AMD RX Vega 64: The Tom's Hardware Liquid Cooled Edition

We know that AMD’s Radeon RX Vega 64 runs hot. There’s no headroom for overclocking and it’s noisy to boot. But can switching to a liquid cooler change the story? And if so, is the payoff worth the cost?

Believe it or not, all-in-one liquid coolers can keep more than 400W of waste heat in check, so long as you have the right radiator. How do you generate that much heat in the first place? Buy a Radeon RX Vega 64 and overclock it, of course! Perhaps because Tom's Hardware is a little more discerning with awards than most publications, AMD didn't send over one of its liquid-cooled cards for review. Instead, we set out to build something better than what AMD sells at a hefty premium.

A true custom loop with a full copper block would be a big step up in thermal performance, of course. But not everyone already owns a water-cooling kit, and many gamers are intimidated by the idea of piping together their own pump, radiator, fan, and tubing.

So, we turned to an all-in-one liquid-cooling prototype that, as luck would have it, we had sitting around anyway. Our Alphacool Eiswolf GPX-Pro came with the necessary radiator, tubing, and quick-release fittings. We'll compare the turnkey kit's performance to our own open-loop setup. As we'll soon see, however, there are some good reasons to suggest we could have saved ourselves the trouble.

For our tests, we're starting with a 120mm radiator (the same size AMD includes with its $700 card). AMD's all-in-one cooler costs $80, which is the price the company's partners have to pay for it. It accommodates one (push) or two (push/pull) fans. But we'll show you why a 240mm radiator is the minimum to consider if you plan on overclocking.

First things first, though. We have to provide some necessary context for our results:

We use The Witcher 3 at Ultra HD for our benchmarks because it produces the highest loads and provides a realistic scenario at the same time. As we’ll see later, your workload has a big impact on the clock rates that can be achieved. Also, The Witcher 3 is an ideal stability test for both the GPU and HBM2.

Note also that AMD’s drivers reported frequency numbers that were way too high in some cases, until AMD fixed this right before launch. Nevertheless, the inflated numbers made their way into some reviews. So, if you think that our results look suspiciously low, please keep this in mind. Here’s AMD’s statement about the issue:

"This new driver also reports the previously advertised GPU boost clock instead of the top DPM peak frequency...We’ve not yet had time to test this functionality with all other aspects of the driver and needs further testing before prime time, but we wanted to enable you to play around with OC!"

Installation & Notes

Don't be bothered by the fact that the thermal paste isn't cleaned up completely. This doesn’t actually make a difference, so long as you're using non-conductive paste. After switching out the cooler more than 10 times, a thorough cleaning job just isn’t possible anymore without endangering the package. In spite of the molding, there’s still some room between the interposer and package substrate. This is why we really don’t want to apply any pressure to the tiny board during cleaning. We only have the one card, after all.

When it comes to picking a thermal paste, stay away from the more viscous options out there. You still want something of high quality, though. Blobs and lines won't work due to Vega’s two HBM modules. Apply a thin and even layer with the help of a spatula instead. We used Kryonaut by Thermal Grizzly, but there are certainly other low-viscosity pastes that work just as well. (See our roundup of 85 thermal pastes for more.

It’s also important to cover the very short voltage converter tops with thermal pads of suitable thickness. Apply them on a larger scale in order to catch some of the board’s waste heat, if you can. And the coils should be included too, if possible, since the voltage converters alone yield a rather small surface area.

Alphacool supplies an active backplate, which is to say that the backplate helps with cooling. It’s not enough to simply cover the parts of the board below the voltage converters with thermal pads. The phase doublers also beg for some cooling attention when they hit more than 110°C in the face of aggressive overclocking. That's why we cut soft pads into large strips to make contact with the backplate in more places than Alphacool originally envisioned.

Conversely, feel free to leave that pad right below the GPU off; if anything, our upcoming infrared measurements suggest it's counterproductive.

The install wraps up with screwing the backplate on (don't over-tighten!). Even if you've never swapped out the cooler on your graphics card before, this shouldn't take more than an hour.

