• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

HD 2900XT Sapphire from ZZF (As well)

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
GTX is longer. The 2900 is actually shorter than the X1900XT suppositly I never measured it though.
 
Rattle said:
the issues in the drivers are blamed on imaturity, but how does a driver perform so well in 3dmarks and then they claim it has issues with AA in games and so forth.

its called driver optmization.. look now and keep track of the results. check again in a couple months.

they did the same thing with the x1900's which I used and loved for 11 months from launch day till novemeber 8th when the 8800gtx hit

All drivers from NV and ATI are optimized over time which is good. What is bad is benchmark specific
optimizations that effect nothing other than a specific benchmark. Doing that can often hurts the scores
in everything else.

It appeared that ATI specifically optimized the drivers for score in 3DM05 with the 1900/1950 cards. That
was obvious in the 3DM05 score in relation to any other benchmark. The 3DM05 scores those cards
produced was so high compared to everything else it looked like ATI was giving the world a middle finger
salute lol.

The 2900 drivers do not appear to have any benchmark specific optimizations at all. Their general gaming
performance issues have been related to the use of AA when it is dialed up and that is almost certainly the
AA handling algorithms in the immature drivers.

NV had similar issues with the 8800's when they first popped out and if you remember NV was releasing
newer beta drivers on darn near a weekly basis for a quite a while to deal with it's gaming issues. ATI is
doing the same thing with it's RC Beta drivers but it doesn't make them available to Joe sixpack like NV
does it's beta's which hurts them IMO as it takes longer for users to get there hands on better drivers.
ATI's once a month Catalyst release schedule doesn't cut it here.

Viper
 
toddm27 said:
they definatly would be cpu limited, as far as bench wise, difference of gtx's on you're setup are around the high 9k mark from what I've seen in the forums, havn't seen any 2900's benched on amd though

well I got

939 Opteron 165@ 2.88ghz (320x9)
Asus A8N32 sli Delux
X2900XT @ 850/1000
2x1gb Adata@200mhz

06 10,210
05 15,4xx
03 37,447
__________________
 
Not bad but you can tell at least in bench mark numbers its highly CPU limited right there. In 05 by alot.
 
deathman20 said:
Not bad but you can tell at least in bench mark numbers its highly CPU limited right there. In 05 by alot.

All these new cards take a ton of system speed to make fly with the eagles. As soon as the C2Q's drop
in price in July I am going to get one. My 3.9Ghz C2D just isn't cutting it anymore lol.

Viper
 
what about high res gaming though ?

titan run the fear test at 1680x1050 and deathman do the same

run max settings in the game and set drivers to app controlled and mipmap high quality CAT AI off
 
Rattle said:
what about high res gaming though ?

titan run the fear test at 1680x1050 and deathman do the same

run max settings in the game and set drivers to app controlled and mipmap high quality CAT AI off

I dont have fear........sorry

The only games I got with me are bf2, bf2142 and NFSC my brother got my Doom3 and Halflife.

I can say I get 60-70 fps with an average of 50-55fps and lows around 32fps no lower in BF2 8xaa 16x af and highest details at 1680x1050......with some instances of 100fps.

also with 8xaa and 8xaf 1680x1050 highest details...I average 85-95fps with lots of 100fps hits.....lows of about 60fps, no lower.

If there are any tests you know of that I could download and do some compares let me know and Ill run them for you.





some problems with the drivers with 16xaf??



 
Titan7170 said:
I dont have fear........sorry

The only games I got with me are bf2, bf2142 and NFSC my brother got my Doom3 and Halflife.

I can say I get 60-70 fps with an average of 50-55fps and lows around 32fps no lower in BF2 8xaa 16x af and highest details at 1680x1050......with some instances of 100fps.

also with 8xaa and 8xaf 1680x1050 highest details...I average 85-95fps with lots of 100fps hits.....lows of about 60fps, no lower.

If there are any tests you know of that I could download and do some compares let me know and Ill run them for you.
download the fear demo, then run the demo
 
I downloaded it but theres no test section in the performance options like its supposed to be??
 
Titan7170 said:
I dont have fear........sorry

The only games I got with me are bf2, bf2142 and NFSC my brother got my Doom3 and Halflife.

I can say I get 60-70 fps with an average of 50-55fps and lows around 32fps no lower in BF2 8xaa 16x af and highest details at 1680x1050......with some instances of 100fps.

also with 8xaa and 8xaf 1680x1050 highest details...I average 85-95fps with lots of 100fps hits.....lows of about 60fps, no lower.

If there are any tests you know of that I could download and do some compares let me know and Ill run them for you.





some problems with the drivers with 16xaf??




This guy says differently, LOL. Amazing the lengths people go through to bash a product unnecessary.
 
Rattle said:
what about high res gaming though ?

titan run the fear test at 1680x1050 and deathman do the same

run max settings in the game and set drivers to app controlled and mipmap high quality CAT AI off

In other words leave the drivers at their defaults lol.

Viper
 
ECH said:
This guy says differently, LOL. Amazing the lengths people go through to bash a product unnecessary.

He may not have been bashing. You really have to look for apples to apples in the compares as in what OS
was being used and what driver version.

Most of the reviews, etc that came right after the cards launched and the NDA lifted were done with 8.370.2
or 3. The .2 came on the CD and were the fastest but were buggier than hell and all of the 8.370 series had
real issues with AA.

DM20 and several people here are using the 3.380.0 RC2 drivers and report much improved gaming and AA
performance.

Viper
 
I have the 8.38 RC7's and the game performance is good. My CS experience was smoother without the lag freezes I had earlier.

BF2 seems to still play like utter garbage on Vista, but that's BF2's bad engine at work. Hence the black shadows that appear to pop up everywhere and the shaky gameplay, you can even see them in the screenshots above when he's in the chopper.

I still have the fog bug in CS:S though, like on de_dust2 if people are far enough away boxes appear blue and so do players of any skin so you can't really tell what is what. IT bothers me and I feel that is some sort of bug with max_dxlevel 95 enabled by default, but when I try to run with mat_dxlevel 90 cs won't let me connect to games or save the setting like it used to. Almost being forced to use 95. (9.0c+)
 
vixro said:
I have the 8.38 RC7's and the game performance is good. My CS experience was smoother without the lag freezes I had earlier.

BF2 seems to still play like utter garbage on Vista, but that's BF2's bad engine at work. Hence the black shadows that appear to pop up everywhere and the shaky gameplay, you can even see them in the screenshots above when he's in the chopper.

I still have the fog bug in CS:S though, like on de_dust2 if people are far enough away boxes appear blue and so do players of any skin so you can't really tell what is what. IT bothers me and I feel that is some sort of bug with max_dxlevel 95 enabled by default, but when I try to run with mat_dxlevel 90 cs won't let me connect to games or save the setting like it used to. Almost being forced to use 95. (9.0c+)

DM20 tried the RC7's on ME-II, said they were crap and went back to the RC2's.

Specific game bugs can be from a than general gaming and AA performance issues. They
can take longer to work out. It took months before Oblivion worked right on a 8800.

Viper
 
Back