Powercolor HD 7990 Devil 13 6 GB Review

[RussianSensation]

8 stores in Europe seem to have it for sale: http://geizhals.at/eu/830039

€ 999

 

[Jeff007245]

I'm waiting for it to pop up on Newegg or even better Amazon (no tax/shipping) so I can give Powercolor, the retailer, and AMD my money already.. hard earned money btw, but well worth the pleasure of having it in the long run... btw I don't care about which card is faster by 5 to 10 fps. As long as I can play my games in 3d eyefinity smoothly, and be more productive with my spreadsheets and projects, I'm happy. I like my money to be spent on stuff that match and are visually pleasing. Pleasureable life > money... And for me AMD > Intel/nVidia... don't bother arguing or asking why, its merely brand preference and knowing I have the best of what the underdog has to offer (which btw is more than enough for anyone with a 3year electronics life cycle).

Thanks for the reviews... and loving the triple monitor setup performance on hardware heaven...

 

[cmdrdredd]

I'm waiting for it to pop up on Newegg or even better Amazon (no tax/shipping) so I can give Powercolor, the retailer, and AMD my money already.. hard earned money btw, but well worth the pleasure of having it in the long run... btw I don't care about which card is faster by 5 to 10 fps. As long as I can play my games in 3d eyefinity smoothly, and be more productive with my spreadsheets and projects, I'm happy. I like my money to be spent on stuff that match and are visually pleasing. Pleasureable life > money... And for me AMD > Intel/nVidia... don't bother arguing or asking why, its merely brand preference and knowing I have the best of what the underdog has to offer (which btw is more than enough for anyone with a 3year electronics life cycle).

Thanks for the reviews... and loving the triple monitor setup performance on hardware heaven...

you are full of crap. dual 7970 or 7950 overclocked will beat thios overpriced piece of garbage.

 

[chimaxi83]

you are full of crap. dual 7970 or 7950 overclocked will beat thios overpriced piece of garbage.

Despite this common opinion (which I share), the card will still sell out, just like every other overpriced specialty dual GPU card.

 

[3DVagabond]

The results of the card are pretty obvious given that AMD chose not to mass-release this card like previous generations.

With the current 7970 ghz pricing, this seems like a HUGE price premium over getting 2x 7970 solutions.

The 690 is a much better dual-GPU card (performance is slighyl better/more consistent and much better thermals) for about the same price.

Even if you need a lot of number-crunching power, it would be cheaper to just build an extra machine with a cheap dualie and put more 7950s in there.

AMD doesn't even make this card. The obvious reason, to me, is power usage.

We've got 2 reviews with varying results. There were issues with TPU, for sure. I'm not sure how much of it might have been the particular card or not. Clearly HH's card performed much better. Trading blows in most games and completely dominating in the newest releases. If you look at the 2 pieces of hardware you can see the cost involved with making the 7990 that justifies the price (If any of the prices this round are justified?). There's also the added VRAM and bandwidth for Eyefinity/Surround resolutions. I'm not so sure that the 690 is really a much better card for the settings/resolutions these cards are needed for. As far as compute power, that is one metric that the 690 isn't even in the same league as GCN.

To be clear, I'm not saying that there aren't situations that the 690 wouldn't be the better choice. It's just not clearly the better card overall.

 

[3DVagabond]

My post:

Reading your review you mention actually playing the games, multiple runs, etc. I'm assuming that if there were any anomalies you would have mentioned it. just to be clear though, did you see anything like micro-stutter, screen tearing, crashes, artifacts, features not working or driver settings not available? Anything like that?

P.S. Thanks for taking the time for posting in response to the concerns voiced. :thumbsup:

Veridian3's response:

Yes absolutely, we/I play through sections of the game which are pre-chosen multiple times. Essentially when reviewing the games themselves we note the demanding sections, save a game there and use it for benching the cards at a later date.

The real world testing is the only way to properly identify all/any issues and while it would be sooooo much easier and quicker to do timedemos, inbuilt benches or just 3DMark and similar they rarely give results which are the same as users will see... or thorough enough to pick up any flickering textures, micro stuttering and the like. (Shogun 2/Total War is a great example, it's inbuilt benchmark mode doesn't give results even close to the framerates when playing.)