Thanks to Alphacool's quick-release fittings, we’re able to swap between cooling solutions without changing tubing or refilling the loop.

Test System & Methodology

We introduced our new test system and methodology in How We Test Graphics Cards. If you'd like more detail about our general approach, check that piece out.

The hardware used in our lab includes:

Test Equipment & Environment
System- Intel Core i7-6900K ($865.60 On Walmart) @ 4.3 GHz
- MSI X99S XPower Gaming Titanium
- Corsair Vengeance DDR4-3200
- 1x Toshiba OCZ RD400 1TB OCZ RD400 (1TB) ( On -)
- 2x Toshiba OCZ Trion 150 960GB OCZ Trion 100 480GB ($361.28 On Amazon)
- be quiet! Dark Power Pro 11 850W ($199.00 On -)
CPU Cooling
- Alphacool Eisblock XPX
- Alphacool Eiszeit 2000 Chiller
- 2x be quiet! Silent Wings 3 PWM (Closed Case Simulation)
- Thermal Grizzly Kryonaut (Used when Switching Coolers)
Environmental
Room Temperature: Constant 22°C
PC Case
- Lian Li PC-T70 with Extension Kit and Mods
- Configurations: Open Benchtable, Closed Case
Monitor- Eizo EV3237-BK
Power Consumption Measurement
- Contact-free DC Measurement at PCIe Slot (Using a Riser Card)
- Contact-free DC Measurement at External Auxiliary Power Supply Cable
- Direct Voltage Measurement at Power Supply
- 2 x Rohde & Schwarz HMO 3054, 500MHz Digital Multi-Channel Oscilloscope with Storage Function
- 4 x Rohde & Schwarz HZO50 Current Probe (1mA - 30A, 100kHz, DC)
- 4 x Rohde & Schwarz HZ355 (10:1 Probes, 500MHz)
- 1 x Rohde & Schwarz HMC 8012 Digital Multimeter with Storage Function
Thermal Measurement- 1 x Optris PI640 80Hz Infrared Camera + PI Connect
- Real-Time Infrared Monitoring and Recording
Noise Measurement- NTI Audio M2211 (with Calibration File, Low Cut at 50Hz)
- Steinberg UR12 (with Phantom Power for Microphones)
- Creative X7, Smaart v.7
- Custom-Made Proprietary Measurement Chamber, 3.5 x 1.8 x 2.2m (L x D x H)
- Perpendicular to Center of Noise Source(s), Measurement Distance of 50cm
- Noise Level in dB(A) (Slow), Real-time Frequency Analyzer (RTA)
- Graphical Frequency Spectrum of Noise
Operating System & Driver
- Windows 10 Pro (Creators Update, All Updates)
- Crimson 17.30.1051 Beta 6a (11.08.2018)

MORE Best Graphics Cards

MORE: Overclocking GeForce GTX 1080 Ti To 2.1 GHz Using Water

MORE: How To: Optimizing Your Graphics Card's Cooling

MORE: AMD Radeon Vega RX 64 8GB Review

MORE: AMD Radeon Vega Frontier Edition 16GB Review

This thread is closed for comments
51 comments
    Your comment
  • AgentLozen
    The conclusion page makes the liquid cooled Vega64 sound unappealing.
    When you guys reviewed Vega64 on Monday, you had a more restrained conclusion.

    ...AMD is ready to show off Radeon RX Vega 64, and it's...alright.

    Your tone seems a bit different today.

    Ultimately, Power Saver mode and undervolting doesn’t make the Radeon RX Vega 64 any more attractive. It just makes it more efficient and less noisy. What doesn't change is the card's high price. And that's a problem.

    I know you guys try to keep your emotions and bias out of your writing, but it seems like you're frustrated with Vega64. I felt the same when I looked at the benchmarks and power consumption results a few days ago. This thing fell way short of my expectations.

    edit: spelling, minor redundancy changes
  • bit_user
    Thanks for your thorough testing of this scenario.
  • zfreak280
    AMD really needs to ditch Global Foundries. These power consumption figures are ridiculous for finfet.
  • artk2219
    I'm honestly curious to see what the other OEM's can do with their own custom board and cooling designs. It seems like there's potential in Vega, especially the 56, but I'm not sure if we will ever see it fully tapped.
  • max0x7ba
    Well, the original Sapphire Vega 64 Liquid easily keeps 1850/1000 clocks, I haven't tried pushing it harder. https://www.reddit.com/r/Amd/comments/6udvyj/sapphire_radeon_rx_vega_64_liquid_cooling_oc/
  • FormatC
    496490 said:
    Your tone seems a bit different today...

