PDA

View Full Version : Conjecture on Ramsinks


moonunit
11-27-01, 11:54 AM
Bear with the ramblings of a newbie.

I recall reading stuff a few months ago about ramsinks, to the effect that they don't boost overclocking capacity noticeably.

Assuming RAM is like CPUs and causes crashes when it gets too hot, ramsinks are a good thing. We know RAM can only access words so fast (eg 4ns) before the data is not ready on time, and BSOD follows. Forget any thoughts about peripherals for a while...

Normally, the computer hangs because the RAM can't keep up, but at what stage does heat become the limiter? 150MHz? 160?
I haven't pushed my cruddy old RAM about, but what if you pack DDR? Clearly that stuff must generate more heat per clock cycle, as will RIMMS (spit, ptooi). Obviously, at the moment RAM is capable of staying cool enough by itself, but what if things get really toasty in the next couple of years? My GF2 already sports nice copper heatsinks as standard.

Will someone reassure me that we're not going to need to ice (figuratively) the whole blooming motherboard before too long?
I'd also like to hear from anyone who has found ramsinks to be really useful.

The Overclocker
11-27-01, 12:02 PM
ramsinks are genrally useless because the ram does not produce enoight heat for them to be usefull, ram speedsc an be a problem but you have to know the timing and amke of the ram to work ou what they could go up to, eg, pc66 (66mhz) ram cannot run at 133MHz (pc133), it is dependant on the ram chips

Thelemac
11-27-01, 12:42 PM
Wild_Andy_C did a test of this here (http://www.wildandyc.co.uk/pieces/ttramsinks/) showing some positive results from the use.