p with ultra settings is going to be the performance sweet spot. Using the aforementioned canned settings, these cards will reach 60 fps on average for most titles. The 6GB is adequate vRAM for 1080p Ultra as well as running at 2560 x 1440 using medium settings.

At 1080p with Ultra settings, our MSI GTX 1660 Super Gaming X ranged from 43.4 fps (on Metro: Exodus, of course) up to 121.8 fps in Forza Horizon 4. Of our 11 titles, eight of them were above 60 FPS with only Metro, Ghost Recon: Breakpoint (56.8 fps) and Borderlands 3  (54.6 fps) falling below that mark. In short, this is a plenty capable 1080p ultra card.

If we compare this to the EVGA GTX 1660 Super SC Ultra, we find a virtual tie between them. The MSI averaged .2 fps more than the slightly lower clocked EVGA. Between them, they split ‘winning’ in games 5/5 with one tie. And the ‘wins’ weren’t by much, with the largest gap about 3 fps. When looking at the EVGA GTX 1660 Ti XC, it is the fastest card between all three, but not by much at all, with1 fps or around 1.5%.

The Division 2

Ghost Recon: Breakpoint

Borderlands 3

Gears of War 5

Strange Brigade

Shadow of the Tomb Raider

Far Cry 5

Metro Exodus

Final Fantasy XIV: Shadowbringers

Forza Horizon 4

Battlefield V

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Going up in resolution to 2560 x 1440 and down to medium settings, we find the MSI GTX 1660 Super Gaming X is again more than capable of running at these settings. Frames per second ranged from 53.3 (Metro again) up to 119.6 in Final Fantasy XIV. Outside of Metro, all titles run at 60 fps or more (yes, I am counting Ghost Recon: Breakpoint as 60 fps).

When comparing the EVGA GTX 1660 Super SC Ultra, we again see a virtual tie with the MSI card averaging 81.8 fps across all titles and the EVGA 81.5 fps. The MSI is negligibly faster in seven of 11 titles, with the largest difference still less than 5 fps in Far Cry 5. Adding the EVGA GTX 1660 Ti XC in, we find it is faster overall, by around 1%--not much to speak of. The 1660 Super cards really rendered the 1660 Ti obsolete, at least these higher clocked models.

As above, when comparing the AMD 5500 XT, it loses out by quite a bit, though it is nearly $100 cheaper. The 5600 XT, on the other hand, is about 23% faster and costs $10-$20 more. Ouch. In today’s market at current pricing, the GTX 1660 Super can be a tough sell unless you are looking at a lower-tier SKU .

The Division 2

Ghost Recon: Breakpoint

Borderlands 3

Gears of War 5

Strange Brigade

Shadow of the Tomb Raider

Far Cry 5

Metro: Exodus

Final Fantasy XIV: Shadowbringers

Forza Horizon 4

Battlefield V

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Power Consumption

We used GPUz’s logging to measure the card’s power consumption with the Metro: Exodus benchmark running at 2560 x 1440 using the default Ultra settings. The card is warmed up prior to testing and benchmarking is started when it settles to a normalized idle temperature (after about 10 minutes). The benchmark is looped a total of five times, which yields around 10 minutes of testing. In the charts you will see a few blips in power use; this is a result of the benchmark starting the next loop.

We also use Furmark to capture worst-case power readings. Although both Nvidia and AMD consider the application to be a “power virus,” or a program that deliberately taxes the components beyond normal limits, the data we gather from it tells us about a card’s capabilities.

Our power consumption values during a gaming scenario were similar, around 125W, for all GTX 1660 cards in the chart (two 1660 Supers and one 1660 Ti). The MSI card averaged 125W, 5W over the reference TDP, while the EVGA card was just under that at 123W. The EVGA GTX 1660 Ti XC was a bit higher, averaging nearly 128W. In this case, the MSI card is a bit more efficient than the EVGA, but you will not notice such small differences running even several hours per day. Nvidia has found a nice sweet spot for performance and efficiency on this mature 12nm process in the 1660 variants.

Moving on to Furmark, what we found here, are the cards power limits. Each card held steady throughout the test while they all worked diligently to maintain themselves under the maximum board power. For the MSI this was 130W, same as the GTX 1660 Ti. The GTX 1660 Super was a bit less at 125W. You’ll see how this affected clock speeds below.

Temperatures, Fan Speeds and Clock Rates

Gaming

MSI’s Twin Frozr 7 cooler managed to keep the card at a cool 67C or less. Compare this to EVGA’s result, a couple of degrees Celsius higher at 70C, and we find that the MSI solution is a bit more effective. That said, both values are well within an acceptable range.

In order to keep the temperatures where they are, the GTX 1660 Super Gaming X's fans settled at 1,200 RPM for an average, which was whisper quiet on the large 90mm fans. EVGA’s cooling solution spun a lot faster at almost 2,300 RPM. And though it was louder than Twin Frozr, it was not intrusive over the other system fans.

In order to determine the average clock speed during the game testing, we averaged the clocks speeds during the 5th loop of the Metro: Exodus test. This way, the card is warmed up and clocks settle a bit (clocks in the first loop average around 30 MHz higher).

In this case, the MSI averaged 1,929 MHz, well over the listed boost clock of 1,830 MHz. The EVGA 1660 Super XC came in at 1,899 MHz, around 30 MHz less than the MSI. The slightly higher clock speed is the reason for the just slightly better performance of the MSI.

Furmark

Switching up to our GPU killer Furmark, temperatures here were similar with the MSI peaking at 65C -- two degrees Celsius less. The reason why we see lower temperatures is due to the card significantly reducing clock speeds and voltage in order to fit within the power budget. The EVGA card behaved in the same manner, peaking at 68C.

Fan speeds when running Furmark ramped up steadily and ended running a little slower averaging 1,000 RPM across the entirety of the test. The EVGA card settled in a little slower as well at 2,250 RPM. During both tests, both cards were quiet, though I could hear the EVGA more clearly over the MSI.

About those clock speeds we mentioned earlier: Here the MSI was running at 1,731 MHz or around 200 MHz slower. Voltage was lowered to around 0.85V compared to 1.03V with higher clocks and gaming. The comparison EVGA ran at 1,603 MHz versus 1,899 MHz -- an almost 300 MHz difference with similar voltage drops. Furmark isn’t a very realistic scenario, but it shows how the card can react to non-gaming loads maxed out.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

MSI Afterburner has been a long-standing member of the GPU overclocking utility landscape going back several years. We’ve seen it improve over time with increasing functionality as well as GUI changes leaving users with one of the most stable to use (with most any card) utilities out there. Afterburner displays pertinent information about the card including clock speeds and voltage as well as temperature from two main dials.

Between the dials are the control sliders for power and temperature limits, fan speeds, as well as core and memory clock adjustment. Simply drag the slider to the right, clocks/limits/speeds go up, and to the left, they go down. Simple. On the bottom portion is a real-time chart that displays temperatures, clocks, voltage, among several other monitoring options a scroll away. The monitoring window can also be detached and expanded to display as many live charts as your screen will fit vertically.

MSI Afterburner is one of my favorite GPU overclocking tools, especially when using the old-school skin (Default MSI Afterburner V2). This version (above) reduces the clutter on the Cyborg interface, with sliders on the left and monitoring on the right with few aesthetic addons, making for a simple-to-use interface.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

MSI’s GTX 1660 Super Gaming X proved that it is a more-than-capable 1080p ultra card, averaging over 80 fps across all 11 games. Most of these were well over 60 fps or very close to it outside of our resident GPU killer, Metro: Exodus. MSI’s card managed to be a hair (not noticeably) faster than the EVGA GTX 1660 Super SC Ultra. Compared to the 5500 XT 8GB, the MSI Gaming X 1660 Super runs about 22% faster. When looking at the next highest RDNA card in the 5600 XT, we found it to be almost 20% faster at 1080p and over 22% faster at 2560 x 1440.

Performance is one thing, but price is another. Nvidia’s GTX 1660 Super cards presented users with a worthwhile performance uptick over the non-super cards. The Super cards came in priced at $229 and up versus the non-super at $203-plus. This was fine upon release --welcomed actually,-- and made good sense in the market. But looking at things today, we have AMD’s 5500 XT at $179 and $199 for the 4GB and 8GB variants respectively, and most recently the 5600 XT starting at $279.

The latter, which is significantly faster than this MSI card, while starting at just $10 more as of this writing, makes the Gaming X --and all high-end GTX 1660 Tis-- a tough sell without a major price drop.

Our MSI card is found online for $269.99 at Newegg, just $10 less than the notably faster 5600 XT, but as mentioned above, quite a bit slower. For a mere $20 more, a highly overclocked 5600 XT is possible. That said, other GTX 1660 Supers, like the EVGA we tested with are found for as little as $229 which at that price, makes it a much better value both inside the Nvidia ecosystem as well as when AMD cards get involved.

MSI’s premium card includes the Twin Frozr 7 cooler which keeps it running cool and quiet throughout our testing as well as some hands-on gaming. The zero fan capability is welcome and yields no noise during desktop work or light loads for the GPU. When the fans do spin up, they are barely heard over the other fans in our test system.

In the end, this review simply came late to the party. A mere two months ago this card only had to compete with its peers, but today the landscape in the sub $300 market is crowded since AMD joined the party. The card does well in performance, power consumption, and acoustics, but the current $269 price is too much for the current market. Until the price comes down on the MSI GTX 1660 Super Gaming X to below $240, there are others priced notably less which makes the value prospect more tolerable.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

We’ve taken a look at a few of Nvidia’s Super cards recently, in particular, the GeForce GTX 1650 Super and GTX 1660 Super. With both of these cards, we were left with an overall positive impression about the GPUs' performance and value. Sitting between these cards is the GTX 1660 -- a video card that is marketed for 1080p gaming, but with budget considerations in mind.

Now up for review: the Zotac GTX 1660 Amp. Zotac takes the reference values of the GTX 1660 and puts its own twist on things including a factory overclock, improved power delivery and the IceStorm 2.0 cooling solution, all of which improve upon the base specifications.

The GTX 1660 occupies one of the budget spaces in the Turing lineup, alongside the GTX 1650 -- and now, of course, their Super versions. Pricing for the GTX 1660 starts at $219, but will vary by board partner depending on the changes to the card. The Zotac Amp version currently sells for $239.99 on Newegg, and other AIB cards are also floating around that price point.

Like many of these budget cards, Nvidia doesn’t have a reference SKU; there are only reference specifications. The Zotac version we have uses a dual-fan cooler, a single 8-pin PCIe power connector, and is compact -- good to use for small form factor builds.

Features

The GTX 1660 is based on the Turing architecture and uses a variant of the TU116 silicon found in the GTX 1660 and 1660 Ti, just like the 1660 Super. As with other GTX-branded cards, there’s no RTX support here for ray tracing. The TU116 die is manufactured on the 12nm FFN (FinFET Nvidia) process by TSMC, consists of 6.6 billion transistors and measures in at 284 mm². The GTX 1660 brings 1,408 cores, 88 TMUs and 40 ROPS which is cut down from the full TU116 variant found in the GTX 1660 Ti. The Ti version is 1,536 cores, 96 TMUs, and 48 ROPs.

The GTX 1660 uses GDDR6 video memory, just as the Super cards do above and below it. The memory sits on a 192-bit bus, with six 32-bit memory controllers and a 6GB capacity -- a sweet spot for 1080p gaming. Where some 4GB cards will choke when running out memory, 6GB cards have 50% more VRAM to store textures and other bits without swapping out to slower storage off the card.

The Zotac GTX 1660 comes with a 1,530 MHz base clock and 2,001 MHz (8,004 MHZ effective) memory clock. The core boost clock is listed as 1,845 MHz, which is the same as the reference clock. TDP is listed at 130W, 10W over the reference value. Most card partners recommend the GTX 1660 paired with a 350W or greater power supply.

Like the GTX 1650 Super we’ve reviewed, the GTX 1660 and the TU116 GPU core includes the Turing NVENC encoder, rather than the Volta encoder included in the original GTX 1650 and the TU117 chip. NVENC encoding on the Turing-based cards decreases CPU utilization and improves performance, especially with higher-resolution encoding and decoding.

Below is a complete list of specifications for the lower-end range of Nvidia's current GPUs.

Design

Zotac’s GTX 1660 Amp Measures 8.3 x 4.7 x 1.61-inches (209.6 x 119.3 x 41mm) and is a true dual-slot card, allowing users to use the slot below (albeit without much room between it and the intake for the card). The card’s relatively small size will allow it to fit in more builds than full-sized GPUs. Like the previous Zotac GTX 1650 Super we looked at, Zotac says this card should also fit in "99%" of systems. As always, double-check your chassis specifications to make sure it will fit.

The design of the card is subtle and standard fare, with a grey shroud surrounding two 90mm fans. The shroud uses sharp angles throughout, almost giving it a stealth-like appearance, like it was trying to avoid radar. On top  is the Zotac Gaming branding in black, along with the GeForce GTX branding, while the back of the card sports a wraparound backplate. This is for rigidity and protection, as it does not use thermal pads to efficiently remove heat from the PCB. If you want RGB lighting in your system, you won't find any from this card.

The Zotac GTX 1660 Amp uses dual 90mm fans to move air into the full-size heatsink. The blades have small stripes/channels on them which the company says breaks the flow of air into smaller channels, enabling more optimized airflow and quieter rotation.

The heatsink below is made of aluminum and uses a wide array/full body for increased heat dissipation.  Three 6mm heat pipes meander their way through the fin stack, starting directly above the GPU. Making direct contact with the video card die is a copper plate which is in direct contact with the heat pipes. There is a metal plate surrounding the copper, which is designed to cool the Samsung memory. The VRMs are also cooled by the heatsink using a thick thermal pad between it and the contact point on the fin array.

The cooling solution proved relatively quiet throughout testing and kept the card running cool. I just wish the fans stopped on idle or low loads for true silent operation.

Looking at the PCB and power delivery, we are able to see a total of eight phases, which are separated into a 6+2 configuration. Power is managed on the board by a uPI 9512 voltage regulator that's able to run up to 8 channels. A single 8-pin PCIe connector sends power to the VRMs, which are plenty capable of delivering the necessary power for stock and overclocked operations.

There are four ports on this card: three DisplayPorts (v1.4) and a single HDMI (2.0b). This is typical of this class of card, though it does vary by vendor. Gone are the legacy ports we've seen in past generations of cards -- sorry DVI!

How We Tested Zotac’s GTX 1660 Amp

We’ve updated our GPU test system to a new platform and swapped from a six-core i7-8086K to an eight-core Core i9-9900K. The CPU sits in an MSI Z390 MEG Ace motherboard, along with 2x16GB Corsair DDR4 3200 MHz CL16 RAM (CMK32GX4M2B3200C16). A Corsair H150i Pro RGB AIO keeps the CPU cool along with a 120mm Sharkoon fan for airflow across the test system. A single 2TB Kingston KC2000 NVMe PCIe 3.0 x4 drive stores our OS and gaming suite.

The motherboard was updated to the latest (at the time of testing) BIOS (version 7B12v16), from August 2019. Optimized defaults were used to set up the system. We then enabled the memory’s XMP profile to get the memory running at the rated 3200 MHz CL16 specification. No other changes or performance enhancements were enabled. Windows 10 (1903) was used.

As time goes on we will build our database of results back up based on this test system. For now we are including GPUs that are close in performance to this GPU. We have a Gigabyte GTX 1650 Gaming OC, EVGA GTX 1660 Super SC Ultra, Zotac GTX 1650 Super, Representing AMD is an XFX RX 590 Fat Boy, Sapphire RX 5500 XT Pulse and Gigabyte RX 5500 XT Gaming OC 8G.

Our list of games test games is currently Tom Clancy’s The Division 2, Ghost Recon: Breakpoint, Borderlands 3, Gears of War 5, Strange Brigade, Shadow of The Tomb Raider, Far Cry 5, Metro: Exodus, Final Fantasy XIV: Shadowbringers, Forza Horizon 4 and Battlefield V. These titles represent a broad spectrum of genres and APIs, which gives us a good idea of the relative performance difference between the cards. We’re using driver build 441.20 for the Nvidia cards and Adrenalin 2019 Edition 19.11.3 for AMD.

We capture our frames per second (fps) and frame time information by running OCAT during our benchmarks. In order to capture clock and fan speed, temperature and power, we u se GPU-Z's logging capabilities. Soon we’ll resume using the Powenetics based system found in previous reviews, as soon as the equipment is ready.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

The 1660 is almost 13% faster than the 1650 Super across this testing. If we look at the AMD side using a 5500 XT (Gigabyte 8GB model), the 1660 is almost 9% faster and similarly priced. 

For this set of cards, we tested at 1080p using ultra settings as well as medium settings. This gives us a good look at the cards working where they were intended to, while still trying to reach that magic 60 fps-plus metric many strive for. Jumping up to a higher resolution, even 2560x1440, uses more memory and would require turning down image quality settings even further in order to maintain reasonable frame rates.

Running at the Ultra settings, the Zotac GTX 1660 Amp’s performance ranged from 38 fps in Metro Exodus, up to 103 fps in the Forza Horizon 4 benchmark. Borderlands 3 and Ghost Recon Breakpoint sat under 60fps, alongside Metro. The rest of the titles were higher -- either very close to 60 fps or well above. Outside of a couple of games, 60 fps using Ultra type settings is certainly possible.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Dropping the settings back to medium, the Zotac GTX 1660 Amp is able to run all of our titles at 60 fps or well above. Even in Metro Exodus, the GTX 1660 managed 62 fps on average. Most titles sat around 100 FPS or so, and the cards averaged over 100 fps across all games we test. You can likely play some of these titles at 2560x1440 on medium settings as well and still be over 60 fps.

When comparing performance to the GTX 1650 Super, we find the GTX 1660 is over 8% faster. Compared to the 5500 XT 8GB, this card is 9% faster.

Zotac’s GTX 1660 Amp proved to be a solid 1080p gamer, able to reach 60 fps or more using Ultra settings. While we didn’t test this card at 2560x1440, it is clear with lowered settings most titles will be playable at the higher resolution.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Power Consumption

We used GPU-Z’s logging to measure the card’s power consumption with the Metro Exodus benchmark running at 2560 x 1440 using the default Ultra settings. The card is warmed up prior to testing and benchmarking is started when it settles to a normalized idle temperature (after about 10 minutes). The benchmark is looped a total of five times, which yields around 10 minutes of testing. In the charts you will see a few blips in power use; this is a result of the benchmark starting the next loop.

We also use Furmark to capture worst-case power readings. Although both Nvidia and AMD consider the application to be a “power virus,” or a program that deliberately taxes the components beyond normal limits, the data we gather from it tells us about a card’s capabilities.

Power consumption during gaming wasn’t crazy with these budget cards. The GTX 1660 averaged around 110W, the GTX 1650 Super about 94W, and the AMD 5500 XT at 90W. In this case, there are peaks above 120W with more power for more performance. Surprisingly, the 4GB variant used more power than the 8GB and was a lot closer to the GTX 1660 in power use.

In Furmark, the Zotac GTX 1660 Amp uses around 127W average. Both the 1650 Super and the GTX 5500 XT OC 8G use less, coming in at 100W and 122W respectively. From the relatively flat line on the chart, we can see the Zotac is power limited to around 130W at default, which matches the 130W TDP on the Zotac website.

Temperatures, Fan Speeds and Clock Rates

Gaming

Temperatures during the gaming test peaked at 63 degrees Celsius with the card's IceStorm 2.0 cooling. While we do not have a direct comparison, the heat sink was able to keep the card around the low 60s. This temperature is quite low and will minimize the drops in boost bins due to temperature.

Fan speeds on the Zotac 1660 Amp ramped up slowly and settled around 2,000-2,100 RPM through our testing. This consistent speed allowed for quiet operation during normal loads. Although the fan doesn’t use an idle off mode, it is whisper quiet at the 1,200 RPM idle speed.

As far as clock speeds go, the Zotac 1660 ran at an average clock speed of 1,922 MHz. Compared to the specifications, it is well over the listed boost clock of 1,845 MHz. This behavior is typical of Nvidia cards, though that story will change below with Furmark.

