Sign in with
Sign up | Sign in

Which Radeon HD 7950 Is Right For You?

Radeon HD 7950 3 GB: Six Cards, Benchmarked And Reviewed
By

Summary And Conclusion

Overall, two very different graphics cards rise to the top. First, we have HIS' HD 7950 IceQ Turbo sporting a tall two-slot thermal design. It manages to edge out the other five boards, taking first place in performance and cooling, along with third place in acoustics. Second, there's Asus' HD 7950 DirectCU II, a heavier three-slot card that generates the least noise, finishes a close second in cooling performance, and still posts competitive benchmark results. PowerColor's HD 7950 PCS runs quietly, but suffers much warmer operating temperatures. And the other two factory-overclocked submissions are both hotter and louder. Finally, AMD's reference design demonstrated the worst performance, the least-attractive acoustic profile, and the highest temperatures. Its only redeeming quality is price.

There's an elephant in the room that keeps us from wholeheartedly recommending any Radeon HD 7950, though: GeForce GTX 670. Although two of the six cards we tested currently cost less than a reference $400 GTX 670, the rest actually cost more. The least-expensive Radeon HD 7950s give you an approximately $20 savings (before rebates), but neither showed particularly well here. Unless you have an application where the 7950's compute potential is absolutely imperative, you'll probably want to hold off until these cards become a little more affordable (which they're already in the process of doing).

Best Cooling Performance: HIS HD 7950 IceQ Turbo

The DHE cooling solution on HIS' HD 7950 IceQ Turbo really works. Not only does it cool well, but it outperforms the thermal solutions of the other cards we tested. It also manages to beat them in the gaming benchmarks without generating that much extra noise.

Although this board won our recommendation in Europe, we're not able to give it an endorsement in the U.S. As we've seen HIS do in the past, it's only selling this higher-clocked Turbo model overseas. North American customers can get the HD 7950 IceQ (no Turbo), but it runs at a reference 800 MHz and costs more than $400. Sure, it's more attractive than AMD's reference design. But it's certainly not a front-runner at that lower speed.

Best Noise Performance: Asus HD 7950 DirectCU II

If you don't mind losing three expansion slots to a single graphics card, and are alright tracking down a bracket to keep this heavy board in place, then Asus' HD 7950 DirectCU II does wonders to the baseline acoustic performance of AMD's reference Radeon HD 7950. Its cooling performance is bested only by HIS' impossible-to-buy HD 7950 IceQ Turbo, and an aggressive 900 MHz core clock rate tops our charts. 

Solid Choices: Sapphire HD 7950 OC, MSI HD 7950 Twin Frozr III, and PowerColor HD 7950 PCS

If you don't feel like stepping down to HIS' HD 7950 IceQ (since the Turbo isn't available here in the U.S.) and the wide/heavy Asus HD 7950 DirectCU II isn't doing it for you, then the other three cards in our round-up might be more interesting. All of them are solid choices capable of delivering admirable performance.

MSI HD 7950 Twin Frozr III: Solid design, top build quality; a recently-released BIOS makes it a great card for overclocking.

PowerColor HD 7950 PCS: Slim enough to enable two cards in a CrossFire configuration; very quiet; unfortunately, unplugging power connectors can be difficult.

Sapphire HD 7950 OC: Not as quiet; again, unplugging power connectors is difficult.

Dead Last: AMD's Reference Card

The only reason to consider the reference design is if you want to attach a third-party cooling solution. Otherwise, the price difference is just not large enough to justify buying a slower, louder, hotter card over any of the models in today's round-up.

Display all 72 comments.
This thread is closed for comments
Top Comments
  • 20 Hide
    rmpumper , June 18, 2012 4:59 AM
    Gigabyte HD7950 missing in action :( 
  • 13 Hide
    mayankleoboy1 , June 18, 2012 5:57 AM
    scrumworksThat nvidia turbo boost is a bit cheating and you should turn it off for the test. It's basically same as overclocking and your review sample cards are most likely binned to get much better than average OC.


    WOW.
    how is it cheating? it is cheating only if the average user will not use the "GPU boost" feature, but reviewers are using it.

    and, the card is faster, cheaper and cooler than the competition. so if you are better, you cheat?
    Fanboi.
  • 10 Hide
    BeWater , June 18, 2012 7:33 AM
    Testing GTX6XX-Cards with enabled GPU Boost is certainly not cheating. Its a just a feature of this boards.

    But everyone who is comparing stock clock benchmarks should keep in mind that these cards most likely wont overclock as much over their gpu boost frequency as a card without boost overclocks over its stock frequency (assuming both cards have the same overclocking capability). Especially since the gpu boost on non-reference cards sometimes already is 200mhz over their base frequency.

    This can make a precise comparision more difficult. Particularly if you compare cards which are very close like the 7970 and 670 for example.

Other Comments
  • 20 Hide
    rmpumper , June 18, 2012 4:59 AM
    Gigabyte HD7950 missing in action :( 
  • 5 Hide
    mayankleoboy1 , June 18, 2012 5:21 AM
    in some of the earlier charts, gtx680 is written as "gtx 680 1536MB".

    also, gtx580 seems to be missing in the crysis2 charts.
  • 6 Hide
    Anonymous , June 18, 2012 5:23 AM
    Just thought you might like to know that the x-axis captions for the noise level graphs are labelled as they would be for temperature, rather than decibels
  • -9 Hide
    Dragh0n , June 18, 2012 5:35 AM
    Nice article. Glad to know I'm not missing much by sticking with brand loyalty.
  • 6 Hide
    jimmysmitty , June 18, 2012 5:50 AM
    Why was there no expanded information on the overclocking ability of the GPUs? Half at 1025MHz and Half at 1050MHz doesn't help anything.