    Call it bandwith. Two authors, two meanings and different lyrics. Ok, this piece was filtered also a little bit during transalation.

    For the launch review we had not enough time to investigate all this things more in detail. Now I got the time to run more things and to measure it, also under better conditions (water cooling).

    For Vega56 (yes, I got one now after one week of waiting) I will do a comparison of all modes, also with power draw and performance with heavier loads. But this software is killing me...

    2018360 said:
    Well, the original Sapphire Vega 64 Liquid easily keeps 1850/1000 clocks, I haven't tried pushing it harder. https://www.reddit.com/r/Amd/comments/6udvyj/sapphire_radeon_rx_vega_64_liquid_cooling_oc/
    Which drivers?

    Only the beta6 press driver can show you the real clock rate. What you see with the normal drivers from AMDs homepage is the frequency of the next DPM state, not the real clock rate. I wrote about this in my review. With this older driver and a chiller I can crack easy the 1.9 GHz barrier (in theory). But in real it were 1.6 to 1.7 GHz (like in the review).
  • AgentLozen
    FormatC said:
    Two authors, two meanings and different lyrics. Ok, this piece was filtered also a little bit during transalation.


    How big is the Tomshardware staff? I know you have a German team, but in America locally how many people do you have? When you give out your mailing address, does it lead to a rented out floor in an office building or does it point to some dude's garage?
  • AndrewJacksonZA
    I really love the in-depth stuff that you guys do, thank you!! :-)

    Also, the buttons for voting comments up and down have disappeared when viewing the comments underneath the article. Is that a known bug for you guys?



    Thank you,
    Andrew
  • redgarl
    All this power and OC arguments makes me laugh... AMD provided a stronger card in DX12 than the 1080 GTX and all I see are complains about power and OC.

    Guess what, the 64 and 56 are still more powerful than their 1080 and 1070 counterparts with no drivers optimization.

    It will take 2 months to know really what to expect from Vega.
  • FormatC
    496490 said:
    I know you have a German team, but in America locally how many people do you have? When you give out your mailing address, does it lead to a rented out floor in an office building or does it point to some dude's garage?
    I'm German and I own a well equipped lab with a lot of high-end stuff and I'm sitting not in a garage. We write mostly the launch things in coop between US and GER, the single VGA comes from me. We are writing in France, Italy, US and have also a franchise guy in Russia. :)
  • InvalidError
    1431429 said:
    AMD really needs to ditch Global Foundries. These power consumption figures are ridiculous for finfet.

    While some of the power consumption may be due to process, architecture can account for a large chunk of it as well. AMD put more pipelining into Vega in a bid to raise clock frequencies, which means more clocked data latches and control circuitry contributing to power consumption without adding useful work of their own.

    Also, since AMD is under contractual obligation to order a certain minimum of wafer starts from GloFo, AMD would owe compensation to GloFo for those unused wafer starts if it quit using them altogether, so that isn't an option until the last of AMD's GloFo commitments expire.
  • AgentLozen
    FormatC said:
    I'm German and I own a a well equipped lab with a lot of high-end stuff and not a garage. We write mostly the launch things in coop between US and GER, the single VGA comes from me. :)


    Maybe I was a little vague. I don't know anything about the internal workings of Tomshardware besides what I see from articles. When I think about the writers doing their work, I don't know what to imagine. So that's why I asked what about what kind of environment you work in and how many people their are involved.

    When you work on gathering results from hardware testing (like recently with Vega), is it just you in your lab for 12 hours a day running tests?

    One more thing. Maybe I'm being silly, but "VGA"? Video graphics array?
  • photonboy
    Redgarl,
    The problem is that VEGA64 air-cooled will throttle severely (blower card. still waiting 3rd party) and the liquid-cooled is so EXPENSIVE that you are better off by far getting a GTX1080Ti which kicks its ass in every way.

    That HEAT also goes into your room and it would make my small office unbearably hot several months of the year (even in the winter due to closed Windows).

    Wishful thinking that VEGA will suddenly get a big boost won't make it happen. We'll see minor improvements and that's it.

    I agree DX12/Vulkan once properly optimized for should utilize VEGA relatively better than NVidia though it's worth noting that usually drives up power usage since it uses more of the transistors at a given time (i.e. ACE) though FP16 should help.

    We don't know yet how much benefit that will be compared to the GTX1070/1080 so can only guess. All things being close to EQUAL I find VEGA56 at least quite interesting due to potential FUTURE performance.

    At this point until we get more information for VEGA I can only recommend a card that doesn't yet exist such as the ASUS STRIX RX-VEGA 56 if the price is roughly $450USD (assuming similar GTX1070 is about the same cost then).
  • FormatC
    496490 said:
    When you work on gathering results from hardware testing (like recently with Vega), is it just you in your lab for 12 hours a day running tests? One more thing. Maybe I'm being silly, but "VGA"? Video graphics array?
    12 Hours? Sounds like vacation. I'm working here in average up to 14 hrs per day and I have two locations. One is the lab with office and the other one a storage and the semi-anechoic chamber. What you see on page One (above the system description) is only the VGA testing lab. But I have here also hi-res infrared (thermal) cams for stills and radiometric videos and a few more places to test nice things like CPUs, thermal grease, PC-Audio, workstations and so on. I'm also the co-author of the latest CPU launches (simply take a look at the authors in bottom line). But I do here not only my job for Tom's. I do here a lot of product development and quality management for a few big companies. It's fulltime und we are all a big team :)
  • AgentLozen
    Thanks for your reply. I enjoy reading about your work.
  • 10tacle
    Sure will be nice when when Tom's development team adds back the "comment from the forums" option.
  • 10tacle
    251426 said:
    All this power and OC arguments makes me laugh... AMD provided a stronger card in DX12 than the 1080 GTX and all I see are complains about power and OC. Guess what, the 64 and 56 are still more powerful than their 1080 and 1070 counterparts with no drivers optimization.


    Sorry to have to break the news, but when you throw in overclocking a factory overclocked GTX 1080 which still consumes far less power than the reference RX 56, that DX12 gap is gone (keywords: still less power consumption even overclocked).

    Guru3D's average FPS of a GTX 1080 STRIX OC overclocked to 1846MHz - 2126MHz core/boost at 1440p:

    Hitman 2016 - 124
    Rise Of The Tomb Raider - 111
    Dues Ex: Mankind Divided - 81

    Guru3D's average FPS of the RX 64:

    Hitman 2016 - 107
    Rise Of The Tomb Raider - 90
    Dues Ex: Mankind Divided - 74

    Guru3D's review of that ASUS GTX 1080 pretty much showed what it's capable of in overclock scaling, and note that the factory overclock baseline of that card straight out of the box is still faster than the RX 64:

    http://www.guru3d.com/index.php?ct=articles&action=file&id=33149

    251426 said:
    It will take 2 months to know really what to expect from Vega.


    And in 5-6 months Nvidia will roll out the consumer level Volta (GTX 2xxx series).
  • redgarl
    Photonboy, you are comparing a 500 USD card with a 750+ USD card. Here in Canada, the 1080 GTX is still more expensive than a Vega 64. It is the launch right now, third party supplier didn't offer their cooling options and PCB design. I see all this negative press about cooling and throttling... guess what, Fender Edition cards were even the same and Nvidia was charging 100$ more to add to the irony.

    My EVGA 1080 FTW died for a second time in less than 10 months, so don't think Nvidia is so much better. In my experience, my Nvidia cards were having way more issues. My BFG 8800 GTX died on me, now this, however my 290x were working fine and my 6850s also.