Nothing is 100% perfect obviously but for this particular product there were no noticeable issues.

I built a system for a friend of mine here with everything matching what I have in my sig except a Maximus V Formula mobo instead of the Gene. We were undecided on GPU setup for 2560x1440 and I searched everywhere to try to get info. Nobody tests microstutter and shows SLi vs crossfire config except a few numbers of 3dmark and game benches etc. That's utterly useless because it doesn't tell you the gameplay experience. So I resorted to finding someone who owns each setup and getting feedback.

To make a long story short, I found someone who owns both a 670SLI and 7950 crossfire setup on the same mobo and CPU config clocked to the same speed. The answers I received was that while most benchmarks show a 7950 averaging higher that doesn't tell the whole story. He told me that he does notice that the crossfire setup has some points in gameplay where the frames literally jump. He can't diagnose it as a hardware issue with the CPU/mobo or anything like that. It's subtle he said but he definitely notices it. He told me the 670 SLI has much less and he can't really see it on that setup. However, the caveat he mentioned with SLI was that it doesn't seem to hold the FPS as high and will be lower overall which creates a slightly less fluid picture than optimal anyway. So he said the game plays smoothly but might not always look it on SLI while crossfire feels jittery and affects gameplay. In his words "crossfire 7950 breaks the gameplay significantly at times."

I have never personally used crossfire but I was considering this setup for the system I built because of the pricing. We decided that the gameplay smoothness was more important than looking at the FPS numbers. So this personal observation was what we based our decision and ultimately decided to go SLI 670s.

Microstutter does indeed matter and should be part of the discussion. Nobody should put up blinders and say "you can't see it anyway...only .0000001% of the users notice it" or whatever. It's like people who claim anything more than 30fps you don't notice, or 60fps. It does make a difference to some so yeah...why not have that discussion and present your findings?

According to the reviewer from Hardware Heaven there were no issues. On his system (admittedly higher end than a typical gaming rig) with current drivers (I'm assuming it wasn't 12.8 on your friend's system?) he noticed no issues. I specifically asked about micro stutter and screen tearing. I did this because I knew people would bring it up, as they always do, even though it was never mentioned as an issue.

 

[cmdrdredd]

My post:
According to the reviewer from Hardware Heaven there were no issues. On his system (admittedly higher end than a typical gaming rig) with current drivers (I'm assuming it wasn't 12.8 on your friend's system?) he noticed no issues. I specifically asked about micro stutter and screen tearing. I did this because I knew people would bring it up, as they always do, even though it was never mentioned as an issue.

I don't trust reviewers, I trust real users. There's many forum posts around the net where people mention micro stutter being more an issue with crossfire. One reviewer doesn't override that.

You need to read my post again. I built a system for a friend with no video card chosen. We asked people who owned each solution for input and I managed to find someone on another forum who had both. His experience with both side by side is what made our decision and yes he had the latest drivers because I asked for that specifically.

 

[SickBeast]

I'm waiting for it to pop up on Newegg or even better Amazon (no tax/shipping) so I can give Powercolor, the retailer, and AMD my money already.. hard earned money btw, but well worth the pleasure of having it in the long run... btw I don't care about which card is faster by 5 to 10 fps. As long as I can play my games in 3d eyefinity smoothly, and be more productive with my spreadsheets and projects, I'm happy. I like my money to be spent on stuff that match and are visually pleasing. Pleasureable life > money... And for me AMD > Intel/nVidia... don't bother arguing or asking why, its merely brand preference and knowing I have the best of what the underdog has to offer (which btw is more than enough for anyone with a 3year electronics life cycle).

Thanks for the reviews... and loving the triple monitor setup performance on hardware heaven...

I just really don't know what you're doing running a card like that on an 1100T CPU. You're really going to be held back.

Not only that, but why not get a pair of 7950s and put the $500 you'll save toward a better CPU and motherboard?

 

[SickBeast]

I don't trust reviewers, I trust real users. There's many forum posts around the net where people mention micro stutter being more an issue with crossfire. One reviewer doesn't override that.

You need to read my post again. I built a system for a friend with no video card chosen. We asked people who owned each solution for input and I managed to find someone on another forum who had both. His experience with both side by side is what made our decision and yes he had the latest drivers because I asked for that specifically.

