View Single Post
  #18 (permalink)  
Old 7th September, 2004, 06:25 PM
cadaveca cadaveca is offline
Member/Contributor/Resident Crystal Ball
Join Date: March 2004
Posts: 7,451

Originally Posted by Staz
My computer is built in a drawer, so my ventalation sucks. I have pushed my card so hard that the heat builds up then after a few hours of 3D gaming the card would lock up. Touching the HSF would result in slightly burnt finger tips. Yet after a reboot and some cool down time the card was always fine. I killed it while adding extra cooling to it.

lockups happen LOOOOOONG before permanant damage takes place. if you use VPU recover, you should never have a problem that will permanently damage your card, as long as you never increase the AGP voltage
although agp volts will allow you to run the agp bus at a faster speed, there are very important reasons as to why you never should. In order for that card to display an image proerly, there are alot of timing issues at play, from the actual rendering pipeline to the vertex pipeline, to the memory bus on the card to the memory bus over the agp bridge, to the refresh rate of monitor...they are all set to a predetermined "safe" level for things to go smoothly. people kill thing when overclocking when generally one of these parts, or another one i have not mentioned has been forgotten.

good reading is the ATI SDK, in which, if you can force it through your head, you'll find proper detailing on to how you can tweak the driver to compensate for certain things being pushed out of spec, and keep your hardware safe.

all the temps i posted above for maxe should cause lockups, with the exception of the 420. this gpu tens to go as far as you want it, to the point that i think it may be possible that we see a new revision of this chip coming out first quarter next year. Generally i get these things right too!
Reply With Quote