ATI and NVIDIA Benchmark Number Problem?

[Apoptosis]

ATI and NVIDIA Benchmark Number Problem?

Ding, Ding, Ding... Ladies and gentlemen. Your attention please. In this corner, fighting in the mainstream market, GeForce 7600 GS of Santa Clara, California. And in this corner, Radeon X1600 XT of Ontario Canada. After our GeForce 7600 GS preview yesterday ATI came running and ready to fight... see why!

ATI contacted Legit Reviews and expressed their concern the benchmark numbers specifically. ATI sent us a chart where they ran the NVIDIA GeForce 7600 GS against the ATI X1600 XT and came up with very different results than what NVIDIA showed us from their testing results.

Article Link: ATI and NVIDIA Benchmark Number Problem?

[pastorjay]

Hehe... this is getting very interesting. I don't think any of us thought that ATI or NV would actually show their card losing, but they should know that we at LR are going to show actual bechmarks and test results from what we find as well as what they supply.

I find it almost humourous to see them so uptight about it. I can't wait to see the results! One of my favorite CSI lines is: The evidence can't lie... We will see the truth soon enough!

[-EVRE-]

I would like to see the numbers on this card lined up against:
6800GT/6800GS
7600GT
7800GT/7900GT
@ 1280x1024 /1024x768 (LCD res)

Im building a HTPC with enough horse power to run WoW nicely on my HDTV. actually Im replacing one that has a 2600+ and 9800 pro with something a little quieter and less power hungry. One thing about a HTPC, it never gets turned off!

[xaustinx]

there's a few things going on in ati's graph that make it hard to compare to nvidia's the ones that stand out the most in my mind are 1. nvidia's graph goes to 230%, ati's goes to 160%, 2. nvidia is smoothing the curves of their thier graph, while ati is just drawing lines to points, 3. the size of the ati graph is smaller, making those straight lines to points apppear significantly more pronounced than they actually are.

If you throw that in, with the fact that they weren't run on identical systems, they both end up being crap. more to the point, you can't compare the two for all the reasons i mentioned above. Also, using the other card as the baseline, is an odd way of doing it. It should be measured agiasnt something like a 3rd GFX card vendor's product, or even 0, like most statistical analysis.

[Apoptosis]

I agree NVIDIA and ATI both benchmark against eachother differently than the review sites. The results don't mean much like you alredy noted.

[Bwall]

The XFX 7600GS arrived this morning...what an awesome looking card! Time to get to work.
I'll be benching each card over the weekend and will have a full story early on next week. I'm very curious to see which card comes out on top as this is a very hot price point for a lot of friends of mine.

[-EVRE-]

I am also curious how these cards stack up against the two variants of the 6600gt. the 128mb ddr3 and the 256mb ddr2 cards.

[Apoptosis]

We only have the GDDR3 cards in stock for the 6600GT I believe.