AMD Radeon HD 7990 Graphics Card Reviews

[onethreehill]

PowerColor Devil 13 HD 7990 Graphics Card Reviews
http://www.hardwareheaven.com/review...roduction.html
http://www.techpowerup.com/reviews/P...7990_Devil_13/
http://www.hardwarecanucks.com/forum...evil-13-a.html
http://hexus.net/tech/reviews/graphi...7990-devil-13/

[tbone8ty]

http://www.techpowerup.com/reviews/P...7990_Devil_13/

shouldn't the title be more specific? as there is only 1 7990 as of right now :o

uh oh...powercolor boo boo

http://www.techpowerup.com/reviews/P...es/tempfix.jpg

[Andrew LB]

I find it interesting that Techpowerup somehow forgot to include the overclocked score made by the GTX 690. If you're going to run OC scores for one, you gotta do it for the other... or risk people questioning your journalistic integrity.



I also enjoy the conclusion where they say....

PowerColor is asking $999 for their card, which is the same as NVIDIA's GTX 690. I find it hard to recommend the PowerColor card at that price-point. NVIDIA's card provides the better overall experience, less power, noise, heat, better max OC performance, in a dual slot form factor, and better drivers.

... then end the article with this:




Seriously? lol.

[Jamesrt2004]

VERY expensive cards for me personally. I think (at least in UK) if you want dual, 2 7870/7950's are the way forward (not too hot on the knowledge of the nvidia cards atm)

[Spectrobozo]

I find it interesting that Techpowerup somehow forgot to include the overclocked score made by the GTX 690. If you're going to run OC scores for one, you gotta do it for the other... or risk people questioning your journalistic integrity.

they only overclock the main card subject of their review, it's nothing new on their tests

http://www.techpowerup.com/reviews/N...TX_690/31.html

the 7990 seems to gain more from overclocking, even if it's not overall faster in this game,

[Motiv]

so it's the single fastest card on the market today then?

[MaddMuppet]

they only overclock the main card subject of their review, it's nothing new on their tests

http://www.techpowerup.com/reviews/N...TX_690/31.html

the 7990 seems to gain more from overclocking, even if it's not overall faster in this game,

That's just silly really. It makes one card look much better than the other.

[BoredByLife]

That's just silly really. It makes one card look much better than the other.

Have you read the review? A table stating the max OC of the GTX690 FPS were shown in comparison to the 7990...

[naokaji]

2x 680 are cheaper than this.
3x 670 barely costs more than this.
2x 7970 are much cheaper than this.
3x 7970 barely costs more than this.
3x 7950 are cheaper than this.

Yet somehow techpowerup calls it highly recommended"?

[MaddMuppet]

Have you read the review? A table stating the max OC of the GTX690 FPS were shown in comparison to the 7990...

I wouldn't comment, if I didn't read it.

[SKYMTL]

I think you guys are mistaking the TPU OC numbers. The PowerColor card provides an automatic switch to run the card at higher clock speeds without the need for end user tweaking and "overclocking". The GTX 690 does not provide a feature like this. As such, I think the comparison is fair.

[Lanek]

I think you guys are mistaking the TPU OC numbers. The PowerColor card provides an automatic switch to run the card at higher clock speeds without the need for end user tweaking and "overclocking". The GTX 690 does not provide a feature like this. As such, I think the comparison is fair.

I was going to respond the same.. they have a dual bios, as many retail oc cards ..one OC, one standard. ( one at 925mhz, the other at 1000mhz ).

Interesting, Hardware heaven look to dont have got the same problem you can find on TPU with the cooler contact.

[BababooeyHTJ]

I don't know what is up with the tpu review. The difference between the power 13 and gtx690 in quite a few games is much larger than I saw with my 7950s and gtx 670s. One that really stood out is hard reset which scales very well for me.

I think that the triple slot cooler was a bad idea. It sort of kills being able to do tri or quad-fire on a standard z77 or something. Maybe EK will develop a block but with such a limited edition card I doubt that it'll happen. That really kills what I feel is the only real pro that these cards have over two normal cards in crossfire.

[boxleitnerb]

Not a word on microstuttering - only half a review. Seriously, what is up with that?

Is it so difficult to record some frametimes AND put 2-3 people in front of the keyboard, let them play a couple of games for a few minutes and have them write down their experiences in terms of smoothness in 1-2 sentences for each game? It's not like multi-GPU is tested on a daily basis. So if you do it 2-3 times a year, at least do it right and as thorougly as possible. Geez, the continued lack of coverage on microstuttering/frametime consistency is becoming really annoying. It is an integral part of AFR, so test it!

[BababooeyHTJ]

Not a word on microstuttering - only half a review. Seriously, what is up with that?

