7970: Another Disappointment from AMD

Dangerous_Dave · 22 Dec 2011 at 19:46

It seems like AMD can do no right recently: their Bulldozer chip was designed for a world that doesn't exist, and consequently is sometimes even outperformed by the old Phenom 2 Quads. Now we have this 7970, which by all account seems to be about 30-35% faster than a 6970, and (in terms of minimums) is only just ahead of the 40nm GTX580!

At 28nm you have roughly double the transistors per area compared with 40nm, yet the 7970, which is almost the same size as the 6970, is only 1/3 faster. It should be at least twice as far ahead of the 6970 even on immature drivers. Given that the 40nm GTX-580 is only just behind the 7970 specifically for minimums (which is really the only thing that matters), AMD is surely going to get destroyed when Nvidia moves to 28nm.

Think about it this way: For 3.4 billion transistors, AMD could have done no research at all and simply integrated two 6870s onto a single die (similar to 5870 vs 4870), ramping the clock speed up 20-30% to somewhere over 1Ghz (28nm would easily allow this). This would have produced performance somewhere close to a 6990, and far ahead of the 7970. Instead, AMD have spent lots of money on research and used 4.1 billion transistors to produce performance far worse than a 6990.

There has to be something wrong with a company when the fruits of their research are far worse than if they hadn't bothered doing any.

arc@css · 22 Dec 2011 at 19:56

If it had been priced lower and had been quieter then I think that would have sorted out the only real negatives that I can see that it brings.

Bobisuruncle54 · 22 Dec 2011 at 19:57

Read other reviews that show a 40-50% gain over the 6970 and 20-30% gain over the GTX 580. In some instances an overclocked 7970 is up there with a 6990, and that's on restricted settings in Overdrive.

Oh and a lot of transistors were spent on Compute performance, not just Gaming performance.

inSilica · 22 Dec 2011 at 20:01

OMG!

Here we go again! There is always one person trying to capture all the headlines.

Mr Paul · 22 Dec 2011 at 20:01

Some reviews its 60~70% faster than 6970, and some 30% faster than overclocked 580.

Not a fail at all.

Dangerous_Dave · 22 Dec 2011 at 20:02

Bobisuruncle54 said:
Read other reviews that show a 40-50% gain over the 6970 and 20-30% gain over the GTX 580. In some instances an overclocked 7970 is up there with a 6990, and that's on restricted settings in Overdrive.

Oh and a lot of transistors were spent on Compute performance, not just Gaming performance.

@Compute performance: Why? Couldn't they build separate subtly modified highly priced chips for that purpose?

@20-30% gain over GTX-580: Did those reviews show the figures for minimums? Can you link?

titaniumx3 · 22 Dec 2011 at 20:10

Dangerous_Dave said:
It seems like AMD can do no right recently: their Bulldozer chip was designed for a world that doesn't exist, and consequently is sometimes even outperformed by the old Phenom 2 Quads. Now we have this 7970, which by all account seems to be about 30-35% faster than a 6970, and (in terms of minimums) is only just ahead of the 40nm GTX580!

At 28nm you have roughly double the transistors per area compared with 40nm, yet the 7970, which is almost the same size as the 6970, is only 1/3 faster. It should be at least twice as far ahead of the 6970 even on immature drivers. Given that the 40nm GTX-580 is only just behind the 7970 specifically for minimums (which is really the only thing that matters), AMD is surely going to get destroyed when Nvidia moves to 28nm.

Think about it this way: For 3.4 billion transistors, AMD could have done no research at all and simply integrated two 6870s onto a single die (similar to 5870 vs 4870), ramping the clock speed up 20-30% to somewhere over 1Ghz (28nm would easily allow this). This would have produced performance somewhere close to a 6990, and far ahead of the 7970. Instead, AMD have spent lots of money on research and used 4.1 billion transistors to produce performance far worse than a 6990.

There has to be something wrong with a company when the fruits of their research are far worse than if they hadn't bothered doing any.

What's with all the armchair engineers these days? Not sure whether to laugh or cry after reading this pile of complete and utter nonsense.

