Sign in with
Sign up | Sign in

PowerColor Devil 13 Dual Core R9 290X 8 GB Review: Dual Hawaii on Air

PowerColor Devil 13 Dual Core R9 290X 8 GB Review: Dual Hawaii on Air
By

PowerColor’s Devil 13 graphics card, with its two Hawaii GPUs and massive heat sink, weighs in at more than two kilograms and exudes luxury. But can it compete with AMD’s dual-GPU reference design with closed-loop water cooling? Let’s find out!

To call PowerColor’s air-cooled dual-GPU project daring would be an understatement. We know darned well how hot AMD's Hawaii GPU can get. Just one is enough to push most thermal solutions to their limits. That's why AMD went with liquid cooling for its Radeon R9 295X2 (a card we found to be well-built, by the way).

Now, PowerColor is trying to improve upon the first workable reference design we've seen from AMD in years with a gigantic air cooler. Is this an act of deft engineering or blind ambition?

Since AMD's sampling policy (at least over in Europe where they tested these cards) is best described as weird, we went the extra mile to make this head-to-head comparison and bought our own Radeon R9 295X2. It performs a little worse than the card we have in the U.S., but the difference is small. There's no reason to suspect anything other than normal variance between them. 

Air versus water. A huge chunk of metal versus hoses and a radiator. We can hardly wait.

The three-slot bracket on the back of PowerColor's Devil 13 Dual Core is the first hint that this card means serious business. Picking it up, the thing appears built like a tank, and you have to wonder if your motherboard's PCI Express slot will take all of the weight. This is a graphics card you need to hold with two hands.

PowerColor is asking the same $1500 that AMD wants for its own Radeon R9 295X2. But the Devil 13 board attempts to differentiate itself by including a Razer Ouroboros in its bundle. That's a nearly $140 add-on. And the mouse is extremely comfortable, too. What's more, at least at the time of writing, Newegg is running a special on the Devil 13 that includes a 120 GB Samsung 840 EVO valued at $100.

Before we get started, let’s take a look at the specifications we're dealing with. After all, PowerColor promises us the world’s fastest graphics card, even though the Devil 13 operates at 18 MHz below the peak clock rate of AMD's Radeon R9 295X2. But company representatives are sure they can compensate for this small disadvantage with a higher memory frequency.

Ask a Category Expert

Create a new thread in the Reviews comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 43 comments.
This thread is closed for comments
Top Comments
  • 16 Hide
    derekullo , July 21, 2014 9:41 AM
    In my house at Russia I use this to stay warm in winter and scare away brown bears.
Other Comments
  • 3 Hide
    fl-gators-fan , July 21, 2014 1:19 AM
    Looking at page 4, that screen is insane!! Love it!
  • 0 Hide
    jlwtech , July 21, 2014 1:37 AM
    Was this card staying at 1000Mhz during those benchmark runs?
    The FPS difference at 1080p is ~7%, yet the clock difference is less than 2%.

    Also,
    Did you find the Devils maximum stable overclock? With all that power available, I'd imagine this thing could achieve better overclocks than the 295x2.
  • 2 Hide
    FormatC , July 21, 2014 1:46 AM
    OC makes no sense, because the card will be really loud. And please read the review attentively! The performance difference between both cards reflects the difference in power consumption nearly 1:1! To handle this cooling by air, the power color card uses a lower power target. Since AMDs Power Tune and Nvidias Boost the pure core clock rates says nothing about the final performance! In my eyes this is also a good study about the limits of an air-cooler.

    Take a look at the page with the HiRes power draw. This card isn't a perpetuum mobile. Less power consumption = less gaming performance. OC brings really nothing. Ok, you can destroy your ears... (or the card). We had to handle this rare card very carefully, so I was not able to break the voltage barrier.
  • 3 Hide
    jlwtech , July 21, 2014 2:08 AM
    Quote:
    OC makes no sense, because the card will be really loud. And please read the review carefully! The performance difference between both cards reflects the difference in power consumption! To handle this cooling by air, the power color card uses a lower power target. Since Power Tune and Boost the pure core clock says nothing abou the final performance!

