Helical Logic Computation

[JohnnyYesterday]

Multi-stage pulse tube refrigerators/cryocoolers have already made it to below 2 Kelvin.

They're simple.

[JohnnyYesterday]

If the technology has only a very limited market appeal, it is likely to remain a niche product at best and so remain relatively rare and expensive. I find it difficult to imagine the need for a domestic cooler that can refrigerate items down to one Kelvin.

If molecular manufacturing is available to individuals, something like a cryocooler would only be rare unless people didn't want to make them--which seems unlikely in a future where people grow their own starships, or they are starships.

[Drashner1]

Regardless of how easy or difficult the creation of the necessary cooling would be, I think that we need to have one of our physics people, probably Adam if he has time, read through the paper in question and give their take on the feasibility of the physics and the basic idea. As mentioned earlier, reversable computing is not something we've played with in the setting as yet, but it could have a role.

If deemed workable, we would need to consider to what degree we would want it to be workable in the setting. A longstanding principle in the setting is that tech is rarely as efficient/capable in practice as it is in theory (although transapients seem to manage it from time to time). As such we might say that the tech can only be built successfully by transapients (although modos tried and failed many times) or that while modos could make the processor they couldn't make a transapient mind from it (in OA, raw processing power by itself is insufficient to create a mind, an appropriate pattern of processing is required and minds can't be created with complete certainty by an entity of equal or lower S-level, especially by a lower S-level), or whatever we end up deciding.

In a similar vein, the cooling requirements of this tech may (or may not) limit it if the builders prefer/need something that can operate at higher temperatures more than they need compact processing. Or may find cooling hot processors easier than cooling cryogenic ones. Or may just invent a room temp reversable computing device and save a lot of bother all around. Basically all of this would need to be discussed.

While OA is certainly adaptable and willing to change with new information, we don't make this sort of change lightly or simply on the basis of the existence of the paper. So before we get too much into this, I think we need Adam or someone to take a look at it.

Todd

[stevebowers]

I can recommend Martyn J Fogg's essay on galactic colonisation as well
http://www.users.globalnet.co.uk/~mfogg/fogg1987.pdf

[Drashner1]

I can recommend Martyn J Fogg's essay on galactic colonisation as well
http://www.users.globalnet.co.uk/~mfogg/fogg1987.pdf

Er. What? Was this post intended to go on this thread? It's a cool paper (I have a Analog article that is either based on it or is it), but I'm not seeing how it would tie into Helical Logic Computation... Or is there something in it that maybe didn't make it into the article I have?

Todd

[stevebowers]

Oops; I posted this in the wrong thread...

[Tachyon]

Those appear to be talking about irreversible computation, so their figures aren't applicable here since reversible computation is thermodynamically reversible.

That's why it is so efficient, and doesn't turn into a hot plasma from its own waste heat.

Sigh. For the nth time, reversible computing isn't a free lunch. TAANSTAAFL.

Reversible computing was rehashed here:

http://movies.groups.yahoo.com/group/ori...sage/54610

Which, if you don't have access to the archives, is:

On 3/15/06, drashner1 <drashner1@yahoo.com> wrote:

> > > a) What about reversible computing? One of my books talks about
> >
> > I've been unable to wrap my mind around it.
>
> Same here. And this bothers me since it gives me this vague feeling
> that we could be missing something here.

Erm, not really much to say about reversible computing.

For an N-step algorithm, which normally expends energy ~ kTN using an
irreversible process, you may reduce the energy use by a factor of N^b
(0<=b<=1), but then the calculation will take a factor of N^b longer.

It just means an archai can throttle down their power use by slowing
their thinking.

No free lunches here. The same total energy is still expended. (And it
would be quite startling if that weren't the case.)

> Todd
>
> > -xx- Damien X-)

Adam
--
"Invincibility is in oneself, vulnerability in the opponent." -- Sun Tzu

[Tachyon]

<snip>
It seems to me that comparing the Ultimate Chip and this device is also somewhat like comparing apples and oranges. The UC is intended to operate at 'room temperature' pretty much while this tech is described as operating at cryogenic temps from what I can gather (haven't had time to read the whole article, sorry). And I also recall Adam stating when the UC was first introduced that it is the best possible within the temp range described and using phonons since they are most efficient in these conditions. That certainly leaves the door open to many other devices that might operate under very different conditions. A super-cooled or actively cooled UC might be much more powerful, although such hasn't been described yet.

To date, we do NOT have reversible computing really addressed in the setting, mainly because we haven't been able to find much solid info on it. If it can be made to work it may have a major impact on computation. That said, one thing I seem to recall is that it is overall somewhat slowed down by the need to perform all its calculations in reverse in order to 'cancel out' the entropy or energy dissipation from doing them initially. So this may also slow things down overall a bit. Although maybe this tech gets around that somehow.

<snip>

Thanks!

Todd

Computation is equivalent to a heat engine. For a good review article on the subject, see:

http://tnt.phys.uniroma1.it/twiki/pub/TN...machta.pdf

As with any heat engine, the Carnot efficiency *increases* as the differential between system and reservoir increases. That is, a computer operating at near zero is typically less efficient than one operating at room temperature, which is less efficient than one operating at neutron star temperatures, i.e.:

Cryogenic << Room Temperature << Fusion temperatures << Big Bang

I added "typically" because there are interesting quantum effects that may change this in some cases, e.g. interstellar chemistry mediated by quantum tunneling which recently appeared in the literature. It's far from certain that this is general.

BTW, this is the reason "cold fusion" i.e. muon-catalyzed fusion didn't work out. You get 10-15MeV of energy per deuteron but you spend 200MeV for the muon, so you need nearly a score of collisions just to break even, but collisions are exponentially unlikely. Raise the temperature and it's more probable, but then it isn't "cold" fusion anymore.

With respect to reversible computing, again, any Turing-complete computation can be broken down into a series of steps (i.e. an algorithm), for which you pay nKT energy to complete that step in a given timeframe according to Landauer's principle (where n is the number of bits you must erase for that computation). You can pay less than nKT energy per step, but then your timeframe for that step increases by the same factor.