    By the way, remember RX 480... 3-6 months later? Where the performances jumped almost 10 percent due to drivers optimization? I am not doing some wishful thinking at all. This is exactly what happened in the past with their Polaris architecture. Not 5 years ago, barely 18 months ago.

    People tend to forget about Infinity Fabric and AMD involvement in the console market that is not going to stop consolidating. Not taking that into consideration is a mistake.

    It is not their best card obviously, however it's not such a monstruosity. They beat Nvidia offering for a little less and did it with HBM2. Power consumption is bad, however it is made for playing games. Your 750 W power supply can easily handle it so i don't see the big fuzz over it. It's a concern, but not to the point of being a deal breaker.
  • redgarl
    10Tacle

    I have seen other bench saying otherwise so please... as of now, the 64 is beating the 1080 GTX without any proper drivers, fact... for cheaper. My whole argument...


    My last experience with Nvidia...


    https://www.newegg.ca/Product/Product.aspx?Item=N82E16814487245&cm_re=evga_1080_ftw-_-14-487-245-_-Product

    The latest 1 egg review is mine. Says everything. Go ahead an OC your card... I was not even doing it and this happened twice.

    There is never a mention about building quality, always performances... this is what you get in the end.
  • redgarl
    Also, Toms did mention they did not know if their sample was bad because they seen better results elsewhere. This whole discussion is pointless.
  • petergreen958
    Until these annoying persistent auto play videos stop I am out of here. Sad really as I am a long time fan.
  • 10tacle
    I provided an across the board comparison from one site using the exact same core hardware in GPU comparisons and benchmarks. Take it as you wish. Bottom line for me: the Vega RX 64 will not be getting a recommend from me when giving advice to people trying to decide between it and a GTX 1080. And here's why:

    https://tpucdn.com/reviews/AMD/Radeon_RX_Vega_64/images/perfwatt_2560_1440.png
    https://tpucdn.com/reviews/AMD/Radeon_RX_Vega_64/images/perfdollar_2560_1440.png

    And the RX 64 being cheaper than the GTX 1080? You sure about that Redgarl?

    https://www.newegg.ca/Product/ProductList.aspx?Submit=ENE&N=100007708%20601301447%20601194948&IsNodeId=1&bop=And&Page=2&PageSize=36&order=BESTMATCH
  • artk2219
    202972 said:
    I provided an across the board comparison from one site using the exact same core hardware in GPU comparisons and benchmarks. Take it as you wish. Bottom line for me: the Vega RX 64 will not be getting a recommend from me when giving advice to people trying to decide between it and a GTX 1080. And here's why: https://tpucdn.com/reviews/AMD/Radeon_RX_Vega_64/images/perfwatt_2560_1440.png https://tpucdn.com/reviews/AMD/Radeon_RX_Vega_64/images/perfdollar_2560_1440.png And the RX 64 being cheaper than the GTX 1080? You sure about that Redgarl? https://www.newegg.ca/Product/ProductList.aspx?Submit=ENE&N=100007708%20601301447%20601194948&IsNodeId=1&bop=And&Page=2&PageSize=36&order=BESTMATCH


    Man I knew prices were bad, but those are horrific. Vega definitely needs another process and core revision, im not sure if we will get one though with Navi not so far away. Hah what am I thinking, I was saying the same crap about Vega a year ago :) . Also, if anyone is looking for the forum link for this thread it's below. For some reason the site designers seem to have gotten it in their heads to strip away comment voting and a simple link that points to the friggin forum. Are they trying to change this to Anandtech's crummy outdated forum design?

    http://www.tomshardware.com/forum/id-3499133/amd-vega-tom-hardware-liquid-cooled-edition.html
  • 10tacle
    ^^Yeah the price gouging is outrageous, but honestly it was to be expected. I'd love to know the ratio of gamers vs. miners snapping up the Vegas causing resellers like NewEgg to jack their prices up.

    Regarding Tom's commenting from articles and no option to comment from the forum format, it's a known issue and they are working on it. It happened after the new login change (and it's not the only problem either they are working on). Until they fix it, the only way we can access the forum commenting format in articles is to post a comment to one with that generic lame interface and then go into tracked threads.