//flex table opened by JP
PDA

Click to See Complete Forum and Search --> : BBA Radeon 9000 64MB=Bad clocker?

causticVapor
05-30-2003, 06:50 AM
This is in one system, newly arrived to replace an MX 200. Much more fluid overall, even though it only gets a 3dmark2001se score of ~6000.

It has a passive heatsink, and I've only been able to get the card up to 275/220 from stock 250/200.

Maybe active cooling would help, but it didn't help at all with the MX 200. :p
PacNW CE
05-30-2003, 06:11 PM
My experience with passive heat sink cards and overclocking is about the same as yours with the MX. I had an ATI 7200 with passive heatsink, that I was able to push up a whopping 5% before SERIOUS artifacts appeared. HL showed sleet like streaks of sliver when I'd turn around.

Take what you have and be happy it's dx8.1 compliant.:D
causticVapor
06-01-2003, 11:03 AM
The thing is, the MX had a much higher % overclock than the 9000. Methinks the 9000 is more close to the limit of an average .15-micron GPU design. I'll try active cooling and see if that helps.

To tail that, I am SERIOUSLY impressed with what ATi can do with .15-micron on their front end cards. Almost 400MHz (and soon above) is the pinnacle of that process.
NDD
06-03-2003, 08:00 AM
It's not about HS - your memory is holding you back. Does your card use the exact reference design from ATI ?
AllGamer
06-03-2003, 10:50 AM
Memory Clock really makes a BIG difference in the overall speed

:t
$1500-P4 gamer
06-03-2003, 11:59 AM
Yep it does.

8500LE stock 250/275 bba I get 9k on 3dmark.;)
causticVapor
06-09-2003, 07:43 PM
It's the BBA reference design...

Haven't checked the mem chips... but they're hynix. 4.5ns would make sense, but I can't really open the case right now..

EDIT - Sorry guys, this makes sense. It's the 9000 np, so the speeds are as follows...
250/200

versus

275/275

for the 9000 pro. I would assume that would make a dramatic difference.

Oh well, it's an OK card. :eek: