|
| quote: | Originally posted by DigiNut
bunch of words here about stuff |
Yes I realize that my CPU is being slowly damaged. Big goddamn deal. If it dies I replace it. Cheap as hell.
You CAN absogoddamnlutely make up for lack of cache memory with clock speed. Take that Q9300 and ramp it up to 3.5 or 3.6GHz, and hell yeah it'll be beating that Q9450 in most (oh holy fuck, exceptions!) applications.
I know this is a synthetic benchmark but it's impossible to really measure real world performance.

Look at that. 3.33GHz on the E2160 and it is within 1 point of the E6850, which has 4 times the amount of L2 cache memory. The E2160, with its mere 1MB of L2 cache is nearly on par with a CPU costing 4 times as much money at their respective times of release and needs just 333MHz more than the E6850 cranks out stock to do so. What's that you say? "Oh but Cale, golly gophers sir that isn't a white paper from Intel themselves! Surely it is flawed and therefore should be taken with a grain of salt!" Maybe, but you go find me another half-decent way of comparing performance.
Again, tell me why it's a bad idea? You'll have a very, very small number of CPUs that really do die prematurely. My Athlon (Thunderbird core) 800 ran at 1.1GHz for almost 5 years before I retired it, and those were about the hottest things ever.
Regardless of whether or not something in the CPU is being damaged, if it does not affect the end user in any way, explain to me why that is detrimental.
Overclocking is absolutely a fine idea if you know what you're doing and realize the consequences (or lack thereof, 99% of the time).
Last edited by VERTiG0 on Aug-28-2008 at 02:37
|