Furmark

With the more consistent, heavier load in Furmark, we see similar results compared to the gaming side. In this case, temperatures peaked at 65C, just two degrees Celsius over the gaming temperatures. The integrated protection built into these cards throttle Furmark loads considerably, protecting your hardware.

Fan speeds during Furmark ramped up steadily and ran a bit faster, around 2,200 RPM, in the last half of Furmark testing. At this speed, the cooling solution was still relatively quiet and mixed in with the noise from other fans attached to the system.

The GTX 1660’s clock speeds in Furmark are lower than what we saw in gaming, but not nearly as much as we saw from other Nvidia GPUs in the past. In this case the card averaged around 1,880 MHz over the duration of the test while under load. Compare that to 1,922 MHz when gaming and we confirmed that the GPU is working hard trying to fit within its power limits.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Zotac, like the other card partners, offers its own in-house software, called FireStorm, to control and monitor the card. This lightweight app displays information about the GPU such as current clock speeds, temperatures, as well as fan speed. The GUI itself is mostly black with some dark and lighter blue highlights. The text is mostly white or a lighter blue and easy to read.

Other options include sliders for raising core and memory clocks, power limits and temperatures. With this card, the power and voltage sliders are unlocked and overclocking is possible on the core and memory. The power limit slider goes up to 111-percent which suggests there is some additional headroom for overclocking available.

FireStorm’s monitoring capabilities include displays for GPU core and memory clocks, voltage and temperatures, along with GPU and memory utilization data. While the software displays pertinent information, it does not have the ability to add others. What you see is what you get here, but it should be informative for the average user.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Zotac’s GTX 1660 Amp showed that it is more than capable of running a 1080p monitor along with games at ultra settings. Nearly all titles were at or over 60 fps and those that were not needed simple adjustments to get there. When compared to the Gigabyte RX 5500 XT Gaming OC 8GB, the results showed the Nvidia card almost 9% faster. Within its own product stack, the GTX 1660 is also several percentage points faster than the GTX 1650 Super we recently reviewed, but still falls short of the GTX 1660 Super.

The card uses the IceStorm 2.0 cooler that is able to keep the card running cool and nearly silent. Though it does not have an idle fan stop, it is nearly inaudible at idle and quiet on load. Heat is exhausted in the chassis but running under 120W, doesn’t take much to mitigate. Proper case airflow goes a long way.

At $239.99 on Newegg, Zotac’s Amp is priced a lot higher than most other GTX 1660s, which start at around $200, like Zotac’s non-Amp version. Looking at the big picture, AMD’s Radeon RX 5500 XT 8GB starts at $199.99 on Newegg as well but performs several percentage points slower -- at least on our PCIe 3.0 test bed. AMD’s recently announced Radeon RX 5600 XT will be an interesting addition to the GPU market, likely performing better though at a higher price point.

Overall, Zotac’s GTX 1660 Amp with the IceStorm 2.0 cooler and factory overclock, performed well at 1080p using Ultra settings. Price to performance may not be the best compared to similar cards, but the cooling and quiet operation may offset that cost -- especially if you need a compact card for an SFF system. 

If you are looking for a card that can surpass 60 fps at 1080p ultra settings in most titles, a GTX 1660 should be on the shortlist. If you can find the Zotac Amp for less than $230, it is a good deal. Otherwise, there are less-expensive cards to be had from Nvidia as well as the AMD RX 5500 XT 8GB variants if you don’t mind slightly less performance.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Asus has shrunk its Dual GeForce GTX 1660 Super graphics card to fit inside Intel's NUC 9 Extreme (codenamed Ghost Canyon) debuted at CES 2020 and other small form factor (SFF) enclosures. The new graphics card carries the Mini branding to denote its miniature footprint.

The Dual GeForce GTX 1660 Super Mini checks in at 19.7cm long with a 2-slot design. The graphics card flaunts a elegant, black shroud with very subtle RGB lighting. A pair of Axial-tech cooling fans are in charge of active cooling duties.

According to Asus, the Axial-tech design utilizes a smaller hub, long blades and a barrier ring that allows the fans to deliver improved downward air pressure. As typical, the fans come with Asus' 0dB feature, meaning they won't start spinning until the graphics card's operating temperature crosses the 55 degrees Celsius (131 degrees Fahrenheit) mark. The fans are also IP5X dust-resistant, which should help with its lifespan.

The Dual GeForce GTX 1660 Super Mini is based on Nvidia's Turing TU116 silicon. It's equipped with 1,408 CUDA cores and 6GB of 14 Gbps GDDR6 memory across a 192-bit memory interface. Asus currently offers the graphics card in two flavors. The vanilla edition features a 1,815 MHz boost clock, while the OC edition rocks a 1,860 MHz boost clock.

The Dual GeForce GTX 1660 Super Mini has a single 8-pin PCIe power connector, so you should be able to get by with a 450W power supply. The graphics card supports up to three displays simultaneously. dDisplay connectors are one DVI-D port, one HDMI 2.0b and one DisplayPort 1.4 output.

Asus didn't reveal the pricing or availability for the Dual GeForce GTX 1660 Super Mini, but we'll keep our eyes peeled. 

EVGA's GTX 1660 Super SC Ultra graphics card is currently selling for its lowest price ever. That's one of the best tech deals we've spotted today. With a $10 rebate and the $10 off with the promo code "VGAPCJP767" you can get it for just $210 at Newegg. 

In our EVGA GTX 1660 Super SC Ultra review, we liked it so much that we awarded it 4.5 stars and our Editor's Choice award. So seeing the card with a lower price makes us very happy, especially considering that many variants of the GTX 1660 Super appear to sell for well above MSRP.

The card's GPU comes packed with 1,408 CUDA cores and a boost frequency of up to 1,830 MHz. Memory is handled by 6GB of GDDR6 memory over a 192-bit memory bus clocked at 14 GHz, which makes it good for a data rate of 336 GBps. EVGA's variant comes with a nifty dual-fan cooler and an all-metal backplate.

EVGA's GTX 1660 Super SC Ultra is a great graphics card for some serious 1080p gameplay, but it's also great for use with 1440p gaming monitors. Be swift if you want to get your hands on it though: the deal ends on Sunday.

AMD recently launched the Radeon RX 5500 XT, which aims to capture the attention of mainstream gamers with a mid-range budget to spend on a GPU. Following our review, we already know it’s not as powerful as Nvidia’s GeForce GTX 1660, but can it offer enough value to make up for what it lacks in performance? Let’s find out.

It’s an exciting time to be in the market for a new GPU, with several new products on the market in the last 12 months. Last year at this time, gamers were eagerly awaiting the reveal of AMD’s Navi architecture and the new line of GPUs that would reinvigorate the company’s Radeon Graphics brand. And Nvidia had not yet launched the RT and Tensor core-free 16-series Turing GPUs for the mid-range market. 

Today, we’ve got myriad options from team red, including the recently launched Radeon RX 5500 XT, which introduces Navi to the high-volume segment of the market, in the $200 price range. AMD is no stranger to this price point. For several generations, the Radeon team has focused its attention on mainstream cards like the RX 480, 580, and 590 (all based on the same Polaris architecture). 

However, Nvidia’s GeForce GTX 1660, which hit store shelves in March of 2019, is a strong contender for your gaming money. That GPU’s successor, the GTX 1660 Super, is somewhat more expensive ($230+), but the original model is still readily available and often sells for $210 or less. 

Now the question becomes, which card should you choose if you’re in the market for a new GPU for roughly $200. We compared the features and benefits of the Radeon RX 5500 XT and the GeForce GTX 1660 in four categories -- featured technology, game performance, power consumption & heat output and Value -- to help you decide which of these two cards is right for you.

Featured Technology 

AMD’s NAVI 14 XTX GPU features the same number of cores (1408) as the version of Nvidia’s Turing TU116-300-A1 chip found in the GTX 1660. Although it operates at a higher frequency and produces higher peak floating-point performance (5.2 TFLOPS vs the GTX 1660’s 5 TFLOPS). 

AMD’s NAVI 14 XTX features 6.4 billion transistors, which is 200 million fewer than the Turing TU116, but AMD crammed its chip into a much smaller package. Thanks to TSMC’s 7nm finFET process, AMD’s GPU fits on a 158 mm² die. Turing GPUs are also manufactured on TSMC’s finFET process, but Nvidia is working with a larger 12nm process. As a result, the TU116-300-A1 is nearly twice the size of NAVI 14 XTX at 284 mm².

The RX 5500 XT also offers faster GDDR6 memory compared to the GTX 1660’s GDDR5 memory modules. Nvidia does provide a larger 192-bit memory bus versus AMD’s 128-bit memory bus configuration, but the faster memory more than makes up for the difference in bus size. Theoretically, the NAVI 14 XTX has 224 GB/s of memory bandwidth to work with, compared to the 192 GB/s that TU116 offers. The Radeon RX 5500 XT is available in 4GB and 8GB configurations. But for this comparison, we’re focusing on the 8GB version. The GeForce GTX 1660 falls right in the middle, with 6GB of graphics memory.

By the numbers, AMD’s RX 5500 XT appears to be the stronger of the two GPUs. It’s manufactured on a smaller, more efficient process, offers a higher base and boost clock speed, and has a higher theoretical performance limit. The higher capacity of faster GDDR6 memory also gives it a significant advantage.

AMD offers a variety of software features for the Radeon RX 5500 XT. The company’s Radeon Software allows you to monitor your GPU’s temperature, fan speed and voltage, adjust the graphics settings for your games, access performance reports and game states, and update your GPU drivers. The Radeon Software Adrenalin 2020 edition also lets launch games directly from the Radeon Software dashboard. It even offers integrated media capture and game streaming features, including screenshots, video, and instant gif capture as well as compatibility with many popular streaming services such as Twitch, YouTube, Mixer, Facebook and more.

AMD also offers a feature called Game Boost, which monitors your framerate and dynamically adjusts the resolution of your games when it won’t impact the visual experience to maximize performance. 

The Radeon RX 5500 XT also supports a technology called Radeon Anti-Lag, which aims to reduce input lag by managing the queue of work sent to the CPU and ensuring that the CPU can’t get ahead of the GPU. Also, Radeon Image Sharpening combines contrast-adaptive sharpening with GPU upscaling to produce crisper in-game visuals. 

Nvidia’s GeForce Experience software is a reasonable equivalent to AMD’s Radeon Software Adrenalin. It enables you to keep your drivers up to date, offers performance-optimized game setting configurations, and allows you to launch your games from the GeForce Experience dashboard. 

Nvidia’s software supports media capture and game streaming features that integrate with popular streaming platforms. The GeForce GTX 1660 takes full advantage of Nvidia’s NVENC hardware encoder technology, which lowers the CPU overhead associated with software encoding. NVENC allows the CPU to send a single set of instructions to the GPU, which the GPU uses to output to your display and broadcast without impeding performance.

GeForce Experience also includes Nvidia’s advanced screenshot capture technology, called Nvidia Ansel, which allows you to capture extremely high-resolution (up to 33x 1080p) images of supported games (for regular OS captures, see how to take screenshots in Windows). Ansel also offers free-camera capture, which allows you to snap angles that aren’t possible in-game, and the ability to capture 360-degree images. Ansel even supports RAW EXR output, which allows you to make changes to the image in post-processing, such as create HDR-enhanced images. 

Both AMD and Nvidia offer support for adaptive display technology. The Radeon RX 5500 XT supports both FreeSync and FreeSync 2, whereas the GeForce 1660 is compatible with Nvidia’s more-expensive G-Sync adaptive display technology. Nvidia also supports a select few FreeSync displays. And as for performance, both AMD and Nvidia offer dynamic clock speed boosting technologies that automatically overclock your GPU to maximize the capabilities of your hardware configuration.

Winner: Tie. Both AMD and Nvidia offer compelling feature sets for their budget-minded options. Each card supports automated game optimization, game streaming, and driver update software. And they both support performance optimization features that maximize your gameplay experience.

Gaming Performance

We tested 11 relatively demanding games on both GPUs with maximum and medium settings enabled. In The Division 2, Gears of War 5, Strange Brigade, Far Cry 5, Shadow of the Tomb Raider, Final Fantasy XIV, Forza Horizon 4 and Battlefield V both cards managed to drive 60 fps average gameplay with the graphics settings maxed right out. 

More demanding titles such as Borderlands 3 and Metro: Exodus, we had to drop the settings to medium to achieve the coveted bare minimum of 60 fps. Ghost Recon: Breakpoint was also very demanding, and our Radeon was unable to crest 50 fps at max settings. However, the GTX 1660 fell just a sliver short with an average framerate of 58.7 fps on max settings. 

Both cards produce acceptable performance for 1080p gaming, but the GeForce outshined the Radeon by a small margin in all but two of our tests (at medium settings), but the two that AMD won were by less than a 1% margin. The GTX 1660 often held a 5% or better lead over the RX 5500 XT in our max settings tests. 

Winner: Nvidia. Despite having the superior on-paper specifications and equivalent software features, in a competition of might, the GTX 1660 outperformed the RX 5500 XT in almost every test we threw at these two cards. Nvidia’s configuration appears much more capable of handling games with maximum graphics settings enabled than AMD’s. Strangely, the RX 5500 XT is a stronger competitor with medium settings, but why would you settle for medium if you don’t need to? 

Power Consumption and Heat Output 

On the budget end of the GPU market, power consumption shouldn’t really be much of concern. None of the GPUs of this calibre draw anywhere near enough power to make a significant difference in your system’s configuration. Both AMD and Nvidia recommend at least a 450-watt power supply for their respective cards.

AMD’s Radeon RX 5500 XT features a TDP rating of 130w, whereas Nvidia’s GeForce GTX 1660 includes a TDP rating of 120w. In practice, the Radeon GPUs actually draw quite a bit less than 130 watts, except under extreme loads. We tested two examples of the RX 5500 XT, a 4GB Sapphire and an 8GB Gigabyte card, and we found that, in-game, the AMD card will average close to 100w. 

The Gigabyte GeForce GTX 1660 sample that we tested in March 2019 needed an average of 126 watts, and peaked at 134 watts, and the Zotac model that we used in the Radeon RX 5500 XT review pulled between 110 and 120 watts in our Metro: Exodus power test. 

Interestingly, the Zotac GeForce GTX 1660 peaked around 100w in our Furmark stress test, whereas the Gigabyte sample drew around 130w for this test. Both the Gigabyte and Sapphire Radeon cards topped 120w in this test, with the Sapphire model touching 136w at its peak. 

You likely won’t be concerned with the power requirements for these cards, but higher power draw does equate to higher heat output, which is likely of higher concern to you. 

Our Gigabyte RX 5500 XT OC 8G proved to be one of the coolest-running GPUs that we’ve seen in this segment. Following three passes of our Metro: Exodus test, the data showed that the GPU never exceeded 63C. Our tests with Gigabyte’s GeForce GTX 1660 OC 6G showed that it peaked at a 68C. It’s also worth noting that the fans on the GeForce were spinning at 2000 rpm, and the Radeon card spun at a leisurely 1800 rpm.

Winner: AMD. The power requirements for either of these cards is low enough to be negligible. However, the low heat output from the Radeon RX 5500 XT, likely aided by its smaller 7nm process, gives AMD a leg up in this category.

Value Proposition 

Comparing the value of the Radeon RX 5500 XT 8GB to the GeForce GTX 1660 is a somewhat challenging task. If you compare these cards at the same price, the GeForce GTX 1660 is the clear winner. 

Nvidia’s offering outperforms the competition handily in gaming tests, which if we’re honest is kind of the point of a graphics card—especially in the $200 range. However, AMD’s option still delivers acceptable performance and costs a few dollars less, which somewhat evens that playing field. Where the Radeon maybe 5% to 10% cheaper, it’s also 5% to 10% slower in many cases.

Winner: Nvidia. We can’t call the Radeon RX 5500 XT a bad deal. It’s far and away a better card than you would have bought for the same price a year ago. However, for just a few dollars more you can get better gaming performance from a GeForce GTX 1660. 

Bottom Line

What a difference a year can make. The last time we did a showdown of the mid-range GPUs, we put the Radeon RX 590 toe to toe with the GeForce GTX 1060 and AMD’s card was the definitive winner. This time around, the tables have turned and Nvidia has captured the mid-tier crown back from AMD. 

But pricing is close and can always change. So if you’re reading this in the weeks and months after first publication (in mid-January 2020), make sure to check your favorite online store to see if the value balance has shifted -- or if something better has come along. Like most areas of PC tech, price cuts and new products are always coming along to change the battle for your hard-earned component dollars.

Turing Without The RTX

11/21/2019 Update: Since the launch of the GTX 1660 Ti in February 2019, the GPU landscape has changed dramatically, with a swath of "Super" cards based on the same Turing architecture, but pushing both higher performance and lower prices than the company's initial Turing lineup. Most relevant to potential buyers of the GTX 1660 Ti is the GeForce GTX 1660 Super, which delivers similar performance to the 1660 Ti, at a lower starting price of $229. At this writing, that's about $30 less than the lowest-price GTX 1660 Ti.

Nvidia GeForce GTX 1660 Ti is built on TU116—an all-new graphics processor that incorporates Turing’s improved shaders, its unified cache architecture, support for adaptive shading, and a full complement of video encode/decode acceleration features. The GPU is paired up to GDDR6 memory, just like the higher-end GeForce RTX 20-series models. But it’s not fast enough to justify tacking on RT cores for accelerated ray tracing or Tensor cores for inferencing in games. As a result, TU116 is a leaner chip with a list of specifications that emphasizes today’s top titles.

Nvidia says that GeForce GTX 1660 Ti will start at $280 and completely replace GeForce GTX 1060 6GB. Although that base price is $30 (or 12 percent) higher than where the Pascal-based 1060 6GB began its journey back in 2016, the company claims GeForce GTX 1660 Ti is up to 1.5 times faster—and at the same 120W board power rating, no less.

Improved performance per dollar isn’t something we’ve seen much of from the Turing generation thus far. Can Nvidia turn that around with a GPU more purpose-built for performance at 1920 x 1080?

Meet TU116: Turing Sans RT and Tensor Cores

We’ve seen Nvidia launch four separate GPUs as it escorts us down the Turing hierarchy. With each, the company peels away resources to target lower price points. But we know it’s trying to maintain balance along the way, minimizing the bottlenecks that’d unnecessarily rob lower-end processors of their peak performance.

GeForce RTX 2060 is equipped with 44 percent of the 2080 Ti’s CUDA cores and texture units, 54 percent of its ROPs and memory bandwidth, and 50 percent of its L2 cache. Before the 2060 launched, we suspected that luxuries like RT and Tensor cores would no longer make sense at those levels. But a series of patches for Battlefield V—the one ray tracing-enabled game available at the time—enabled big performance gains, proving that Turing’s signature features could still be utilized at playable frame rates.

It turns out we were off by one tier. Nvidia considers TU116 the boundary where shading horsepower drops low enough to preclude Turing’s future-looking capabilities from serving much purpose. After stripping away the RT and Tensor cores, we’re left with a 284mm² chip composed of 6.6 billion transistors manufactured using TSMC’s 12nm FinFET process. But despite its smaller transistors, TU116 is still 42 percent larger than the GP106 processor that preceded it.

Some of the growth is attributable to Turing’s more sophisticated shaders. Like the higher-end GeForce RTX 20-series cards, GeForce GTX 1660 Ti supports simultaneous execution of FP32 arithmetic instructions, which constitute most shader workloads, and INT32 operations (for addressing/fetching data, floating-point min/max, compare, etc.). When you hear about Turing cores achieving better performance than Pascal at a given clock rate, this capability largely explains why.

Turing’s Streaming Multiprocessors are composed of fewer CUDA cores than Pascal’s, but the design compensates in part by spreading more SMs across each GPU. The newer architecture assigns one scheduler to each set of 16 CUDA cores (2x Pascal), along with one dispatch unit per 16 CUDA cores (same as Pascal). Four of those 16-core groupings comprise the SM, along with 96KB of cache that can be configured as 64KB L1/32KB shared memory or vice versa, and four texture units. Because Turing doubles up on schedulers, it only needs to issue an instruction to the CUDA cores every other clock cycle to keep them full. In between, it's free to issue a different instruction to any other unit, including the INT32 cores.

