For many years NVIDIA has been much superior to AMD when it comes to compute performance, though now AMD is on par with Tahiti. Not that it matters too much, since 99% of these cards are used for gaming.
I am guessing the 1/4 DP speed is ruled out for GK104. That's how they are able to keep the chip so lean/small. If you look at HD7870, that's exactly what AMD has done. They made HD7870 a full-blown gaming chip. It only has 1/16th DP performance which makes it worthless for any real world computing work. But then they are able to have a 212mm^2 chip that can come close to HD7950 with some overclocking.
I wouldn't doubt it if GTX680 is a straight up gaming chip with neutered DP performance. Nvidia also likes to sell full featured computing chips in their Tesla and Quadro lines and sell them for $3-5k+. They artificially neuter their consumer line. For years AMD has been the brand to go to for consumer DP performance.
Not sure how this is relevant to G80 in context. Cypress was slower than GTX480/570/580, while G80 dominated 2900 and 3800 series entirely. Either way, his implication is obvious: why would Nvidia spend 2 years to launch a "high-end" Kepler (whatever the name will be GTX780/880, etc.) and then that card end up losing to the HD7970 that's just 25% faster than a GTX580? Doubtful.
Is it just me or are the posts out of order or what? I'm seeing posts that are quoted before the person even posted it and so on...
Please run the heaven bencmark at identical res and settings as he has in the link on a stock 7970 and provide the results here please. Heaven is actually a decent indicator of gaming performance to a certain extent.
Not sure how this is relevant to G80 in context. Cypress was slower than GTX480/570/580, while G80 dominated 2900 and 3800 series entirely. Either way, his implication is obvious: why would Nvidia spend 2 years to launch a "high-end" Kepler (whatever the name will be GTX780/880, etc.) and then that card end up losing to the HD7970 that's just 25% faster than a GTX580? Doubtful.
Could be Nvidia locked out the full functionality until the launch date drivers.
Could be GTX680 is not as fast as predicted.
Could be user error (didn't activate some +20% TDP maximum boost / some other feature in the driver panel)
Food for thought: Why is 28nm GTX680 that's supposedly high-end Kepler clocked at 706 mhz when GTX580 was clocked at 772mhz on 40nm process? GTX680 also has lower ROP count, same memory bandwidth.
Any of these are possible:
1) NV flopped and abandoned the large die strategy, there is no GK110;
2) GTX670 --> GTX680 is like HD5870 --> HD6870 (where the marketing name is not reflective of the true standing in the Kepler lineup)
3) Nvidia couldn't get the flagship out on time so they had to resort to using GTX680; GK110 will be GTX780. Perhaps the plan was always to counter HD7970 with a card +/-5% within HD7970 until they sort out whatever problems they are having with power, yields on the large die flagship.
4) Nvidia is loving the profit margins on a 294mm^2 chip. Milk the market/consumer. It seems consumers are eager to pay $550 for HD7970, why not pay $500 for GTX680?
5) Nvidia is having yield issues. It wouldn't' be feasible to get out GK110 right now at reasonable profit levels without pricing it at $800 (good luck with that).
Either way, looking more and more like this entire 28nm generation is lackluster. Even if GTX680 is 10% faster than HD7970, that's still not fast enough after 2 years since GTX480 (my opinion).
Same CPU, same settings on a stock HD7970.
Russian, it could be so simple as the GPUz screen was captured while the card was in one of its idle states.
I'm no engineer, but I'm not seeing the 680 providing the sort of performance advantage typically seen from nvidia's vs AMD's flagship. I don't think it's pessimistic, just realistic, to think that just getting the same performance of Tahiti with a few wins and some losses is impressive enough with a smaller die for nvidia.
I think there is a GK110 coming too, but don't believe nvidia is sitting on it because they can. This whole launch looks like nvidia having had to adapt their strategy because of having the same problems they had GF100. Considering how small the market is for high end cards they could bring out GK100 for $800 today if they had it ready and likely sell as many as they would if they launched it at $500.
I think a good answer to what this card is, is going to be how much faster is it than a GTX580 compared to their previous flagships compared to the prior. Given nvidia's prior history with flagships, I think 680 is going to give dissapointing peformance contrasted against their past flagship deliveries.
Since no one has offered to yet.. why ?!?!
Please run the heaven bencmark at identical res and settings as he has in the link on a stock 7970 and provide the results here please. Heaven is actually a decent indicator of gaming performance to a certain extent.
Only care about performance here myself, rest is just peripheral. :hmm:
Well like I said we have no official benchmarks. The whole debate of mid-range vs. high-end: impossible to say. For me, I keep in mind that since GeForce 3 Nvidia's next generation mid-range card has always either beaten or tied the previous Nvidia high-end chip. (GeForce 3 ti 500 < GF4 4200Ti, 5950U < 6600GT, 6800 U < 7800GT, 7900GTX < 8800GT/S, 8800GTX/U < GTS250/GTX260, GTX285 < GTX460 1GB). Nvidia can call GTX680, GTX980 for all I care. If it performs like a GTX660Ti (i.e, 20% faster than GTX580), then some people will view it as a "next generation" mid-range card priced at $550. Someone buying next week might not even care about this classification. To them, if it's faster than GTX580, it's Nvidia's "high-end" chip at that moment in time.
