[ / / / / / / / / / / / / / ] [ dir / ask / general / hikki / htg / jewess / maka / radcorp / startrek ][Options][ watchlist ]

/tech/ - Technology

You can now write text to your AI-generated image at https://aiproto.com It is currently free to use for Proto members.
Name
Email
Subject
Comment *
File
Select/drop/paste files here
* = required field[▶ Show post options & limits]
Confused? See the FAQ.
Expand all images

File (hide): ee9d49dbd71317e⋯.jpeg (185.43 KB, 1050x1050, 1:1, images.duckduckgo.com.jpeg) (h) (u)

[–]

 No.805045[Watch Thread][Show All Posts]

https://spectrum.ieee.org/computing/hardware/the-future-of-computing-depends-on-making-it-reversible

Can somebody explain reversible computing to me? I understand power dissipation results in heat, so are the two states on and on on? If so, do two ons become on on, and two on ons become on on on on? What happens when the transistors aren't being used? Are they just ignored, and the power consumption the exact same all the time? It's very confusing to me.

 No.805049>>805059 >>805322

It's all about state change. As it is now, a state is associate with an amount of energy and to change the state this energy is transformed into heat.

This, just like everything else, causes a net increase in entropy.

Now, you can never decrease entropy, but you can leave it the same.

That's the idea behind a reversible process:

https://en.wikipedia.org/wiki/Reversible_process_(thermodynamics)

So to apply this to computing you'd have to return your transistor to its initial state before you change its state.

That's where the problem comes in. Strictly speaking, a reversible process takes an infinite amount of time to reverse. In order to make it feasible you'd have to be able to this at a rate of at least 2 GHz, and that's where you'd have to compromise and still increase the entropy by a little.


 No.805054>>805064

dropped at 'it's time to embrace'

fucking popsci trash


 No.805059>>805082

>>805049

Transistors only have an off state and on state, do they not? Would returning a transistor to its original state before changing states not dissipate twice as much heat? This is the bit that doesn't make sense to me.


 No.805064

>>805054

He's right tho, family. Heat is the reason you don't see stock 6GHz CPUs.


 No.805082

>>805059

It depends how you do it.

If the process is "adiabatic" (as described in the article) you don't transfer any heat out of the system.


 No.805322

File (hide): fb0c11302ee8cc9⋯.jpg (152.09 KB, 508x800, 127:200, im ok.jpg) (h) (u)

>>805049

Interesting.




[Return][Go to top][Catalog][Screencap][Nerve Center][Cancer][Update] ( Scroll to new posts) ( Auto) 5
6 replies | 1 images | Page ?
[Post a Reply]
[ / / / / / / / / / / / / / ] [ dir / ask / general / hikki / htg / jewess / maka / radcorp / startrek ][ watchlist ]