• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

7970: Another Disappointment from AMD

Bear in mind that the 6970 will already have had redundancy built into its area usage, so the 7970 must have massive redundancy to fall so far off even the 6970 figure in terms of transistors per mm^2. I guess we'd all been reading about the troubles at TMSC though.
 
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.

Good point, maybe because people don't really expect it from AMD with the way they have gone with their CPUs and GPUs currently and in the past few generations.

Also there is an argument as to which generation this really is (in terms of performance). I guess it's "supposed" to be next gen, but perhaps people feel as though it's not groundbreaking enough. Especially as most people are expecting nvidia's kepler to be just that. Although now that AMD have shown their hand first, nvidia can now play things how they want.
 
I want to see what the OC headroom is like with decent cooling and vGPU increase. Imagine how quick it will be if it can hit 1.3+GHz
 
Don't worry guys, drunkenmaster will be along shortly to tell you you're all wrong.

You must have missed his one post on the subject since the reviews came out.

It's not like him to be so quiet.

I wonder why?

ACtually I've said many many times in the past that generally speaking you'd be comparing non refresh products, you can go back and Look, I've been saying that for 5 years, 4870 vs 5870 is a fairer comparison and gave you around the 80% mark. 7970 over 5870 is both, double the transistors, 5870 is 2.15billion, 7970 is 4.3billion, Cayman is 2.64billion.

It IS 80% faster than the 5870, its even more than 80% faster in many cases.

Its actually hugely faster than the 6970 in many cases, unfortunately its best speed comes in eyefinity, which will always be a very niche product simply because most people won't buy 3 screens, won't set up 3 screens with an angle, have a desk that can comftably use 3 screens or want to, I don't want to either. 3d performance I will be very interested in as one screen + 3d is FAR more appealing than 3 screens in the space I don't have for them, though 3d isn't as appealing as 120hz.

Where AMD went wrong is ignoring performance at 1080p/1600p for performance less than 5% of top end kit buyers will be using.

Also, the price.......... oh well, I was wrong on the price, but personally I think it hasn't been helped by peoples attitude........ we've had 6 months of people talking about new gen and how it will be expensive. If we had 6 months of people saying on every forum worldwide you'd never pay more than £350 for a card, it would likely never have gone beyond £350.

We'll also have to see where prices end up in a month or two, and when the GK104 launches.

As for yields, I don't think that has much to do with it, AMD are advertising 1Ghz clocks with ease, stock heatsink is hitting 1.1Ghz+ good heatsinks will likely hit well beyond that, it could be a monster at higher clock speeds without extreme cooling. I've yet to see proof yields are bad, capacity on a process and yields are NOT the same thing, not even close.

Price though, at £300 its a genuine 80% faster than 5870 card at the same price point on a new process, randomly being sold at £420-500(seen pretty much all prices listed now in reviews) its a bit of a joke. This is coming across as a new CEO with a new plan and that is to milk as much money as they can, but I get the feeling in doing so they'll be losing more customers and sales than they make with higher prices.

Oh, and techpowerup have been AWFUL for donkeys years now, COD 4, the first one, as a performance benchmark which just luckily pushes the 6990 right into the 580gtx, removed all of AMD's best games, disabled xfire in the one game AMD excelled in then finally removed that game.

Their % comparisons are literally worthless as they've added more and more completely CPU limited games and come up with both the least fair and the least relevant benchmarking suite of essentially any website out there.

A 590gtx is less than 20% ahead of a 580gtx, a 6990 the same, its laughable how badly they've ruined the site with CPU limited rubbish, COD 4 and World of warcraft.

If you simply go through and throw out the CPU limited benchmarks to get a fairer comparison you won't get anything even remotely close to those final numbers.
 
ha,
i'm not going to buy one !
But no doubting its the faster then the 580GTX :p

Seriously you kids need to grow some ball shaped lemons. When your Student grants need to be paid back you need to ask yourself, was it really worth it? :p (that is right before you start to advise others they should protest)
I do wonder how good half you guys are playing games anyway with your 60+ FPS in the poor lagfest PS3 ports.

