Florian Charpentier :
Thank you.
For those who complain about the commas, yes, the article was originally written in French, which is, why. More seriously: we'll work on that next time.
Regarding the accusation of being an nvidia-fanboy it reassure's me as for the last article some of you said I was an AMD-fanboy.
But be more specific: quote the paragraphs that are not based on facts.
Regarding the presentation of the graphs: does everyone want us to split each game in two? Because that's not a problem.
My criticisms were for the following reasons:
First, it was not necessary or productive to include 5 pages of fluff about Direct3D10.1 that could have been summarized in one page. More importantly, most of the data here was somewhat off-topic and belongs in a separate article on Direct3D10.1.
There are certain statements that I took issue with, and I think the article would have been better off without them:
"Contrary to what their name may imply, let's right away cut through the marketing smoke screen, which AMD couldn't resist to pull once more;" - statements like this right at the beginning automatically make you appear biased and undermine your credibility.
"The difference between the two Radeon HD 3800s, even though they're based on the same chip, is important and, first and foremost, physical." - I would think the difference in clocks and memory types would be more important to most people than the slot and fan size. I guess not.
There are some good statements that are the important things we should be seeing in a hardware review, such as:
"There are still only 16 texturing units (what a shame not to take the opportunity to correct this limitation!) and 16 ROP."
"AMD steps up by enabling the activation of the PowerPlay of the Radeon Mobility on the HD 3800."
"Also on stage, the UVD, yet absent from the HD 2900 XT if you recall, but the GeForce 8800 GT gaining back PureVideo 2, the Canadian didn't have a choice and it's all for the best."
"It's not really surprising, the HD 3800 brings about a huge drop in consumption compared to the HD 2900 XT, but the extent of it surprises us; the drop in peak total consumption amounts to 109 W when going from a 2900 XT to the 3870 with performances that are slightly superior!"
THESE are the types of things we want to see in reviews, not the needless fluff that undermines your credibility as a reviewer. I am not trying to be critical, but rather offer constructive feedback so that the reviews can improve.