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

First attempt at overclocking

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Artyboy

Member
Joined
Mar 12, 2004
Location
colorado
My rig is in my sig. My processor voltage is 1.7. I have 2 ceramic core exhaust fans, a raidmax that came with the case in the side panel and a pair of coolermaster intakes in the front. I'm using an aspire (not sure on model....it's big and it's blue) for the cpu cooler w/ as3. My temps are 51c idle and up to 62c under load. I don't do anything to stress the cpu except for benchmarks for now so I'm not worried about the high load temps. By the time I start gaming with this machine a lot I'll improve the cooling.

For some reason I get a hardware failure within a few minutes of running the prime95 torture test. Before I got this new stick of RAM I could run it for hours. I was using pc2100 with my fsb at 166. My processor was set at 12x166. When I first installed this new RAM I hadn't upped the fsb on my processor yet so it was still at 166x12. I got 100 checksum errors within a few minutes when I tried the torture test then. After I upped the fsb to 200 I started getting the hardware failures. When I run the small FFT's test (the cpu intensive one) my cpu gets hot but I don't get any errors. I can run 3dmark01 looped for hours with no problems. I ran memtest86 for about 10 hours last night and got no errors. My OS is winME.

I'm just wondering if I can fix the problems with my hardware failure in Prime95. At first I thought it was my memory...until I passed memtest86. Is this something I need to worry about? Thanks in advance.
 
Your Vdimm and Vcore voltage may be too low + an all copper heatsink like Thermalright will allow you to OC further too.

You may need to relax your memory timings, but I doubt that's the reason. Passing 12 hours of Prime95 Torture Test without the program displaying any errors would be a good start...
 
My vdimm voltage is still stock and I haven't messed with the memory timings at all. They're set at 3/3/3/8 or something close. At 166 fsb the cs was 2.5 but when I upped the fsb of my processor to 200 it defaulted to 3. If I start messing with the timings and upping the voltages on my memory will it effect my cpu or system temp much? Are the above timings pretty standard for pc3200 mushkin blue line? They come with heatsinks installed and I really don't want to mess with them if there's something wrong with the memory (warranty). I assume there isn't since I passed memtest86. I just haven't messed with memory much yet and my processor is running so hot I'm nervous to change anything else. What do you guys think?
 
If you haven't touched your memory timings, forget about that for a moment, just up the Vdimm and Vcore voltage and see if that helps you pass Prime95 Torture Test.

1.7 Vcore is frequently way too low for major overclocks of 2000+.
 
I'll give it a shot and hope it doesn't raise my temps too much. 60c+ just makes me nervous :p. Ah well. I'll try and track down a thermalright in town this week. Thanks for the tip.
 
I would ride it out into the 60's without any sweat so long as you are nearby the system to keep an eye on it. During a heatwave when I lived in columbus a couple years back I ran with my processor in the mid 60C's commonly without even getting errors of any sort. The temps peaked at around 70C. I definetly paid attention to them, and backed down on things when I wasn't going to be around... But so long as you are around to watch for errors or lock-ups I think the 60's are no problem.

If your chip is getting a little hot under the collar, it will tell you about. So long as you are there to notice, I wouldn't worry about it.
 
Whenever hitechjb1 discussed voltage he'd say "Remember to keep the on die CPU temperature under 65 C."

Personally I'd heed his advice about not going over 65 C.

When we discussed how 2 volts is not to blame for killing CPUs, just shortening their life, 65 C was used as a cut off temperature point.
 
I chickened out and backed my cpu settings to default for now. I've been running prime95 now for 2 hours with no errors so who knows what it could be. I still think that the memory is the culprit since I never saw a single error in about 30+ hours of prime95 with my old crucial pc2100 stick.

I went and checked DFI's memory compatibility list and mushkin wasn't even on the list so that might be the problem :(. Then again, as I said, I get no errors on memtest86 and since I backed off to default settings on my processor I've been able to run prime95 np. I'll have to start doing some serious tweaking when I get better cooling.
T
hanks for the help guys. I'll be asking for a lot more advice here in a couple of days :p.
 
Back