>>>either way its a risk - especially with video cards ...
>>>a lot of extra heat and in some cases a cooling fan requirement...
>>>overheat the CPU ...
>> Actually, 450 MHz is the best way to OC the Celeron 300A, IMHO,
>>since you run the CPU at a high (but stable) speed, the memory at
>>100MHz (stable for PC100 memory), and the PCI/AGP bus at 66 MHz, so
>>you don't risk data loss. YMMV. From a happy Celeron 300A/450A
>>user...
>Please don't bite me, I'm a complete newbie at this overclocking
>stuff. I'm just happy my computer runs at all. As suggested, I will
>check out what's being posted in alt.comp.hardware.overclocking.
>But please do explain to me, if overclocking results in such a
>significant speed increase for your computer without risk of serious
>harm to the hardware, why don't they come overclocked that way
>directly from the manufacturer?
>What's the catch?
Overclocking isn't guaranteed to work - for the manufacturer, or the
end-user. The definition (mine) of overclocking is specifically running the
processor (or other system clock-driven component) beyond manufacturer's
specifications. Example: Intel designed the Celeron 300A to run reliably on
a 66Mhz bus with a CPU multiplier of 4.5. I happen to be lucky enough to
have a Celeron 300A that is happy to run on a 100Mhz system bus, yielding
450Mhz - and costing about $450 less (at time of purchase) than a 'true'
PII-450. This is not true for every 300A system. There's more detail to
this story, but this is the gist. And this is r.a.s.
--
Philip M. D'Amato
"You can observe a lot just by watchin'"
--The Inimitable Yogi Berra