PowerColor PCS+ R9 290X Review: Cool, Quiet, And Priced Right

Power Draw: Gaming, Web Browsing, And Idle

Measuring the Power Draw

The following diagrams contain a text header, which is followed by the actual measurement curves. In that header, you'll find the average and peak power draw on all three rails, along with the sum of those averages, giving us the total average and peak power consumption.

Total peak wattage is not simply the sum of the three individual peaks, but rather the total peak wattage observed within the 120-second sample window. That makes more sense than adding the peak wattages together, since it's unlikely that all three max out at the same moment.

Power Consumption during Gaming

These diagrams employ a logarithmic scale. The thick red line represents average total power draw at a time resolution of six seconds.

The average power consumption demonstrated by PowerColor's PCS+ R9 290X is 262 W, in line with what we expected. That's 3 W higher than the 30 MHz-faster MSI R9 290X Lightning. Approximately half of a watt can be attributed to the PowerColor card's memory, which operates at 1350 MHz, rather than 1250 MHz. But a few watts could be lost by less complex voltage regulation circuitry. 

Still, the differences are small. And while the PCS+ R9 290X may exhibit slightly higher power use, it also pulls about 10 W less from the motherboard slot than MSI's card.

Power Draw During Internet Surfing

People don’t spend all of their time gaming; typically, much more is spent browsing the Web. We simulate this workload with a static version of our home page, scrolling all the way down and back up again. While scrolling does make use of hardware acceleration, the power draw for that is quite benign.

Power Draw at Idle Load

Many people leave their PCs on all the time, making the power consumption of components at idle quite important. Measuring draw from the wall doesn't cut it if we want to isolate the graphics card, since there's no way to reliably factor out the rest of the platform.

Our state-of-the-art test equipment shows that this graphics card only draws 13.7 W at idle, which is less than our previous generation of lab instruments indicated:

  • combine1237
    For some reason I thought the pcs+ used hynix.
    Reply
  • CaptainTom
    I just want to point out that this and most 290X's beat a stock 780 Ti. The fact is both 780 Ti and the 290X are trade blows and belong on the same GPU tier. However only one does cost $150 less and come with 4GB VRAM...
    Reply
  • Memnarchon
    13232466 said:
    I just want to point out that this and most 290X's beat a stock 780 Ti. The fact is both 780 Ti and the 290X are trade blows and belong on the same GPU tier. However only one does cost $150 less and come with 4GB VRAM...

    Well, since you are comparing a non reference GPU, you should take also a non-reference GPU to compare.
    Tom's using Gigabyte's 780ti OC which costs the same as the reference card...

    So even the non-reference models are on a different tier as well as their reference...
    Reply
  • bemused_fred
    Any chance of measurements in metric as well as imperial?
    Reply
  • FormatC
    13232896 said:
    Any chance of measurements in metric as well as imperial?

    Typical translation errors, the original is in metric ;)
    http://www.tomshardware.de/powercolor-r9-290x-pcs-review,testberichte-241519-3.html4

    I will clearify with Chris, that we use both in the future. Metric is worldwide more common :D
    Reply
  • dave_trimble
    Surprised the benchmark graph show performance at 1080p. Aren't the 290 series kind of wasted at that resolution? I would love to see the results at 1440p or even 4k. I have a feeling the 780ti might not look quite as good in comparison at higher resolution.
    Reply
  • photonboy
    A stock GTX780Ti is 7% faster at 2560x1440, and 8% faster at 1920x1080 (18 games averaged):
    http://www.techpowerup.com/reviews/EVGA/GTX_780_Ti_SC_ACX_Cooler/26.html

    With such a HUGE difference in prices, quality, and performance on all top-end cards you really have to do your research.

    The R9-290X prices vary from $550 to $780 USD!
    Reply
  • dave_trimble
    13233877 said:
    A stock GTX780Ti is 7% faster at 2560x1440, and 8% faster at 1920x1080 (18 games averaged):
    http://www.techpowerup.com/reviews/EVGA/GTX_780_Ti_SC_ACX_Cooler/26.html

    With such a HUGE difference in prices, quality, and performance on all top-end cards you really have to do your research.

    The R9-290X prices vary from $550 to $780 USD!

    Thanks for the response! I thought I had seen reviews elsewhere that showed the 290 series really closing the gap, or even surpassing the 780ti at higher resolutions, but perhaps I was remembering wrong. I recently decided to upgrade to a dual-290 setup, but if I had gone with a single card, the 780ti was at or near the top of my list (until the 290x prices came crashing down, that is).

    Reply
  • That_Guy88
    So is the difference between 290x's almost entirely due to cooling (and some OC)? I have someone who wants to sell me a reference 290x for $350, but I have a gtx 770, so it would seem that I would need to buy an after market cooler as well to make it worth my while. Thoughts?
    Reply
  • vertexx
    13234604 said:
    So is the difference between 290x's almost entirely due to cooling (and some OC)? I have someone who wants to sell me a reference 290x for $350, but I have a gtx 770, so it would seem that I would need to buy an after market cooler as well to make it worth my while. Thoughts?
    See these articles for after-market cooling options:
    Air cooling:
    http://www.tomshardware.com/reviews/r9-290-accelero-xtreme-290,3671.html

    Liquid Cooling:
    How to:
    http://www.pcper.com/reviews/Graphics-Cards/Using-NZXT-Kraken-G10-Watercool-Radeon-R9-290
    And results:
    http://www.tomshardware.co.uk/radeon-r9-290-and-290x,review-32872.html
    Reply