The bit where you said you disagreed with my contention that drivers don't bring tangible performance gains, but
did agree with my point
regarding games on a game-by game basis.
So on what part of performance do you disagree if games are taken out of the equation?
Well done on wasting two seconds. At 1080, the performance difference is a whopping 2.44%, and at 1440p is 3.36% -margin of error territory. And just for the record, the reason that Star Swarm (the ONLY title (demo) to show any tangible improvement) shows an increase is because 337.50b was the FIRST Nvidia driver to have Star Swarm profile support.
Don't believe me? Check the release notes for 334.89 and 335.23 - the two drivers released between Star Swarm's launch and 337.50.
Anandtech actually compared 337.50b to a twelve month old 331 series. The only two games to show an improvement were M:LL and Bioshock Infinite - both of which shockingly enough were new releases at the time
²
Well, it was more of a pricing comparison than anything else since the thread is likely being read by a reasonable cross section of people. I'd also note that judging the knowledge base (and emotional sensitivity
) level of randoms on the forum is an inexact process at best. Judging by your THW link and its flawed argumentation I don't think I need reassess, but you have my best wishes on a full and speedy recovery. Causing you emotional stress wasn't my intention.