Is it so difficult to record some frametimes AND put 2-3 people in front of the keyboard, let them play a couple of games for a few minutes and have them write down their experiences in terms of smoothness in 1-2 sentences for each game? It's not like multi-GPU is tested on a daily basis. So if you do it 2-3 times a year, at least do it right and as thorougly as possible. Geez, the continued lack of coverage on microstuttering/frametime consistency is becoming really annoying. It is an integral part of AFR, so test it!

Most people don't even notice the blatantly obvious gamebyro stutter (64hz bug) in Oblivion, Fallout 3, and New Vegas. Most people don't notice the input lag in the HP LP2465 or Acer GD235hz that I noticed right away. I don't trust most people's opinions when it comes to microstutter. Most people on the forums don't even know what microstutter is, this thread is a great example of that.

How is someone like Anand going to test for that when he says that he doesn't notice 40ms of input lag in his lcd reviews?

[boxleitnerb]

That is why multiple opinions are necessary. To completely ignore this subject is no solution either. At least reviewers should make an effort.

[BababooeyHTJ]

That is why multiple opinions are necessary. To completely ignore this subject is no solution either. At least reviewers should make an effort.

Thats why you have to read about people's experiences on the forums. Reviews seem to miss driver issues all of the time. A good example is the vsync issue with Kepler. I know damn well that certain sites had that issue since it wasn't working in their 3d vision benchmarks. Reviews only tell you half of the story. It really is almost like most reviewers just run a benchmark without ever looking at the game.

Even if you are going to do a subjective microstutter comparison two or three people is no where near large enough a sample size to come to any reasonable conclusion. Techreport seems to make an effort in examining frametimes in their reviews. I guess that is one way to do it but I'll be honest I don't put too much stock into those numbers. They don't seem to tell the whole story either.

[boxleitnerb]

It really is almost like most reviewers just run a benchmark without ever looking at the game.

So true!
The problem I see is this:
Many people are not aware of microstuttering. They read a review, say "amazing, the 7990 does 50fps, the 690 only 45 and the 680 only 25! I get the 7990". In the end they may be disappointed because the experience may even be worse than with the 680@25fps (all hypothetical) and nobody told them in the first place. Not with a single word were they informed about the drawbacks of AFR. That VRAM is not doubled, that it relies on profiles etc. This is basic stuff and a must-know and I find it incompetent at best that SLI/CF are tested the same way single-GPU products are.

[Greg83]

what i found interesting is 7990 beating crossfire 7970 in that one test. here
was the oc button pressed , were they clocked the same or did somehow the 7990 beat crossfire 7970's at the same clocks in games too. even at lower clocks?
some benefit after all to 7990 over crossfire 7970's?

[SKYMTL]

People don't seem to realize this "microstutter" can be highly evident on single GPUs as well and its perception is highly subjective.

[Lanek]

what i found interesting is 7990 beating crossfire 7970 in that one test. here
was the oc button pressed , were they clocked the same or did somehow the 7990 beat crossfire 7970's at the same clocks in games too. even at lower clocks?
some benefit after all to 7990 over crossfire 7970's?

As they put in the temp, noise, overclocking page, i can imagine it was working at 1000mhz, including for their temp, noise, TDP tests. ( OC only with the sliders without voltage bump to 1158mhz ( cant say fully stable or not)
Then ofc, dont all reviewers redo tests of old gpu's as they dont have them and so reuse old numbers.

Like allways, take 4-5 different reviews, compare, and find what is in the middle between them.


about different thing, i have a little bit of problem to understand how the 7990 Little devil have an higher TDP in idle when the second core should normally be off as said by both review. The memory should play a role maybe.

[solofly]

People don't seem to realize this "microstutter" can be highly evident on single GPUs as well and its perception is highly subjective.

100% correct...(all I do is game, day in day out)

[BababooeyHTJ]

People don't seem to realize this "microstutter" can be highly evident on single GPUs as well and its perception is highly subjective.

Then its not microstutter. I think that the general consensus is that microstutter is the side effect of AFR that can make games appear to run more choppy than it should at a given framerate. If I'm not mistaken that is what the term microsutter was originally used to describe. I think that the problem is that most people don't really know much about the subject and use it to describe other things like a normal stutter that you might see with a single card.

[Cold Fussion]

People don't seem to realize this "microstutter" can be highly evident on single GPUs as well and its perception is highly subjective.

Microstutter is pretty noticeable. This is why techreport do the best reviews as you can see the rendering time traces, which makes it easy to spot whether it will be smooth or not.

[BababooeyHTJ]

Microstutter is pretty noticeable. This is why techreport do the best reviews as you can see the rendering time traces, which makes it easy to spot whether it will be smooth or not.

It may be noticeable to you and me but not everyone notices it for some reason. So he is right that it is subjective.