Geforce 4 Ti4200 8X Shootout - PAGE 1

, Peter Judson
- Monday, February 17th, 2003
Like Share






Get updates when we publish new articles

Comments

Sort by date: ascending descending
0 thumbs!
^
CleavesF Feb 17, 03
When are you guys going start using the new 3dMark2003?

Obviously since it's brand new I'm pretty sure it's not optimized for any cards yet, but I always wondered to see the scores on the latest of Nvidia's offerings.

Consirering I heard the program was optimized for NV chips.

As for the GeForceFX, since it's limited production isn't nessesarily aimed towards a "mainstream" market; should we expect a benchmark from you guys?
0 thumbs!
^
Redemption Feb 17, 03
Entity will have a better idea whether or not we will review the GFFX. I think our choice will be to review one, but in such a limited production run review samples become very scarce and I am not sure whether manufacturers will be willing to part with enough to allocate us something.

We're still evaluating 3DMark2003 as a benchmark for our uses. I've long since wanted 2003 to come out, since I feel that 3DMark2001SE no longer represents the cutting edge of 3D testing, and I think that many new features for cards aren't being properly tested without a newer benchmark like 3DMark2003 - but we're concerned by comments made by other reviewers that the benchmark may not be suitable for their needs.
0 thumbs!
^
CleavesF Feb 17, 03
I'd have to agree with both points.

Since 3dmark2003 is basically Doom III today But the problem is many games don't use any of the features included in the benchmark. But then it is DirectX 9 heavy, which is the future anyways.

I find it a bit intensive too, I ran the thing on my rig and my 2003 score was 4 times lower than my 2001SE score.

3887 vs ~12000 (upper 11000 something)

I also found the CPU benchmark part to be very inaccurate. I got 1 FPS on my CPU benchmark on the Troll Demo (the second time around). I was like a 2000+ XP gets only 1 FPS!

Time to upgrade to the barton

I can see where you're coming from, but later on it will have to be an adjustment when teh DX9 games start coming. Just wondering when
0 thumbs!
^
Redemption Mar 7, 03
Well there are other concerns too regarding 3DMark2003. My main concern is that there really isn't any real gameplay simulation in 3DMark2003... in the 2001SE version, at least the begining was a good simulation of actual gameplay playback.

The second concern, which is actually more important, is that the test results will be heavily skewed towards DX9.0 cards. While there's nothing wrong with this, it WILL make non-compliant cards (and here I'm mostly thinking about the GF4 line ) appear to perform really poorly. So at this current time, 3DMark2003 is not so much a gaming benchmark, as a DX9.0 gaming benchmark. Because if you went and benchmarked a GF4 Ti4600 vs a RADEON 9500PRO on 3DMark2003, the performance ratio of the two cards will not actually reflect actual gaming performance ratios when running current games.
Sort by date: ascending descending
Add your comment:
Name *:  Members, please LOGIN
Email:  We use this to display your Gravatar.

Sign in with
Comment *:
(0.1720/d/web8)