[rc5] Dangers of overclocking
Rebecca and Rowland
rebecca at astrid.u-net.com
Sat Jul 12 23:50:21 EDT 1997
>Date: Fri, 11 Jul 1997 22:46:59 -0400 (EDT)
>From: Vincent Janelle <random at avara.com>
>Subject: Re: [rc5] Dangers of overclocking
>On Sat, 12 Jul 1997, Rebecca and Rowland wrote:
>Most of the time, the processors are shiiped which actually run at a
>higher speed than advertised.
The fact that virtually all microprocessors are quite happy at higher
speeds is irrelevant - there's a whole computer out there, and overclocking
affects *all* of it. What about the extra power line glitches? Is there
enough decoupling for the higher speeds? What about the extra power
consumption - are the circuit board tracks dropping to much voltage? And
that's just for starters - there's an awful lot of components and
connections that have nothing to do with the microprocessor, and you're
speeding all of it up to a speed beyond its design criteria.
Sometimes it works, sometimes it doesn't.
>A friend of mine actually has a P166 that
>runs at 183 mhz, and he iis thinking about reseting it up as a P-180.
>Another thing about heat problems. The new TX motherboards will put the
>processor to sleep if teh heat gets too high(don't ask me how this is
>done, I just found out about it today), so, the heat factor is less of a
>problem. And I agree, Sticking a high quality heat sink is necessary to
>keep a processor running, overclocked or not..
Rhubarb - my 68LC040 Mac doesn't need any heatsinks. You only need
heatsinks if you're using high power consumption processors of the sort
Intel and DEC produce - PPC 603e processors don't need heatsinks at 300MHz.
However, if you need a heatsink, it's vital to use heatsink compound to
ensure good thermal contact between the heatsink and the IC - a thin layer
of gas is an amlost perfect insulator, so not using heatsink compound can
make overheating worse.
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.
More information about the rc5