Power, Temperature, & Noise

Moving on from performance metrics, we’ll touch upon power, temperature, and noise. This is also normally where we’d discuss voltages, but as Vega is a new chip on a new architecture, nothing seems to read Vega 64 and 56 correctly.

In terms of average game clockspeeds, neither card maintains its boost specification at 100% with prolonged usage. Vega 64 tends to stay closer to its boost clocks, which is in line with its additional power overhead and higher temperature target over Vega 56.

Radeon RX Vega Average Clockspeeds
  Radeon RX Vega 64 Air Radeon RX Vega 56
Boost Clocks
1546MHz
1471MHz
Max Boost (DPM7)
1630MHz
1590MHz
 
Battlefield 1
1512MHz
1337MHz
Ashes: Escalation
1542MHz
1354MHz
DOOM
1479MHz
1334MHz
Ghost Recon: Wildlands
1547MHz
1388MHz
Dawn of War III
1526MHz
1335MHz
Deus Ex: Mankind Divided
1498MHz
1348MHz
GTA V
1557MHz
1404MHz
F1 2016
1526MHz
1394MHz
 
FurMark
1230MHz
HBM2: 868MHz
1099MHz
HBM2: 773MHz

With games, the HBM2 clocks ramp up and stay at their highest clock state. Expectedly, the strains of FurMark cause the cards to oscillate memory clocks: between 945MHz and 800MHZ for Vega 64, and between 800MHz and 700MHz for Vega 56. On that note, HBM2 comes with an idle power state (167MHz), an improvement on Fiji's HBM1 single power state. Unfortunately, the direct power savings are a little obscured since, as we will soon see, Vega 10 is a particularly power hungry chip.

As mentioned earlier, we used the default out-of-the-box configuration for power: Balanced, with the corresponding 220W GPU power limit. And under load, Vega needs power badly.

Idle Power ConsumptionLoad Power Consumption - Battlefield 1Load Power Consumption - FurMark

The performance of both Vega cards comes at a significant power cost. For the RX 500 series, we mused that load consumption is where AMD paid the piper. Here, the piper has taken AMD to the cleaners. In Battlefield 1, Vega 64 consumes 150W more system-wide power than the GTX 1080, its direct competitor. To be clear, additional power draw is expected, since Vega 64 is larger in both shader count (4096 vs. 2560) and die size (486mm2 vs. 314mm2) to the GTX 1080. But in that sense, when compared with the 1080 Ti, powered by the 471mm2 GP102, Vega 64 still consumes more power. 

As for Vega 64's cut-down sibling, Vega 56's lower temperature target, lower clocks, and lower board power make its consumption look much more reasonable, although it is still well above the 1070.

In any case, the cooling solutions are able to do the job without severe effects on temperature and noise. As far as blowers go, RX Vega 64 and 56 are comparable to the 1080 Ti FE blower.

Idle GPU TemperatureLoad GPU Temperature - Battlefield 1Load GPU Temperature - FurMark
Not Graphed: Temperature of the actual Vega (Star): 9329C

Noise-testing equipment and methodology differ from past results, with a more sensitive noise meter and closer distance to the graphics card. Readings were also taken with an open case. As such, the noise levels may appear higher than expected.

Idle Noise LevelsLoad Noise Levels - Battlefield 1Load Noise Levels - FurMark

 

Synthetics Final Words
Comments Locked

213 Comments

View All Comments

  • FourEyedGeek - Tuesday, August 22, 2017 - link

    Your reflexes aren't fast enough
  • Aldaris - Monday, August 14, 2017 - link

    NV fanboy alert.

    Tell me, in what world did those results suggest to you it's slower?
  • Manch - Tuesday, August 15, 2017 - link

    ddriver calls them an Intel/Nvidia shill.
    Vladx calls them an AMD/Apple shill

    I think it was fair and balanced :D
  • sor - Monday, August 14, 2017 - link

    Performance wise it actually seems pretty good. People were worried it wouldn't even be able to compete with a 1080, but in many cases it slots between the 1080 and the Ti. The killer though is that power consumption. Burning 100+ more watts is insane. Otherwise, seems like it was a nice, competitive card.
  • blublub - Monday, August 14, 2017 - link

    This excessive power draw is, and many ppl forget that, node related.

    It's the same as with Ryzen:
    GloFo's 14nm is low power plus! Meaning it's very power efficient up to a certain frequency but once it surpasses it it drinks electricity like an elephant in steroids.

    It can be seen with Ryzen and Polaris, drop frequency and voltage and power goes down more than proportionally.

    AMD just didn't have enough money and was bound to GloFo so they couldn't take out different GPU sizes and on a different process
  • FreckledTrout - Monday, August 14, 2017 - link

    Yeah but they do have a shining light in that IBM bought 7nm process, its high frequency should really help both AMD's GPU and CPU's a lot.
  • Manch - Tuesday, August 15, 2017 - link

    You can't drink electricity. I get your point though.

    Make like a tree and get the out of here!
  • Yojimbo - Monday, August 14, 2017 - link

    It'll be interesting to see how much game developers take advantage of double rate FP16. Maybe there are some bottlenecks that can be alleviated without impacting quality much.
  • beck2050 - Monday, August 14, 2017 - link

    Over clocking seems very limited with that power draw. Custom 1080s are often 10 to 15% faster out of the box and still cooler and less power hungry.
    A bit disappointing.
  • mapesdhs - Monday, August 14, 2017 - link

    I mentioned that elsewhere, in the UK a 1080 with a 1759MHz base is 60 UKP cheaper than a Vega64/Air, and one can get a 1080 Ti for the price of a Vega64/Liquid.

Log in

Don't have an account? Sign up now