Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Some comments

Author: Slater Wold

Date: 22:27:42 03/28/02

Go up one level in this thread


On March 28, 2002 at 20:32:40, Tom Kerrigan wrote:

>Some things come to mind, thinking a little more about this.
>
>First is electromigration. When current goes through circuits, the paths are
>eroded by miniscule amounts. This basically only matters within dense chips like
>microprocessors, where the connections are only a few hundred molecules wide.
>The point is that these chips wear out when you use them, but they're still
>designed/manufactured for 10+ years of continuous use, so it doesn't really
>matter. The problem is when you start overclocking and upping the current
>through the chips--some people claim that with high current, CPUs can be
>destroyed because of electromigration within 2 years. Again, not a problem if
>you don't overclock.
>
>The only other parts of a computer that wear out are the capacitors and the
>moving parts, i.e., the fans and hard drive. Old capacitors leak, but I don't
>know how usage patterns affect this leakage. I also don't know if fans wear out
>faster if they're left on or if they're cycled on and off. I know that hard
>drives used to wear when they were turned on and off, but now that we have
>autoparking heads and so forth, I don't think that's an issue. Laptop
>power-saving software is constantly turning hard drives on and off, and I
>haven't heard complains frop laptop users about hard drive lifespans.
>
>Really, I don't think it matters. I know people who leave their computers on all
>day and I know other people who turn theirs on and off many times per day, and
>the compuers all last a darn long time.
>
>-Tom

When I bought my AMD 2x1.53Ghz, my first benches were my *slowest*.  Of course,
I booted up these CPU's for the first time ever, and went straight to the
benchmarks.  They were the slowest I've ever got.  2 hours later the benches
were 3% higher, constantly.  Therefore, that would blow that theory right outta
the water.  No CPU is cooler than one that's never been used before.  :D

Also, I just looked, and my computer has been running for > 20 days.  The
benches are within .5% of my benches I ran 12 hours after getting the system.

If you have a memory leak (best way to tell is to check resources, open program,
use program, close program, and then check resources) your system performance is
of course going to degrade *very* fast.  (CM used to have a pretty bad leak.)

I have no idea what lead this guy to write such a bogus article, but it's pretty
silly.  When I turn on my system, my CPU's almost immediatly jump to 20 C or so,
and then never go above about 45 C (after 18 hours of running WAC).  25 C is
*not* enough heat to cause *any* kind of problems.  In *any* computer.




This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.