[orangeandblack]

OK here's the deal. Every now and again when using my computer the entire machine will lock up for 10 seconds or so then unlock and everthing resumes as normal. Usually the lock ups will occur in pairs and within a few seconds of each other. After that the time between another set of lock ups can be anything from 10 minutes to 6 hours.

Using Process Explorer I discovered that the lock ups/freezes were caused by a hardware interrupt that hogged the CPU - effectively halting the computer for around 10 seconds. When it released control other processes could run again and the system continued as normal.

I have only had this problem since migrating to XP from Windows ME (quite possibly the most bug ridden OS in existence) and initially manged to cure it by uninstalling and disabling my mobo's onboard LAN support.

However, since then I have upgraded my RAM from DDR 266 512 MB (Kingston + generic) to some DDR 400 1 GB Corsair XMS.

As soon as I did this the lock ups returned, occuring exactly as they had before. Needless to say this did not make me very happy. At a loss as to what to do (the LAN controller was still disabled and re-enabling it then disabling it again had no effect), I stumbled across an article explaining the horrors of IRQs.

As a result of reading the article I disabled everthing unnecessary in the BIOS in order to free up IRQs and disabled a USB port through Windows in order to prevent IRQ sharing between the sound and USB systems, both of which were sharing IRQ 22.

None of this solved my problems and I am now at my wits end as it is the only glitch on what is otherwise a flawless PC. I'm tempted to just ignore the problem but as it was somehow solved before and the PC is used by my family (there's nothing more annoying then suddenly hearing music in a looped stutter for 10 seconds) im reluctant to just let it go.

If anyone out there has any suggestions, I'll be more than happy to hear them.

Edited by orangeandblack, 15 September 2005 - 07:02 PM.

[Advertisements]

automatic updates from micrsoft?

Thats just a guess from me...im sure there are others here that could help more

[OneCool]

automatic updates from micrsoft?

Thats just a guess from me...im sure there are others here that could help more

[Doby]

Hi,

Please list your system specs

When you "migrated" from ME to XP did you do a clean install or upgrade?

Rick

[orangeandblack]

Yep computer is fully updated and I couldn't find any viruses, spyware, malware etc with a variety of scanners.

[orangeandblack]

Sorry I thought my specs were included at the bottom of my post.

Here goes:

AMD 2500 XP-M Oc'd to 2.4 GHz
1GB DDR 400 Corsair XMS
Abit NF-7 v2.0
Geforce 3 Ti 500

All temps are fine and CPU disconnect is disabled.

The XP installation was completely clean - no way was i going to upgrade over the glitched mess that was my ME installation.

[OneCool]

Have you tried it at stock clocks..no OC

[Doby]

Well that things scream'in for a 2500, I see its a mobile witch I have never overclocked but how much vcore and vdimm have you added?

does the same behavior occur if you backoff the OC of somewhat?

[orangeandblack]

The processor's an Athlon Mobile so it has no real stock speed. When the problem initially occurred I was running an Athlon Thoroughbred 2200+ and that never went above stock speeds. I upgraded that to the Athlon Mobile I have now and the lock ups didn't show again until I upgraded the memory. There's no doubt in my mind that the memory upgrade triggered the return of the freezes but as to why I have no idea.

[Doby]

Whats the fsb and vcore set at?

[orangeandblack]

vcore is set to 1.75 prime 95 stable

fsb is set to 200

vdimm was set to 2.65 but ive just upped it to 2.75 as thats what the Corsair specification recommends

CPU temp is around 39 to 43 C

[Doby]

I think the vdimm was to low causing the problem let me know how the new setting works

[orangeandblack]

Thanks for that, the RAM installed with the proper timings so it never crossed my mind to check the vdimm. Its raised to 2.75 now (Corsair specification) and ill report back once the computers been on for a few hours.
I would expect a low vdimm to crash the computer outright though, rather than just lock it up for a few seconds.

This doesn't explain why disabling the onboard LAN cured the problem with the old RAM either, maybe turning off the LAN controller discreetly raised the vdimm slightly or something.

[orangeandblack]

UPDATE!!

About 5 minutes after posting the post above another pair of lock ups came along. So it can't be the vdimm although raising that was probably a good idea anyway.

[orangeandblack]

I know that the problem is caused by a hardware interrupt completely hogging CPU cycles for a length of time, however I do not know what interrupt is responsible and I have no idea why the problem has re-occurred when i installed my new memory.