System temps go up/down according to clock speed??

[Bonka]

I'm not sure if this is suppose to happen, but on my ST6, system temperatures will vary at whatever clock speed I'm running at. For example, at my overclocked speed of 1.553GHz 1.8v max, my system temp averages around 32c. However, right now, I'm underclocked at 600MHz (1.8v) and my system temp is 24c. I'm not sure if these temps are suppose to drop so dramatically even though I have pretty decent case cooling. From what I've seen, case temps are move only about 1-2c at different speeds, yet mine fluctuates like a psychotic cow. Anyone know what the cause of this is?

Hope I posted in the correct forum since I didn't seem to get very much response in the Intel forum.

 

[Lt. Max]

yes they are sposed to drop cause your cpu creates a lot less heat when its underclocked... your temps are very good btw.

 

[Bonka]

Actually, when I was running at [email protected], I just managed to get my system temps around 26-27c under load. Strange how it jumps up to the lower to mid 30's by adding only 3MHz to the fsb and maxing out my voltage.

I've heard of speculation that my boar'ds system temp sensor is located inside the winbond sensor, therefore giving me higher temps. Has anyone heard of this, even on a different board? Does this even hold any water?

 

[Lt. Max]

voltage increase creates a lot of heat, so the lower the voltage the better, as long as your comp is stable.

 

[smilingcrow]

The power output of a chip is I believe a function of the square of the voltage.

Gimme that again but in English brother!

Okay, if you double the voltage the power output quadruples.

i.e. 1 gHz @ 1.5 volts is not the same as 1.5 gHz @ 1 volt.

Hope that's clear.

Smiling Crow

 

[Voodoo Rufus]

Smiling crow is right. But you didn't decrease your CPU voltage. Your dramatic decrease in CPU speed could result in it. I think clock speed is proportional o heat output. Still, you have a very dramatic temp difference that is weird. You might not have good enough airflow.

 

[mx-6*]

Dude, his temps are great.

Clock speed has nothing to do with heat. What makes heat is the current drawn by the processor x the voltage.

He maxxed out his voltage so you have to know that his heat increased by a large amount.

P=I2R, P=V2/R, or P=VI


Current is directly proportional to Voltage

 

[Josef]

Dude, his temps are great.

Clock speed has nothing to do with heat. What makes heat is the current drawn by the processor x the voltage.

He maxxed out his voltage so you have to know that his heat increased by a large amount.

P=I2R, P=V2/R, or P=VI


Current is directly proportional to Voltage

True.

 

[Cooler666]

Current is directly proportional to Voltage

well unless resistance increases

 

[mx-6*]

Your point is?

 

[smilingcrow]

P = V * I

Yeah, of course, I don't know where I got the idea that power is related to the square of the voltage from?
Glad I aired it so I could clear up my misconception.

Thanks All.

<<Clock speed has nothing to do with heat>>

Well you may or may not be tehnically correct on this point in theory, I don't know, but in real world scenarios I'd say that the two metrics do seem to have some kinda relationship and it 'aint platonic. The relationship produces too much heat for it to be platonic



Smiling Crow

 

[Cooler666]

actually its V=I*R is what they were talking about, those rearranged power formulas have nothing to do with it

but who cares, i wish i had those kind of temperatures

 

[Silver]

Get radiate off the front of the site and play with it some. It will give you a better idea of what voltage and cpu speed do to heat.

 

[mx-6*]

actually its V=I*R is what they were talking about, those rearranged power formulas have nothing to do with it

but who cares, i wish i had those kind of temperatures

How do you get power from V=IxR? That gives you voltage, which he can easily check in the bios.

And Smilingcrow, figuring out why his heat went up is no mystery. His CPU is drawing less current at the same voltage. If you want, get out a meter and check it yourself.

 

[Voodoo Rufus]

The higher the clock rate for a given voltage the more current a processor will pull. It is proportional I think, too.