I can stably run this cpu (@1.65v) with this setup at 160fsb and ddr400specs 2880mhz..cpu temps range from 32-35C under loads....Ok this is a good thing not the problem...the problem is that when I go to benchmark using 3dmark or any other application in 3d it will reboot after about the first 15seconds....it doesnt do this when it is running at stock speeds 2400mhz and ddr333....but in other benchmarks such as sandra etc it works fine and runs for days...it is just the 3d enviroment that is killing me.....i upped the memory voltage to 1.7v and I still get the problems.....then I tried to run the benchmarks at 145fsb and I still get the reboots!....this makes no sense at all because it should easily be able to do this....to me it sounds like the power supply since reboots during games etc is usually due to power supply shortage but this psu supplies 18amps to the 12v rail which is sufficient (i think)....i checked and im using total 11amps off the 12v rail with my setup....cpu temps seem to be chilly so I dont think it is that...I flashed to the latest bios, and am using the latest ati display drivers (catylst 3), direct 9X......after the reboot I get a message in winxp that says system has recovered from a serious error due to the display drivers....I have tried all the drivers and still get this problem....what do you guys think?....I also tried lowering the cas latency to 2.5-3-3-7
My guess is that it is power related. The Radeon 9700 Pro is a sap sucker. So is the Audigy2. I'd try checking the 3.3V rail. (And while you're at it, the 5V rail.) I bet it's just simply not enough power across the board ... so to speak.
PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?
yes I have the agp/pci frequencies locked at 66/33....so it cant be that....I havent payed attention to those rails and can check it when I get in this evening.....a new psu might be on the way :smile: ....thanks for the suggestions
but in other benchmarks such as sandra etc it works fine and runs for days...
Try Prime95, in my experience nothing beats it when it comes to non-graphic load. If your problems continue to occur under heavy graphics load, I think slv is right about the PSU being the problem otherwise you will simply have to lower your FSB until you can complete 3DMark2001 SE.
<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
Wow his watercooling system must suck a big one.....air cooled im running around 32-35C and he is in the 50's!!...lol...I will check out what he was talking about....I will keep you tuned...I see that 160fsb is the wall on most pe's but why would it run benchmarks fine at this fsb and even set on 145fsb crash in a 3d enviroment...strange, but I will figure this out :smile:
It doesn't help that the temperature sensors are out of whack on this board (CPU temperatures are at least 20C too low), but my finger is calibrated well enough to find out that the memory hub, the back of the 9700 and everything else inbetween (including the sensor) were getting awfully hot.
hahaha! Im actually quite good at doing the same, but i wanted a bit more accuracy so i forked out 60 Ozzbux ($30 US) for a nice digital multimeter with added thermistor probe.
Its very useful. The tip of the thermistor is very small, so can be squeezed into tight places, and is very quick to equilabrate when placed on a hot surface. Within seconds i get a accurate temp reading of whatever i want, to within 1C.
My nothbridge is only running at 37C, despite only having a simple 1" square of finned aluminium with no fan and 166fsb.
The edge of my mcx-462 copper block is also only 38C, indicating that cooling is most adequate, ever with the 25% OC. And the back of the GPU is a balmy 44C under full load.
<b><i>The Very Hungry Caterpillar</i> - George W. Bush's favorite childhood book.
Note: This book was first published a year after Mr Bush graduated from College.</b>
I have to agree with Crashman, it does sound like a power supply problem. You are running some pretty serious hardware and only have 400W to push it...(I can't believe I just said ONLY 400W. Wow hardware is moving fast). I would suggest unplugging any hardware you don't absolutely need running (such as your sound card, any extra CD-ROM's and so forth) and test it again.
Hey, overclocking is just perhaps a passing fancy!
Better read the warnings on overclocking or you'll live to regret it.
You want your system to go faster, then upgrade!
Its better than having your system destroyed in the long run!
What the hell is this guy talking about reading warnings blah blah blah ...You are too sensitive to be posting in this forum :smile: ....besides if you ask me it is a growing trend not a dying one since manufacturers are making overclocking such an easy thing to do with jumperless design, many more overclocking features..
Btw I think it has to be the psu with the way it is acting...I figured a 400w psu would be plenty, here is the psu specs: +5v (40A) , +12V (18A), +3.3V (28A)...if a better power supply would make a difference make suggestions of what would be "enough"...what are you guys using with comparable systems?
I finally got a chance to check out my rails....and I can say that I think I have found that the problem does lie in my power supply.....all of my rails are above the ratings except for my 12v rail which is running at 11.78V...I could see this causing alot of my instability when overclocking....what do you guys think?....I went ahead and ordered a new enermax 433W psu that supplys alot more power to all the rails thaN my current psu and of course has the fan adjustment =]....I was going to opt for the antec 550 true but the volt rails were all way below the enermax and the enermax 550 was even lower on some which was strange but anyways...