    As well, what kind of memory overclocks did you get? Did you have to change the voltage of the GPUs to get said clock speeds?

    This article was missing a bit too much info there to really be able to get the full picture of these GPUs. The HD7970 6 card shootout is how it should have been done for the HD7950s. Each cards overclocking ability (core/memory) and what they got with stock voltage and overvolted.
  • 13 Hide
    mayankleoboy1 , June 18, 2012 5:57 AM
    scrumworksThat nvidia turbo boost is a bit cheating and you should turn it off for the test. It's basically same as overclocking and your review sample cards are most likely binned to get much better than average OC.


    WOW.
    how is it cheating? it is cheating only if the average user will not use the "GPU boost" feature, but reviewers are using it.

    and, the card is faster, cheaper and cooler than the competition. so if you are better, you cheat?
    Fanboi.
  • 2 Hide
    SessouXFX , June 18, 2012 6:10 AM
    XFX HD7950 didn't make the list?

    You know what this is like? Imagine yourself as a concert prompter, trying to put a lineup together. You have Megadeth, White Snake, Poison, Guns -n- Roses, AC DC, you think you're doing just fine...What? Metallica wanted nothing to do with that roundup of Who's Who? Did they turn you down, or did you guys turn them down?

    Because XFX matters. XFX always matter! And I'm pretty sure they weren't scared to step into the ring and beat some heads in with a comparo with the best in the biz. You guys need to get in touch with them and test their HD7950 and see if their card has the potential to blast that "elephant in the room" and hang it's head on the wall.

    This is Tom's Hardware after all...
  • 5 Hide
    we_san , June 18, 2012 6:28 AM
    scrumworksThat nvidia turbo boost is a bit cheating and you should turn it off for the test. It's basically same as overclocking and your review sample cards are most likely binned to get much better than average OC.

    You want to add 7970 OC into the crowd ? Just ask ...
  • 4 Hide
    Deemo13 , June 18, 2012 6:29 AM
    I really like the charts posted for the benchmarking! Gives a much wider range of cards.
  • 0 Hide
    A Bad Day , June 18, 2012 6:38 AM
    HIS HD 7950's fan at load reminds me of one of those machinery rooms in a manufacturing plant.
  • 7 Hide
    keigo_kanzaki , June 18, 2012 6:40 AM
    This is the first time I heard about buying a "graphic card bracket". For the ASUS DCII card, don't you usually just put screws (3 screws for this card) on THE (usual) bracket to hold the card firmly in the case?
  • 3 Hide
    hellfire24 , June 18, 2012 7:10 AM
    DCII,i am impressed!
  • 1 Hide
    userngalan , June 18, 2012 7:19 AM
    wow! the msi and powercolor is 50$ cheeper than asus..
  • -9 Hide
    Sonny73N , June 18, 2012 7:21 AM
    Dear Mr Wallossek, we buy our cards based on performance per price as recommended by Tom's monthly best graphic cards for the money guide. This is the time of the GTX 670, simply because it has the best value in its class. Since they're available everywhere, why don't you give us some benchmark on those 670s instead of these over-priced (still) 7950 cards with nothing left to talk about?
  • 10 Hide
    BeWater , June 18, 2012 7:33 AM
    Testing GTX6XX-Cards with enabled GPU Boost is certainly not cheating. Its a just a feature of this boards.

    But everyone who is comparing stock clock benchmarks should keep in mind that these cards most likely wont overclock as much over their gpu boost frequency as a card without boost overclocks over its stock frequency (assuming both cards have the same overclocking capability). Especially since the gpu boost on non-reference cards sometimes already is 200mhz over their base frequency.

    This can make a precise comparision more difficult. Particularly if you compare cards which are very close like the 7970 and 670 for example.

  • 0 Hide
    xxplosiv88 , June 18, 2012 8:02 AM
    Wow, I feel even more awesome now for putting my order in for a GTX 670.
  • 2 Hide
    DataGrave , June 18, 2012 8:02 AM
    Quote:
    Dear Mr Wallossek, we buy our cards based on performance per price as recommended by Tom's monthly best graphic cards for the money guide. This is the time of the GTX 670, simply because it has the best value in its class. Since they're available everywhere, why don't you give us some benchmark on those 670s instead of these over-priced (still) 7950 cards with nothing left to talk about?


    Dear Sonny73N,
    Why? Because the translation is slower than a crappy GT 430 :D 

    The 7950 roundup was published in Germany: May 12, 2012
    The GTX670 roundup was published in Germany: May 22, 2012

    http://www.tomshardware.de/Nvidia-Geforce-GTX-670-Roundup,testberichte-241027.html


  • 0 Hide
    unksol , June 18, 2012 8:10 AM
    Quote:
    Why was there no expanded information on the overclocking ability of the GPUs? Half at 1025MHz and Half at 1050MHz doesn't help anything.

    As well, what kind of memory overclocks did you get? Did you have to change the voltage of the GPUs to get said clock speeds?


    So you read the 2nd sentence of the overclocking but not the first?

    Quote:
    We overclocked each card as much as possible without increasing their voltage settings. Half of them managed to hit 1050 MHz, while the other half maxed out at 1025 MHz.


    sorry couldnt help myself. the fact they all hit 1025Mhz is relevant. Doing it at stock voltage is the "safe" way. and a way to show what the "average Joe" can do safely with just the sliders in CCC.

    Its clearly just a puff piece for these cards and nothing to do any real comparison with when its only game benchmark is crysis 2
  • 1 Hide
    zander1983 , June 18, 2012 8:28 AM
    *looks at current pc setup, opens local pc shop webpage, see great deal no upgrade only to find out gadget budget is empty*
Display more comments