• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

The relationship between clock speed and heat output

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

microfire

Member
Joined
Oct 8, 2001
I been thinking and could it be ture that when you increase the clock speed that the heat generate becomes much than the increase in speed, I try explain what I mean with the following for example:
Say if you had a 3ghz cpu at default voltage cpu and it generated 90watts of heat, then would the same cpu at 6ghz default voltage would generate 360watts of heat.
I know that that high speed and voltage is not possible, its only intented to give you an idea of what im trying to say.
 
You are thinking about a very reasonable mathematical relationship.

Essentially what you are asking/proposing is that the relationship between FSB (in effect, clock speed) and heat output may not be a 1 to 1 (1:1) relationship.

Example: If you simplify the relations between clockspeed and wattage they are easier to compare. I will use the numbers you proposed:

3Ghz/6Ghz = 1/2; 90watts/360watts = 1/4;

With a 1:1 relationship between clockspeed setting and resultant wattage then both fractions would equal 1/2 - double clockspeed, double heat output. Your example however, has double the clockspeed and quadruple the wattage - a 1:4 relationship.
End example

The relationship between FSB and heat output behaves as somewhat of a direct relationship - as you increase FSB a little, heat output will increase a little. I think people call the relationship between FSB and heat output a 1:1 relationship.

The relationship between a voltage increase and heat output on the other hand could not be considered 1 to 1. A very slight increase in voltage will mean a considerably larger increase in heat output.
 
Last edited:
Ive always wondered what sort of relationship speed and thermal dissipation had, for the most part i dont think that it is a linear relationship, i.e. a linear function x^1, i always thought that it was more of a 10^X type of relationship, also dont forget that the thermal dissipation differs when process size is changed, so its sort of like a stepped function. Grrr, hard to explain. Ill try plugging in some numbers to may calc and fit a curve.
xb-70
 
Back