Overclocking problems with 2.4c

[kalrith]

First of all, even though I have few posts on this forum I have been a pretty avid reader for the past year. I've been running my system rock solid at 250 fsb 1:1 with the RAM timings at 2.5-3-3-6 and vdimm at 2.85v for the past 9 months or so. I didn't try for more, because my cpu was hitting 59 degrees at load with stock cooling and voltage. I recently installed the Scythe Ninja and Thermalright NB-1, thinking that I would get a lot more overclocking headroom, especially with my cpu's SL6WF Malay stepping. My load temps now don't even hit 40. I hit a wall with stock voltage at 257 fsb. I increased the voltage to 1.65, but that won't even get me to 265fsb. I loosened my memory timings all the way and also tried 5:4 ratio but neither worked.

The strange thing is that increasing the voltage causes random reboots, and when I send an error report, it says that I received the windows stop error because of a device driver. However, this never happens when I run the voltage at auto. When I increase the fsb past 257 at stock voltage, prime95 fails, but I experience no reboots. When I run at 1.65v and 265fsb, prime95 does not fail, but the computer reboots every 2-3 hours with the device-driver error. I'm pretty sure that I have some decent headroom, because at 1.65v the system will boot to windows all the way up to 290fsb. At that speed and 1:1, memtest86 ran with 0 errors, so I'm pretty sure my memory can handle a measley 265fsb.

I have most of the settings in bios on auto, other than what I have stated above. I get the same results in either turbo or standard mode. I have the graphics aperture size set to 128mb, and my agp/pci locked to 66/33. I can't think of any other bios settings to mention. You can see my sig for the rest of my hardware.

Thanks in advance!

 

[kalrith]

Any takers?

 

[JustStarting]

probably the cpu's limit. SL6WF was a pretty good chip, but not all were guaranteed to OC farther than you have taken it already.

3100 is pretty respectable. I've seen a few users here had them up to 290+ FSB but very seldom.

 

[kalrith]

But, does it make sense that Windows is saying the reboots are caused by a device driver? I've heard of random reboots from overclocking a system too much, but I've never heard of Windows saying they were caused by a device driver. I guess I'm just asking if the reboots are caused by a limited cpu or if there's something else that just doesn't like the extra voltage and is causing Windows to reboot.

 

[secretanchitman]

i have your same cpu and mobo (non deluxe), and i never get any reboots unless i clock too high. maybe your windows is screwed or some component that you installed is messed? i have no idea. i doubt your reboots are caused by a limited cpu, as you have good cooling and ram.

 

[kalrith]

Originally posted by: secretanchitman
i have your same cpu and mobo (non deluxe), and i never get any reboots unless i clock too high. maybe your windows is screwed or some component that you installed is messed? i have no idea. i doubt your reboots are caused by a limited cpu, as you have good cooling and ram.

Thanks for the insight. I think I'll run Windows repair tonight to see if that fixes the driver issue.

 

[mcMikeyT]

Hi there kalrith,

I know I'm a bit late on this one but bear with me, first post in this forum, joined yesterday.

Your device problem that Windows is talking about IS the CPU.

To Windows, the CPU is just another device (take a look in DevMgr) so repairing Windows won't help, this is all down to o/cing the processor.

I think you've hit the limit :-(

 

[letdown427]

Hope this isn't completely wrong, but it could be the power supply? As in, if you're running the cpu at a higher voltage, perhaps the supply isn't as stable as it was? If it fluctuates and drops too low that would almost definately cause a bit of a mess. Just a suggestion. You probably have hit the limit of the cpu though.

 

[kalrith]

Well, I'll have to look at the power supply voltages when the system is under load, just to rule that out. I never thought of the cpu as being a device, but I guess that makes sense. I did try to repair Windows, but it would only get about 5 minutes into it and then restart, so it was due for a reinstall anyways. And, of course, that didn't fix the problem.

So, I have a couple of questions about upgrading. Would I notice a big difference from upgrading to a 3.0c? Someone has one for sale for $140 right now, which is a really good price on it. He's only had it to 3.2, but he said he never tried to overclock further, so hopefully I could hit around 3.4-3.6. My memory right now, though, is running at DDR 516, and it would be much slower with the 3.0c. I would just go out and get a whole new board, video card, and opteron 165, but I definitely want to wait until after AM2 and Conroe are released to make my major upgrade.

Another thing I thought of was doing was the droop mod for my motherboard. My motherboard doesn't reboot when I put it at 1.6v. At that voltage it stays around 1.61-1.62 at idle. At load it droops all the way down to 1.55 or stock voltage. If I did the mod, I might be able to get a little bit more out of the processor, but it might be more trouble than it's worth.