[COOLN] [PF] ATI HD 7970 GHZ Edition has been reviewed (1075MHZ GPU) - Page 20

Well, 70% more expensive than a 7970 would have been a launch price of ~$935. Of course, I know those numbers aren't serious. We'd probably be looking at something like ~30-40% more performance for a GK100 680, and ~15-20% more performance for a GK100 based 670. The thing is, x70 cards of the past have traditionally competed with AMD's high end offering (going back to Fermi that is). A GK100 680 would definitely cost a lot more than the 7970, but a hypothetical GK100 670 would probably aim to compete with the 7970 on price, at least initially.

Regardless of what Nvidia would have priced things had GK100 made its way to the market, the performance per dollar curve would still be a in a better state.
The issue with this is that these chips are significantly smaller than their last generation counterparts. Yes, these next generation cards are better value, but only barely so, without looking at power consumption and increased overclocking headroom. If I had to choose between getting a current generation and last generation card, I'd easily pick this generation, but the prices are still not interesting.

Let's look at GK104. I can't find the exact die dimensions without looking harder than I'd like to put the effort into right now, but according to this calculator here, we're looking at roughly 200 dies per wafer if we assume the die is a perfect square (17.1mm*17.1mm). GF110 is 520mm², or 22.8*22.8 (once again assuming it's a perfect square out of laziness), and around 100 fit on a wafer. Obviously, 28nm yields are going to be worse than 40nm, and we're also looking at higher wafer costs, but that is simply a lot more dies per wafer -- roughly double. Smaller dies yield better than larger dies as well.

Right now, the GTX 680 is retailing for $500 MSRP. The 580 was launched at $500 as well. At release, the 680 was only 20% faster than the 580. So despite being able to cram twice as many dies on a wafer, we're paying the same amount for 20% more performance and ½GB more of vRAM. It's a step forward, but knowing that there's so much more room for significant price drops in the future makes this generation a pass so far.

Not to pick on only Nvidia... AMD is no less guilty.

680 was definitely the exception to the rule as was the 7970. If GK110 does end up being a GeForce card and the 8970 ends up being a fairly standard second card on the same process node type deal I'd imagine you see prices that are much more similar to the 6970 and 580 however.

Also Homeless I don't think you are taking into account the awful yields Nvidia has had with GK104. Maybe you'll dispute that this is even a known thing but perusing newegg right now I'm pretty confident that GK104 has been pretty hard on production and is likely not as cheap as you are making it out to be. This is getting off topic but I also believe that Nvidia had to jack up clocks to compete with Tahiti with their 'lesser' chip and this is probabally the root of the issue.

Your basic point I get however and I agree to a certain extent because even if the 680 was widely availble and yields were great we'd probabally be paying the same prices.
Edited by jtom320 - 6/7/12 at 8:04pm

I'm not convinced that it's yielding any worse than Tahiti. Nvidia's just using that part for every market under the sun. It's also freaking Nvidia... if AMD and Nvidia switched places as far as performance this generation, Nvidia would still be out of stock because the fanboys can't get enough of their addiction. I'm really hoping that AMD can squash the "bad drivers" cloud that's been raining on their head forever.

This is exactly what I believe is going on with this generation. Nvidia had to lock voltages on Kepler because it was trying to compete with Tahiti with its midrange card so it clocked the tits off of it and used GPU boost as well to make the 680 seem on par but I believe there is very little headroom above stock on Kepler...

The reason AMD charged 150$ more is because the 580 was still a $500+ card & the 7970 out performed it. The 580 wasnt the Single GPU beast anymore & AMD wanted everybody to know that.

The 570 was 150$ cheaper than a 580. nVidia charged the heck out of people for 580's all because of the reason I just explained. It was the fastest single GPU card on the market. They can charge w/e they want for it & people will buy em because they want the "top" card.
Edited by WizrdSleevz - 6/7/12 at 9:15pm

This. Exactly.

People complaining about the prices this generation just don't get it. You're not paying for die sizes but for performance relative to your competition. The 680 and 7970 (especially now) are the two fastest gpu's on the planet and they are priced accordingly. If you are building a high end rig today and want the best in your system you will not be getting cheated buying either of these two cards as nothing will topple their performance for the rest of the year...

I know this is a bit out of context but this is the only live thread I could find on it. Will the 7970M work with CAD software? I know it uses OPENLC which is not widely supported by CAD. Will it run with programs like COMSOL, and MatLab? Thanks for the help!