    Take a look at the page with the HiRes power draw. This card isn't a perpetuum mobile. Less power consumption = less gaming performance. OC brings really nothing. Ok, you can destroy your ears...


    "Destroy my ears"?
    "OC makes no sense because the card will be really loud"? Are you serious??????
    Since when has that stopped anyone?
    This thing is quiet compared to high end cards from 5+ years ago....

    Have we become so spoiled by the advances in technology, that has enabled higher performance at lower noise levels, that we will not push the limits in fear of a little noise!?


    It's an ultra high end GFX card made for the kind of people who like to push the limits. It should absolutely be overclocked and benchmarked. With a big fat mind-blowing power usage chart with figures higher than any card has ever pushed!

    Also, clock rates still directly correlate to performance. Lowering the power tune limit will limit clock rates, and vice versa. Lower clock rates equals lower performance, but lower power does not always equal lower performance.
  • 1 Hide
    Mac266 , July 21, 2014 3:27 AM
    EDIT: Typo Fixed :D 
  • 2 Hide
    jlwtech , July 21, 2014 3:40 AM
    I just got done reading 3 other reviews for this card, and each of those reviews had this card slightly above the 295x2 in their gaming benchmarks. (despite the Devil's boost clock being slightly lower).
    That seems a little odd.

    The games, game settings, drivers, average clock rates, and Bios mode used for the benchmark figures/comparison, are not listed in this review. (unusual for Toms) That information would be very helpful.
    I suspect that the 295x2 was maintaining a higher average clock rate, in this comparison. (higher than 2%, anyways)
    PowerColor, almost certainly made sure that this card meets-or-beats the performance of the 295x2, before sending it off for review.
  • 7 Hide
    Menigmand , July 21, 2014 3:56 AM
    It's interesting to see this kind of enormous, powerful, noisy card being developed in a market where most games are designed to fit the limits of console hardware.
  • 9 Hide
    mapesdhs , July 21, 2014 3:57 AM

    Why have they called this a 290X? Rather confusing, it should be called a 295X.
    Having it listed in the 290X section on seller sites is dumb. Also, it's not an 8GB
    card, it's a 2x4GB card. I really wish tech sites would stop GPU vendors from
    getting away with this inaccurate product spec PR. Call it for what it is, 2x4GB,
    and if vendors don't like it, say tough cookies. The user will never see '8GB' so
    the phrase should not be used as if they could (though PowerColor seems happy
    to have such misleading info on its product page). I'm assuming you agree with this
    Igor, because the table on pp. 11 does refer to the Devil 13 as a 2x4GB... ;) 

    Btw, checking a typical seller site here (UK), the cheapest 290X is 1040MHz core,
    so given the Devil 13 uses 3 slots anyway, IMO two factory oc'd 290Xs make more
    sense, and would save more than 300 UKP.

    Ian.

    PS. The typo Mac266 mentioned is still present.

  • -4 Hide
    jlwtech , July 21, 2014 4:44 AM
    Why did the person who wrote this article focus primarily on power consumption and efficiency?

    This review has 4 pages of power consumption/efficiency data, with some impressively detailed information. But, it only has 2 pages of actual performance data, with almost no details at all.

    Who wouldn't want to see this card overclocked to a ridiculous extent, with plumes of smoke coming off of it, and the only power consumption figures showing that it's consuming more power than any other card ever made?


    (I had to edit this comment. That first revision was a little crazy.)

  • 6 Hide
    bemused_fred , July 21, 2014 5:38 AM
    Quote:
    The primary focus of this review is absurd!
    Why did the person who wrote this article focus primarily on power consumption and efficiency?

    This review has 4 pages of power consumption/efficiency data, with some impressively detailed information. But, it only has 2 pages of actual performance data, with almost no details at all.



    The performance is so close to the performance of the liquid-cooled R9-295X that it would basically be a repeat of that review. If you want an idea of its performance, just re-read that review and maybe reduce each frame rate by 3%-ish.

    OT: 60Db? Into the trash it goes. I don't care how expensive a card is, if it's too loud that I can't have a goddamn normal conversation near my computer, it's going in the skip. I don't want to surround myself in an anti-social bubble of noise-induced hearing loss every time I want to game.