In TU116 specifically, Nvidia says it replaces Turing’s Tensor cores with 128 dedicated FP16 cores per SM, which allow GeForce GTX 1660 Ti to process half-precision operations at 2x the rate of FP32. The other Turing-based GPUs boast double-rate FP16 as well though, so it’s unclear how GeForce GTX 1660 Ti is unique within its family. More obvious, based on the chart below, is that the 1660 Ti delivers a massive improvement to half-precision throughput compared to GeForce GTX 1060 and its Pascal-based GP106 chip.

But when we run Sandra's Scientific Analysis module, which tests general matrix multiplies, we see how much more FP16 throughput TU106's Tensor cores achieve compared to TU116. GeForce GTX 1060, which only supported FP16 symbolically, barely registers on the chart at all.

In addition to the Turing architecture’s shaders and unified cache, TU116 also supports a pair of algorithms called Content Adaptive Shading and Motion Adaptive Shading, together referred to as Variable Rate Shading. We covered this technology in Nvidia’s Turing Architecture Explored: Inside the GeForce RTX 2080. That story also introduced Turing's accelerated video encode and decode capabilities, which carry over to GeForce GTX 1660 Ti as well.

Putting It All Together...

Nvidia packs 24 SMs into TU116, splitting them between three Graphics Processing Clusters. With 64 FP32 cores per SM, that’s 1,536 CUDA cores and 96 texture units across the entire GPU. Board partners will undoubtedly target a range of frequencies to fill the gap between GTX 1660 Ti and RTX 2060. However, the official base clock rate is 1,500 MHz with a GPU Boost specification of 1,770 MHz. Our EVGA GeForce GTX 1660 Ti XC Black Gaming sample topped out around 1,845 MHz through three runs of Metro: Last Light, while other cards we’ve seen readily exceed 2,000 MHz. On paper, then, GeForce GTX 1660 Ti offers up to 5.4 TFLOPS of FP32 performance and 10.9 TFLOPS of FP16 throughput.

Six 32-bit memory controllers give TU116 an aggregate 192-bit bus, which is populated by 12 Gb/s GDDR6 modules (Micron MT61K256M32JE-12:A) that push up to 288 GB/s. That’s 50% more memory bandwidth than GeForce GTX 1060 gets, helping GeForce GTX 1660 Ti maintain its performance advantage at 2560 x 1440 with anti-aliasing enabled.

Each memory controller is associated with eight ROPs and a 256KB slice of L2 cache. In total, TU116 exposes 48 ROPs and 1.5MB of L2. GeForce GTX 1660 Ti’s ROP count compares favorably to RTX 2060, which also utilizes 48 render outputs. But its L2 cache slices are half as large.

Despite a larger die, a 50%-higher transistor count, and a more aggressive GPU Boost clock rate, GeForce GTX 1660 Ti is rated for the same 120W as GeForce GTX 1060. Unfortunately, neither graphics card includes multi-GPU support. Nvidia continues pushing the narrative that SLI is meant to drive higher absolute performance, rather than give gamers a way to match single-GPU configurations.

EVGA’s GeForce GTX 1660 Ti XC Black Gaming

The GeForce GTX 1060 Founders Edition was also a 120W card and it squeaked by with one six-pin auxiliary connector. EVGA’s GeForce GTX 1660 Ti XC Black Gaming, on the other hand, employs an eight-pin input, giving it quite a bit of additional headroom. As we’ll see in our per-rail power testing, the card draws 3A of current over its PCIe slot during our stress test—the rest comes from its eight-pin connector.

EVGA utilizes four power phases for TU116. The GPU’s phases are controlled by an older ON Semiconductor NCP81276 on the back of the PCB, which is attached to a quartet of ON Semiconductor NCP302155s.

Those four components integrate the high- and low-side MOSFETs, a driver, and the bootstrap diode. They’re the same parts used on GeForce RTX 2070 Founders Edition, capable of average currents up to 55A.

Ubiq Semiconductor’s familiar dual-phase uP1666Q controls the memory’s voltage regulation circuitry by way of two QM3816N6 dual N-channel MOSFETs.

More interesting than the GeForce GTX 1660 Ti XC Black Gaming’s fairly simple power supply, perhaps, is the fact that EVGA’s PCB has vacant pads for an additional two GPU phases. There’s also a pair of emplacements for two more GDDR6 memory modules. Nvidia did something similar with GeForce GTX 1060, leaving a couple of blank spots on its Founders Edition card that were never populated. This is a time- and cost-saving measure, which allows the company to use one PCB for multiple products.

A metal plate sits on top of the PCB, sandwiching thermal pads between the integrated driver/MOSFETs, GDDR6 memory modules, and current sense resistor. More thermal pads on top of the plate keep heat moving into the main sink assembly, which is mounted around the GPU at four points and screwed in through the PCB’s back side.

The thermal solution itself is composed of a fairly thin copper pad that makes direct contact with TU116. Three flattened pipes are soldered to the top of it, and an array of aluminum fins are, in turn, soldered to the heat pipes. A relatively thick fin stack is exaggerated by the shroud, which houses a single 85mm fan and adds even more depth. All told, EVGA’s GeForce GTX 1660 Ti XC Black Gaming eats up three expansion slots on your motherboard.

EVGA ends up trading thickness for length. The GeForce GTX 1660 Ti XC Black Gaming may be 2” deep, but it’s only about 7.5” (~190mm) long and 4 ⅜” (111mm) tall. Moreover, compared to the beefy Founders Edition cards we’ve been reviewing, a total weight of 1 lb. 7 oz. (656g) feels downright light.

Up front, the GeForce GTX 1660 Ti XC Black Gaming exposes one dual-link DVI connector, an HDMI port, and a DisplayPort interface. The USB Type-C-based VirtualLink connector seen on every other Turing-class card so far is gone, a sign that we’re getting down to a performance level unconducive of smooth VR gameplay (even on the best VR headsets). Board partners that choose to add VirtualLink to their designs are free to do so; EVGA simply didn’t implement it on this model.

How We Tested EVGA’s GeForce GTX 1660 Ti XC Black Gaming

Obviously, GeForce GTX 1660 Ti is more mainstream than the other Turing-based boards we’ve reviewed. As such, our graphics workstation, based on an MSI Z170 Gaming M7 motherboard and Intel Core i7-7700K CPU at 4.2 GHz, is apropos. The processor is complemented by G.Skill’s F4-3000C15Q-16GRR memory kit. Crucial’s MX200 SSD is here, joined by a 1.6TB Intel DC P3700 loaded down with games.

As far as competition goes, the 1660 Ti mostly goes up against GeForce GTX 1070, though we include 1070 Ti as well. Of course, comparisons to GeForce GTX 1060 are inevitable. All of those cards are included in our line-up, along with GeForce RTX 2060 and GeForce RTX 2070. On the AMD side, we’re mostly interested in Radeon RX 590, although Radeon RX Vega 64 and Radeon RX Vega 56 make for interesting additions, too.

Our benchmark selection includes Ashes of the Singularity: Escalation, Battlefield V, Destiny 2, Far Cry 5, Forza Horizon 4, Grand Theft Auto V, Metro: Last Light Redux, Shadow of the Tomb Raider, Tom Clancy’s The Division, Tom Clancy’s Ghost Recon Wildlands, The Witcher 3 and Wolfenstein II: The New Colossus.

The testing methodology we're using comes from PresentMon: Performance In DirectX, OpenGL, And Vulkan. In short, these games are evaluated using a combination of OCAT and our own in-house GUI for PresentMon, with logging via GPU-Z.

We're using driver version 418.91 to test GeForce GTX 1660 Ti and build 417.54 for everything else. AMD’s cards utilize Crimson Adrenalin 2019 Edition 18.12.3.

MORE: Best Graphics Cards

MORE: GPU Benchmarks

MORE: How to Buy the Right GPU

MORE: All Graphics Content

Gaming at 1920 x 1080

By Nvidia’s own suggestion, GeForce GTX 1660 Ti is designed for high performance at 1920 x 1080 with anti-aliasing enabled. Although we found the card’s frame rates to be perfectly playable using higher resolutions, an emphasis on 120+ FPS would keep most gamers using FHD displays at this price point.

On average, GeForce GTX 1660 Ti lands a couple of percentage points away from GeForce GTX 1070 across our suite. We’d make the claim that newer games put the two cards closer together in the benchmarks. However, GeForce GTX 1660 Ti is slightly faster in The Witcher 3, and that’s an almost four-year-old title.

AMD’s Radeon RX Vega 56 beats the new GeForce on average. But it’s also typically quite a bit more expensive.

Ashes of the Singularity: Escalation (DX12)

Battlefield V (DX12)

Destiny 2 (DX11)

Far Cry 5 (DX11)

Forza Horizon 4 (DX12)

Grand Theft Auto V (DX11)

Metro: Last Light Redux (DX11)

Shadow of the Tomb Raider (DX12)

Tom Clancy’s The Division (DX12)

Tom Clancy’s Ghost Recon (DX11)

The Witcher 3 (DX11)

Wolfenstein II: The New Colossus (Vulkan)

MORE: Best Graphics Cards

MORE: GPU Benchmarks

MORE: All Graphics Content

Gaming at 2560 x 1440

We also ran the GeForce GTX 1660 Ti XC Black Gaming through our suite at 2560 x 1440. In some games, quality settings that worked well at 1920 x 1080 were still viable. In others, we needed to dial back the detail.

After a bit of tuning, we ended up with a suite-wide average of 66.4 FPS from EVGA’s entry-level GeForce GTX 1660 Ti. Nvidia’s GeForce GTX 1070 Founders Edition, for comparison, achieved 68.1 FPS—a lead of less than 3 percent.

Ashes of the Singularity: Escalation (DX12)

Battlefield V (DX12)

Destiny 2 (DX11)

Far Cry 5 (DX11)

Forza Horizon 4 (DX12)

Grand Theft Auto V (DX11)

Metro: Last Light Redux (DX11)

Shadow of the Tomb Raider (DX12)

Tom Clancy’s The Division (DX12)

Tom Clancy’s Ghost Recon (DX11)

The Witcher 3 (DX11)

Wolfenstein II: The New Colossus (Vulkan)

MORE: Best Graphics Cards

MORE: GPU Benchmarks

MORE: All Graphics Content

Power Consumption

Slowly but surely, we’re spinning up multiple Tom’s Hardware labs with Cybenetics’ Powenetics hardware/software solution for accurately measuring power consumption.

Powenetics, In Depth

For a closer look at our U.S. lab’s power consumption measurement platform, check out Powenetics: A Better Way To Measure Power Draw for CPUs, GPUs & Storage.

In brief, Powenetics utilizes Tinkerforge Master Bricks, to which Voltage/Current bricklets are attached. The bricklets are installed between the load and power supply, and they monitor consumption through each of the modified PSU’s auxiliary power connectors and through the PCIe slot by way of a PCIe riser. Custom software logs the readings, allowing us to dial in a sampling rate, pull that data into Excel, and very accurately chart everything from average power across a benchmark run to instantaneous spikes.

The software is set up to log the power consumption of graphics cards, storage devices, and CPUs. However, we’re only using the bricklets relevant to graphics card testing. EVGA's GeForce GTX 1660 Ti XC Black Gaming gets all of its power from the PCIe slot and one eight-pin PCIe connector.

Idle

Sub-10W power consumption is a big improvement compared to Nvidia’s higher-end Turing-based cards that idled closer to 20W at launch.

Gaming: Metro: Last Light

Three runs of the Metro: Last Light benchmark give us consistent, repeatable results, making it easier to compare the power consumption of graphics cards.

An average of 124W is slightly higher than Nvidia’s specification. At least the 131W peak we recorded is not egregious.

EVGA uses the PCI Express x16 slot sparingly, instead leaning on the eight-pin auxiliary connector. No wonder this card eschewed the six-pin connector found on many GeForce GTX 1060s.

Charting power consumption across our three-run benchmark shows the narrow range we’re working with. For the most part, GeForce GTX 1660 Ti behaves itself, which is more than we can say for certain other mid-range graphics cards.

GeForce GTX 1660 Ti uses slightly more power than GeForce GTX 1060 Founders Edition, despite their similar board power ratings. Meanwhile, GeForce GTX 1070, which is the closer performance match, hovers closer to 150W.

Current draw over the PCIe slot varies a bit between 2A and 3A. However, the instantaneous dips and spikes are incredibly small. In comparison, GeForce RTX 2060 pulls closer to 5A from its 16-lane slot, demonstrating a much straighter line across our three-loop Metro benchmark.

There’s quite a bit more noise through the eight-pin connector’s line. Average draw on that rail lands just under 8A.

FurMark

FurMark is a steadier workload, resulting in less variation across our test run. Average power is almost identical, while less room for dramatic swings brings the peak reading down under 128W.

There’s no room for performance to vary in FurMark, so the wavy line chart we saw under Metro: Last Light is gone, replaced by much more consistent power consumption through more than 15 minutes of recording.

GeForce GTX 1660 Ti and 1060 land near each other, yet again. But it’s the GeForce GTX 1070 that EVGA’s XC Black Gaming card more closely resembles in our benchmarks. As you might imagine, the GTX 1660 Ti achieves a better performance/watt ratio.

MORE: Best Graphics Cards

MORE: GPU Benchmarks

MORE: All Graphics Content

Temperatures, Fan Speeds and Clock Rates

Gaming: Metro: Last Light

Both previous-gen Founders Edition models ramp up slowly during our Metro: Last Light benchmark. Given their centrifugal fans, we’d expect them to spin faster and make more noise to achieve similar thermal performance as EVGA’s GeForce GTX 1660 Ti XC Black Gaming. Instead, it’s the 1660 Ti that quickly crests 2,250 RPM. It’s not as loud as AMD’s Radeon RX Vega 56, despite a higher rotational speed. But it’s not as quiet as the Founders Edition cards, either.

At least the GeForce GTX 1660 Ti XC Black Gaming’s fan keeps TU116 nice and cool. Whereas GeForce GTX 1070 Founders Edition butts up against 80°C, EVGA’s card doesn’t break 70°C.

Voltage and clock rate do fall slightly over the course of three Metro: Last Light benchmark loops. The 1660 Ti starts around 1,900 MHz and ends the sequence closer to 1,800 MHz. That’s higher than Nvidia’s 1,770 MHz GPU Boost rating. However, we’ve seen other 1660 Tis capable of maintaining more aggressive settings.

FurMark

Radeon RX Vega 56 and GeForce GTX 1070 FE come closer to GeForce GTX 1660 Ti XC Black Gaming’s fan speed under FurMark. But EVGA’s card still spins fastest, matching our readings in Metro.

Our GeForce GTX 1660 Ti actually runs cooler through 15 minutes of FurMark than it did in our real-world gaming workload.

Bumping right up against its power limit, the GeForce GTX 1660 Ti XC Black Gaming has no choice but to throttle back voltage and clock rate. By the end of our 15-minute test, TU116 is under 1,600 MHz at just over 0.800V.

MORE: Best Graphics Cards

MORE: GPU Benchmarks

MORE: All Graphics Content

Conclusion

If you’re looking for a new mainstream graphics card today, there’s a fair chance you missed the Pascal generation altogether. Three or four years have passed since your last upgrade, and that GeForce GTX 960 or Radeon R9 380 is starting to feel a little slow. There are still a few GeForce GTX 10-series cards floating around. But as Nvidia fills its portfolio with Turing-based boards, previous-gen products like the GeForce GTX 1070 will disappear altogether, joining the now-unavailable 1080 and 1080 Ti. Today, GeForce GTX 1060 gets added to the endangered species list as GeForce GTX 1660 Ti replaces it.

The 1060 had a good run. It launched at $250 and served up excellent frame rates at 1920 x 1080, gingerly stepping on Radeon RX 480’s toes in the process. However, GeForce GTX 1660 Ti blows right past it in the benchmarks. Our results show the 1660 Ti averaging about 100 FPS across our suite, beating Radeon RX 590, roughly tying the old GeForce GTX 1070, and losing slightly to Radeon RX Vega 56. And that’s at a price point just $30 higher than the 1060 6GB in 2016.

Step up to 2560 x 1440 and the GeForce GTX 1660 Ti continues delivering playable performance with an average of more than 60 FPS between 12 different games. That’s 19% faster than Radeon RX 590 and darn near even with GeForce GTX 1070. Radeon RX Vega 56’s lead grows to 10%, likely due to a massive memory bandwidth advantage that staves off bottlenecks at higher resolutions. But the cheapest model in stock on Newegg at the time of writing was $400. A 43% price premium on a card rated for 75%-higher power consumption just doesn’t make sense for 10%-higher frame rates.  

AMD does appear to be chipping away at Radeon RX 590’s price. Entry-level models still cost $260, though. They need to be lower to justify 85% of GeForce GTX 1660 Ti’s performance at 188% of its power consumption.

Previous-gen GeForce cards get their fair share of shade, too. Remaining GeForce GTX 1060s fail to shine with GeForce GTX 1660 Ti beating them by 30%+, despite the fact you can find a 6GB model in the $260 range. And obviously you’re going to want to avoid the few GeForce GTX 1070s left on shelves priced at $310 and up.

Taking a step back, then, it looks like GeForce GTX 1660 Ti is the card to beat for fast-paced gaming at 1920 x 1080 and solid performance at 2560 x 1440. Our only hesitation in recommending it comes from GeForce RTX 2060, which doesn’t look as good in our performance per dollar charts but does include Nvidia’s Tensor/RT cores. Do you make every dollar count by buying the GPU focused on accelerating today’s games or spend a little more in the hopes that ray tracing/DLSS gains momentum in the months to come? Let us know on the forums!

MORE: Best Graphics Cards

MORE: GPU Benchmarks

MORE: All Graphics Content

After months of rumors, Nvidia finally announced the GTX 1660 Super and GTX 1650 Super (coming in November) graphics cards today. And Gigabyte is right there with it, unveiling three of its custom GeForce GTX 1660 Super models. 

Gigabyte's offerings span from the gigantic GTX 1660 Super Gaming OC 6G to the itty-bitty GTX 1660 Super Mini ITX OC 6G. Like usual, the graphics cards will carry many of the brand's exclusive features, such as 3D Active Fan, Alternate Spinning, Ultra Durable components and RGB Fusion 2.0 lighting (when applicable).

Gigabyte GeForce GTX 1660 Super Gaming OC 6G

Gigabyte's GeForce GTX 1660 Super Gaming OC 6G is the flagship model. It's 280mm long and utilizes the WindForce 3X cooling system, which is comprised of three 80mm cooling fans and three composite copper heat pipes. Gigabyte has included a sleek backplate as well.

While Nvidia's reference specification points to a 3+2-phase design, the Gigabyte GeForce GTX 1660 Super Gaming OC 6G features a 4+2-phase power delivery subsystem. The graphics card has a 1,860 MHz boost clock and is equipped with one 8-pin PCIe power connector, three DisplayPort 1.4 outputs and one HDMI 2.0b port. 

The Gigabyte GeForce GTX 1660 Super Gaming OC 6G is available for $249.99 at Amazon and Newegg.

Gigabyte GeForce GTX 1660 Super OC 6G

The Gigabyte GeForce GTX 1660 Super OC 6G employs the more compact WindForce 2X cooler, which consists of two 90mm cooling fans and a single composite copper heat pipe. This model measures 225.65mm long and also comes with a backplate.

The graphics card comes with a 1,830 MHz boost clock. There's a single 8-pin PCIe power connector for external power. The display outputs are identical to those found with the Gigabyte GeForce GTX 1660 Super Gaming OC 6G.

The Gigabyte GeForce GTX 1660 Super OC 6G is available for $239.99 on Amazon and Newegg.

