No it won't..it will actually run cooler..think about this-when you overclock it gets hotter, its the opposite when you downclock (assuming you don't mess with voltage at all). On top of that, you can undervolt when you're running the processor slower, though there's a limit to how far you can go depending on processor technology (too low-processor will be unstable, overclockers know this too well) and what you can set in your mobo. Underclocking & undervolting is nowhere as dangerous as overclocking, I'll say there is little that can go wrong if you just stick to adjusting settings from BIOS (DON'T be ambitious and pull off your HSF thinking you don't need it anymore since your proc is cooler-those pre-pentium days are looooong over). I haven't tried this on VGA, but it should apply since the logic is the same.

If I'm not mistaken most silent PC sites speak of this technique to run PCs cooler so that you can use quieter, but less efficient cooling (preferably fanless). Just google for silent PC. I've underclocked and undervolted for the reasons above (and a few more obscure ones), if I've time I'll explain further with pics if possible.

PS-someone PM me how to insert pics in posts..im a noob..still stuck with text

This post has been edited by lohwenli: Sep 7 2006, 09:39 PM

Check this article from silentpcreview, its about undervolting the processor to save power

http://www.silentpcreview.com/article313-page1.html

the numbers are quite surprising, even the Athlon 64 X2 is pretty amazing (Conroe wasn't out yet for another few months when the article waas written). Never expected AMD's CPUs to take up so little power when tweaked right. Of course, this article makes no sense to overclocking addicts..

Maybe can hold competition for lowest power consuming PC (and probably most silent & coolest running) that can meet a certain level, say, 3000 3Dmarks in 3DMark2003? Should be interesting..

I said "Lowest POWER consuming". Taking the graphics card into account by including a graphics benchmark would make things even more interesting. Whats the point of saving 20W on the processor only to have it spent by the graphics card? Can set a few categories based on minimum required benchmark performance (can set a variety) and find the best tweaked system for lowest power consumption. Think its about time competitions like this came up, considering how insanely high system power requirements have gotten these days and the current performance-per-watt craze.

Only catch is that they'll need equipment to measure the power consumption, preferably from the wall socket so can also take PSU efficiency into account. For active pfc PSUs with power factors of >0.99 measuring from wall socket is easy (shunt resistors), but for those non-active pfc with lower power factors need slightly fancier equipment which I'm not sure where to get.

Well I just had a few ideas on how to measure system power, however all have their weaknesses. Clamp meters are not cheap, a cheap alternative is to use a shunt resistor in series in the power cable and measure the voltage across it (using a DMM) to calculate the current and power drawn. Problem is when the PSU has a power factor below 0.99, the current measured by this method or with a clamp meter will be higher than the actual amount. Its also possible to use a UPS and measured by the time it lasts before the UPS runs out of battery power-but I heard this is also affected by power factor although I'm not sure.

Btw, power factor and power efficiency are 2 different things. Very few PSUs can boast efficiency better than 85%, even 80% is achieved by only half of the quality brand PSUs while any active PFC PSU can easily achieve PF >0.95. For more details, read the article link below.

PFC decoded
http://www.dansdata.com/gz028.htm