Happy Christmas,
maybe think of those that have sod all this time of year rather then a £500+ GFX and argue about that.
 
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!

Are you s*****g me? To compare the 7970 to Bulldozer in any sense is going beyond hyperbole and enters the realm of fanboy rhetoric and out-right lying.

Is the 7970 as fast as we'd hoped? No, of course not, but expectations are often unrealistic. Is it faster that the nearest competition? Yes. Is it a performance and architectural improvement on AMD's previous generation? Again, yes. If you were given one tomorrow would you immediately replace any single unit with it? Hell Yes! All four of the previous questions would be a resounding No with Bulldozer.

Driver improvements will work on minimum and maximum FPS values, just as they did with the 480 and other cards based on a new generation and architecture. But people need to stop making out that the card is worthless, just as elements of this forum did with the 480 last year. It's not, and it's a bloody good card.

Christ you guys are jaded sometimes :P
 
OP is trolling!! :o

They have done what was required and released a card that is better than the current king of the hill.

Why is it a fail?

Flip it upside down, and havent Nvidia failed for allowing AMD to take the performance crown? I doubt Nvidia are anywhere near ready with their next Gen.

I wouldn't call it fail, but given the size of the die shrink and the historical performance jump on a major generation revision like this its a bit underwhelming, especially when the changes to the shader architecture should have seen decent increases in efficency even in older DX10 titles. Its a bit telling when pro-AMD sites are reviewing it against the "latest" nvidia card... the GTX560Ti 448 instead of the GTX580 as much as they can get away with.

I suspect most people wouldn't care if they priced it competitively against the GTX580 but the suggested pricing is more of that of a card that decimated the 580 instead of one that often struggles to convincingly beat it (tho it does generally beat it).
 
Was anyone really expecting it to be cheaper than a 580? I mean really. The only reason ATI gpus have been less expensive than NVIDIA lately is because they are slower! Remember how much the X1900XTX cost?

Whether it's 5% or 80% faster, it's going to be more expensive.

I'm just hoping for a used 6970 or 580 for cheap, because I don't want to wait until prices start dropping on the new cards...because it's probably not going to happen until kepler comes out
 
This is an interesting response, and something that seemed odd to me too. As the poster below it indicates, AMD has only increased (useful?) transistor count by 64%, yet we would have expected around 100% by consideration of area alone. This seems to suggest that there is a lot of redundant silicon on the 7970. I read an article a while back that said that in the face of poor manufacturing processes, designers will double or triple up important elements that might not fab right, leading to redundant silicon. Not doing enough of this on an immature process is what led to nvidia's problems with the GTX480/470 release. So it could well be that 28nm is still not working at all well. By the time Nvidia come to market it might be working rather better, which would be bad news for AMD.

Against the most favourable reviews, I suppose an increase of ~40% on 64% more transistors isn't all that bad, especially if they can improve it with better drivers.

So perhaps it's a combination of a slight design fail and a big process fail.

Two things, if there was lots of redundancy, transistor count would most likely have been up, and 5870 already had what was likely similar levels of redundancy, you're not really talking about SP's or anything major but some via's and other things doubled up, this cost 10-15% die size on the 5870, but its likely the proportion would likely stick.

Other things worth noting, 40nm and 28nm are just PR, the real numbers aren't that, I have no idea what they are, but I've heard from multiple people these are rounded up/down. LIkewise HKMG for AMD at least costs around 10% die size, and its quite possible 28nm is really bigger than that, significantly so. They have 28nm varients that have no HKMG afaik, so its perfectly possible 28nm is for low power non HKMG, and the HP process is really pushing 31-32nm.

Secondly, transistor count "usually" doubles from the initial core on a process to the initial core on the next process. 4870/5870, 5870/7970..

It also seems Cayman took up rather a large chunk of the efficiency improvement between 5870/7970, moving to VLIW4 improved the shader efficiency dramatically.

Personally I've also thought and we always seem to see AMD underspec ROP's, which seems to somewhat limit their "high end" performance. THose several games where not cpu limited on older games Nvidia can run away off to 200fps in something old while AMD tends to be limited.