Gigabyte GeForce GTX 1660 Super Mini ITX OC 6G

Next is the Giagbyte GeForce GTX 1660 Super Mini ITX OC 6G. It's 170mm in length and sports Gigabyte's custom cooling system that's made up of a single 90mm cooling fan and composite heat pipe.

The graphics card flaunts a 1,800 MHz boost clock and requires a single 8-pin PCIe power connector to function properly. Like its siblings, the it provides three DisplayPort 1.4 outputs and a single HDMI 2.0b port.

Gigabyte didn't reveal the graphics card's pricing but said it will hit hardware shelves in North America in the coming weeks.

After weeks of rumors, Nvidia has released the GTX 1660 Super graphics card today. Meanwhile, rival AMD cards have received price cuts on the Radeon RX 590 graphics card, as spotted by HotHardware.

The RX 590 is currently around or below $200 mark on Newegg for a good handful of variants. When it first came out, cards started at $280. That's a fair price, offering similar frames per second per dollar as the GTX 1660 and GTX 1660 Super.

Despite the RX 590 being about a year old, it does compete with the GTX 1660 Super. Although, in our testing, AMD's card didn't perform quite as well as Nvidia’s new card. We've seem many publications name the GTX 1660 Super as the reason for the price drops; however, the GTX 1650 Super arriving November 22 might have something to do with it as well. The GTX 1650 Super is expected to bring both memory and GPU upgrades over the GTX 1650 and cost under $200. That'd pose a serious threat to the RX 590.

Additionally, the AMD Radeon RX 5500 is also coming soon (release date TBD). The RX 590 price drops could be in anticipation for that Navi-based mid-tier card.

AMD's RX 590 is based on Polaris, which despite being an older architecture still performs amicably well. It works great for gaming at 1080p and high settings and does well at 1440p. It can even work at higher resolutions but only at the expense of buttery framerates and graphics quality.

However, if you're thinking about buying a new graphics card but aren't in a rush, you may want to wait and see what the RX 5500 and GTX 1650 Super bring to the table. 

In time with the launch of its new GTX 1660 Super and GTX 1650 Super graphics cards, Nvidia is launching its latest Game Ready Driver. That driver is not only for the launch of the new Super cards, but also has new features and supports some enhancements.

Among the improvements are an image sharpening filter in the Nvidia control panel, which the company said will allow for faster performance in games using DirectX 9, 11 and 12. This includes a slider for sharpening that can be changed per-game.

Additionally, Nvidia is improving its Nvidia Ultra Low Latency or "NULL" mode to work with G-Sync. The two will work together, allowing for both fast response times and gaming without visual tearing.

And for fans of ReShade, a program that allows you to filter your games, Nvidia is integrating ReShade filters into its Ansel and Freestyle features with the new update.

New G-Sync Compatible Monitors

Beyond the drivers, Nvidia also today announced new G-Sync Compatible monitors. Those are:

Over the last few weeks we have seen lots of rumors about the alleged Nvidia GeForce GTX 1660 Super. Today, we found the last bits of information that we needed to know, which includes the card’s specifications and pricing. The latest rumor (via VideoCardz) includes that Nvidia will officially launch the GeForce GTX 1660 Super on October 29th, which means that by this time next week, we will finally know how much truth there is to all these rumors.

As such, it’s time for a ‘Here is all we know’ post, detailing all the information we know about the GeForce GTX 1660 Super to date.

Nvidia GTX 1660 Super TLDR

·         The GTX 1660 Super’s only upgrade will be from GDDR5 to GDDR6 memory, thereby increasing the memory bandwidth from 192 GB/s to 336 GB/s – a 75% increase

·         The GTX 1660 Super is rumored to cost $10 more than the non-super variant.

·         The GTX 1660 Super is expected to be announced on October 29

Nvidia GTX 1660 Super Specifications

As far as the rumors are concerned, the GTX 1660 Super will be identical to the ‘ordinary’ GTX 1660, with only one exception: instead of GDDR5 memory, it will come with GDDR6 memory.

This one change can have quite the impact, however. The GDDR5 memory on the standard card ran at an effective speed of 8,000 MHz, whereas in the GDDR6 configuration it will run at an effective speed of 14,000 MHz. This change presents a 75 percent increase in memory clock, and given that both cards run over a 192-bit memory interface, also a 75 percent increase in memory bandwidth, jumping from 192 GB/s to 336 GB/s. This figure is higher than the GTX 1660 Ti’s 288 GB/s memory bandwidth.

Exactly what this change will mean for performance is unclear. Despite the higher memory bandwidth, we still expect the GTX 1660 Super to perform between the GTX 1660 and the GTX 1660 Ti. Where on the spectrum the performance difference will be will depend on two primary factors: the extent to which the GTX 1660’s GPU was bottlenecked by the memory bandwidth, and the application’s need for high memory bandwidth.

High-resolution gameplay with larger textures and more complicated graphics will benefit more from the increase in memory bandwidth than games with simpler graphical features. Because of this, we expect to see the biggest performance gains from today's demanding AAA titles.

The GTX 1660 Super’s GPU is said to be the same 12nm Turing-based TU116 chip with 1408 CUDA cores, running at between 1530 and 1785 MHz in the reference specification – identical to the standard GTX 1660. Of course, these clocks might be slightly higher from the board partner’s cards.

Spotted Board Partner Models

Thus far, we’ve seen leaks of GTX 1660 Super graphics cards from the makes of MSI, Zotac, and Maxsun. VideoCardz.com has also leaked two EVGA cards. All are pictured below.

The thing to note about these AIB card leaks is that there often isn't anything new that happens to their design, and the same is the case here. The cards from the AIBs come with the usual design offered by their manufacturers in their segment -- for example, MSI's Gaming X card (the first card pictured), is the company's higher-end offering in terms of cooling, and it comes with addressable RGB. The second MSI card pictured is the Ventus XS OC, which doesn't come with the fancy Twin Frozr cooler or RGB, bringing it closer to Nvidia's recommended price point.

The most interesting thing about the card leaks is often specifications that are visible on the box art, such as the memory capacity, type, and on the rare occasion, clock speeds.

Nvidia GTX 1660 Super Pricing & Release Date

As mentioned, Nvidia is expected to announce the GeForce GTX 1660 Super on October 29th – this time next week. Rumor has it that the Super variant of the GTX 1660 will cost $229 --$10 more than the standard version. The higher-end offerings from Nvidia, the RTX 2060 Super, RTX 2070 Super, and RTX 2080 Super, all kept their pricing identical or lower than the non-Super variants, which led to the non-Super cards being phased out.

Consequently, it’s a little odd that Nvidia might be raising the price of the GTX 1660 Super from its non-Super counterpart, but there is a simple explanation for this: The higher-end offerings had been on the market for a much longer time than the GTX 1660, meaning it was time for a price drop anyway. This, paired with the arrival of AMD’s very competitive Navi cards, likely prompted Nvidia to drop the prices of its higher-end chips.

It’s also well-known that the higher-end cards, although not sold in huge volumes, have higher profit margins, meaning Nvidia has more room to play with their pricing or to absorb slightly higher production costs. This is not the case in the mid-tier segment, where the manufacturer’s goal is to sell the cards in large volumes, accepting lower profit margins per card.

Moreover, AMD’s RX 5500 is also coming soon. With AMD expecting its performance to beat the GTX 1650 by a solid margin, we expect the RX 5500 to rival the GTX 1660 – which is probably why Nvidia is launching its GTX 1660 Super – so that Nvidia comes out on top of the performance charts.

A listing on JD.com lists specifications for the unreleased Nvidia GeForce GTX 1660 Super graphics card, as spotted by longtime hardware leaker @momomo_us.

There have many leaks about the GTX 1660 Super already, so expected specs have been out for a while. However, this is the first official listing of a custom model, and it helps corroborate early reports.

Let's start with what hasn't changed. The GTX 1660 Super is expected to retain the same number of CUDA cores, 1,408, as the original GTX 1660. The clock speeds are reportedly the same too, which means we're looking at a 1,530 MHz base clock and 1,785 MHz boost clock. Like the GTX 1660, the GTX 1660 Super's TDP (thermal design power) is rated for 120W, and both use a single 8-pin PCIe power connector, according to the JD listing. 

The most obvious difference between the two graphics cards is the memory type. The GTX 1660 debuted with GDDR5 memory clocked at 2,000 MHz (8,000 MHz effective). The Super version will have seemingly have speedier GDDR6 memory at 1,750 MHz (14,000 MHz effective). Although both graphics cards are still restricted to a 192-bit memory interface, the memory swap should let the GTX 1660 Super deliver 75% more memory bandwidth than its sibling. 

JD lists the Maxsun GeForce GTX 1660 Super Terminator for 1,899 yuan (~$268.36). The non-Super Terminator version sells for 1,499 yuan (~$211.83). That's about a 26.7% premium over the GTX 1660. 

With October 29 being the rumored launch date for the GTX 1660 Super, hopefully we'll soon find out if the upgrade to GDDR6 memory is enough deserve our consideration.

With rumors of Nvidia's GeForce GTX 1660 Super series launch in full force, we're hearing murmurs of third-party GTX 1660 Super graphics cards as well. Today, VideoCardz.com reported that MSI has got two GTX 1660 Super cards in the works, the GTX 1660 Super Ventus XS OC and the GTX 1660 Super Gaming X.

MSI's Gaming X line is its higher-end brand, packing a fancy cooler with RGB lighting and a neat overclock on the GPUs. 

The Ventus XS OC isn't expected to be as fancy, But it will reportedly arrive with a dual-fan cooler and handsome backplate. 

Display outputs for both the cards consist of three DisplayPort connectors and one HDMI connection. Both the cards are powered by a single 8-pin PCIe power connector

That’s all the information we have about these cards for now. There was no information shared on exact clock speeds or similar.

However, Nvidia’s GTX 1660 Super is expected to have 1,408 CUDA cores that run at a boost clock of up to 1,785 MHz, so MSI's two cards will likely clock higher. The reference card will also reportedly pack pack 6GB of GDDR6 memory, as opposed to the 6GB of GDDR5 in the Nvidia GeForce GTX 1660. 

Word around town is that the GeForce GTX 1660 Super will arrive with high-speed GDDR6 memory next month. VideoCardz's leak today seemingly lends credence to the early rumors.

With the exception of the memory upgrade, the GTX 1660 Super is rumored to carry over the same main specs as the GeForce GTX 1660. The updated Super version is expected to use the Turing TU116 die, meaning it should have 1,408 CUDA cores. It's unknown if Nvidia will give the GTX 1660 Super higher operating clocks.

GTX 1660 Super Specs

*Specifications aren't confirmed.

The GTX 1660 Super will reportedly maintain a 6GB memory configuration, but faster. The regular GTX 1660 utilizes GDDR5 memory clocked at 2,000 MHz (8,000 MHz effective) across a 192-bit memory interface. The product has a memory bandwidth up to 192 GBbps.

On the other hand, the GTX 1660 Super purportedly uses GDDR6 memory operating at 1,750 MHz (14,000 MHz effective). Under the same 192-bit memory bus, the GTX 1660 Super could deliver a memory bandwidth of 336 GBps, which is a 75% improvement over the standard GTX 1660. The performance gain won't be as significant as having more CUDA cores, but the memory upgrade should -- or at least Nvidia hopes -- put the graphics card in the same class as the AMD Radeon RX 5500, which targets 1080p gaming. 

VideoCardz posted what it said are product images for the regular and AMP version of the Zotac Gaming GeForce GTX 1660 Super. Visually, both graphics cards resemble their non-Super counterparts. If it wasn't for the Super moniker on the packaging, we could swear we were looking at the GTX 1660.

Zotac's custom GTX 1660 Super offerings seemingly conserve the compact, dual-slot cooling solution, and the AMP version seems to rock the IceStorm 2.0 cooler. We expect the GTX 1660 Super AMP to come with a very generous factory overclock, like other AMP cards. 

The GTX 1660 Super is expected to launch sometime this month, along with the GTX 1650 Super.

Nvidia GeForce GTX 1660 Super cards may be coming in October. A report from Videocardz claims that sources from Asus have confirmed the card is coming, including models from the Dual Evo, Phoenix and Tuf3 lines.

According to the report, the card will have the same GPU as the existing 1660, but with 6GB of GDDR6 memory at 14 Gbps. The existing GTX 1660 has 6GB of GDDR5 at 8 Gbps. This would make the new Super's memory faster than the 1660 Ti with 6GB of GDDR6 at 12 Gbps.

Update: 2:15 p.m. ET: Nvidia told Tom's Hardware that "we do not comment on rumors."

We reached out Asus about this story but did not hear back before publication. This story will be updated if and when the company responds.

Earlier this year, we heard of rumors about a GTX 1650 Ti, but haven't heard anything further on the matter.

Acer's Predator Helios 300 is on sale on Amazon Prime Day for $999, with a GTX 1660 Ti along with an Intel Core i7-9750H and 16GB of DDR4 storage.

This laptop usually goes for $1,199, so there's savings of $200. It's rare to see these specs, including a 144Hz display, for less than $1,000. A lot of deals have also had older GPUs and CPUs, so this is a more future-proof budget notebook.

• See the Acer Predator Helios 300 for $999

For more savings, check out our list of best Amazon Prime Day deals and best pc hardware deals overall as well as dedicated lists of current sales on ssds, cpus, gpus, gaming laptops and Raspberry Pi stuff.

eBay is currently selling the Zotac Gaming GeForce GTX 1660 Ti AMP graphics card for $240, with a 15% discount applied when you add it to your cart. That's down from its typical retail price of around $270. Be quick though; the graphics card is selling fast and has already been in and out of stock a few times.

• Get the Zotac Gaming GeForce GTX 1660 Ti AMP for $240.

The Zotac Gaming GeForce GTX 1660 Ti AMP is a compact dual-slot graphics card that has a 209.6 x 119.3 x 41mm footprint. You should have no problems fitting the graphics card inside your your favorite PC case. It's equipped with Zotac's IceStorm 2.0 cooling solution, which depends on a pair of 90mm cooling fans, and a sleek backplate too.

Based on Nvidia's latest Turing architecture, the Zotac Gaming GeForce GTX 1660 Ti AMP has 1,536 CUDA cores that feature a 1,860 MHz boost clock. The graphics card also has 6GB of GDDR6 memory running at 1,500 MHz (12,000 MHz effective) across a 192-bit memory interface. If you need a reference for performance, the GTX 1660 Ti has been shown to perform on a similar level as the GTX 1070. It delivers an excellent 1920 x 1080 gaming experience and a very acceptable 2560 x 1440 experience.

The Zotac Gaming GeForce GTX 1660 Ti AMP has a power consumption of 130W and requires at least one 8-pin PCIe power cable to function correctly. We recommend you have at least a 450W power supply from a reputable brand to feed this little baby (you can find our recommendations on our Best Power Supplies page). The graphics card has three DisplayPort 1.4 outputs and one HDMI 2.0b port for connecting your monitor.

Should You Buy This Graphics Card?

To see if this graphics card will fit your needs, we recommend checking out our review of the Nvidia GeForce GTX 1660 Ti. You may also want to see our face-off between reference cards and third-party graphics cards.

For more help picking the best graphics card for you, there's our graphics card buying guide, the AMD and Nvidia GPU hierarchy and our breakdown of the best graphics cards we've tested.

Gigabyte GeForce GTX 1660 Gaming OC 6G Review

If you play your favorite games at 1920 x 1080, Nvidia’s GeForce GTX 1660 graphics card offers some of the best performance per dollar out there, along with excellent performance per watt. In our launch review, Nvidia GeForce GTX 1660 Review: The Turing Onslaught Continues, we found Gigabyte’s GeForce GTX 1660 OC 6G to be more than 16% faster than a previous-gen GeForce GTX 1060 6GB at a similar 120W TDP (and a lower price).

It takes a much more expensive Radeon RX Vega 56 to outperform the GeForce GTX 1660, and a slower Radeon RX 580 to rival its value. That’s why we were quick to praise the 1660 for its combination of playable frame rates, conservative power consumption, and a $220 starting price.

Gigabyte’s GeForce GTX 1660 Gaming OC 6G is the company’s flagship implementation of the 1660, offering higher core clock rates, more aggressive cooling, and a bit of RGB flair. Typically, we’d expect those extras to impose a hefty premium. But Gigabyte only asks for an extra $10/£10 compared to its base-level models. That seems like a small price to pay for the benefits of a truly beefed-up board.

Meet Gigabyte’s GeForce GTX 1660 Gaming OC 6G

We don’t often get our hands on multiple models from the same vendor in a given product class. The 1660 Gaming OC 6G is one step up from the 1660 OC 6G we already reviewed though, giving us a nice foundation for quantifying the reasons you’d want to spend a few dollars more.

Right out of the gate, we know that the 1660 Gaming OC 6G sports a GPU Boost rating of 1,860 MHz, which is 75 MHz higher than Nvidia’s reference specification. Its TU116 processor is topped with a larger three-piece sink cooled by a trio of 80mm fans. We expect the big Windforce 3x thermal solution to yield measurable advantages when it comes time to take temperature readings, though it’ll be interesting to see how the 1660 OC 6G’s two 90mm fans stack up in comparison.

You will, of course, need more room in your case to accommodate the 1660 Gaming OC 6G. It’s similar in size to the premium GeForce RTX 2060 Gaming OC Pro 6G and GeForce RTX 2070 Gaming OC 8G, which are more than 11 inches (28cm) long, 4.4 inches (11.3cm) tall, and 1.5 inches (3.9cm) thick.

The aforementioned Windforce 3X cooler hangs off the PCB’s back edge by more than 2”. As you might imagine, a thermal solution designed to cope with RTX 2070’s 175W TDP and RTX 2060’s 160W power limit would be more than ample for GTX 1660’s 120W ceiling. Under the fan shroud, however, Gigabyte does cut costs with a less dense sink that weighs 1.49lb. (677g) instead of the RTX 2060 card’s 1.79lb. (811g). And although both designs employ a multi-part configuration, the 1660 Gaming OC 6G utilizes three heat pipes between them, whereas the higher-end RTX 2060 employs four. Those pipes are flattened in the middle and placed directly over Nvidia’s GPU, where they draw heat away before dissipating it through the fin arrays.

On the cooler’s left side, closest to the display outputs, one part of the sink contacts voltage regulation circuitry underneath through thermal pads. On the right side, another piece of the sink doesn’t touch the board at all. It simply helps the middle section keep TU116 and its surrounding memory modules running at the lowest temperature possible.

Three 80mm fans sit on top of the thermal solution. The outside fans spin counter-clockwise, while the middle fan rotates clockwise. Turbulence is purportedly kept to a minimum, generating less competing airflow from adjacent fans. Then, at idle, the fans stop spinning altogether by virtue of a feature that Gigabyte calls 3D Active Fan. Enthusiasts who prefer to maintain lower idle temperatures can disable 3D Active Fan using downloadable Aorus Engine software. Frankly, the fans make so little noise that we prefer to keep them spinning (even though the semi-passive mode is one of this card’s competitive advantages).

Around the fans, Gigabyte uses plastic gratuitously to form a shroud. Sharp angles and silver accents do look nice, though folds over the card’s top and bottom block air moving through the vertically-oriented fins. A freer-flowing arrangement, particularly above the heat sink, might have helped the already-impressive thermal performance. Instead, that’s where we find this card’s RGB-backlit Gigabyte logo.

Still, the company’s relatively light cooler performs admirably. In Metro: Last Light, we observed the Gaming OC 6G consuming about 129W. Although that’s about 3W more than the 1660 OC 6G, Gigabyte’s flagship model topped out 3°C lower. Moreover, it maintained 1,980 MHz through our run, whereas the base-level card demonstrated a range of 1,920 to 1,935 MHz.

Gigabyte’s 1660 Gaming OC 6G sports a plastic backplate. This plate doesn’t do anything for rigidity, and it certainly doesn’t improve the card’s cooling. The plate may, however, help protect the PCB in case something gets dropped on it.

We approve of Gigabyte’s display output choices. Three DisplayPort connectors and a single HDMI interface cover the most common multi-monitor configurations. Down at this level, VirtualLink support isn’t really necessary. And doing away with DVI yields a freer-flowing grille. Unfortunately, the potential benefit of additional airflow is lost since the cooler’s fins move air perpendicular to the bracket.

