29 comments

[ 5.0 ms ] story [ 70.2 ms ] thread
There's a problem with this approach: black holes evaporate, which would yield a useful-life ceiling of 10^100 years. I know I'd like my machines to work longer than that.
Photons in general relativity actually do have gravitational fields--the stress-energy tensor, which determines spacetime curvature, is a function in part of the local four-momentum which photons do possess. Assuming four-momentum is conserved in the radiator's frame, the photons emitted by the radiator are absorbed (or rather, accumulate on the event horizon surface) by the black hole and contribute to its mass. I think the net mass balance is positive, but someone who actually understands black holes may want to correct me on that.

Edit: I suppose you could always throw in the maintenance engineers when you're done with them. :)

I think 10,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 years would be a pretty usable amount of up time.
you're thinking like a human, I reckon 640kb of memory would be enough too ;)
That's a really good idea.

Come to think of it, could black holes actually be used as an energy source? A setup like the one he describes would create a temperature gradient, and it's easy to extract power from that.

That would only make sense if this process somehow "depleted" the black hole; otherwise one could reverse entropy. IANA theoretical physicist so I don't know which mechanisms would be at work here.

Non-relativistically speaking (danger will robinson!), the photons emitted by the radiator have a huge gravitational potential associated with them. You're free, I believe, to extract that energy through a heat engine if you like, but I'm not sure the how feasible or efficient it would be.

I'm totally ignorant on the GR front; where's RobotRollCall?

I wonder if black holes might be usable as cosmic-scale batteries. Suppose your civ was preparing to settle in for long-term survival into the super distant future. You might not need high amounts of energy, just a low constant supply for incredibly long amounts of time. You could drop a good percentage of a galaxy's mass into a black hole, then use the evaporated Hawking radiation over cosmological time scales to power your civ.

Suppose that the entities living inside this civ are simulations. A second to us might be tens of thousands of years to them, because their hardware processing runs very slowly because of the low amounts of available energy.

There's a lot of energy to be had simply in pushing black holes around 'til they crash into each other.
Or for young whipper-snapper civilizations who want to live faster, you could turn that black hole + galaxy-sized mass into a really big hydro-electric plant -- that is, power your civ from the potential energy.
If your blackhole is spinning, you can use its magnetic field as a source of energy. Eventually you will have robbed it of its spin, but that will take ... some time.
I'm confused. How is radiating your heat into a black hole better than radiating your heat into space?
Agreed. Indeed, if you follow the concept of Matrioshka shells or brains (http://en.wikipedia.org/wiki/Matrioshka_brain), when energy is released as heat after computation another layer sitting outside the previous one takes advantage of the heat to power further computation. By placing enough shells going outward, you effectively release very little energy. Possibly less than or equal to the CMB.
The black hole is colder than the average temperature of space. Computation costs less at low temperatures, because temperature is defined as the ratio between the change in the energy and the change in the entropy of a system.
But assuming your radiator is going to be hot anyway (we've got a Dyson sphere to produce all this problematic energy, right?) then surely it's not that big a deal?

Still, it gets me thinking science-fictionally. Are there any stories where aliens invade the Earth just to use it as a heatsink? Travelling around from star to star, dumping all your waste heat into whatever planets you can find -- sounds like a great lifestyle.

This was actually explained in the article, which is about cooling things down close to absolute zero. I haven't thought about this enough to see if it makes sense, but the claim is that:

It's not too difficult to cool a system down to the temperature of the cosmic background radiation. All you need to do is build a radiator in interstellar space with a very large surface area, and connect it with the system you're trying to cool with some high thermal-conductance material.

However, even at the cosmic background temperature of T=3K, erasing a bit still costs a minimum of kTln 2 = 2.87e-23 J. What is needed is a way to efficiently cool a system down to near absolute zero. I think the only way to do it is with black holes.

The theoretical limit for your thermal efficiency is going to be based on the temperature of whatever you are cooling and the temperature of the environment you exhaust the waste heat into. Minimizing the latter is what he seems to be talking about here.

Since these equipment cannot operate with perfect efficiency, they will need to eliminate waste heat.

Who says it has to operate at perfect efficiency? I would imagine that any slight improvement you get by incorporating a black hole into your cooling system (however the hell that would work) would be negated by the fact that a cooling system that involves a black hole is probably a bit expensive.

It might not be. Assuming you could find a black hole which is not actively engaged in swallowing something (and hence spewing out huge amounts of heat and relativistic particle jets), you could surround it with mirrors to create an internal volume with a very low photon temperature at relatively (in astronomical terms) mild expense.

Obviously a spherical shell would be unstable, but you don't need a contiguous shell to lower the internal temperature. A rotating ring (or easier, an orbiting series of small, overlapping satellites) would be stable (not in the orbital sense, but in that they wouldn't need to hold themselves up) and could use the thermal gradient between their sides to power their station-keeping. You could construct additional rings at greater distances, each with a different inclination, to cover more of the angular area.

That's like saying making a star go supernova is a great way to generate energy for an advanced civilization — there are some huge leaps in logic in making sweeping assumptions of that scale. It would be like someone in the Victorian era predicting that we'll farm whales for their oil...
It would be like someone in the Victorian era predicting that we'll farm whales for their oil...

That actually sounds like a pretty cool idea. Would it be possible to keep whales alive on land indefinitely?

They won't be actual whales — we'll use their genetics "or something like that"
No need to keep them on land. Arthur C Clarke's novel 'The Deep Range' had whales being kept in herds, like dairy cows, for their milk. Not implausible at all!
Didn't they do exactly that at the time?
They hunted whales, but "farm" implies some kind of active breeding.
I think science fiction writer Arthur C. Clarke said it best "Any sufficiently advanced technology will indistinguishable from magic." I paraphrased so don't hurt me if I didn't get it word for word. I have to agree with him that any civilization that was advanced as the essay speculates, would seem like real magic to our relatively young civilization.
When you cool something down to near absolute zero it can turn into a Bose–Einstein condensate (BEC). http://en.wikipedia.org/wiki/Bose-Einstein_condensate For example at 1.7×10^-7K rbidium turns into a BEC. At 10^-8K perhaps that would turn normal matter into a BEC. Then you would have to be able to get useful information out of a state of matter where every possible value exists at once, and the tool your using to read values warms the condensate past the point of being one. Hey - I think you just invented a quantum computer. Congrats.

On a site note I think lasers are the advanced cooling system of the future!

You're going to have some major practical trade-offs with this choice of cooling system. For one thing, any data you get out of this computer when you're positioned further away from the event horizon is going to be red-shifted down to a lower data rate.

For another thing, the electrical and magnetic fields generated by your computing and data transmitting hardware are going to induce a Lorentz force on the event horizon and torque it into rotation, which in turn produces both gravitational frame dragging and dissipation. Now it's a lot hotter than if it were just a Schwarzchild black hole emitting only Hawking radiation.