gtx570vs6970vs6950vs5870 all overclocked review> gtx570 wins

[scooterlibby]

obligatory "sample size = 1" quip

Haven't read reviews at hardwareheaven in ages, they get +1 style points for the supercool graph layout.

Back to the point of sample-size, does anyone out there do reviews where they secure say 10 cards of the same SKU and generate any kind of statistics regarding the distribution in OC results?

Thank you IDC for this reminder. As a stats guy the n=1 reviews do kind of drive me crazy. The reason I started to really like AT was that I read an article where they ran regression for testing. It's probably a time/resource issue, but I too would like multiple observations and a bit more statistical sophistication.

 

[Shmee]

I am kinda disappointed in the lack of better overclocking support for the 6900 series. Makes a 570 or 580 solution sound more interesting. However, since i game at 2560x1600, 2GB would be nice. For now I will stick with the 5870's though.

 

[3DVagabond]

Do you know what sour grapes means? All that my post stated was that I look forward to a 'full out' review, eventually, where the physical limitations of the reviewed card are reached, as opposed to a software limitation. Sour grapes would mean that I bought one of these cards (I haven't) and am thus bitter about Nvidia's cards (for now) overclocking better (I'm not). Where the heck did you get sour grapes from? Please reread my post.

It's sour grapes because you didn't heap praises and bow down to the god of graphics to pay proper homage. You now need to say 10 Hail Mary and 5 Acts of Contrition for your penance. Now, go and sin no more.