If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Well all I can say is the drivers work perfect in quake II, 97% perfect in quake III. I haven't seen 3D this good and fast ever. The NT4 drivers never got good enough before I switched to win2K.
I get the same screen corruption when exiting UT under Win2k. I'm not oc'ing the card either.
I run UT at 16 bit color, and setting the desktop to 16 or 32 makes no difference. Logging off resets the screen, but it is pretty annoying.
Unless you have the AGP clock divider set correctly, you may be unwittingly overclocking your G400 via the AGP bus. Have you tried running your system at 366?
Scytale-
I dropped back to 366 and fired up UT...now I get EXACTLY the same ting as vanguardian1. The Matrox Tweak utility that I was using under 98 doesn't work with Win2k...so I have no idea how I would underclock the card.
...and my OGL doesn't work either ...SOF doesn't recognize OGL and neither does Homeworld.
...and I don't have an AGP divider jumper. The rev 1.02 boards are supposed to switch automatically. Is there any way I can confirm what the AGP speed is?
DarkSyd - PowerStrip works with Win2K (for trying underclocking). AFAIK, there's no program for determining AGP speed for W2K, but there's SiSoft Sandra for W98.
Strange enough under W2K I'm able to o'clock my G400DH32 more than under W98. It now runs stable at 154/209 (previously 145/193). I really tried Mhz per Mhz so W2K is definately better (or powerstrip messes around under w2k).
Only thing I can add is my personal experience - avoiding this type of corruption in respect to core clockspeed was a matter of delicate tuning. When overlocking my MAX core to 162, the corruption would occur in minutes or hours, but eventually it would occur. At 160 it never occurred, at least I haven't experienced it yet in months.
Comment