  • -5 Hide
    jlwtech , July 21, 2014 6:14 AM
    Quote:
    Quote:
    The primary focus of this review is absurd!
    Why did the person who wrote this article focus primarily on power consumption and efficiency?

    This review has 4 pages of power consumption/efficiency data, with some impressively detailed information. But, it only has 2 pages of actual performance data, with almost no details at all.



    The performance is so close to the performance of the liquid-cooled R9-295X that it would basically be a repeat of that review. If you want an idea of its performance, just re-read that review and maybe reduce each frame rate by 3%-ish.

    OT: 60Db? Into the trash it goes. I don't care how expensive a card is, if it's too loud that I can't have a goddamn normal conversation near my computer, it's going in the skip. I don't want to surround myself in an anti-social bubble of noise-induced hearing loss every time I want to game.



    So what if the performance is close? EVERY OTHER review Toms has done on aftermarket flagship cards has been performance-based, why would that be a reason to stop now?
    This is the first custom version of the fastest GFX card on earth, why wouldn't we want to see a performance-based review?
    Why would anyone want to see power efficiency numbers opposed to performance numbers, on a card like this? Does that really matter to the kind of person who would actually buy a $1500 GFX card? Of course not.

    "anti-social bubble"? That's pretty thin...
    Do you tend to make conversation with people in the room, while gaming at full tilt? Sounds kinda distracting...
    5+ years ago, 60DB+ was par-for-the-course for any high-end card like this. GTX480. GTX295. HD6990. HD5970. HD4870x2, the list goes on and on....
    Besides, 60DB is on an open test bench at close range. Real-world usage will be much quieter. Check some other reviews to see more realistic figures.

    Everyone is getting spoiled these days.....
  • -4 Hide
    jlwtech , July 21, 2014 6:33 AM
    Quote:
    OT: 60Db? Into the trash it goes.

    Would you really throw a $1500 card in the trash, simply because it was loud?
    I'd just wear headphones....
  • 1 Hide
    anthony8989 , July 21, 2014 8:26 AM
    Quote:

    "Destroy my ears"?
    "OC makes no sense because the card will be really loud"? Are you serious??????
    Since when has that stopped anyone?
    This thing is quiet compared to high end cards from 5+ years ago....

    Have we become so spoiled by the advances in technology, that has enabled higher performance at lower noise levels, that we will not push the limits in fear of a little noise!?


    It's an ultra high end GFX card made for the kind of people who like to push the limits. It should absolutely be overclocked and benchmarked. With a big fat mind-blowing power usage chart with figures higher than any card has ever pushed!

    Also, clock rates still directly correlate to performance. Lowering the power tune limit will limit clock rates, and vice versa. Lower clock rates equals lower performance, but lower power does not always equal lower performance.


    You seem to have missed the part where he said ``We had to handle this rare card very carefully, so I was not able to break the voltage barrier.`` Hence no overclocking.

    Quote:
    Also, clock rates still directly correlate to performance.


    You`ve not tested many cards I can see. Clock rates have diminishing returns once you pass a certain point. This point varies from GPU to GPU and even from individual cards. At that point, it`s not worth the ridiculous amount of excess noise to gain 1-5 %. In addition to instability that would definitely follow due to a lack of reliable voltage.
  • 1 Hide
    anthony8989 , July 21, 2014 8:27 AM
    Quote:

    Why have they called this a 290X? Rather confusing, it should be called a 295X.
    Having it listed in the 290X section on seller sites is dumb. Also, it's not an 8GB
    card, it's a 2x4GB card. I really wish tech sites would stop GPU vendors from
    getting away with this inaccurate product spec PR. Call it for what it is, 2x4GB,
    and if vendors don't like it, say tough cookies. The user will never see '8GB' so
    the phrase should not be used as if they could (though PowerColor seems happy
    to have such misleading info on its product page). I'm assuming you agree with this
    Igor, because the table on pp. 11 does refer to the Devil 13 as a 2x4GB... ;) 

    Btw, checking a typical seller site here (UK), the cheapest 290X is 1040MHz core,
    so given the Devil 13 uses 3 slots anyway, IMO two factory oc'd 290Xs make more
    sense, and would save more than 300 UKP.