I used to feel that way but as time goes on there seem to be more and more shills on the forums. I find that the best policy is to read several reviews before you buy. I find that the users on the forums can sometimes bring up interesting things like overclocking potential, but there are really way too many heavily biased posters lately.

 

[Keysplayr]

I just really don't know what you're doing running a card like that on an 1100T CPU. You're really going to be held back.

Not only that, but why not get a pair of 7950s and put the $500 you'll save toward a better CPU and motherboard?

Won't work. He only buys from the underdog. As he stated.

 

[TakeNoPrisoners]

Those are some seriously messed up drivers. For some games where the drivers actually work the card does pretty well.

But then you see stuff like this:

Pathetic.

 

[Jeff007245]

Won't work. He only buys from the underdog. As he stated.

Good job. Also I already have a good motherboard. Just waiting for the right time to drop a new CPU to upgrade it. For the time being, this setup should hold me off for a while until I will need to upgrade anything.

Btw, running 3x Samsung S27A950D monitors. I only play Diablo 3 and Guild Wars 2. And i use my computer for music listening, office suites, browsing, network storage, and most importantly projects prrojects and projects... I don't benchmark anymore like I used to when I was 12 to 16. You grow out of it after a while... And now 10 years later, i just appreciate what I have. The single XFX 7970 Ghz edition was doing fine, until it decided to crap on me a week ago. So I decided to return it and hold out for this card, which will only be better.

System:

Case: Lian-Li PC-50R AMD Dragon Platform Edition
Motherboard: Asus Crosshair V/Thunderbolt
CPU: AMD Phenom II 1100t (needs future replacement)
Cooler: Corsair h80 (2x Aerocool Shark Red Devil Edition in push/pull)
Memory: 16GB (4 x 4GB) Mushkin Redline 2133
PSW: XFX 1000 Watts Platinum Black Edition
SDD: Samsung 830 256mb
HDD: Western Digital Red 1TB x 2 Raid 5
Optical: LG BD-R

I have 3 to 5 years product life cycle on my main build. My next major overhaul will be when DDR4 becomes mainstream by 2014 to 2015. Until then i'll upgrade as needed... And even then I will still build an AMD platform. Why you ask? Because I simply would not feel the difference between the best system AMD and Intel offers... That is just the bottom line and real-time user experience...

 

[3DVagabond]

I don't trust reviewers, I trust real users. There's many forum posts around the net where people mention micro stutter being more an issue with crossfire. One reviewer doesn't override that.

You need to read my post again. I built a system for a friend with no video card chosen. We asked people who owned each solution for input and I managed to find someone on another forum who had both. His experience with both side by side is what made our decision and yes he had the latest drivers because I asked for that specifically.

You can choose to believe who you want, and I'll believe reviewers before I believe anonymous forum posters.

 

[Borealis7]

crossfire scaling is not impressive with this one...but it can play Crysis 1200p 4AA at 60 FPS (average, don't know about the minimum)

 

[3DVagabond]

crossfire scaling is not impressive with this one...but it can play Crysis 1200p 4AA at 60 FPS (average, don't know about the minimum)

We are going to need more reviews. In the review at TPU the card definitely had issues. Different outcome though at Hardware Heaven. I'd say the jury is still out on this one.

Now, with that said, I'd opt for crossfire 7950's, and keep $300-$400 in my pocket, before I bought either the 690 or 7990.

 

[Borealis7]

i think the TPU Performance Summary tables are not relevant in dual GPU cases since not all game run Xfire/SLI correctly and that skews the end result. so a 7990 is not 11% slower than a 690 in my opinion, it's just broken in more games (Batman? really?)

 

[-Slacker-]

Once again, please read all the reviews in the thread, which were all posted several times each:







http://www.hardwareheaven.com/revie...rd-review-power-temps-noise-overclocking.html

 

[RussianSensation]

You need to read my post again. I built a system for a friend with no video card chosen. We asked people who owned each solution for input and I managed to find someone on another forum who had both. His experience with both side by side is what made our decision and yes he had the latest drivers because I asked for that specifically.

