Like I'm surprised...

[Sicna]

Am I the only one who thinks that Tom's Hardware is really biased in his choice.

He's so sold out to to Nvidia it's not even funny:

"Matrox
I am sorry to say it, but Matrox can't offer even one 3D-chip that could threaten NVIDIA's good-old TNT2-Ultra. Therefore I won't include Matrox in the competition this time. I am seriously hoping that Matrox will be able to design a competitive 3D-chip before it's too late for them. It would be sad to see Matrox leaving the competition as well. "

Quote from his new article on detonator 3 in Linux.

Come on. G400 MAX is much better than a TNT2. The TNT2 is a good card maybe but they are not in the same league.

Come on Matrox, show these "lost souls" what real quality is. I'd hate to see then leave the market of video cards too. The only difference is thatI'm sincere!

Sicna

------------------
Asus K7V
Athlon 750Mhz
1 X 256 megs Kingston PC 133 Mhz CAS 3-2-2
Matrox G400 MAX
Soundblaster Live
Windows ME & 2000 Professionnal

What kind of party is this. No booze and only one hooker. -Bender

[Thundrchez]

Tom's Hardware is very scientific, but completely lost. Many other hardware sites are in the exact same boat.

fps is not the only judge of a video card! Who cares if you can do 132,233,423 fps if each frame looks like crap. Sure, nVidia TNT2 looked better than Voodoo3, but neither of them are anywhere close to Matrox G400 video quality.

When will these sites clue in that video quality matters, and they should review the quality as well as the speed.

Some sites are completely obvious who they are sold out to. They try to cover up their conclusions with scientific evidence, but they leave a hole so big you could drive a Mack truck through it. And then they try to rewrite history.

How many hardware sites am I NOT boycotting for such shoddy reviews?

[Barbarella]

hi,

you can find something here. It's in french so i will make a little translation.
http://www.onversity.com/cgi-bin/pro...mots&P=N200009

First what is not said, is that the left picture come from a G400 and the other from a ASUS 3800 Pro TV. (TNT ultra)

In openGL you can choice between block tansfer mode or page flipping. the first give the best quality the second the best performance. But nobody care to check in witch mode is the graphic card tested !

Nvidia product are set to use page flipping. But what is funny is that even in page flipping (look at the tech utilities 2.0) the G400 get better quality than nvidia product in block transfer mode

In Q3 the ASUS 3800 lost about 10 to 20 fps in Q3. depend on the settings.

[Strider]

Very interesting. I wonder how many more such tricks are out there (visual quality vs speed) that we haven't heard about. I'm going to put my high school french to the test and check out that site.

[Snake-Eyes]

Hehe. I sent Tom an e-mail about his 'inflammatory comment' in that article, explaining that I think he's basically full of it, since the G400 series is every bit a direct competitor for the TNT2 series in performance, and even better in quality. I also made mention of the driver support Matrox has in Linux, and how it might've been smart of him to compare the Viper770Ultra against a G400Max both running on the latest drivers in Linux and under Win9x and Win2K (oh wait.. THAT's RIGHT! The TNT2U wouldn't work properly under Win2K! hehe).

------------------
Ace

[DosFreak]

Nope, sorry. Nvidia drivers are set to block transfer by default.

[Jeff]

I don't think the blurb is that bad. Sure my Max looks good, but it is at least 2 generations behind in chip technology. I would not mind a new release every year or so but it has been longer than that and the G800 has not even been announced. Maybe Matrox needs to listen to what he is saying and keep a little more current with their card releases.

JP

[Barbarella]

be carrefull,

the link don't comprare G400 to TNT2 only the best you can get from block transfer and the worse from page flipping.

Dos freak, witch driver did you use on which GC ?

But the must important thing is that you have to control in which mode you are because this give diffrent result when you benchmark.

[Salmonius]

It's true that the comment about the g400 and the tnt2 is a bit inflammatory. However, I don't see anything wrong with his article other than that. The point was to see what kind of performance you can get under linux compared to Windows. Since the g400 is relatively old and not really considered a performance part compared to the newer cards like the Radeon and the geforce2, he just chose not to include it. The Radeon has no support for Linux and according to Tom, the voodoo5 drivers aren't that good or something. So he felt that he had no other performance card to test than the geforce2.

I think it's very fair except for the fact that performance is always perceived as being the 3D performance in two games. Personnally I like RTS games and I hate when they lag, so 2D is at least as important. Not to mention visual quality.

[Sicna]

I know the G400 can't keep up with the new cards but nevertheless good Linux support from a manufacturer is hard to come by.

Not mentionning the G400 only because it can't keep up the performance is ridiculous. I mean how often are you going to need 3d in Linux. It's mainly meant to be a server not a workstation. I know this is changing but for everyday use, I'd take a G400 in Linux than a Geforce.

But people in not too long will not even know that this card is greatly supported because of reviewers who choose to put speed before usefullness.

Just my 2 cents anyway

Sicna