    Ian.

    PS. The typo Mac266 mentioned is still present.



    Legally they are not allowed. AMD wants that moniker reserved only for their reference model.
  • 1 Hide
    CaptainTom , July 21, 2014 8:43 AM
    LOL The Titan Z also uses 400w just like this 3-slot 295X2. Kepler is so much more power efficient. /s
  • 3 Hide
    anthony8989 , July 21, 2014 8:51 AM
    Quote:

    "anti-social bubble"? That's pretty thin...
    Do you tend to make conversation with people in the room, while gaming at full tilt? Sounds kinda distracting...
    5+ years ago, 60DB+ was par-for-the-course for any high-end card like this. GTX480. GTX295. HD6990. HD5970. HD4870x2, the list goes on and on....
    Besides, 60DB is on an open test bench at close range. Real-world usage will be much quieter. Check some other reviews to see more realistic figures.

    Everyone is getting spoiled these days.....


    60db is too loud when quieter options are available for the same price or less. Most people get around 58db from the stock 290x blower when it hits its "über" mode @ %55 or w/e :

    http://www.youtube.com/watch?v=QQhqOKKAq7o - Uhhh no thx.
  • 1 Hide
    FormatC , July 21, 2014 9:07 AM
    Quote:
    The primary focus of this review is absurd!
    Why did the person who wrote this article focus primarily on power consumption and efficiency?

    This review has 4 pages of power consumption/efficiency data, with some impressively detailed information. But, it only has 2 pages of actual performance data, with almost no details at all.


    Read the review!
    Quote:
    We're using the 2014 VGA Charts database for comparative benchmarking. This gives us a great basis for creating an index with all games and tests taken into consideration. If you want the individual test results, check out the charts section itself, where you'll find all of the individual numbers composing the index and have the option to create comparison tables.


    For all the blinds again as separate link:
    http://www.tomshardware.com/charts/2014-vga-charts/benchmarks,175.html

  • 0 Hide
    ingtar33 , July 21, 2014 9:09 AM
    Quote:
    I just got done reading 3 other reviews for this card, and each of those reviews had this card slightly above the 295x2 in their gaming benchmarks. (despite the Devil's boost clock being slightly lower).
    That seems a little odd.

    The games, game settings, drivers, average clock rates, and Bios mode used for the benchmark figures/comparison, are not listed in this review. (unusual for Toms) That information would be very helpful.
    I suspect that the 295x2 was maintaining a higher average clock rate, in this comparison. (higher than 2%, anyways)
    PowerColor, almost certainly made sure that this card meets-or-beats the performance of the 295x2, before sending it off for review.



    I haven't seen a review yet that this card outperforms the r9-295x2. In every review this card goes into a case it temp throttles pretty badly. the only time it seems to not temp throttle is when it's on an open bench.

    I think everyone is missing the point of this card. This card isn't meant to be used with the stock cooler. With 4x 8 pin power plugs, this card is meant to be put on a water block and then overclocked to the moon. this card was designed to shove as much power as you can into it. the air cooler is an afterthought.

    they likely saved more money sticking the air cooler onto it, and giving out that mouse, then the stock heatsink/waterloop would have cost them.

    What they need to do is review this card like it was intended to be used. on a waterblock.
  • 2 Hide
    FormatC , July 21, 2014 9:33 AM
    Quote:
    What they need to do is review this card like it was intended to be used. on a waterblock.

    Prohibited by vendor. After the thirth reviewer has the card disassembed, stressed and "repaired" this piece is simply waste. OC with this hoover? Beware...

    And for the others: the card is not able to hold ther stock clock rates stable without increasing the fan rpm to a horrible level. Have fun with this orchestra. My conclusion was simply and short: this cooler ist limited.
  • 16 Hide
    derekullo , July 21, 2014 9:41 AM
    In my house at Russia I use this to stay warm in winter and scare away brown bears.
Display more comments