Time to underclock?

[wju425]

Hi gang.

I have this old xp1700 tba that I have been running at normal clock ( 11 x 133 ) for about 3 years. Lately I've been getting BSODs and formatting and reinstalling windows don't seem to fix them. My firewall and AVG are updated so my system's staying clean. I decided maybe my cpu or mems ( 2 x pc133s ) are going so I tried underclocking them ( 12.5 x 100 ) and my system quit crashing. So the cpu and the mems gets worn out? Does that happens to you?

\o/ Billy

EDIT: By the way, my mobo's Asus A7V133-c so I'm sure that's a good brand.

[proffit]

well i would try to reset the cmos before you try any thing else.

[paul9]

Ram chips are notoriously sensitive to damage from static electricity. Maybe that is your problem? The cpu and memory have done a lot of work in an average three year old pc, and it is not unheard of for them to die in that time. If you have a spare hard drive, try installing to that, to try to reduce the chances that your hard drive is causing these problems.

[excuzzzeme]

Did you check to see if maybe a heat issue? Clogged up cooling fins on heatsinks or fans starting to slow? (Just another possibility).....

[#43 fan]

That's what I had to do w/ my Athlon XP 2600+ on an Asus A7V600. I upgraded the memory, and it fixed it mostly, except when it's on for hours and I play a game and it overheats.
Which BSOD do you see?

[wju425]

I've done all the above.

Right now it's 43c for the cpu and 33c for the mobo. I reflashed my cmos one time to get away from the annoying alarm going off. The BSOD are giving me vxd errors. My hard drive has been thoroughly scanned and found 8k of bad sectors last year. It's been scanned again with clean results. It's been about a week since I underclocked and no problems so far. I don't seem to notice any performance loss from underclocking. In fact it's doing better. Go figure.

\o/ Billy

[paul9]

CPU stressing out?

[Redwolf]

43 degrees is high for a 1700+ (esp. a revision B chip), but it shouldn't be overheating at that point. If I remember right, I've run a 2100 rev. A to 120 degrees before (about 49 degrees Celsius) before it overheated, so I don't understand why yours would be doing that.

As paul9 said, I'd try checking the RAM first (it's far more likely to be that). Get Memtest86 3.2 here:
http://www.memtest86.com/
and see if your RAM checks out. With your problems, it should be apparent in the program if the RAM is bad or not (you'll see alot of errors!).

If it's not the RAM, it could also the the hard drive or cable. I've had similar problems to the one's you're reporting when one of my HDD cables broke, and replacing it seemed to fix it. Had to reinstall windows to completely fix it though. If scandisk is throwing lots of bad sectors on what should be a good drive, suspect the cable (or the drive circutry).

[wju425]

Ran memtest and it passed. Of course I'm running at 100fsb so I'll keep it that way. It's running fine as it is.

\o/ Billy

[sr71000]

isn't there a program to test the proc? like prime95 or something like that?

[Redwolf]

Wierd. Maybe the chip was too stressed or something. 1700+s were some of the coolest running chips around.

[wju425]

Mine's a t-bred "A" revision. Not a very popular chip but it's what I "won" at the newegg lottery.

\o/ Billy

[samwichse]

Out of curiosity, have you checked the motherboard capacitors? I'd be willing to bet that at least one of them has a bulging top and/or some leaked juice.

Reduced power cleanliness + underclock = stable again?

Sam

[wju425]

Out of curiosity, have you checked the motherboard capacitors? I'd be willing to bet that at least one of them has a bulging top and/or some leaked juice.

Reduced power cleanliness + underclock = stable again?

Sam

You may be right but I haven't looked.

\o/ Billy