Higher Temps = Higher Power Consumption?

Aris · Post by **Aris** » Wed Sep 19, 2007 5:50 pm

i was over at toms reading about a solar powered pc, when i noticed this graph they did.

http://www.tomshardware.com/2007/09/13/ ... page8.html

it shows that when they tried to passivly cool the processor, that their power consumption increased? By a quite a noticable amount too.

I didnt think method of cooling and the temp of the cpu had anything to do with the power consumed. Are they wrong in their analisys or have i missed something?

cmthomson · Post by **cmthomson** » Wed Sep 19, 2007 6:06 pm

I forget the formula, but yes, hotter chips do consume more power than cooler ones. The numbers quoted there are surprising though.

dhanson865 · Post by **dhanson865** » Wed Sep 19, 2007 7:43 pm

It was over double the load temp. You don't see people with Ninjas running a CPU at 86c. I'm assuming their "passive" test was with the stock HS not with a HS designed for passive use.

mad_man · Post by **mad_man** » Thu Sep 20, 2007 5:58 am

well, if you try to apply some non_silent_practise to this

given a active cooling, or water, at given Vcore your cpu can go to certain mghz.
if you drop the temp by a LN2, at the same Vcore it can go higher. its not because it likes cold, its because of Si. the higher the temp, the higher the electrical resistance. the higher the resistance at given Vcore, the higher the current draw...

when i switched from air to water, it aloved me to use 0.1V less Vcore at same speed.

so it can be true

bobkoure · Post by **bobkoure** » Fri Sep 21, 2007 4:46 am

mad_man wrote: when i switched from air to water, it aloved me to use 0.1V less Vcore at same speed.

Same here - or o/c a bit at stock voltage.