Anyway, so far, still as far as we know, the only failure is price. it should be cheaper, simple as that. I don't like Nvidia pricing, full stop, but they ARE making 530mm2 cores with frankly awful yields, still. It would take something along the lines of, 40-50% cost increase of wafers to make a 360mm2 28nm core cost more to produce than a 530mm2.

The ONLY comparison I would make to Bulldozer is, have AMD designed a 28nm core for higher speed's to help bring latency down for compute work, but missing the higher speed bins of a not really ready 28nm could potentially have hurt it? Maybe, who knows.

I think there is more to come from drivers, this is an entirely different architecture, meaning NO games are optimised for it, not a single one, and that shouldn't be ignored either. But driver wise, VLIW 4 drivers have come on quite a long way, there are a lot of situations a 6970 is significantly ahead of a 5870 now, something that wasn't necessarily the case on launch. But VLIW 4 drivers weren't drastically different from VLIW 5. GCN will be utterly different, though also simplified, still takes time to get used to it and it will take time for AMD drivers to be optimised for games on a new architecture(where possible) and it will take time for games to come out with dedicated optimisations to work better with GCN.

That could be anything from 5% to god knows what, time will tell.

Still, even considering all that performance in extreme situations is stupid faster. Frequently(from the minimal reviews I've seen including results) 80-100% ahead in eyefinity situations which also again suggests to me there isn't enough raw horsepower in the front end to push the frames out, but in the backend when in silly power requireing situations and lower framerates, the front end is less limited and the shader power really comes out, and its crazy good.

At 300 this card would rock.
 
Still, even considering all that performance in extreme situations is stupid faster. Frequently(from the minimal reviews I've seen including results) 80-100% ahead in eyefinity situations which also again suggests to me there isn't enough raw horsepower in the front end to push the frames out, but in the backend when in silly power requireing situations and lower framerates, the front end is less limited and the shader power really comes out, and its crazy good.

At 300 this card would rock.

80-100% ahead in Eyefinity situations? What review(s) have you seen this on? I am genuinely interested as I run Eyefinity and the likes of HardOCP (important as it tests real gameplay) and WSG are only showing ~ 50% improvements once averaged across the board.

I agree on your ROP comment though. I can't quite get my head around why this card only has 32 ROPS. Given that it has the same memory controller layout as the GTX580 (high level look of course) I would have expected 6 back ends with 8 ROPs each for a total of 48 ROPs. The structure of the back end also seems unbalanced considering the 6 memory controllers.
 
Was anyone really expecting it to be cheaper than a 580? I mean really.
Nope, people were expecting it to be only as high price as GTX580, not HIGHER than it, so it would push the price of GTX580 down. But at the moment, all they've done is stack the price on top of an already overpriced card (GTX580).

7970 is on average only 8-12% faster than the GTX580 at 1920 res, and they are asking for 12-14% higher in price so people are simply paying more money for more performance, and not a single bit more on the "bang for bucks", despited moved from 40nm to 28nm. I think this is what people not happy about. If the 7970 was only 5-8% more expensive than the GTX580, then it would be better bang for bucks.

Alternatively...AMD should release 7970 on 1.5GB at a lower price toward the "single 1920 res monitor users crowd that prefer single card solution" to truely slap GTX580's 1.5GB's face. Honestly, the performance in terms of having 3GB VRAM vs 1.5GB VRAM has little to none difference in 98% of the situations/games on only 1920 res. I'm sure pretty most people from that category would rather pay £350~£380 for a 7970 1.5GB, rather than £420~£450 for 7970 3GB.

On the other hand, AMD's top end card (not talking about dual-GPU cards) were never good value/bang for bucks comparing to its mid-range (i.e. 5870 vs 5850, 6970 vs 6950), so we should wait and see what 7950 delivers, and at what price point.
 
Last edited:
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.

+1

The amount of stupidity that some members here show is simply shocking.

