O/C'd Athlon XP won't reset. Upping core voltage doesn't help. What else to try?

[Arschloch]

First, the specs:
Abit KG7-RAID
Arkua 7528 HS/fan
Athlon XP 1600+ (AGOIA "Y" stepping)
256 MB Crucial PC2100 DDR RAM

Right now, I can run it completely stable at 1909+ speed (10.5x153) at even a little bit less than default voltage (1.73V). Yes, that is a pretty significant increase from the default speed, but as I'm sure some of you can relate to, that isn't enough.

Anyway, I tried pushing it up to 2000+ speed (10.5x159). It runs completely stable except for the fact that the computer does NOT reset correctly -- either by pushing the reset button, or by selecting Start->Shut down->Restart. Basically, it closes out of Windows correctly, but the computer never makes it back to POST. If I then completely power down and turn back on, it boots up without a problem.

Someone suggested that this is probably a core voltage issue -- that is, that I wasn't giving it enough. Well, I tried pushing the core voltage up to 1.89V, and that did not solve the problem -- everything is still stable in Windows, but it does not restart correctly.

I've concluded one of the two is true:
1) It isn't core voltage related, and I need to try changing something else.
2) The CPU just can't do anything faster than my current 1909+ speed. Or at least, it can't restart at any faster speed.

Why would this be? What else can I try? Any ideas or suggestions?

Thanks again.

-Arschloch

 

[Duvie]

It is a mobo issue...

Many here can relate this to early i845 mobos with their p4's...classic restart problem....The mobo does not intiate vcore set in bios early enough to get cpu to fire up properly....NO matter what you set in bios for vcore would not help...The boards would only give default vcore initially then boost it shortly to what it was set at. Once you got to that point of getting it to post at default core you were destinied to have this issue...

Way around it??? Vmod..."wire trick"...connecting pins at the bottom of the cpu to trick the mobo into think the default for the p4 was 1.7 to 1.85v...

way around it for xp chip??? Maybe...I am not 100percent familar anymore but I believe you can possible close a bridge on the cpu to cahnge the vcore...bios settings likley not to work like above p4 example...

also...
just to be sure set ram timings to conservative...

 

[VirginiaDonkey]

I believe there is a set of bridges you can close so the CPU will always defualt to 1.85v. L4`s??

there should be plenty of articles on the web that address closing these bridges. BUT....... if youre gonna start closing bridges, maybe you should consider closing the L1`s as well, so you have multiplier control as well.

good luck!!

Chuck

 

[Arschloch]

Thank you for two of the more helpful replies I've ever received here. I -never- would have thought it was a motherboard issue, but now that you guys explained it to me, I can easily see how that would be the problem!

I actually feel much better about it now. I assumed it was an issue with the CPU itself, and that's why I was very apprehensive about pushing it any further (because I thought it was already near its limit). However, if it is a motherboard issue, perhaps I will try to go a little further and just live with not being able to reset.

Thanks!