Originally Posted by tpi2007
The question is that they don't draw the logical conclusion from what they have in front of them when comparing the architectures: Turing in that specific configuration that they chose to compare is viable at 12nm whereas Navi needs 7nm to exist at a similar TDP, and thus Turing is the superior architecture, not even counting the RTX features. If it's an article that's about the architectures, that is.
since there is no navi on other than 7nm and turing on 12nm, what are they going to do?
in a perfect world they would be on the same node but well, i think we both know that world doesn't exist. in simple straight forward testing, you use what you have.
but having said all that, i have really no problem with what you're stating; pointing out differences is part of a peer review. but calling an article biased because it doesn't include pricing, that is irrelevant for it's purposes
, is a major misunderstanding.
fwiw, i'll concede my overstated great to your good since you have some good points.
watching tv while typing so missed your edit w/quote. i personally don't see that as a conclusion but understand if one does.
editII to be clear:
aren't conclusive enough to make any definitive judgment.
tells me no conclusion can be drawn . .