Dangerous_Dave · 22 Dec 2011 at 20:12

titaniumx3 said:
What's with all the armchair engineers these days? Not sure whether to laugh or cry after reading this pile of complete and utter nonsense.

The other option would be to offer an intelligent rebuttal. Unless of course you're trying to audition for a job at AMD.

OpenToSuggestions · 22 Dec 2011 at 20:12

Card looks good, but too much £ for me I am afraid.

I am keeping my unlocked 6950 2GB.

nitram100 · 22 Dec 2011 at 20:38

It's just the price, wait and see what it sells for before making a judgement.

Marine-RX179 · 22 Dec 2011 at 20:42

Mr Paul said:
30% faster than overclocked 580.

Only scenerios I saw it came close to that are at 2560 res, AA applied and comparing to a GTX580 1.5GB rather than GTX580 3GB.

For 1920 res, the 7970 is merely 8-12% faster than the GTX580 1.5GB/3GB. Basically it's not really any better bang for bucks, considering people have to pay 12% higher in price than GTX580 for extra 8-12% speed.

The card itself is no slow-poke and it is indeed the current fastest single GPU card, but for a 28nm process card, it's extra performance is simply too little over the 40nm GTX580...so clearly AMD is not making use of the 28nm process anywhere near effectively.

Dangerous_Dave · 22 Dec 2011 at 20:46

Can somebody please link me all these great results for 7970 compared with GTX580? I would hate to think that people were plucking single results out at implausible resolutions/settings and/or not comparing minimums as in original post.

Dangerous_Dave · 22 Dec 2011 at 20:48

nitram100 said:
It's just the price, wait and see what it sells for before making a judgement.

Regardless of the price, considering the amount of silicon and the technology (28nm vs 40nm) involved, as per my original post, this has to be considered a massive disappointment. Unless somebody can provide links to the contrary.

ishoeyi · 22 Dec 2011 at 20:50

i just don't understand why everybody seem to think its ok for nvidia to charge a high price for their card and yet amd can't do that, when their card is now faster.

so amd will forever seen as a budget range.

Duff-Man · 22 Dec 2011 at 20:56

Dangerous_Dave said:
@Compute performance: Why? Couldn't they build separate subtly modified highly priced chips for that purpose?

It's not a subtle modification though... Both AMD and Nvidia went through dramatic architectural changes to produce their compute-oriented GPUs (Fermi and GCN). These are necessary in order to overcome the inflexibility of "pure graphics" VLIW-type designs for general parallel computing purposes.

That being said, I'd very much like to see AMD and Nvidia producing entirely separate GPU designs for gaming and compute markets. Clearly this would increase R+D costs, but it would certainly give them an advantage in terms of performance-per-transistor.

mjones1 · 22 Dec 2011 at 20:57

We will see what these cards really can do once a decent set of drivers ar eout, really do love how well it scales in crossfire!

:(Rhodri:) · 22 Dec 2011 at 20:58

Dangerous_Dave said:
Regardless of the price, considering the amount of silicon and the technology (28nm vs 40nm) involved, as per my original post, this has to be considered a massive disappointment. Unless somebody can provide links to the contrary.

The performance is pretty impressive from what I've seen so far, also bare in mind the drivers will bring further performance enhancements as time goes by. Just because you've doubled the number of transistors you should in no way expect double the performance, it just doesn't work like that.

555BUK · 22 Dec 2011 at 20:58

It's a bit disappointing, but definately not a failure because NVidia's response will not arrive for 6 months or so. Given the specifications for NVidia's Kepler, it should be a true upgrade when it eventually arrives.

Marine-RX179 · 22 Dec 2011 at 21:12

ishoeyi said:
i just don't understand why everybody seem to think its ok for nvidia to charge a high price for their card and yet amd can't do that, when their card is now faster.

Because this kind of performance is disappointing for going from 40nm to 28nm?

Nvidia managed to pull off close to similar performance on the old 40nm process at a year ago.

gareth170 · 22 Dec 2011 at 21:12

people are forgetting about the drivers it's running on, when the drivers are more optimized for the new series we most likely see more performance