Joker114w and CPU Temperatures

[bob]

This is very interesting.

On my I7-920 running at stock speed, quad core engines get my cpu to between 44 and 49 degrees celcius. Single core engines get it between 41 and 42 degrees celcius. Joker 114w (single core) consistently gets my cpu to 45 degrees celcius!

Is this an example of efficient programming, getting the most out of a core?

I'm curious as to what causes the extra heat?

Thanks

I just fired up Crafty with 2 threads on my core-2 duo laptop. After 60 seconds, the CPU temp is 95C and slowly inching up toward the boiling point of water. I've not seen it pass 100C but I have seen it get there. I terminated Crafty and the temp dropped 20 degrees in about 2 seconds... and was below 70C in 15 seconds.

[hgm]

You don't seem to understand the issue. Of course a CPU gets hotter when you start using it heavily. And if your cooling is lousy, it gets hotter at the same clockspeed and mask technology (65nm, 45nm) than when your cooling is good. And if you use 4 cores at full speed, it gets hotter then when you use 1. But that has nothing to do with the issue.

The question was why one single-core Chess engine heats up given CPU significntly more than all others.

[bob]

You don't seem to understand the issue. Of course a CPU gets hotter when you start using it heavily. And if your cooling is lousy, it gets hotter at the same clockspeed and mask technology (65nm, 45nm) than when your cooling is good. And if you use 4 cores at full speed, it gets hotter then when you use 1. But that has nothing to do with the issue.

The question was why one single-core Chess engine heats up given CPU significntly more than all others.

And it _could_ be better cache utilization. Or it _could_ be using different instructions in the CPU. It _could_ be a lot of things that turn gates on and off. As an example, is Joker supposedly faster than Crafty, NPS-wise? Very few programs can actually claim that. So does Crafty heat up the CPU more than others? Don't know the answer to that, although it definitely heats it up. But clearly, cache is not the _only_ possible explanation. And a program that uses SSE stuff might take this a few degrees higher by using a part of the CPU that most are not touching.