Thanks. Well if this is consistent across benches, what we can hope for are cheaper prices with two cards that perform the same trying to differentiate themselves. That or a premium and lots of discussion of fluff features like physx.
I am probably stuck waiting till year's end for an upgrade now. Blargh.
Thanks. Well if this is consistent across benches, what we can hope for are cheaper prices with two cards that perform the same trying to differentiate themselves. That or a premium and lots of discussion of fluff features like physx.
I am probably stuck waiting till year's end for an upgrade now. Blargh.
I remember I presented that scenario to you previously. It seems that is the way they are going.
I noticed a lot of trends have been breaking which I blame the overall finacial state of the world mixed in with the resource scare and the overall shift in industry dominace (consoles vs PC/mobile vs desktop/USD vs World).
Nothing last forever.
maybe that run is at 700 instead of 1006?
Unlikely IMO. The card is going to run it's clocks correctly when a bench or game is active. If there was anything wrong it would be that GPU-Z was run without loading the card(meaning the chip is running on power saving clocks) or that GPU-z is not reading the card correctly at all.
Did you see how many GTX570/580s are still sold on Newegg? It seems NV is going to use those cards to compete against HD7850/7870/7950 for the time being. I am not seeing any indication of any other GTX6xx variants launching until April even May. Unless Nvidia pulled the biggest secret launch in its history and reveals 3-5 new desktop GPUs this week, it looks like their top-to-bottom desktop Kepler line-up is simply not ready. GTX580 is going for $370-380 on Newegg. I think that's a reasonable price drop vs. $450 HD7950. GTX570 is undercutting HD7870 by almost $100. Remember GeForce 5? AMD had a 6 months of dominant position while NV had to sell their GeForce 4s.
I don't think the world is changing per say. It's just taking them longer to perfect the new nodes. I believe it's harder to migrate to lower manufacturing nodes than it ever was. I think we'll still get a real flagship (so to speak), but it will just come much later because of 28nm yield issues, etc. Imagine if NV has been having trouble getting a 294mm^2 die chip out in volumes on 28nm, what kind of problems they must be going through if a 500mm^2 die chip exists (interconnects, clock speeds, power issues, horrible yields). Instead of getting huge performance increases every generation, we might be seeing 20-30% incremental gains every 8-10 months instead of 50-75% gains every 24 months as we've seen before.
In 12 months, AMD could wait until 28nm process matures and launch a 2304 SP @ 1200mhz HD8790 and call it a new generation. Just a conjecture.
I would guess it is also although some things could possibly benefit from double precision, like emulation and maybe really complex physics. I think that eventually single fp precision won't be enough for graphics.I am guessing the 1/4 DP speed is ruled out for GK104. That's how they are able to keep the chip so lean/small.
No I thought DX10.1 more easily allowed for MSAA with deferred rendering or that regular DX10 had trouble with MSAA and float depth buffers. I could be wrong about that though.Sarcasm?
(SNIP)I don't think the world is changing per say. It's just taking them longer to perfect the new nodes. I believe it's harder to migrate to lower manufacturing nodes than it ever was.
Isn't 700mhz kind of high for idle? That's like, what, double wha the HD 7xxx does?
I'd just figure GPU-Z isn't capable of reading the clocks accurately.
I don't care for "monikers" on cards.
I care for the GPU on the board.
GK104 has the same performance no matter if they call it 680GTX, 780GTX or 880GTX...the latter being a useless marketing name.
You are eating up all the PR-FUD...but neglecting the devil in the details.
GF104 -> GK104.
Now try to add this:
GF110 -> GK110
You might think that 680GTX is highend.
I think it's NVIDIA's planned midrange (GK104 A2) GPU rebranded from a 660Ti to 680GTX...due to the (lack of) performance range of the 7970.
Hence why NVIDIA has been so quiet...it's another G80 round.
Another few pics:
Did you see how many GTX570/580s are still sold on Newegg? It seems NV is going to use those cards to compete against HD7850/7870/7950 for the time being. I am not seeing any indication of any other GTX6xx variants launching until April even May. Unless Nvidia pulled the biggest secret launch in its history and reveals 3-5 new desktop GPUs this week, it looks like their top-to-bottom desktop Kepler line-up is simply not ready. GTX580 is going for $370-380 on Newegg. I think that's a reasonable price drop vs. $450 HD7950. GTX570 is undercutting HD7870 by almost $100. Remember GeForce 5? AMD had a 6 months of dominant position while NV had to sell their GeForce 4s.
Prob. On the bright side, that's not a bad "return on investment" with those 480s. If those were 5870's with 1GB of VRAM, you would have been on 7970s right now! :biggrin:
I know you did . I don't think the world is changing per say. It's just taking them longer to perfect the new nodes. I believe it's harder to migrate to lower manufacturing nodes than it ever was. I think we'll still get a real flagship (so to speak), but it will just come much later because of 28nm yield issues, etc. Imagine if NV has been having trouble getting a 294mm^2 die chip out in volumes on 28nm, what kind of problems they must be going through if a 500mm^2 die chip exists (interconnects, clock speeds, power issues, horrible yields). Instead of getting huge performance increases every generation, we might be seeing 20-30% incremental gains every 8-10 months instead of 50-75% gains every 24 months as we've seen before.
In 12 months, AMD could wait until 28nm process matures and launch a 2304 SP @ 1200mhz HD8790 and call it a new generation. Just a conjecture.