The GPU at the heart of GeForce GTX 1660 is specifically named TU116-300-A1. It’s a close relative of the GeForce GTX 1660 Ti’s TU116-400-A1, trimmed from 24 Streaming Multiprocessors to 22. We’re obviously still dealing with a processor devoid of Nvidia’s future-looking RT and Tensor cores, measuring 284mm² and composed of 6.6 billion transistors manufactured using TSMC’s 12nm FinFET process.

Gigabyte takes this chip, with 1,408 CUDA cores, and bumps the reference 1,785 MHz GPU Boost frequency up to 1,860 MHz—a 4% increase. Six 32-bit memory controllers give TU116 an aggregate 192-bit bus, which is populated by 8 Gb/s GDDR5 modules pushing up to 192 GB/s. That’s comparable to GeForce GTX 1060 6GB, and a 33% reduction compared to GeForce GTX 1660 Ti. Combined with the loss of two SMs, dropping from GDDR6 to GDDR5 memory accounts for GeForce GTX 1660’s lower performance versus 1660 Ti.

Each memory controller is associated with eight ROPs and a 256KB slice of L2 cache. In total, TU116 exposes 48 ROPs and 1.5MB of L2. GeForce GTX 1660’s ROP count compares favorably to RTX 2060, which also utilizes 48 render outputs. But TU116’s L2 cache slices are half as large compared to TU106.

All of Gigabyte's graphics cards include three years of warranty coverage, and some of them add a fourth year when you register on the company's website. The GeForce GTX 1660 Gaming OC 6G is not one of those special models. In its segment, however, three years of coverage is fairly standard.

How We Tested Gigabyte’s GeForce GTX 1660 Gaming OC 6G

Obviously, GeForce GTX 1660 is more mainstream than the other Turing-based cards we’ve reviewed. As such, our graphics workstation, based on an MSI Z170 Gaming M7 motherboard and Intel Core i7-7700K CPU at 4.2 GHz, is apropos. The processor is complemented by G.Skill’s F4-3000C15Q-16GRR memory kit. Crucial’s MX200 SSD is included, joined by a 1.6TB Intel DC P3700 loaded down with games.

As far as competition goes, the 1660 mostly goes up against GeForce GTX 1060 6GB, though we include GeForce GTX 1070 and 1070 Ti as well. All of those cards are included in our line-up, along with GeForce RTX 2060 and GeForce RTX 2070. On the AMD side, we’re mostly interested in Radeon RX 590, although Radeon RX Vega 64 and Radeon RX Vega 56 make for interesting additions, too.

Our benchmark selection includes Ashes of the Singularity: Escalation, Battlefield V, Destiny 2, Far Cry 5, Forza Horizon 4, Grand Theft Auto V, Metro: Last Light Redux, Shadow of the Tomb Raider, Tom Clancy’s The Division 2, Tom Clancy’s Ghost Recon Wildlands, The Witcher 3 and Wolfenstein II: The New Colossus.

The testing methodology we're using comes from PresentMon: Performance In DirectX, OpenGL, And Vulkan. In short, these games are evaluated using a combination of OCAT and our own in-house GUI for PresentMon, with logging via GPU-Z.

We're using driver version 430.39 to test Gigabyte’s GeForce GTX 1660 and build 417.54 for everything else. AMD’s cards utilize Crimson Adrenalin 2019 Edition 18.12.3.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Gaming at 1920 x 1080

According to Nvidia, the GeForce GTX 1660 Ti we were introduced to back in February was designed for high frame rates in popular games at 1920 x 1080. Naturally, then, GeForce GTX 1660 is meant to drive the same resolution…just a bit slower.

Despite its trimmed-down specifications, the 1660 is still about four percent faster than Radeon RX 590 and 16% quicker than GeForce GTX 1060 6GB, on average. Meanwhile, it offers roughly 88% of the GeForce GTX 1660 Ti’s frame rate through our benchmark suite. With the exception of Metro: Last Light, where we enable SSAA, and Ashes of the Singularity, where we dial in the Crazy quality preset, GeForce GTX 1660 can be counted on to average 60 FPS or more in the games we test (almost all of which run at their top detail settings).

There’s very little performance difference between Gigabyte’s GeForce GTX 1660 Gaming OC 6G and GeForce GTX 1660 OC 6G, though. If you’re going to spend an extra $10 on the Gaming OC 6G model, let it be for the beefier cooling or RGB lighting, not a higher GPU Boost rating.

Ashes of the Singularity: Escalation (DX12)

Battlefield V (DX12)

Destiny 2 (DX11)

Far Cry 5 (DX11)

Forza Horizon 4 (DX12)

Grand Theft Auto V (DX11)

Metro: Last Light Redux (DX11)

Shadow of the Tomb Raider (DX12)

Tom Clancy’s The Division 2 (DX12)

Tom Clancy’s Ghost Recon (DX11)

The Witcher 3 (DX11)

Wolfenstein II: The New Colossus (Vulkan)

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Gaming at 2560 x 1440

We established this in our GeForce GTX 1660 OC 6G review, but when you pay $220 (or $230) for a graphics card, you can’t really expect smooth frame rates at 2560 x 1440. It’s certainly possible to achieve playable performance by dialing back graphics quality a notch or two. But if you’re gaming on a 2560 x 1440 monitor, consider something in the Radeon RX Vega 56/GeForce GTX 1070 Ti/GeForce RTX 2060 range at least.

Ashes of the Singularity: Escalation (DX12)

Battlefield V (DX12)

Destiny 2 (DX11)

Far Cry 5 (DX11)

Forza Horizon 4 (DX12)

Grand Theft Auto V (DX11)

Metro: Last Light Redux (DX11)

Shadow of the Tomb Raider (DX12)

Tom Clancy’s The Division 2 (DX12)

Tom Clancy’s Ghost Recon (DX11)

The Witcher 3 (DX11)

Wolfenstein II: The New Colossus (Vulkan)

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Power Consumption

Slowly but surely, we’re spinning up multiple Tom’s Hardware labs with Cybenetics’ Powenetics hardware/software solution for accurately measuring power consumption.

Powenetics, In Depth

For a closer look at our U.S. lab’s power consumption measurement platform, check out Powenetics: A Better Way To Measure Power Draw for CPUs, GPUs & Storage.

In brief, Powenetics utilizes Tinkerforge Master Bricks, to which Voltage/Current bricklets are attached. The bricklets are installed between the load and power supply, and they monitor consumption through each of the modified PSU’s auxiliary power connectors and through the PCIe slot by way of a PCIe riser. Custom software logs the readings, allowing us to dial in a sampling rate, pull that data into Excel, and very accurately chart everything from average power across a benchmark run to instantaneous spikes.

The software is set up to log the power consumption of graphics cards, storage devices, and CPUs. However, we’re only using the bricklets relevant to graphics card testing. Gigabyte's GeForce GTX 1660 Gaming OC 6G gets all of its power from the PCIe slot and one eight-pin PCIe connector.

Idle

Gigabyte’s GeForce GTX 1660 Gaming OC 6G uses a couple more watts at idle than the company’s 1660 OC 6G. Both cards employ a passive fan mode, so the difference must be attributable to the higher-end model’s RGB lighting.  

Gaming: Metro: Last Light

The 1660 Gaming OC 6G also uses more power through our Metro benchmark sequence, though only by a few watts. Despite a significantly different fan configuration (3x 80mm versus 2x 90mm), both boards behave similarly.

Charting power consumption across our three-run test illustrates quite a bit more variance in the eight-pin auxiliary cable’s dips and peaks compared to GeForce GTX 1660 OC 6G. This behavior is repeatable, too—we see the same results each time the Metro benchmark plays through.

But overlaying Gigabyte’s 1660 Gaming OC 6G over other cards in its class assures us that power consumption isn’t something to worry about. The line charts all overlap quite a bit.

Gigabyte keeps current draw over the PCIe slot under 2A, well below the PCI-SIG’s ceiling.

FurMark

The 1660 Gaming OC 6G’s power consumption in FurMark is 13W higher than Gigabyte’s 1660 OC 6G, most of which is delivered by the eight-pin auxiliary connector.

There’s no room for performance to vary in FurMark, so the wavy line chart we saw under Metro: Last Light is gone, replaced by much more consistent power consumption through more than 15 minutes of recording.

Now, the 1660 Gaming OC 6G sits quite a ways above the comparison cards, including Gigabyte’s own GeForce GTX 1660 OC 6G. It acts more like a GeForce GTX 1070 through our test (though it’s nowhere near as fast).

Current draw from the PCIe slot exceeds 2A but remains well below its rated limit.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Temperatures, Fan Speeds and Clock Rates

Gaming: Metro: Last Light

Although the GeForce GTX 1660 Gaming OC 6G uses smaller fans than the 1660 OC 6G, the fact that it sports three of them means they spin slower under load. After gently pulsing to life, Gigabyte’s Windforce 3X cooler ramps up gracefully, peaking just over 1,900 RPM.

The 1660 Gaming OC 6G’s big cooler helps keep temperatures down, even as its fans spin slower than what we see from competing cards. TU116 tops out at 67°C—one degree lower than Gigabyte’s 1660 OC 6G. As a point of reference, last generation’s GeForce GTX 1060 Founders Edition tops out at 73°C in the same workload.

For anyone keeping score, the GeForce GTX 1660 Gaming OC 6G maintains higher clock rates than the 1660 OC 6G through our real-world gaming workload while its TU116 GPU operates at lower temperatures under a heat sink topped with slower-spinning fans.

FurMark

Firing up a synthetic stress test puts more pressure on the 1660 Gaming OC 6G’s cooling solution. Its fans crest above 2,000 RPM to help cope with the added heat of a few more watts.

The faster-spinning fans keep GPU temperatures under control. Gigabyte’s 1660 Gaming OC 6G doesn’t run much hotter in FurMark than it did under Metro, topping out at 68°C. Only EVGA’s GeForce GTX 1660 Ti XC Black Gaming is cooler, and that’s because EVGA’s one fan rotates at a comparatively blistering 2,250 RPM.

Gigabyte’s GeForce GTX 1660s maintain aggressive frequencies through 15 minutes of FurMark. And as we’d expect, the higher-end 1660 Gaming OC 6G enjoys a quantifiable advantage versus the base-level model.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Software

Gigabyte’s software bundle includes one utility for overclocking and fan control called Aorus Engine, and another one for lighting called RGB Fusion 2.0. The latest version of Aorus Engine at the time of writing was 1.60.

On some cards, Aorus Engine allows you to toggle between different presets, optimizing for performance or acoustics. The GeForce GTX 1660 Gaming OC 6G only gave us the options typically available in “Professional Mode,” though. Custom GPU Boost settings, memory clock rates, GPU voltage adjustments, fan speeds, power targets and target temperatures are all configurable.

A set of icons in the bottom-right allow you to turn the 3D Active Fan (semi-passive mode) feature on or off, open the hardware monitoring pane, or set up RGB Fusion 2.0.

RGB Fusion 2.0 opens in a separate window. The software makes it possible to synchronize the Gaming OC 6G with other compatible products like motherboards and peripherals. If you don’t already have any of those complementary extras, this card’s effects come across as underwhelming since only the Gigabyte logo up top illuminates. But we can see how the subtle lighting would be additive in a case full of other RGB Fusion-capable components.

At the time of writing, Gigabyte’s GeForce GTX 1660 Gaming OC 6G also included a Fortnite bundle with 2,000 V-Bucks and a Fortnite counterattack set. According to Newegg reviews, the cost of the bundle is discounted from the card’s price at checkout. We don’t put any additional value on the card based on this promotion.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Conclusion

GeForce GTX 1660 represents one of the best values available for gaming at 1920 x 1080. It’s certainly the most efficient, offering excellent performance per watt. And with a starting price of $220/£200, we have a hard time recommending that you spend more money on the GeForce GTX 1660 Ti.

You might say that we’re especially cost-conscious down at the mainstream price points, where even a few dollars can tip the scales in favor of one product or another. This would seemingly put Gigabyte’s GeForce GTX 1660 Gaming OC 6G at a big disadvantage, since it’s the most opulent implementation of GeForce GTX 1660 in the company’s portfolio. But Gigabyte only charges a $10 premium for its flagship compared to the entry-level 1660 OC 6G. Stepping up gets you higher clock rates, a beefier thermal solution, and RGB lighting.

The overclocked GPU frequency didn’t do much in our benchmarks, so that’s a bit of a gimmick. However, the Windforce 3X cooler made a more appreciable difference, enabling slower-spinning fans and lower temperatures in real-world games and synthetic stress tests. Although RGB lighting is a matter of personal taste, the 1660 Gaming OC 6G has it while the 1660 OC 6G does not. So, if you own other RGB Fusion-compatible components, its inclusion undoubtedly means something to you. As we do every time we review a Gigabyte graphics card, we played with the software and verified that all of its effects, colors, and settings function the way they’re supposed to.

As of this writing, AMD’s Radeon RX 580 is quite a bit less expensive than any GeForce GTX 1660, including the 1660 Gaming OC 6G. It’s also slower and way more power-hungry. But while we’re not sure how much longer the fire sale on Radeon RX 580s will last, we’d be remiss in ignoring it. The AMD card is worth considering if you’re satisfied with its performance at 1920 x 1080 and ambivalent about its 185W TDP.

Mainstream gamers willing to pay for slightly higher frame rates and better efficiency should skip straight to GeForce GTX 1660, though. And although we’re perfectly happy recommending one of the many $220 models, Gigabyte’s GeForce GTX 1660 Gaming OC 6G is improved enough that its modest $10 premium doesn’t bother us.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

When the Nvidia GeForce GTX 1660 Ti first came out, Asus released four different versions of the graphics card with the familiar ROG Strix, TUF Gaming, Phoenix and Dual branding. The three new Asus Dual GeForce GTX 1660 Ti Evo graphics cards unveiled this week feature a different cooling mechanism, as well as the Axial-tech fan design found in their high-end ROG video cards. The Axial-tech fans use longer blades and a barrier ring said to increase downward air pressure while improving cooling and lowering noise levels. 

The fans push air through a newly designed heatsink that is “2.7” slots wide to increase the surface area for lower fan speeds at light loads and increased overclocking potential, via more thermal headroom. The black plastic shroud on the heatsink includes one RGB LED illumination element, a single line on the top of the card located on top of the card where the heat pipes stick out. The implementation looks tasteful for those who may not like the disco light show many of today's cards cast. Last, the PCB uses an aluminum back plate to add rigidity, as well as a physical barrier to protect the back of the cards. 

The Axial-tech fans use a smaller hub to make room for the longer blades. There's also a ring around the outside connecting the edges of the fan blades, which Asus says increases air pressure. If noise is a concern, particularly while at the desktop or light loads, the cards all feature 0dB technology, which disables the fans when the graphics card's temperature is under 55 degrees Celsius.

The new Dual Evo Series is hitting the scene in three flavors. The base model is clocked at 1,500 / 1,770 MHz (base / boost clocks, respectively), the Asus Dual GeForce GTX 1660 Ti EVO Advanced Edition is clocked at 1,500 / 1,785 MHz and the fastest of the three, the Asus Dual GeForce GTX 1660 Ti EVO OC Edition, runs at 1500/1845 MHz. Memory speeds on all three cards come in at 12,002 MHz. If the stock clocks are not enough, all three cards offer an overclock mode with slightly increased core speeds. A single 8-pin PCIe connector feeds the cards power.  

All Dual GTX 1660 Ti Evo cards use a single DisplayPort, two HDMI ports and a DVI-D connector for outputs to your monitor. The I/O plate has large holes to help exhaust air out of your PC case.

Neither pricing nor availability was listed. However, we expect pricing to small somewhere between that of the ROG Strix and Dual GTX 1660 Ti series' pricing.

Photo Credit: Asus

Gigabyte GeForce GTX 1660 Ti Gaming OC 6G Review

Nvidia’s GeForce GTX 1660 Ti was the first Turing-based graphics card lacking the architecture’s signature RT and Tensor cores. Instead, it relied exclusively on performance and efficiency improvements in today’s games to stand up against AMD’s Radeon RX 590 and the previous-generation GeForce GTX 1070. With prices starting at $280, the 1660 Ti certainly offered more value than any of the GeForce RTX cards. But the subsequent introduction of GeForce GTX 1660 showed that gamers could spend even less (as little as $220) and still get great frame rates at 1920 x 1080 with quality settings cranked up.

Now, Gigabyte’s GeForce GTX 1660 Ti Gaming OC 6G challenges us to think in the other direction. Is it worth spending a bit more than the 1660 Ti’s starting price for slightly higher frame rates on high-refresh monitors, over-built cooling, a 200W maximum power limit, and a bit of RGB lighting?

Meet Gigabyte’s GeForce GTX 1660 Ti Gaming OC 6G

Our GeForce GTX 1660 Ti launch review was based on EVGA’s XC Black Gaming model, which represents the 1660 Ti’s base price and 1,770 MHz GPU Boost reference spec. Gigabyte’s Gaming OC 6G, on the other hand, is the company’s highest-end version. Its $300 price tag pays for an 1,860 MHz GPU Boost rating.

Further down the stack, GeForce GTX 1660 Ti Windforce OC 6G sells for $290. A small form factor-oriented Mini ITX OC 6G is listed at $280. And the entry-level OC 6G version sports a similar $280 price.

Aside from a more aggressive GPU Boost frequency, why else would you pay an extra $20 for the Gaming OC 6G? The most obvious reason is its oversized PCB and thermal solution. From the front, Gigabyte’s flagship GeForce GTX 1660 Ti looks a lot like its premium GeForce RTX 2060 Gaming OC Pro 6G and GeForce RTX 2070 Gaming OC 8G. They’re all more than 11” (28cm) long, 4.4” (11.3cm) tall, and 1.5” (3.9cm) thick.

Much of this design’s extra length is attributable to the familiar Windforce 3X cooler, which hangs off the PCB’s back edge by more than 2”. As you might imagine, a thermal solution designed to cope with RTX 2070’s 175W TDP and RTX 2060’s 160W power limit would be more than ample for GTX 1660 Ti’s 120W ceiling. Under the fan shroud, however, Gigabyte does cut costs with a less dense sink that weighs 1.49lb. (674g) instead of the RTX 2060 card’s 1.79lb. (811g). And although both designs employ a multi-part configuration, the 1660 Ti Gaming OC 6G utilizes three heat pipes between them, whereas the RTX 2060 employs four. Those pipes are flattened in the middle and placed directly over Nvidia’s GPU, where they draw heat away before dissipating it through the fin arrays.

On the cooler’s left side, closest to the display outputs, one part of the sink contacts voltage regulation circuitry underneath through thermal pads. On the right side, another piece of the sink doesn’t touch the board at all. It simply helps the middle section keep TU116 and its surrounding memory modules running at the lowest temperature possible.

Three 80mm fans sit on top of the thermal solution. The outside fans spin counter-clockwise, while the middle fan rotates clockwise. Turbulence is purportedly kept to a minimum, generating less competing airflow from adjacent fans. Then, at idle, the fans stop spinning altogether by virtue of a feature that Gigabyte calls 3D Active Fan. Enthusiasts who prefer to maintain lower idle temperatures can disable 3D Active Fan using downloadable Aorus Engine software. Frankly, the fans make so little noise that we prefer to keep them spinning (even though the semi-passive mode is one of this card’s competitive advantages).

Around the fans, Gigabyte uses plastic gratuitously to form a shroud. Sharp angles and silver accents do look nice, though folds over the card’s top and bottom block air moving through the vertically-oriented fins. A freer-flowing arrangement, particularly above the heat sink, might have helped the already-impressive thermal performance. Instead, that’s where we find this card’s RGB-backlit Gigabyte logo.

Regardless, the company’s relatively light cooler performs admirably. In Metro: Last Light, we observed the Gaming OC 6G consuming about 133W. Although that’s about 9W more than EVGA’s GeForce GTX 1660 Ti XC Black Gaming, Gigabyte’s card ran a couple of degrees cooler. Moreover, it maintained between 1,890 and 1,920 MHz through our run, whereas EVGA's XC Black Gaming demonstrated a range of 1,785 to 1,815 MHz.

