"Computational Energy"

J03

~~~~~~~~
Messages
5,558
Location
Wales
Is there such thing?

This is quite theoretical here...

But imagine a processor that gave out no heat, no sound, no light etc.

The processor could do super-complex mathematical equations, with 10000 times the processing power of the current day supercomputers.

Would the energy input always equal the energy output, even though the machine is performing strenuous equations?

Everything that I know about physics (conservation of energy & mass etc) is telling me that no power is needed to do maths... so I don't think there is.

I just find it interesting that energy is not needed to do computations.
 
You mean the laws of thermodynamics wouldn't allow a 100% efficient (wrt heat) processor? Or wouldn't allow "Computational Energy" to exist?
 
You mean the laws of thermodynamics wouldn't allow a 100% efficient (wrt heat) processor? Or wouldn't allow "Computational Energy" to exist?
I don't think they would allow a 100% efficient processor, but I'm not really sure. I know a 100% efficient engine is impossible, not sure about a processor to be honest. I would think impossible though.
 
Everything that I know about physics (conservation of energy & mass etc) is telling me that no power is needed to do maths... so I don't think there is.
Problem is "doing maths" isn't really a physical concept - the key is how you do the maths. However the implication there is that to do maths you need to do work to get the result, and doing work certainly does require power in some form or another (or at least a transfer of energy.)

100% in terms of efficiency is theoretically possible using classical mechanics laws, but practically it'll never happen, or ever get close (and it's a long time since I've looked at it but I'm pretty sure effects such as quantum tunelling will prevent 100% efficiency even in theory on the level processors work at.)

Instead if you're looking at the next huge leap in processor design, check out quantum processors. In the next few decades or so I reckon it'll go mainstream and give us an order of magnitude better processing capability than we have at present (it's actually a lot closer than people realise.) Thing is unlike computers at present you'll never really understand it unless you've got a pretty good grasp at vector maths well beyond 3 dimensions and a solid grasp of quantum theory. The potential however, is huge!
 
It is impossible to build a processor that does not give off heat as it is in direct violation of the Second Law of Thermodynamics.
The second law says no heat engine can be 100% efficient. But a processor does what? Turns electricity into heat? Work?

But I guess with the same law, no process in the real world can be 100% reversible, so it would be impossible to achieve 100% efficiency. But there is no limit on how close to reversible is achievable.
 
Moving electrons through transistors releases heat due to the mechanical make up of the device.
so its not possible.
 
Back
Top Bottom