I'm just a little impressed

[ben6]

It really is too bad I only get to use this card for a day or two. I've discovered something very interesting . If I turn on FSAA+8x anisotropic filtering, I get in the 80~ fps range (found out bilinear filtering was the cause of some of the extra performance now corrected) now what's interesting is if I turn on FSAA without Anisotropic filtering I get exactly what Mike gets . Turn on aniso, get 80 fps turn off aniso 65 fps very strange behavior...

<font size=-1>[ This Message was edited by: ben6 on 2002-03-01 04:31 ]</font>

 

[Matt]

I noticed at first my scores were lower than what I had with a GF3 Ti500 in 3DMark 2001, so I started looking into what could be the problem. I installed RivaTuner (was using NVMax) and used the Power User section and scanned through it for anything that seems out of place. Then I found the EnableALiAGP key was set to 0. Since I'm using a ALi Magick 1 chipset (Soyo K7ADA mobo), I enabled it.

My 3DMark 2001 scores went from 6100 to 8700, which sounds a lot more what I'm supposed to get. So I went back and reran Quake 3 at the settings you set Ben, and went from 39.9fps that I stated previous to 49.8 (default clock and no tweaking done). I decided to overclock to 330/715 and almost hit the 60fps mark! Got 55.2fps instead, which is playable in my books.

Just thought I'd throw that correction up. It's nice to see the Nature scene move smoother, it still takes my breath away with it's beauty. In case anyone is interested in details, here my 3DMark details: http://gamershq.madonion.com/compare2k1.shtml?2883338

Edit: Forgot to add, I'm running at 1.36GHz now.

<font size=-1>[ This Message was edited by: Matt on 2002-03-01 07:52 ]</font>

 

[Mephisto]

Another example of bad online journalism. Thew in the card for some minutes, rush together some benchmark scores and put them on the front page of a website without double checking them or even think about it for a second ... oh dear.

 

[ben6]

Whelp more weirdness. I've freshly formatted my hdd, reinstalled Windows via Ghost, and installed a new build of 27.50 drivers. Using the drivers I get an identical 80 fps with 4x FSAA and 8x anisotropic filtering enabled at max settings . However , when using NVMAX I get 50.6 fps enabling both. So either 1. The drivers are disabling something with the default drivers or 2. Something else is going on..

Mephisto, heh, honestly I don't really give a flying what fps I get in a Quake3 benchmark. However, and I mean this in a good way, when I play Quake3 this resolution is playable to me with FSAA and anisotropic filtering..

In any event, I'm still getting the exact same result. And just thought it would be of some interest as the 4600 was just released yesterday.

 

[Matt]

Mephisto:

Wrong. Another example of human error. I'm sure you've made mistakes in your life too. Bad journalism would be to post the results, find the mistake, and not either a) update the post with the fixed results or b) delete the post altogether til further testing proves no mistakes were made. As you can see at NVNews, the post was taken down.

This is a new card, using new drivers, we're bound to find bugs and mistakes in the process of reviewing it, and it's our job to let the world know about that.