Gigabyte’s 1660 Ti Gaming OC 6G sports a plastic backplate. This plate doesn’t do anything for rigidity, and it certainly doesn’t improve the card’s cooling. The plate may, however, help protect the PCB in case something gets dropped on it.

We approve of Gigabyte’s display output choices. Three DisplayPort connectors and a single HDMI interface cover the most common multi-monitor configurations. Down at this level, VirtualLink support isn’t really necessary. And doing away with DVI yields a freer-flowing grille. Unfortunately, the potential benefit of additional airflow is lost since the cooler’s fins move air perpendicular to the bracket.

What lives under the GeForce GTX 1660 Ti Gaming OC 6G’s hood is already well-known. We dissected the TU116 graphics processor in Nvidia GeForce GTX 1660 Ti 6GB Review: Turing Without The RTX. Gigabyte takes the same chip, with 1,536 CUDA cores, and bumps the typical GPU Boost rating up to 1,860 MHz—a 5% increase compared to the 1,770 MHz we tested from EVGA’s XC Black Gaming board.

The Gigabyte card's 6GB of GDDR6 memory moves data at 14 Gb/s, yielding 288 GB/s of aggregate bandwidth on a 192-bit bus.

All of Gigabyte's graphics cards include three years of warranty coverage, and some of them add a fourth year when you register on the company's website. The GeForce GTX 1660 Ti Gaming OC 6G is not one of those special models. In its segment, however, three years of coverage is fairly standard.

How We Tested Gigabyte’s GeForce GTX 1660 Ti Gaming OC 6G

Obviously, GeForce GTX 1660 Ti is more mainstream than the other Turing-based cards we’ve reviewed. As such, our graphics workstation, based on an MSI Z170 Gaming M7 motherboard and Intel Core i7-7700K CPU at 4.2 GHz, is apropos. The processor is complemented by G.Skill’s F4-3000C15Q-16GRR memory kit. Crucial’s MX200 SSD is here, joined by a 1.6TB Intel DC P3700 loaded down with games.

As far as competition goes, the 1660 Ti mostly goes up against GeForce GTX 1070, though we include 1070 Ti as well. Of course, comparisons to GeForce GTX 1060 are inevitable. All of those cards are included in our line-up, along with GeForce RTX 2060 and GeForce RTX 2070. On the AMD side, we’re mostly interested in Radeon RX 590, although Radeon RX Vega 64 and Radeon RX Vega 56 make for interesting additions, too.

Our benchmark selection includes Ashes of the Singularity: Escalation, Battlefield V, Destiny 2, Far Cry 5, Forza Horizon 4, Grand Theft Auto V, Metro: Last Light Redux, Shadow of the Tomb Raider, Tom Clancy’s The Division 2, Tom Clancy’s Ghost Recon Wildlands, The Witcher 3 and Wolfenstein II: The New Colossus.

The testing methodology we're using comes from PresentMon: Performance In DirectX, OpenGL, And Vulkan. In short, these games are evaluated using a combination of OCAT and our own in-house GUI for PresentMon, with logging via GPU-Z.

We're using driver version 430.39 to test Gigabyte’s GeForce GTX 1660 Ti, 418.91 to test EVGA’s competing 1660 Ti, and build 417.54 for everything else. AMD’s cards utilize Crimson Adrenalin 2019 Edition 18.12.3.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: How to Stress-Test Graphics Cards (Like We Do)

MORE: All Graphics Content

Gaming at 1920 x 1080

By Nvidia’s own suggestion, GeForce GTX 1660 Ti is designed for high performance at 1920 x 1080 with anti-aliasing enabled. Although we found the card’s frame rates to be perfectly playable using higher resolutions, an emphasis on 120+ FPS would keep most gamers using FHD displays at this price point.

On average, EVGA’s base-level GeForce GTX 1660 Ti lands a couple of percentage points away from GeForce GTX 1070 across our suite. We’d make the claim that newer games put the two cards closer together in the benchmarks. However, GeForce GTX 1660 Ti is slightly faster in The Witcher 3, and that’s an almost four-year-old title.

Gigabyte’s GeForce GTX 1660 Ti Gaming OC 6G is clearly faster than the EVGA model, though not in any meaningful way. A few frames per second here or there won’t change the way you set up any of your favorite games.

AMD’s Radeon RX Vega 56 beats the GeForce GTX 1660 Ti on average. What’s more, the least-expensive versions now start at $300. Unless you’re hyper-focused on power consumption or efficiency, that makes it a little more difficult to get behind a premium 1660 Ti card like the Gaming OC 6G.

Ashes of the Singularity: Escalation (DX12)

Battlefield V (DX12)

Destiny 2 (DX11)

Far Cry 5 (DX11)

Forza Horizon 4 (DX12)

Grand Theft Auto V (DX11)

Metro: Last Light Redux (DX11)

Shadow of the Tomb Raider (DX12)

Tom Clancy’s The Division 2 (DX12)

Tom Clancy’s Ghost Recon (DX11)

The Witcher 3 (DX11)

Wolfenstein II: The New Colossus (Vulkan)

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Gaming at 2560 x 1440

We also ran the GeForce GTX 1660 Ti Gaming OC 6G through our suite at 2560 x 1440. In some games, quality settings that worked well at 1920 x 1080 were still viable. In others, we needed to dial back the detail.

Again, although Gigabyte’s card does outperform our EVGA sample from the 1660 Ti launch, paying 7% more for the Gaming OC 6G doesn’t affect playability in an appreciable way. If you can forgive its egregious use of power, AMD’s Radeon RX Vega 56 is definitely better able to handle gaming at 2560 x 1440 for the same amount of money.

Ashes of the Singularity: Escalation (DX12)

Battlefield V (DX12)

Destiny 2 (DX11)

Far Cry 5 (DX11)

Forza Horizon 4 (DX12)

Grand Theft Auto V (DX11)

Metro: Last Light Redux (DX11)

Shadow of the Tomb Raider (DX12)

Tom Clancy’s The Division 2 (DX12)

Tom Clancy’s Ghost Recon (DX11)

The Witcher 3 (DX11)

Wolfenstein II: The New Colossus (Vulkan)

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Power Consumption

Slowly but surely, we’re spinning up multiple Tom’s Hardware labs with Cybenetics’ Powenetics hardware/software solution for accurately measuring power consumption.

Powenetics, In Depth

For a closer look at our U.S. lab’s power consumption measurement platform, check out Powenetics: A Better Way To Measure Power Draw for CPUs, GPUs & Storage.

In brief, Powenetics utilizes Tinkerforge Master Bricks, to which Voltage/Current bricklets are attached. The bricklets are installed between the load and power supply, and they monitor consumption through each of the modified PSU’s auxiliary power connectors and through the PCIe slot by way of a PCIe riser. Custom software logs the readings, allowing us to dial in a sampling rate, pull that data into Excel, and very accurately chart everything from average power across a benchmark run to instantaneous spikes.

The software is set up to log the power consumption of graphics cards, storage devices, and CPUs. However, we’re only using the bricklets relevant to graphics card testing. Gigabyte's GeForce GTX 1660 Ti Gaming OC 6G gets all of its power from the PCIe slot and one eight-pin PCIe connector.

Idle

The overclocked GeForce GTX 1660 Ti Gaming OC 6G uses an extra 1W at idle compared to EVGA’s GeForce GTX 1660 Ti XC Black Gaming. Still, that’s a big improvement compared to when the Turing architecture first launched.

Gaming: Metro: Last Light

Gigabyte’s card uses about 9W more than EVGA’s base-level GeForce GTX 1660 Ti through three runs of the Metro: Last Light benchmark. It also peaks nearly 12W higher.

Both companies use the PCI Express slot sparingly, opting instead to utilize the eight-pin auxiliary connector most heavily.

Charting power consumption across our three-run benchmark shows the narrow range we’re working with. For the most part, GeForce GTX 1660 Ti behaves itself, though Gigabyte’s flagship 1660 Ti demonstrates power dips and spikes more often than EVGA’s 1660 Ti XC Black Gaming.

GeForce GTX 1660 Ti uses slightly more power than GeForce GTX 1060 Founders Edition, despite their similar board power ratings. Meanwhile, GeForce GTX 1070, which is the closer performance match, hovers around 150W.

We also see the difference between Gigabyte’s 1660 Ti (red) and EVGA’s (blue). A higher power limit facilitates better performance, but simultaneously dissipates more heat.

Current draw over the PCIe slot varies a bit between 2A and 3A. However, the instantaneous dips and spikes are fairly small. In comparison, GeForce RTX 2060 pulls closer to 5A from its 16-lane slot, demonstrating a much straighter line across our three-loop Metro benchmark.

There’s quite a bit more noise through the eight-pin connector’s line. Average draw on that rail lands between 8A and 9A.

FurMark

FurMark is a steadier workload, resulting in less variation across our test run. Average power is almost identical, while less room for dramatic swings brings the peak reading down to about 138W.

There’s no room for performance to vary in FurMark, so the wavy line chart we saw under Metro: Last Light is gone, replaced by much more consistent power consumption through 15 minutes of recording.

As a result of an aggressive GPU Boost rating, Gigabyte’s GeForce GTX 1660 Ti Gaming OC 6G uses more power than any other card in our comparison, except for the 1070 Founders Edition. But because the Gigabyte card is also generally faster than the 1070, GeForce GTX 1660 Ti reasserts itself as the more efficient solution.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Temperatures, Fan Speeds and Clock Rates

Gaming: Metro: Last Light

The Gigabyte card’s king-sized thermal solution keeps Nvidia’s TU116 processor cool without requiring the fans to spin quickly. We observe a brief period of passive operation, a quick pulse as the fans jump into action, and a slow ramp that crests under 1,900 RPM.

In comparison, the competing EVGA GeForce GTX 1660 Ti XC Black Gaming gets all the way to 2,250 RPM because of its smaller sink and single fan responsible for dissipating heat.

Despite the faster-spinning fan, EVGA’s 1660 Ti doesn’t stay as cool as Gigabyte’s. Our GeForce GTX 1660 Ti Gaming OC 6G operates at up to 65°C, whereas the XC Black Gaming hits 67°C.

Nvidia’s GeForce GTX 1070 Founders Edition, which offers comparable performance to the 1660 Ti, nearly rises to 80°C through the same workload.

Gigabyte’s GeForce GTX 1660 OC 6G hits the highest clock rates—a necessity, given the number of on-die resources that Nvidia disables.

But when it comes to 1660 Ti cards, a glance at the red and blue lines shows how much more aggressive Gigabyte gets with its flagship model than EVGA can be at the baseline. This takes a slightly higher voltage, which is allowed by a more generous power limit.

FurMark

Even under the duress of FurMark, the GeForce GTX 1660 Ti Gaming OC 6G doesn’t break a sweat. Its ~1,890 RPM fan speed looks a lot like what we saw while running Metro.

EVGA’s single fan must spin a lot faster to keep up, but both GeForce GTX 1660 Tis settle at fairly similar temperatures after 15 minutes of running FurMark.

FurMark forces Gigabyte’s top-end GeForce GTX 1660 Ti well below its rated GPU Boost frequency. Still, the card’s roughly 1,665 MHz is better than the ~1,575 MHz EVGA maintains at the end of our FurMark sequence.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Software

Gigabyte’s software bundle includes one utility for overclocking and fan control called Aorus Engine, and another one for lighting called RGB Fusion 2.0. The latest version of Aorus Engine at the time of writing was 1.60.

On some cards, Aorus Engine allows you to toggle between different presets, optimizing for performance or acoustics. The GeForce GTX 1660 Ti Gaming OC 6G only gave us the options typically available in “Professional Mode,” though. Custom GPU Boost settings, memory clock rates, GPU voltage adjustments, fan speeds, power targets and target temperatures are all configurable.

We used Aorus Engine to peg the Gaming OC 6G card at a +143% power target (200W), which allowed the card to hold 1,920 MHz through our Metro: Last Light benchmark. Attempts to hold clock rates in excess of 2 GHz resulting in lock-ups, though.

A set of icons in the bottom-right allow you to turn the 3D Active Fan (semi-passive mode) feature on or off, open the hardware monitoring pane, or set up RGB Fusion 2.0.

RGB Fusion 2.0 opens in a separate window. The software makes it possible to synchronize the Gaming OC6G with other compatible products like motherboards and peripherals. If you don’t already have any of those complementary extras, this card’s effects come across as underwhelming since only the Gigabyte logo up top illuminates. But we can see how the subtle lighting would be additive in a case full of other RGB Fusion-capable components.

At the time of writing, Gigabyte’s GeForce GTX 1660 Ti Gaming OC 6G also included a Fortnite bundle with 2,000 V-Bucks and a Fortnite counterattack set. According to Newegg reviews, the cost of the bundle is discounted from the card’s price at checkout. We don’t put any additional value on the card based on this promotion.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Conclusion

GeForce GTX 1660 Ti starts at $280/£220. Gigabyte’s GeForce GTX 1660 Ti Gaming OC 6G costs an extra $20/£16. For that small premium, you get higher clock rates, a more flexible power ceiling, an extra-large thermal solution that facilitates lower fan speeds and temperatures, a bit of customizable RGB lighting, and plenty of display outputs. On the surface, that seems like a perfectly reasonable list of value-added extras.

Here’s the thing, though: down where GeForce GTX 1660 Ti does battle, the competition is much more intense than what Nvidia’s GeForce RTX cards face.

When the GeForce GTX 1660 Ti first launched, we looked at EVGA’s XC Black Gaming version of the card and observed that it bested the more expensive (and power-hungry) GeForce GTX 1070, along with AMD’s Radeon RX 590. It also served up better value than the remaining GeForce GTX 1060s and higher-end Radeon RX Vega 56. But today’s landscape looks a little different. The 1060s are almost gone, the remaining 1070s make no sense at all, Radeon RX 590s now start in the $210 range ($50 cheaper than before), and Vega 56 boards dip as low as $300.

In that context, asking an extra $20 for a premium GeForce GTX 1660 Ti may be difficult to justify. After all, the Vega 56 is faster, and Radeon RX 590 is quite a bit cheaper. Nvidia recently followed the 1660 Ti up with a vanilla 1660, compounding the Ti’s predicament by starting down at $220 and still serving up playable FHD performance. Or, for $50 more, you can get your hands on a GeForce RTX 2060 with dedicated RT and Tensor cores, plus enough performance in today’s games to make 2560 x 1440 a playable resolution.

In the end, then, GeForce GTX 1660 Ti’s biggest victory happens when you focus most intently on performance per watt and efficiency. The competition from AMD can’t touch it in those disciplines. If you instead turn to absolute performance, the Radeon RX Vega 56 is the faster choice, despite gobbling up more than 200W of power. Or, if you’re on a budget, the Radeon RX 580 sells for less than $200 and fares really well at 1920 x 1080.

Gigabyte’s GeForce GTX 1660 Ti Gaming OC 6G is built well. Its big heat sink and Windforce 3X thermal solution work in relative slow motion to keep the TU116 processor operating at low temperatures. We prefer its dual-slot form factor to EVGA’s triple-slot design and appreciate the uncorked 200W power limit. With all of that said, we’d love to see its price come down a bit to contend with faster Radeon RX Vega 56 cards selling for the same price.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: How to Stress-Test Graphics Cards (Like We Do)

MORE: All Graphics Content

Image Credits: Tom's Hardware, Nvidia

Nvidia GeForce GTX 1660 Review

It was only a matter of time before Nvidia took the pristine TU116 graphics processor in its GeForce GTX 1660 Ti and carved it up a bit to create a lower-cost derivative. The new GeForce GTX 1660 is, unsurprisingly, very similar to the higher-end model in that it lacks the Turing architecture’s signature RT and Tensor cores. Instead, it aims on-die resources at accelerating today’s rasterized games.

Nvidia doesn’t even cut much from TU116's resource pool in the creation of GeForce GTX 1660: a pair of Streaming Multiprocessors are excised, taking 128 CUDA cores and eight texture units with them. But the GPU is otherwise quite complete. This card’s biggest loss is its lack of GDDR6 memory. By swapping in 8 Gb/s GDDR5 instead, bandwidth drops from the 1660 Ti’s 288 GB/s to a mere 192 GB/s.

Naturally, the GeForce GTX 1660 is primarily aimed at FHD gaming, where 6GB of slower memory won’t hurt performance as much as it would at higher resolutions. Can the $220/£200 board maintain fast enough frame rates to stave off AMD’s Radeon RX 590 with more GDDR5 on a wider bus, though?

TU116 Recap: Turing Without the RT and Tensor Cores

The GPU at the heart of GeForce GTX 1660 is specifically named TU116-300-A1. It’s a close relative of the GeForce GTX 1660 Ti’s TU116-400-A1, trimmed from 24 Streaming Multiprocessors to 22. We’re obviously still dealing with a processor devoid of Nvidia’s future-looking RT and Tensor cores, measuring 284mm² and composed of 6.6 billion transistors manufactured using TSMC’s 12nm FinFET process.

Despite its smaller transistors, TU116 is 42 percent larger than the GP106 processor that preceded it. Some of that growth is attributable to the Turing architecture’s more sophisticated shaders. Like the higher-end GeForce RTX 20-series cards, GeForce GTX 1660 supports simultaneous execution of FP32 arithmetic instructions, which constitute most shader workloads, and INT32 operations (for addressing/fetching data, floating-point min/max, compare, etc.). When you hear about Turing cores achieving better performance than Pascal at a given clock rate, this capability largely explains why.

Turing’s Streaming Multiprocessors are composed of fewer CUDA cores than Pascal’s, but the design compensates in part by spreading more SMs across each GPU. The newer architecture assigns one scheduler to each set of 16 CUDA cores (2x Pascal), along with one dispatch unit per 16 CUDA cores (same as Pascal). Four of those 16-core groupings comprise the SM, along with 96KB of cache that can be configured as 64KB L1/32KB shared memory or vice versa, and four texture units. Because Turing doubles up on schedulers, it only needs to issue an instruction to the CUDA cores every other clock cycle to keep them full. In between, it's free to issue a different instruction to any other unit, including the INT32 cores.

In TU116, Nvidia replaces Turing’s Tensor cores with 128 dedicated FP16 cores per SM, which allow GeForce GTX 1660 to process half-precision operations at 2x the rate of FP32. The other Turing-based GPUs boast double-rate FP16 as well through their Tensor cores, so TU116’s configuration serves to maintain that standard through hardware put in place specifically for this GPU. The following chart is an updated version of the one published in our GeForce GTX 1660 Ti review, which illustrates TU116’s massive improvement to half-precision throughput compared to GeForce GTX 1060 and its Pascal-based GP106 chip.

When we ran Sandra's Scientific Analysis module, which tests general matrix multiplies, we see how much more FP16 throughput TU106's Tensor cores achieve compared to TU116. GeForce GTX 1060, which only supported FP16 symbolically, barely registers on the chart at all.

In addition to the Turing architecture’s shaders and unified cache, TU116 also supports a pair of algorithms called Content Adaptive Shading and Motion Adaptive Shading, together referred to as Variable Rate Shading. We covered this technology in Nvidia’s Turing Architecture Explored: Inside the GeForce RTX 2080. That story also introduced Turing's accelerated video encode and decode capabilities, which carry over to GeForce GTX 1660 as well.

Putting It All Together…

Nvidia packs 24 SMs into TU116, splitting them between three Graphics Processing Clusters. With 64 FP32 cores per SM, that’s 1,536 CUDA cores and 96 texture units across the entire GPU. In losing two SMs, GeForce GTX 1660 ends up with 1,408 active CUDA cores and 88 usable texture units.

Board partners will undoubtedly target a range of frequencies to differentiate their cards. However, the official base clock rate is 1,530 MHz with a GPU Boost specification of 1,785 MHz. Both of those numbers are slightly higher than GeForce GTX 1660 Ti’s clocks, though they cannot compensate entirely for the missing SMs.

