https://i12.tinypic.com/2py29hz.png
"Do I really have to express myself here ?
If these guys are so called my colleagues ( as of being reviewers like me ) then I should feel ashamed, really Evil or Very Mad
What are we looking at here ?
a) They used a better test system today, better drivers ( supposed to be ) and they managed to get the 2900XT to perform worse than their previews bench session with a worse test system & worse drivers ?
How come ?
On April 24 they got 84FPS on F.E.A.R. with the 2900XT with a QX6700 and pre-release drivers, and they got 79FPS today with a QX6800 and retail drivers ? Oh really ? Very Happy
b) Where's the result for the 2900XT in Company Of Heroes today ? Why N/A ?
c) On Half-Life 2:E1 today they got the 2900XT to score 1FPS more than the 2900XTX.
What could've caused this ? A typo ? Quite angelic.
Something else ? Using a CPU limited resolution which would cause both cards to behave like they're the same.And then there's the GTX surpassing the R600s by ~40 FPS. Quite the real leap over the Radeon X1950XTX at that game. Evil ? heh
d) Now, the best part...they scored ~48FPS in Oblivion on the 24th, and now they present us a 54FPS gain by a move from the QX6700 to the QX6800 and the small gain from running the RAM at 1T Command Rate ?
e) A reviewer in order to conduct comparable and a valuable & trustworthy review must use the same testbed, quite they opposite is what they did ( if they really went through a performance testing process )
f) And for what reason would somebody present unclear results in combination with unclear drivers & filters ( AF & AA ) settings ?
***** right boy Very Happy
My two cents ( oh wait, I have another one ) [ I'll save it for later ]
P.S. The stock core clock for the 2900XTX as of current is 800MHz and not 745MHz as they state.
P.S.2. That's pretty much all I can say at the moment.
P.S.3. Now I have to finish a memory roundup and then pack my stuff for a trip.
المفضلات