From: dlou@sdcc3.ucsd.edu (Dennis Lou) Subject: Re: CPU Temperature vs CPU Activity ? Organization: University of California, San Diego Lines: 40 Nntp-Posting-Host: sdcc3.ucsd.edu In article guyd@austin.ibm.com (Guy Dawson) writes: > >In article <1993Apr21.061246.11363@ucc.su.OZ.AU>, montuno@physics.su.OZ.AU (Lino Montuno) writes: >> This may be a very naive question but is there any basis for the >> claim that a CPU will get hotter when a computationally intensive >> job is running? My friend claims that there will be little difference >> in the temperature of an idle CPU and a CPU running a computationally >> intensive job. > >It first depends on what an idle cpu is doing! > >I'm not sure about DOS, but many multitasking OSs have an loop like this > > >loop: > is there anything to do? > YES -> do it; goto loop > NO -> goto loop > > >The CPU is not doing any work but it is still processing instructions... I've done some ASIC and digital design, but not any CPU design. It would seem to me that on a 486, the FPU is not being used, most of the cache is not being accessed, the off chip buffers/drivers are idle, the multiplier isn't multiplying, the barrel shifters aren't shifting, microcode isn't microcoding, etc. This means transistors aren't switching which means less power dissipated (in CMOS), which means less heat.\ From what I understand, the Pentium shuts down those sections of the CPU which aren't being used in order to cut down on heat/power. -- Dennis Lou || "But Yossarian, what if everyone thought that way?" dlou@ucsd.edu || "Then I'd be crazy to think any other way!" [backbone]!ucsd!dlou |+==================================================== dlou@ucsd.BITNET |Steve Jobs and Steve Wozniak went to my high school.