Plus, these cards overclock extremely well. It looks like AMD were purposely very conservative with the clock speeds (probably leaving room for a highly clocked 7975 of some sort?) as mosts reviewers got a 200Mhz + OC with stock volts. Scaling is real good, too.
 
Last edited:
Compare GTX580 3GB vs 7970 is going to be pretty close above 1920x1200 with max IQ. Real world useage will be virtually identical gaming experience as Nvidia cards usually offer a smoother minimum FPS.

GTX590 or 6990 are still both faster solutions. Nvidia will do what they always do wait for AMD pricing then drop their highend pricing in early Jan to hurt the launch as usually they suddenly find GPU stock from somewhere ;) I would expect Nvidia to do something silly in Jan like take £100 off the GTX580 price if they really want to hurt AMD's launch!

Nvidia have a few months to tweak their next-gen to beat this as no games come close to stressing current highend its pretty pointless AMD launching ahead of Nvidia & handing an advantage to Nvidia every single time.

Few gamers have the funds burning a hole in their pockets right now for 7970 you can max every single current game on current GPU's. This is an overkill product launched at the wrong time AMD should have spent much longer on performance & launched same time as Nvidia to maximise potential. Not having silent fans means add another £70+ for a silent solution which makes this even more expensive :rolleyes:

UK pricing will be the usual price gouge event $ for £ lunacy so expect it around the £550 mark then perhaps higher until stock sells out. If you believe different you also believe in santa claus ;) or have not witnessed enough UK high end GPU launches!!

If you conducted a Pepsi challenge blind gaming test with current highend GPU's most gamers would struggle to tell them apart. AMD would have slightly better IQ but Nvidia would provide smoother minimum FPS & superior drivers so its a draw if you want to be very generous to AMD & all down to buyer brand preference/price ;)
 
Compare GTX580 3GB vs 7970 is going to be pretty close above 1920x1200 with max IQ. Real world useage will be virtually identical gaming experience as Nvidia cards usually offer a smoother minimum FPS.

GTX590 or 6990 are still both faster solutions. Nvidia will do what they always do wait for AMD pricing then drop their highend pricing in early Jan to hurt the launch as usually they suddenly find GPU stock from somewhere ;) I would expect Nvidia to do something silly in Jan like take £100 off the GTX580 price if they really want to hurt AMD's launch!

Nvidia have a few months to tweak their next-gen to beat this as no games come close to stressing current highend its pretty pointless AMD launching ahead of Nvidia & handing an advantage to Nvidia every single time.

Few gamers have the funds burning a hole in their pockets right now for 7970 you can max every single current game on current GPU's. This is an overkill product launched at the wrong time AMD should have spent much longer on performance & launched same time as Nvidia to maximise potential. Not having silent fans means add another £70+ for a silent solution which makes this even more expensive :rolleyes:

UK pricing will be the usual price gouge event $ for £ lunacy so expect it around the £550 mark then perhaps higher until stock sells out. If you believe different you also believe in santa claus ;) or have not witnessed enough UK high end GPU launches!!

If you conducted a Pepsi challenge blind gaming test with current highend GPU's most gamers would struggle to tell them apart. AMD would have slightly better IQ but Nvidia would provide smoother minimum FPS & superior drivers so its a draw if you want to be very generous to AMD & all down to buyer brand preference/price ;)

Apart from metro 2033.
 
How can you say that when the card isnt even out yet lol. Not to mention the drivers.
Draw your own conclusions from why AMD have allowed certain reviews to leak which do not compare GTX580 3GB vs 7970 ;) Metro 2033 is a pretty poor gameplay experience & does not even look that good anyway BF3 looks better if you want a high end GPU eye candy title!
 
People forget that nVIDIA's 580GTX is a hugely bigger chip. For Cayman to have been so close to it was incredible, as was for Cypress and RV770 before.

Now, with AMD shifting focus to compute, it might be the first time that NVIDIA's die size will actually translate to decent performance increases over AMD.

Anyway, the only thing really wrong with the new cards is the price. Its too high, but people usually buy the fastest card anyway, especially to avoid the dual card issues.
 
Back
Top Bottom