l i n u x - u s e r s - g r o u p - o f - d a v i s
Next Meeting:
July 7: Social gathering
Next Installfest:
Latest News:
Jun. 14: June LUGOD meeting cancelled
Page last updated:
2005 Mar 12 16:58

The following is an archive of a post made to our 'vox-tech mailing list' by one of its subscribers.

Report this post as spam:

(Enter your email address)
Re: [vox-tech] Persistant hardware problem kicking my butt
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [vox-tech] Persistant hardware problem kicking my butt

I have been overclocking since those famous/infamous celeron 300A to
450. Overclocking can destroy CPU and cause other problems so when you
overclock, please keep that in mind.

I will give you an example.  Back in 2001, I wanted to help my wife's
PhD project.  I built 3 systems to help her run monte carlo with
matlab.  One of the primary number cruncher was an Asus KT133 chipset
system.  I overclocked an OEM thunderbird core Athlon 1Ghz to 1.2Ghz
and the system ran 24/7 for several months without one crash. 
However, after about an year, I can no longer stay at 1.2Ghz without
system crash.  After I switched back to 1Ghz, system stopped crashing.

Based on your description, your overclocking may have damaged some
components, but unless I know exactly what you did, it's hard to tell.
 Overclocking tend to adjust FSB (I was able to unlock my athlon's
multiplier lock with a pencil but I don't think that's possible now). 
When you adjust FSB, depending on chipset (your nforce2 chipset should
have some locks), you may inadvertently adjust AGP (need to stay 66mhz
or below, depending on FSB & divider value), PCI (need to stay at
33mhz or below, again depending on FSB & divider), IDE speed (you can
corrupt your IDE devices, especially hard drives) can all be affected.
 Of course, CPU and memory speed are affected and you may have boosted
voltage to increase stability.  All these has the potential of shorten
your hardware life.  I overclock knowing all those because I retired
my celeron 300A not because they failed, but because they were too
slow.  I also retired my athlon 1GHZ as it developed other hardware
problems.  I believe my system will become obsolete before hardware
failure.  In general, when you overclock, you need to pay special
attention to motherboard, RAM, heatsink and fans, case, and power
supply.  You need to use quality components because you are pushing
your system to the limit, any of those component fail can cause the
whole system to fail.

I would check RAM, don't buy Fry's generic RAM please.  I used to do
that to save money until 1998, I purchased mushkin memory and
suddenly, 90% of my system's random crashes and hang went away.  Ever
since that experience, I stick with only quality RAM.

Power supply is another problem, you need to use good power supplies
that deliver a clean stable current.  Bad power supplies may cause
some of those symptoms you are describing as well as random system
reboots, etc.  Good power supply companies are PC Power & Cooling
(expensive though), Antec, Enermax (about two years ago, some of their
top end PSU had some issues, they have since fixed it), and Sparkle. 
There are some new companies who seem to produce decent power supplies
but since their track record is short, I hesitate to recommend them. 
Even companies on my list has had periodic problems with some batches
of their power supply.  You can find people's complaints against power
supply by google.  Does your power supply have enough juice for your
12V rail?  Your IDE devices consume your PSU's 12V rail.  Not enough
can cause problems.  Power supply are given a total wattage rating but
you really need to pay attention to its composition as not enough
juice in any of the 3.3V, 5V, 12V rail can cause system problems.

A good case will help with your system's cooling but your problem does
not seem to be heat related.

If you can, borrow some good quality RAM and a good PSU and proceed to
test by leaving only one HDD, video card, CPU, 1 stick of RAM inside
your system, run memtest for a few hours.  If it passes, put another
stick of RAM in and memtest for another few hours again.  If possible,
run memtest for 24 hours everytime to get a larger sample size.  When
all those pass, then you can add components back one by one, perhaps
DVD burner after your RAM to make sure, then other components back in.

Since you use good RAM and PSU, that eliminate those two possibility. 
If you cannot pass memtest even with just CPU, memory, video, and 1
HDD, then it's likely it's your motherboard.  Best way to figure out
what caused your problem is through process of elimination.
vox-tech mailing list

LUGOD Group on LinkedIn
Sign up for LUGOD event announcements
Your email address:
LUGOD Group on Facebook
'Like' LUGOD on Facebook:

Hosting provided by:
Sunset Systems
Sunset Systems offers preconfigured Linux systems, remote system administration and custom software development.

LUGOD: Linux Users' Group of Davis
PO Box 2082, Davis, CA 95617
Contact Us

LUGOD is a 501(c)7 non-profit organization
based in Davis, California
and serving the Sacramento area.
"Linux" is a trademark of Linus Torvalds.

Sponsored in part by:
Sunset Systems
Who graciously hosts our website & mailing lists!