Our Gigabyte GeForce GTX 1660 OC 6G sample maintained a steady 1,935 MHz through three runs of Metro: Last Light, operating about 90 MHz faster than the 1660 Ti we reviewed a few weeks ago. On paper, then, GeForce GTX 1660 offers up to 5 TFLOPS of FP32 performance and 10 TFLOPS of FP16 throughput.

Six 32-bit memory controllers give TU116 an aggregate 192-bit bus, which is populated by 8 Gb/s GDDR5 modules that push up to 192 GB/s. That’s comparable to GeForce GTX 1060 6GB, and a 33% reduction compared to GeForce GTX 1660 Ti. Combined with the loss of two SMs, dropping from GDDR6 to GDDR5 memory accounts for GeForce GTX 1660’s lower performance versus 1660 Ti.

Each memory controller is associated with eight ROPs and a 256KB slice of L2 cache. In total, TU116 exposes 48 ROPs and 1.5MB of L2. GeForce GTX 1660’s ROP count compares favorably to RTX 2060, which also utilizes 48 render outputs. But TU116’s L2 cache slices are half as large compared to TU106.

Given the similarities to GeForce GTX 1660 Ti, it’s no surprise that GeForce GTX 1660 is rated for the same 120W. Unfortunately, neither graphics card includes multi-GPU support. Nvidia continues pushing the narrative that SLI is meant to drive higher absolute performance, rather than give gamers a way to match single-GPU configurations.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: How to Buy the Right GPU

MORE: All Graphics Content

Meet Gigabyte’s GeForce GTX 1660 OC 6G

The GeForce GTX 1660 launch is driven by Nvidia’s board partners. There won’t be a Founder’s Edition card. Rather, companies like Gigabyte will introduce a range of models representing multiple price and performance points. Today’s review covers the GeForce GTX 1660 OC 6G, Gigabyte’s entry-level $220 board.

As we continue making our way down Nvidia’s Turing-based product stack, it should come as no surprise that lower prices are accompanied by more cost-conscious designs. The GeForce GTX 1660 OC 6G is no slouch, though. From the front, it’s apparent that Gigabyte employs an aluminum heat sink topped by the company’s Windforce 2X cooler. A plastic fan shroud houses two 90mm fans spinning in opposite directions.

Gigabyte’s PCB is about 6 ¾” (17cm) long. However, the fan shroud and backplate extend beyond its edge, resulting in a ~9” (22.5cm) length measurement. The GeForce GTX 1660 OC 6G is also relatively tall, measuring 4 ¾” (~12cm) from the bottom of the PCI Express slot to its top edge. A double-slot form factor is ideal for compact enclosures. There’s little reason, in our opinion, for a 120W card to take up any more room than that.

Flipping the card over, a plastic backplate bears Gigabyte’s logo and adds a decorative touch over the PCB. However, there is no contact between the plate and any of the components underneath, nor is there any ventilation. This piece of plastic is merely decorative, meeting the fan shroud’s bezel in the back. A quick glance makes it seem like the cooler is completely enclosed.

Up top, the tall PCB dips down to keep your eight-pin power connector from protruding too far. The connector is also rotated 180 degrees to avoid interfering with overhanging plastic pieces. Another silver Gigabyte logo is painted onto the shroud, flanked by Nvidia’s GeForce GTX branding.

By choosing not to expose dual-link DVI, Gigabyte is able to fit three DisplayPort outputs and an HDMI connector on its slot bracket, plus leave tons of room for ventilation. That turns out to be smart in this card’s case, since horizontally-oriented fins allow some heated air to exhaust out the front.

With the backplate and sink removed, we see that Gigabyte’s cooler is more than just aluminum. A single, flattened, copper pipe runs through the center, right above TU116. It bends to the right and left, dissipating heat evenly through the array of fins. Strategically-placed plates cover the voltage regulation circuitry and memory modules, sandwiching thermal pads between them.

Peeling away those pads allows us to identify on-board components. TU116 is rotated 90 degrees and surrounded by six Micron MT51J256M32HF-80:B memory modules rated for 8 Gb/s data rates.

The GPU’s power supply is composed of four phases controlled by a uP9512R on the PCB’s back side. It’s attached to four Alpha & Omega Semiconductor AOZ5332QI DrMos power modules, which include high- and low-side MOSFETs, a driver, and the bootstrap diode. Each is capable of a continuous 50A.

Ubiq Semiconductor’s familiar dual-phase uP1666Q controls the memory’s voltage regulation circuitry by way of two AON6994 dual N-channel MOSFETs.

As we saw from EVGA’s GeForce GTX 1660 Ti XC Black Gaming, Gigabyte leaves vacant pads for two additional GPU phases. There’s also a pair of emplacements for extra GDDR5 memory modules. Nvidia did something similar with GeForce GTX 1060, leaving a couple of blank spots on its Founders Edition card that were never populated. This is a time- and cost-saving measure, which allows the company to use one PCB for multiple products.

How We Tested Gigabyte’s GeForce GTX 1660 OC 6G

Obviously, GeForce GTX 1660 is more mainstream than the other Turing-based boards we’ve reviewed. As such, our graphics workstation, based on an MSI Z170 Gaming M7 motherboard and Intel Core i7-7700K CPU at 4.2 GHz, is apropos. The processor is complemented by G.Skill’s F4-3000C15Q-16GRR memory kit. Crucial’s MX200 SSD is here, joined by a 1.6TB Intel DC P3700 loaded down with games.

As far as competition goes, the 1660 mostly goes up against GeForce GTX 1060 6GB from Nvidia’s line-up, though we want the GeForce GTX 1660 Ti in there as well. GeForce GTX 1070 and 1070 Ti are also relevant from the previous generation. All of those mainstream cards are included, plus GeForce RTX 2060. On the AMD side, we’re mostly interested in Radeon RX 590, along with Radeon RX Vega 64 and Radeon RX Vega 56.

Our benchmark selection includes Ashes of the Singularity: Escalation, Battlefield V, Destiny 2, Far Cry 5, Forza Horizon 4, Grand Theft Auto V, Metro: Last Light Redux, Shadow of the Tomb Raider, Tom Clancy’s The Division, Tom Clancy’s Ghost Recon Wildlands, The Witcher 3 and Wolfenstein II: The New Colossus.

The testing methodology we're using comes from PresentMon: Performance In DirectX, OpenGL, And Vulkan. In short, these games are evaluated using a combination of OCAT and our own in-house GUI for PresentMon, with logging via GPU-Z.

We're using driver version 419.35 to test GeForce GTX 1660, 418.91 for GeForce GTX 1660 Ti, and build 417.54 for everything else. AMD’s cards utilize Crimson Adrenalin 2019 Edition 18.12.3.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Gaming at 1920 x 1080

According to Nvidia, the GeForce GTX 1660 Ti we were recently introduced to was designed for high frame rates in popular games at 1920 x 1080. Naturally, then, GeForce GTX 1660 is meant to drive the same resolution…just a bit slower.

Despite its trimmed-down specifications, the 1660 is still about four percent faster than Radeon RX 590 and 16% quicker than GeForce GTX 1060 6GB, on average. Meanwhile, it offers roughly 88% of the GeForce GTX 1660 Ti’s frame rate through our benchmark suite.

With the exception of Metro: Last Light, where we enable SSAA, and Ashes of the Singularity, where we dial in the Crazy quality preset, GeForce GTX 1660 can be counted on to average 60 FPS or more in the games we test (almost all of which run at their top detail settings). If you don’t have the budget for a $280 GeForce GTX 1660 Ti, then a $220 1660 gets the job done as well.

Ashes of the Singularity: Escalation (DX12)

Battlefield V (DX12)

Destiny 2 (DX11)

Far Cry 5 (DX11)

Forza Horizon 4 (DX12)

Grand Theft Auto V (DX11)

Metro: Last Light Redux (DX11)

Shadow of the Tomb Raider (DX12)

Tom Clancy’s The Division (DX12)

Tom Clancy’s Ghost Recon (DX11)

The Witcher 3 (DX11)

Wolfenstein II: The New Colossus (Vulkan)

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Gaming at 2560 x 1440

In all frankness, GeForce GTX 1660 isn’t a QHD card. It’s certainly possible to achieve smooth frame rates by dialing back graphics quality a notch or two. But if you’re gaming on a 2560 x 1440 monitor, consider something in the Radeon RX Vega 56/GeForce GTX 1070 Ti/GeForce RTX 2060 range at least.

Ashes of the Singularity: Escalation (DX12)

Battlefield V (DX12)

Destiny 2 (DX11)

Far Cry 5 (DX11)

Forza Horizon 4 (DX12)

Grand Theft Auto V (DX11)

Metro: Last Light Redux (DX11)

Shadow of the Tomb Raider (DX12)

Tom Clancy’s The Division (DX12)

Tom Clancy’s Ghost Recon (DX11)

The Witcher 3 (DX11)

Wolfenstein II: The New Colossus (Vulkan)

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Power Consumption

Slowly but surely, we’re spinning up multiple Tom’s Hardware labs with Cybenetics’ Powenetics hardware/software solution for accurately measuring power consumption.

Powenetics, In Depth

For a closer look at our U.S. lab’s power consumption measurement platform, check out Powenetics: A Better Way To Measure Power Draw for CPUs, GPUs & Storage.

In brief, Powenetics utilizes Tinkerforge Master Bricks, to which Voltage/Current bricklets are attached. The bricklets are installed between the load and power supply, and they monitor consumption through each of the modified PSU’s auxiliary power connectors and through the PCIe slot by way of a PCIe riser. Custom software logs the readings, allowing us to dial in a sampling rate, pull that data into Excel, and very accurately chart everything from average power across a benchmark run to instantaneous spikes.

The software is set up to log the power consumption of graphics cards, storage devices, and CPUs. However, we’re only using the bricklets relevant to graphics card testing. Gigabyte's GeForce GTX 1660 OC 6G gets all of its power from the PCIe slot and one eight-pin PCIe connector.

Idle

Although GeForce GTX 1660 is rated for the same 120W as Nvidia’s GeForce GTX 1660 Ti, our Gigabyte GeForce GTX 1660 OC 6G sample uses almost 1W less at idle than EVGA’s GeForce GTX 1660 Ti XC Black Gaming.

Gaming: Metro: Last Light

A 126W average measurement is slightly higher than Nvidia’s specification for GeForce GTX 1660, and a couple of watts higher than what we saw from GeForce GTX 1660 Ti. The peak to nearly 135W isn’t egregious, at least.

Gigabyte uses the PCI Express x16 slot even more sparingly than EVGA’s GeForce GTX 1660 Ti, consuming just 21W under load. Instead, the GeForce GTX 1660 OC 6G leans on its eight-pin auxiliary connector. No wonder this card eschews the six-pin connector found on many GeForce GTX 1060s.

Charting power consumption across our three-run benchmark shows the narrow range we’re working with. This card jumps around a bit more than GeForce GTX 1660 Ti. However, its dips and spikes are well-controlled, giving us no reason to believe TU116 is limited by thermals or power.

A direct comparison between Gigabyte’s GeForce GTX 1660 and EVGA’s GeForce GTX 1660 Ti shows the EVGA card using slightly less power, perhaps because of its lower voltage and clock rate, which we’ll see on the following page.

Current draw over the PCIe slot remains under 2A through our Metro benchmark sequence. Obviously, that’s way below the PCI-SIG’s 5.5A maximum.

Roughly 9A from the eight-pin auxiliary connector is no problem either, especially compared to the ASRock Radeon RX 590 we recently reviewed, which repeatedly crested 16A in the same benchmark.

FurMark

FurMark is a steadier workload, resulting in less variation across our test run. Average power is a couple of watts higher, while less room for dramatic swings brings the peak reading down under 134W.

There’s no room for performance to vary in FurMark, so the wavy line chart we saw under Metro: Last Light is gone, replaced by much more consistent power consumption through more than 15 minutes of recording.

Slightly higher power consumption and lower frame rates combine to ding GeForce GTX 1660 in our performance per watt measurements. But that’s only in a comparison to GeForce GTX 1660 Ti. Versus GeForce GTX 1060, 1070, 1070 Ti or any of the AMD cards, GeForce GTX 1660 is incredibly efficient.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Temperatures, Fan Speeds, and Clock Rates

Gaming: Metro: Last Light

When Gigabyte’s card transitions out of passive mode, its fans pulse on and bounce around a bit before steadily accelerating. Once the GeForce GTX 1660 OC 6G warms up, a constant 2,050 RPM rotational speed is nice and quiet.

Despite a relatively conservative fan curve, the 1660 OC 6G’s TU116 processor tops out at 68 degrees through three loops of the Metro benchmark, making it the coolest-running card in our chart.

Plenty of power headroom and ample cooling allow the GeForce GTX 1660 OC 6G to run at a stable 1,935 MHz with a voltage setting of 1.043V. When cards like the GeForce GTX 1070 Founders Edition show signs of throttling down, Gigabyte’s GeForce 1660 hardly dips at all.

FurMark

Once again, Gigabyte’s Windforce system pulses high as it shifts from semi-passive to active cooling. The two fans then settle into a curve that responds quickly to warmer temperatures, but never makes much noise.

The fan’s gradual ramp allows temperatures under FurMark to spike a little faster than competing cards from Nvidia. However, the GeForce GTX 1070 and 1060 Founders Edition boards eventually break 70°C, whereas Gigabyte’s GeForce GTX 1660 OC 6G never exceeds 69°.

Bouncing between 1,845 and 1,860 MHz, the GeForce GTX 1660 OC 6G slows down under FurMark compared to the frequencies observed through Metro: Last Light. But it definitely takes less of a hit than other cards we tested. Just look at the white space between the red and yellow lines in this graph, and then the trend from Metro, where the field is much closer together.

Although voltage to the TU116 processor falls under 1V, Gigabyte’s card sustains much higher voltage levels than even EVGA’s GeForce GTX 1660 Ti based on the same GPU.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Conclusion

With GeForce RTX 2080 Ti, 2080, and 2070, Nvidia began painting a picture of its Turing-based line-up that concerned many enthusiasts. The cards were fast, yes. But high prices took the sheen off grandiose promises of real-time ray tracing and AI-assisted performance boosts.

Next out of the gate was GeForce RTX 2060, a card that showed us the Turing architecture’s signature features could still be viable from a mid-range model. Because GeForce RTX 2060 hunted in similar territory as GeForce GTX 1070 Ti and the Radeon RX Vega duo, it needed a more compelling price point. Consequently, Nvidia’s value story started sounding compelling.

The GeForce GTX 1660 Ti was even more warmly received. Nvidia’s first Turing-based GPU without RT or Tensor cores relied exclusively on performance in today’s games, just like the Pascal and Graphics Core Next architectures. Its redesigned Streaming Multiprocessors, larger caches, and GDDR6 memory pushed frame rates higher while hitting all the right budget notes.

If you’re gaming at 1920 x 1080, the 1660 Ti’s speed may seem a bit superfluous in light of the vanilla 1660’s smooth frame rates at an even more attractive $220/£200 price point. In fact, according to our calculations, GeForce GTX 1660 would seem to take the crown in a comparison of average frame rate per dollar, usurping GeForce GTX 1660 Ti, the respectable GeForce GTX 1070 Ti, and GeForce RTX 2060.

But does it really? When we fold in benchmark results from the older Radeon RX 580, generated by Tom’s Hardware Germany using a similar test system, and then apply U.S. pricing, AMD’s card scores just a bit better. Granted, it gets wrecked when we switch to a performance per watt comparison. If you’re not really concerned about power consumption, though, it’s really hard to pass on an 8GB Radeon RX 580, currently available at $190, for 1080p gaming.

Even if AMD’s Radeon RX 580 blowout comes to an end, the GeForce GTX 1660 will remain a solid pairing with high-refresh FHD displays. It’s less QHD-friendly than GeForce GTX 1660 Ti, and we’d suggest stepping up to at least a GeForce RTX 2060 if you’re playing at 2560 x 1440. However, when push comes to shove, you can dial down your favorite game’s detail settings to achieve respectable frame rates at higher resolutions on GTX 1660, too.

We approved of GeForce GTX 1660 Ti for its GeForce GTX 1070-like performance at a price point under $300 (and 120W power consumption). Similarly, we like that GeForce GTX 1660 drives great 1080p frame rates at $220. And we’re even more excited that there’s a competing card from AMD with real value chops we can mention in the same breath.

MORE: Best Graphics Cards

MORE: Desktop GPU Performance Hierarchy Table

MORE: All Graphics Content

Well-known graphics card leaker VideoCardz today released alleged specifications for the rumored Nvidia GeForce GTX 1660 graphics card, along with images of MSI's custom offerings.

*=unconfirmed

The Nvidia GeForce GTX 1660 is expected to employ the Turing TU116 silicon, which is manufactured by TSMC on its 12nm node. Unlike the GTX 1660 Ti, which is based on the TU116-400-A1, the GeForce GTX 1660 will utilize the TU116-300 variant, according to VideoCardz. The upcoming graphics card reportedly sports 1,408 CUDA running at a 1,530MHz base clock and with a boost clock that scales up to 1,785MHz.

Nvidia will allegedly equip the GTX 1660 with 6GB of GDDR5 memory. The memory clocks in at 2,000MHz (8,000MHz effective) across a 192-bit memory interface, which would allow the GTX 1660 to deliver a memory bandwidth of 192 GB/s.

VideoCardz claimed in today's report the GTX 1660 will launch on March 14, but DigiTimes has reported that date as March 15. It's expected to cost $219 (~£168).

MSI GeForce GTX 1660 Gaming X

The MSI GeForce GTX 1660 Gaming X is said to feature MSI's 7th-generation Twin Frozr cooling solution. The black and gunmetal grey cooler is decked with the brand's Mystic Light RGB lighting, according to the report. The graphics card is said to use a pair of the latest Torx 3.0 cooling fans with Zero Frozr technology. Being MSI's flagship model, the GTX 1660 Gaming X is expected to boast a 1,860MHz boost clock.

The graphics card draws its power from a single 8-pin PCIe power connector. Video outputs include three DisplayPort 1.4 outputs and a single HDMI 2.0b port.

MSI GeForce GTX 1660 Armor

The MSI GeForce GTX 1660 Armor will seemingly continue to retain the classic black and white Armor cooler. The RGB eye candy is kept to a minimum on this model. Only the MSI logo on the side of the shroud has RGB lighting. The graphics card is rumored to use a pair of Torx 2.0 fans with the Zero Frozr feature for active cooling.

Like the Armor model, the MSI GeForce GTX 1660 Armor should also feed off an 8-pin PCIe power connector, plus provide three DisplayPort 1.4 outputs and a HDMI 2.0b port for connectivity.

MSI GeForce GTX 1660 Ventus XS

The MSI GeForce GTX 1660 Ventus XS look like it rocks a simple black and white cooler without any RGB elements. In the leaked image, the shroud carries an industrial design with twin Torx 2.0 cooling fans. However, the fans apparently don't possess MSI's Zero Frozr technology, which only turns them on when the GPU temperature goes beyond a certain threshold.

Like its the siblings, the MSI GeForce GTX 1660 Ventus XS should rely on an 8-pin PCIe power connector for external power and has three DisplayPort 1.4 outputs and one HDMI 2.0b port.

Want to comment on this story? Let us know what you think in the Tom's Hardware Forums.

According to a DigiTimes report today, Nvidia is going to release one medium and one low-end GPU by the end of 2019, as an attempt to lower AMD's market share below 25 percent or even 20 percent, as well as to clear its inventory and maintain its revenues.

Nvidia Strikes Back?

According to DigiTimes’ sources, Nvidia will launch the GTX 1660 on March 15 and GTX 1650 on April 30, for a minimum price point of $229 and $179, respectively. However, the release date for the 1650 conflicts with a report from VideoCardz last week and HardOCP in January claiming the GTX 1650 will come out in March. 

The release of these GPUs would follow the company’s launch of the RTX 2060 last month at an MSRP of $350 / £330 and GTX 1660 Ti last week for a minimum price point of $280.

