Nvidia have missed the boat - this is how good Fermi could have been

[Freddie]

Now it's fair to say the GTX480 & GTX470 have come into hail storm of criticism on tech forums concerning it's performance and power usage and with that in mind I wanted to find out just how good Fermi could have been IF it were as efficient as the HD5970, HD5870 & HD5850 in terms of power usage.

This first chart shows the average frame rate of the cards in the games below and also the load power draw in Furmark. I got these numbers from Anandtech who are perhaps the most reputable review site on the web.



Here you can see the cards listed in performance order HD5970, GTX480, HD5870, GTX470, HD5850 and at the bottom there power draw and the % increase in performance they have over each other. Note that despite the huge power draws on the two Fermi cards that there lead over the respective ATI cards can be measured in single digits.

Now put that to one side and lets examine just how good Fermi could have been if the chipset was as efficient as ATI's. I've increased the FPS by the % difference in power load between the GTX480 and HD5870 and GTX470 and HD5850.



I can't help and feel Nvidia has missed the boat here, the GTX480 could have been as fast (by cats whisker) as the HD5970 and a lot cheaper to boot and the GTX470 would have forced ATI to make serious price cut backs on it's two other cards.

[Top-Tier-Tech]

Those charts don't reveal the graphics settings for the games. Stuff like AA and physics change the FPS difference.

I clicked on the link you but up and google blocked it saying it is a reported attack site and listed several harmful programs discovered on that website.

[Wheelchair]

ATi definitely won this round no matter what way Nvidia plays this game.

Onto the next round!

[Freddie]

Those charts don't reveal the graphics settings for the games. Stuff like AA and physics change the FPS difference.

I clicked on the link you but up and google blocked it saying it is a reported attack site and listed several harmful programs discovered on that website.

It looks like Anandtech site is being attacked by some ad/malware software, if your using FF with adblocker plus you will be safe.

All the cards were tested using the same settings so your point is mute so long as you all the cards are being tested on level playing field I can work out how efficient the cards are. I could have used Anandtech results or Bitechs results or anyone else's that's credible the point is to examine just how efficient the cards could have been.

What the charts do prove without any doubt is that more transistors doesn't translate into extra performance, the laws physics comes into play here and Nvidia just didn't do there homework they ploughed ahead making a large, expensive and power hungry chip. Sure it's quick but at what cost? The cost of electricity isn't cheap any more and the importance of power to performance is playing a more important factor with each new generation that comes along.

[Top-Tier-Tech]

All the cards were tested using the same settings so your point is mute

Not at all, without AA the results between the cards is closer, with AA the GTX480 pulls a sizeable lead in many games. Seems how most people playing at 1920x1200+ use AA it must be brought into account.

[Freddie]

Not at all, without AA the results between the cards is closer, with AA the GTX480 pulls a sizeable lead in many games. Seems how most people playing at 1920x1200+ use AA it must be brought into account.

Ok then lets look at some results with AA at 1920x1200 but I'll also inculde 1680x1050 and 2560x1600 as well, I'll be using Bi-tech for this one.

I have the spreadsheet here in front of me and the results when AA is used as follows.

The HD5970 is 16% faster then the GTX480 - **** Note results blured by DOW2 as it's doesn't use Crossfire
The HD5970 is 27% faster then the GTX480 (when DOW2 results are not inculded)
The GTX480 is 12% faster then the HD5870
The HD5870 is 9% faster then the GTX470
The GTX470 is 9% faster then the HD5850

These results are in-line with the results from Anandtech. As for the effect of AA (this will all be based of 4xAA) in Battlefield BC 2 Fermi does claw back some lost ground with the GTX470 actually matching the HD5870 however in Dirt 2 Cypress benefits from AA enabled, in Crysis the GTX480 is 5% faster then the HD5870 with AA enabled but the GTX470 is no quicker then the HD5850.

I'll upload the spreadsheet if you really want or you could just take my word for it.