:thumbsup: Sometimes frames per second do not convey the actual smoothness of playing a game. The other factor is that GTX670 is faster than 7950 out of the gate. So unless those 7950s were also overclocked to 1100mhz+, from the get go GTX670 SLI would have been faster.

But then you see stuff like this:

Pathetic.

Ya, that's a fair point. I wonder how @ HH, BF3 on 7990 ran faster and they got ~ 49 fps average and TPU shows 30.

 

[raghu78]

I think techpowerup got a faulty card.

http://www.techpowerup.com/reviews/Powercolor/HD_7990_Devil_13/

"Important: All data in this review was obtained after I repaired a major design error of the card. When I received it, the card ran extremely high temperatures, reaching beyond 100°C, causing instability and black screen hangs. The fan would also run at 100% almost instantly. This is caused by screws with integrated stop that resulted in too little mounting pressure between GPU and heatsink, clearly an engineering oversight. I added three metal washers on each screw by removing the screws, adding the washers and putting the screws back. The cooler was removed for this procedure, and thermal paste was replaced with a thin layer. Only this procedure enabled the card to run stable. The card was unusable out of the box."

BF3 in the hardwareheaven works perfectly on the HD 7990. At 5760 x 1080 the GTX 690 has poor min fps and the HD 7990 is definitely better. techpowerup should have returned the sample when its not functioning out of the box. the techpowerup review is not a true reflection of the hardware's performance.

http://www.hardwareheaven.com/revie...-7990-graphics-card-review-battlefield-3.html

 

[The Alias]

it seems to me the only reason the 690 won was because cf scaling sucked horribly in a few game

 

[RussianSensation]

Based on Hardware Heaven's videos, both the 690 and HD7990 sound too loud for $1000 videocards. I can't see any reason at all to pick either card unless you have a small case. I'd take 2x GTX670 SLI Asus DCUII cards, or Sapphire Vapor-X 7970s, or spend $100 more and get GTX680 MSI Lightning SLI over the 690/7990.

 

[3DVagabond]

Based on Hardware Heaven's videos, both the 690 and HD7990 sound too loud for $1000 videocards. I can't see any reason at all to pick either card unless you have a small case. I'd take 2x GTX670 SLI Asus DCUII cards, or Sapphire Vapor-X 7970s, or spend $100 more and get GTX680 MSI Lightning SLI over the 690/7990.

It's impossible to tell from a recording when the only test criteria given is, "Here is a comparison of the fan tone from 1ft."

They also state, "Temperatures were also lower on the PowerColor card when gaming although the fan noise is louder." It would have been nice if they had attempted to set up a fan profile as well that kept temps equal and checked noise. It's not like the difference in temps was trivial. It was 8-15 degrees cooler for the 7990. What would the 690 sound like if it was running 68-73*c instead of 81-83*c?

Don't get me wrong, unless it was either cheaper, or I didn't have room/setup/mobo for individual cards, I would opt for 2x cards myself. It's just that these tests aren't very scientific and could be more inclusive, IMO.

 

[RussianSensation]

Ya, that's a good point. You can set up a custom fan curve on the 7990 card and let the temperatures rise to 77-80*C to reduce the noise levels.

 

[f1sherman]

Ya, that's a good point. You can set up a custom fan curve on the 7990 card and let the temperatures rise to 77-80*C to reduce the noise levels.

WOW!

So now we have 7990 Devil actually producing LESS NOISE than GTX 690??

That is some pretty wild stuff right there.
Particularly if you keep in mind that we're talking about likely the 2nd most loudest card ever built (1st being 6990),

vs GTX 690 which is arguably the most silent dualie in GPU history.

And keep in mind that unlike 690, EVERY single Watt that Devil spits out adds to the heat inside the case.

 

[ViRGE]

WOW!

So now we have 7990 Devil actually producing LESS NOISE than GTX 690??

That is some pretty wild stuff right there.
Particularly if you keep in mind that we're talking about likely the 2nd most loudest card ever built (1st being 6990),

vs GTX 690 which is arguably the most silent dualie in GPU history.

And keep in mind that unlike 690, EVERY single Watt that Devil spits out adds to the heat inside the case.

Triple-slot card. The wider the card, the larger you can make the heatsinks and fans, the quieter you can make the card overall. This is the same reason Asus has been favoring that design for their high-end DirectCU II cards.