According to the DigiTimes report today, the GTX 1660 and GTX 1650 "are expected to bring heavy pricing pressure on AMD's mid-tier and low-end offerings such as Vega 64 / 65 and RX 590 / 580 / 570." That remains to be seen, as AMD is also expected to continue to lower prices on the Radeon RX 590 and 580 in response to Nvidia's GTX 1660 Ti moves. 

Crypto Mining Decline Is Still Hurting Nvidia

According to DigiTimes' sources, Nvidia will refocus its GPU business on the mid-end and low-end in the first half of this year. However, Nvidia seems to want to lower prices not just to steal market share from AMD, but also because it’s looking to clear its high inventory, caused by a shrinkage in demand in the cryptocurrency mining market. Furthermore, the lower price points may help stabilize the company's revenues, which have been in a declining trend lately.

In the fiscal quarter from November 2018 to January 2019, Nvidia saw a significant drop in both revenues and profits. The company’s revenue fell 24 percent on year and 31 percent on quarter to $2.205 billion, while its profits fell 49 percent on year and 54 percent on quarter to $567 million.

According to DigiTimes’ sources, it will take another quarter for Nvidia to clear its old inventory, and the company’s revenues are expected to stay flat compared to the previous quarter. Market watchers estimated that Nvidia’s revenue in 2019 will fall by 15-20 percent.

Nvidia today released a new GeForce Game Ready driver to accommodate the GeForce GTX 1660 Ti GPU and EA game Anthem, both of which hit the market today. 

The GTX 1660 Ti is essentially a Turing product without the Tensor cores (used for inferencing in games) and RT cores (used to offer real-time ray tracing support) found in the company's RTX lineup. Nvidia is positioning the GTX 1660 Ti as a replacement for the GTX 1060 6GB that's meant primarily for 1080p gaming in modern titles.

In addition to getting the GTX 1660 Ti up and running, the GeForce Game Ready 419.17 WHQL driver also improves support for Anthem, a cooperative online shooter that lets up to four players team up to explore the game's world. (And in the process level up the "exosuits" each character wears to get new abilities, increase stats and more.) The game is available now on PC, PlayStation 4 and Xbox One.

Nvidia's also using Anthem's launch to try to help shore up sales of its RTX 2080 and RTX 2080 Ti graphics cards. The company's partners are offering a variety of bundles that include a copy of Anthem or Battlefield V with the purchase of specific graphics cards. That way both Nvidia and EA can benefit from Anthem launch hype.

The GeForce Game Ready 419.17 WHQL drivers are available now via the GeForce Experience software and the GeForce website.

Want to comment on this story? Let us know what you think in the Tom's Hardware Forums.

Although Nvidia isn't rumored to announce the GeForce GTX 1660 Ti graphics card until tomorrow, prolific leaker VideoCardz today delivered new images claiming to be Asus' custom models of the GPU.

As previously reported, Asus registered numerous models with the Eurasian Economic Commission (EEC) not long ago. The listing also revealed Nvidia's plan to release an alternate variant of the GTX 1660 Ti with 3GB of GDDR6 memory.

According to the new report from VideoCardz, Asus is slated to launch various personalized GTX 1660 Ti graphics cards with its Republic of Gamers (ROG) Strix, The Ultimate Force (TUF), Dual, Expedition, Turbo and Phoenix product lines. We seemingly now have a glimpse of the ROG Strix, Dual and Phoenix GeForce GTX 1660 Ti models, courtesy of VideoCardz.

Asus ROG Strix GeForce GTX 1660 Ti OC

The ROG Strix GeForce GTX 1660 Ti OC is said to be the highest-end model. It sports the brand's beefy triple-fan Strix cooling system. There are a couple of RGB LEDs embedded into the shroud, presumably controllable via the Asus Aura software. Three patented Wing-Blade fans with 0dB technology (for silence) reportedly provide active cooling. Those fans only start spinning when the GPU temperature excels a certain threshold, depending on the model.

The reference GeForce GTX 1660 Ti is expected to operate with a 1,500MHz base clock and 1,770MHz boost clock. Being the crème de la crème model, the Asus ROG Strix GeForce GTX 1660 Ti OC should come with a considerable out-of-the-box factory overclock.

The graphics card will seemingly have one HDMI port and three DisplayPort outputs.

Asus Dual GeForce GTX 1660 Ti OC

Doing justice to its name, the Dual GeForce GTX 1660 Ti OC reportedly employs a dual-slot, no-frills cooling system. The shroud looks painted in pure black with a couple of white stripes and zero RGB eye candy. The two Wing-Blade fans are also supposed to be IP5X-certified, which would make them dust-resistant with a longer lifespan.

Regarding display connectivity, the Dual GeForce GTX 1660 Ti OC is outfitted with two DisplayPort outputs, one HDMI port and one DVI-D port.

Asus Phoenix GeForce GTX 1660 Ti OC

The Phoenix GeForce GTX 1660 Ti OC appears to be a mini version of the Dual GeForce GTX 1660 Ti OC. Thanks to its compact design, the Phoenix GeForce GTX 1660 Ti OC will seemingly target small PC builds. The dual-slot cooler reportedly utilizes a single Wing-Blade, IP5X-certified cooling fan. The display outputs on the Phoenix GeForce GTX 1660 Ti OC are said to be identical to those of the Dual model.

iBuyPower has listed a Nvidia GTX 1660 Ti-based system on Amazon in what seems to be an accidental listing, as the GTX (not RTX) 1660Ti has not been announced yet. The GTX 1660 Ti is expected to be a mid-range performer at a much lower price point than other Turing GPUs thanks to its lack of RTX-compatible hardware.

The GTX 1660 Ti’s MSRP could be quite low considering that the iBuyPower listing has the entire PC, which also includes a Core i7-8700, 480 GB SSD, and 16 GB of RAM, priced at $1200. Similar PCs from iBuyPower featuring the RTX 2060 retail for roughly $100 more. HardOCP and later VideoCardz reported that the GTX 1660 Ti's MSRP is $279 ($70 lower than the RTX 2060).

The 1660 Ti is expected not to contain any RTX-compatible hardware, and even if it did, its 1536 cores probably wouldn’t be enough considering that the 2060 struggles at 1920 cores. As a result, the size of the TU116 GPU (which the 1660 Ti is reportedly based on) is about 70% of the size of the TU106 GPU that powers the RTX 2060 and 2070, according to a leaked die shot from VideoCardz. Coincidentally, that makes the 1660 Ti almost the same size as the 1070 and 1070 Ti, and according to the Final Fantasy XV benchmark database, a leaked 1660 Ti also performs much like GTX 1070. However, considering the 1660 Ti has far fewer cores, it is very impressive.

The upcoming 1660 Ti could be crucial to Nvidia’s financial success after the company reported a staggeringly poor performance during the last half of the year. Revenues, margins, and income dropped intensely while operating expenses shot up. While Nvidia primarily blames China’s economy, the company admitted that RTX sales didn’t meet expectations. While Nvidia’s gaming revenue (Nvidia’s largest source of money) impressed for Q3 of last year, sales plummeted in Q4 which saw gaming revenue barely exceed half of what the company made in Q4 of 2017. Being a more mainstream GPU at a more reasonable price, the GTX 1660 Ti could give Nvidia the boost it needs after such disappointing sales.

EDIT: Clarified that the Titan X used for comparison is from the Maxwell generation.

As per notorious leaker TUM_APISAK's tip, Nvidia's upcoming GeForce GTX 1660 Ti has been benchmarked by an unknown source with Square Enix's Final Fantasy XV benchmark tool. The GeForce GTX 1660 Ti certainly appears to give the GeForce GTX Titan X (Maxwell) a run for its money.

As a quick recap, the GeForce GTX 1660 Ti is going to be Nvidia's first Turing-powered 12nm gaming graphics card that will not come with the real-time ray tracing capability. Built around the brand-new Turing TU116 graphics processor, the GeForce GTX 1660 Ti has been rumored to come with 1,536 CUDA cores operating at an alleged 1,500MHz base clock and 1,770MHz boost clock. The 6GB or 3GB of GDDR6 memory, on the other hand, runs at 1,500MHz (12,000MHz effective) across a 192-bit memory bus.

According to a leaked Ashes of the Singularity (AotS) result last month, the GeForce GTX 1660 Ti outperformed the GeForce GTX 1060 by 19.35 percent.

While many might dispute the reliability and stability of the Final Fantasy XV benchmark, it's still interesting to get a sneak peek of how the GeForce GTX 1660 Ti fares against its present rivals. The Final Fantasy XV result reveals that the GeForce GTX 1660 Ti was tested at the 2560 x 1440 resolution with the graphics on the High Quality preset. The GeForce GTX 1660 Ti scored an impressive 5,000 points, making it just a mere 0.5 percent slower than the $1,000 GeForce Titan X (Maxwell) but up to 0.9 percent faster than the GeForce GTX 1070. With regards to the competition, the GeForce GTX 1660 Ti falls behind AMD's recently launched $700 Radeon VII by 5.28 percent.

Nvidia GeForce GTX 1660 Ti Pricing and Availability

Several rumors point to a February 22nd release date. In terms of pricing, media outlet HardOCP's sources claim that the GeForce GTX 1660 Ti will have an MSRP (Manufacturer's Suggested Retail Price) price of $279 (~£216.34). Meanwhile, VideoCardz has gotten word that Nvidia has reportedly confirmed the GeForce GTX 1660 Ti's $279 price tag with computer hardware reviewers.

Want to comment on this story? Let us know what you think in the Tom's Hardware Forums.

It's no secret that Nvidia isn't selling as many RTX graphics cards as it hoped. The company revised its revenue guidance in January in response to weaker-than-expected sales that were at least partly blamed on how much Turing architecture GPUs cost. Enter rumor, after rumor, after rumor about Nvidia planning a 1660 Ti product that would allow Turing to reach bargain hunters. Today's rumor arrives courtesy of VideoCardz, which shared images of what it said were an MSI GeForce GTX 1660 Ti Ventus XS and the TU116 GPU around which it was built. We haven't confirmed the veracity of these images, but at this point, it's harder to imagine a future in which Nvidia doesn't release a (relatively) low-price Turing GPU than one where it does.

VideoCardz said there are two versions of the Ventus coming, the XS and the not-XS, with the former sharing a PCB with MSI's GTX 1660 Ti Gaming X. The publication didn't share other details about these particular cards, so it's not clear when the MSI cards are expected to debut or at what price.

Album credit: VideoCardz

Yet, VideoCardz did offer some alleged information about the GTX 1660 Ti itself, as well as some images of what it said is its TU116 GPU (above). The GTX 1660 Ti is reportedly set to debut on February 22 with a non-Ti version arriving "probably next month." That lines up with previous reports indicating that Nvidia planned to reveal the GTX 1660 Ti in mid-February, as does the claim that these cards will feature 1,536 CUDA cores and 6GB of GDDR6 memory. VideoCardz also claimed  the 1660 Ti will lack ray tracing support.

Is the GTX 1660 Ti Real?

All of these factors could allow Nvidia to release a mid-price Turing product to appeal to all those people who can't afford the RTX cards that debuted in late 2018. With fewer people upgrading their systems at the moment, especially amid the drought of Intel processors, having a lower-priced offering could help Nvidia persuade hesitant buyers.

Still, the company has yet to officially confirm that it's even making the GTX 1660 Ti. But over the last few months we've gone from "this might exist" to "we might have the specs" to "we're pretty sure we have the specs" to "yeah, it'd really be a surprise if these weren't the specs." At this point, alleged products from EVGA, Palit and MSI have reportedly leaked, as have have plans for a 3GB variant.

Meanwhile, the company's financials already don't look as good as shareholders would like--even before the realization that Nvidia might actually be underplaying its reliance on cryptocurrency miners--and its biggest investor recently sold its entire stake. 

Releasing the GTX 1660 Ti could help, though, especially if people start buying more parts for their PCs again. At the very least it would allow Nvidia to compete in the lower-end of the market instead of letting AMD dominate with its RX 400 and 500-series cards. Combine that with all the leaks, and it's not hard to imagine there's something to these rumors.

Nvidia has yet to announce the GeForce GTX 1660 Ti gaming graphics card, but a recent Eurasian Economic Commission (EEC) listing seemingly points to the chipmaker working on a 3GB variant.

The GeForce GTX 1060 initially launched with 6GB of GDDR5 memory. Months later, Nvidia released a cut-down version of the graphics card with the memory sliced in half. Apparently, old habits die hard, and Nvidia is about to pull a Deja-vu move with the upcoming GeForce GTX 1660 Ti.

As per the models listed in the EEC entry, Asus is preparing a plethora of aftermarket GeForce GTX 1660 Ti graphics cards–eleven with the 6GB configuration and nine with the 3GB configuration. As usual, enthusiasts can expect to find products from Asus's different product lines, such as the Republic of Gamers (ROG) Strix, The Ultimate Force (TUF), Dual, Expedition, Turbo, and Phoenix series. However, Asus apparently reserved the ROG Strix branding for 6GB models.

Asus might have up to three variants of a few selected models. The normal version speaks for itself, while Advanced versions feature slightly higher clock speeds and the OC versions should come with the highest out-of-the-box overclock.

Asus GeForce GTX 1660 Ti 6GB Models

Asus GeForce GTX 1660 Ti 3GB Models

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.

Everything seems to indicate that Nvidia might be launching its GeForce GTX 1660 Ti gaming graphics card very soon. Custom models from Palit and MSI have been listed on various Russian online computer hardware retailers.

Albasoft listed the Palit GeForce GTX 1660 Ti StormX for 26,700.02 ₽ and the Palit GeForce GTX 1660 Ti StormX OC for 27,093.54 ₽, which translates to around $406.11 and $412.09, respectively.

The latest revision of popular hardware monitoring tool AIDA64 added support for the Nvidia TU116 silicon, which confirms its existence. If you've been following the Nvidia leaks, TU116 is the purported graphics processor that will be powering the GeForce GTX 1660 Ti. The TU116 die is a close relative to the TU106 die that Nvidia uses inside the GeForce RTX 2060 and RTX 2070. However, it allegedly lacks Tensor Cores and RT Cores.

Although rumors of the GeForce GTX 1660 Ti have been floating around the Internet for months now, this is the first time we're seeing the specifications from an official listing. DNS Shop briefly listed the specifications for the Palit GeForce GTX 1660 Ti StormX (NE6166T018J9-161F) and Palit GeForce GTX 1660 Ti StormX OC (NE6166TS18J9-161F). The graphics card is outfitted with 1,536 CUDA cores and 96 TMUs (Texture Mapping Units).

The normal Palit GeForce GTX 1660 Ti StormX comes with a 1,500MHz base clock and 1,770MHz boost clock, which is believed to be the reference clock speeds for the GeForce GTX 1660 Ti. On the other hand, the overclocked version has a 1,815MHz boost clock.

The GeForce GTX 1660 Ti also features 6GB of GDDR6 memory operating at 1,500MHz (12,000MHz effective) across a 192-bit memory interface resulting in a memory bandwidth of 288 GB/s. The RTX 20-series graphics cards have their memory clocked at 1,750MHz (14,000MHz effective). Video outputs on the Palit models include a DisplayPort output, an HDMI port, and a DVI-D port. They rely on a single 8-pin PCIe power connector.

Albasoft, another Russian retailer, listed two more GeForce GTX 1660 Ti models from MSI. Both the MSI GeForce GTX 1660 Ti Armor 6G OC and MSI GeForce GTX 1660 Ti Gaming X 6G comes equipped with a 1,536MHz base clock, 36MHz higher than Palit's offering. They also have different display connectors. The MSI graphics cards sport one HDMI port and three DisplayPort outputs.

A little less than a week ago, the existence of the Nvidia GeForce GTX 1660 Ti leaked thanks to a couple of VideoCardz's trusted sources. Today, famous leaker TUM_APISAK discovered one of the first benchmarks for the upcoming GeForce GTX 1660 Ti.

The GeForce GTX 1660 Ti was tested in the popular Ashes of the Singularity (AotS) benchmark on the High quality preset at the 1920x1080 resolution. TUM_APÏSAK noted that this was the mobile version of the GeForce GTX 1660 Ti as the graphics card was housed inside a laptop alongside an octa-core Coffee Lake H (CFL-H) processor.

The mobile GeForce GTX 1660 Ti put up a score of 7,400 points, which makes it approximately 19.35 percent faster than the GeForce GTX 1060. The mobile Nvidia GeForce GTX 1060 scored 6,200 points in the same benchmark with the same settings.

Very little is known about the GeForce GTX 1660 Ti at the moment except the rumor that it'll probably be the first Turing-powered graphics card to launch with the GTX branding. The GeForce GTX 1660 Ti reportedly sports 1,536 CUDA cores and 6GB of GDDR6 memory across a 192-bit memory interface. The graphics card employs Nvidia's TU116 silicon, which is still based on the Turing graphics architecture but allegedly lacks the RT cores for real-time ray tracing. 

According to what VideoCardZ has heard, Nvidia could announce the GeForce GTX 1660 Ti as soon as next month. There is a rumor that a GeForce GTX 1660, which uses GDDR5 or GDDR5X memory, is also in the works, and it might be unveiled at the same time frame.

Three industry sources, including one of Nvidia's board partners, told VideoCardz about the existence of the Turing-based GeForce GTX 1660 Ti graphics card. However, the final product name is still subject to change. In the meantime, the graphics card is being labeled as the GeForce GTX 1660 Ti. 

The GeForce GTX 1660 Ti is reportedly the first Turing product to be released under the existing GTX branding. The GTX 1660 Ti is based on the TU116 die that TSMC produces with the 12nm FinFET manufacturing process. The card is said to lack ray tracing capabilities. In a perfect world, this should result in the graphics card having a lower price than its RTX brethren.

*=unconfirmed

According to VideoCardz's sources, Nvidia will equip the GeForce GTX 1660 Ti with 1,536 CUDA cores so it won't eat into the recently launched GeForce RTX 2060's sales. That would mean the GeForce GTX 1660 Ti sports 24 streaming multiprocessors (SMs) with 96 texture mapping units (TMUs) and 48 render output units (ROPs). The GeForce GTX 1660 Ti is also rumored to come equipped with 6GB of GDDR6 memory that communicates across a 192-bit memory bus. 

There is no information about pricing or availability for the GTX 1660 Ti. However, Nvidia CEO Jensen Huang commented to VentureBeat a few days ago that Nvidia is almost done clearing its GeForce GTX 10-series inventory, and that the GeForce GTX 1060s are the last ones to go. Therefore, it's possible that Nvidia might announce the GeForce GTX 1660 Ti, or whatever Nvidia decides to name it, once the last batch of GeForce GTX 1060 graphics cards are depleted.

Nvidia GeForce GTX 1660 Ti Spotted

Chinese overclocker ShinChao Tou, who commonly goes by Uing07, has snapped a photograph at a closed Chinese media event that seemingly confirms the existence of the GeForce GTX 1660 Ti. The words "New GTX Turing" validates the rumor that the GeForce GTX 1660 Ti is based on Nvidia's Turing silicon.

Gigabyte and MSI GeForce GTX 1660 Ti Graphics Cards

The Eurasian Economic Comission (EEC) has turned into a popular place to find unreleased computer hardware. Two entries revealed multiple GeForce GTX 1660 Ti models from popular manufacturers, such as Gigabyte and MSI. The listing for MSI references four models.

• MSI GeForce GTX 1660 Ti Gaming Z 6G
• MSI GeForce GTX 1660 Ti Armor 6G OC
• MSI GeForce GTX 1660 Ti Ventus XS 6G OC
• MSI GeForce GTX 1660 Ti Gaming X 6G

As for Gigabyte, the second listing refers to eleven different GeForce GTX 1660 Ti graphics cards.

• GV-N166TAORUS-6GD
• GV-N166TGAMING OC-6GD
• GV-N166TWF2OC-6GD
• GV-N166TOC-6GD
• GV-N166TIXOC-6GD
• GV-N166TGAMINGOC PRO
• GV-N166TGAMINGOC WHITE
• GV-N166TGAMING-6GD
• GV-N166TWF2-6GD
• GV-N166TD5-6GD
• GV-N166TIX-6GD