-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Temperature Vs Performance
- Thread starter Zephyrus
- Start date
Don
The rough and ready figure in my experience for A64s is that if you drop the load temp by 10C you can overclock an additional 100MHz or so.
Jokester
Jokester
I know its daft but i was thinking higher temps would make the metals in the chips expand and affect performance that way.
Soldato
A CPU either works, or doesn't, depending on the temperature. Too hot and it stops working, and sometimes too cold and it stops working. But if your temperature is within the specified limits of the processor, no matter what that temperature is, it'll perform the same.
Soldato
I remember reading the the specs for a mates vapo chill... it said when it was below zero it worked faster at the same speed as an aircooled chip.
I have my doubts
I have my doubts
Associate
you could clock to ~ the equivalent of his chip, and test that theory using superPi maybe? would be interesting.
Associate
They will overclock further when cooler but, MHZ for MHZ temp makes no difference.
Don
Goksly said:I remember reading the the specs for a mates vapo chill... it said when it was below zero it worked faster at the same speed as an aircooled chip.
I have my doubts
There's a certain element of truth to that but, though the logic works faster once it finishes what it's doing it just sits there until the start of the next clock. On air it takes slightly longer for the signal states to propagate through any logic than if it was 50C colder. A CPU is stable if this propagation delay is always less than a clock cycle (or however many clock cycles it's supposed to take).
It's that property that lets you clock the CPU higher as you're taking advantage of the quicker transistors.
Jokester
Electrical resistance drops as the temperature drops, so in theory pushing more current (overclocking) through a chilled CPU should cause a smaller increase in temp than the same increase in current for a non chilled CPU, and as you would be starting from a lower temp you would even greater scope for overclocking. As to whether a signal would pass across non overclocked CPU any faster at a lower temp I don't think the difference would be measureable.