by kokojumbo on 1/27/23, 1:14 PM with 40 comments
1. Ease of manufacturing – it ain't rocket science to make cathode ray tube one compared to other currently known displays. It's a macroscopic process with big parts if you are fine with analog oscilloscope/radar type display. Good luck producing LCD array with same level of tech in some remote colony light years from advanced foundries. There is a reason why we are using predominantly LCD just for less than 20 years really.
2. Lowering complicated integrated chips count. Good luck powering matrices of a lot of dots w/o actually attaching a separate computer to it.
3. High refresh rates – good for emergency situations or combat.
4. They integrate well with analog computers which might be used per point 0 with big enough precision (and possibly responsiveness if there were some advancements in OpAmps) speeding up calculus stuff which means simpler Arithmetic Units compared to tons of transistors needed to implement logic gates, ram etc. needed to do calculus in non analog way. It might actually save energy to calculate a lot of that stuff in analog way.
5. Per points 2 and 4 you get a space ship that is more robust against radiation threat that might cause malfunction of very highly integrated circuits due to high density of transistors.
Please feel free to criticize in the most detailed way possible.
P.S. Do you have time to talk about our Lord and Data Savior magnetic tape?
by PaulHoule on 1/27/23, 2:38 PM
A key technology for space colonization is advanced manufacturing that lets a small population be self-sufficient. An outpost on the moon can probably be supplied from Earth but even to colonize Mars you would need an advanced industrial base even when the population is 10,000 so you need something similar to Eric Drexler's assemblers even if it is based on different principles. (The existence of life proves that something like this is feasible... Even if you can't assemble molecules with little robot arms or mills it is definitely possible to build arbitrary complex things by composing a set of building blocks.)
The most feasible path to interstellar travel seems to be to use deuterium-deuterium fusion to be self-sufficient off comets between the stars. Recent discoveries about artificial magnetospheres make this seem more feasible than it did 30 years ago because we know think it is possible to slow down using the interstellar medium which greatly conserves fuel.
Interstellar travelers might take 10,000 years to get to the next star but might have a comfortable lifestyle the whole time. If they found a Pluto-sized world floating between the stars they could even establish a colony and build some more ships. The big contradiction is that people like that might not really care to visit star systems at all, which I think is part of the resolution of the 'Fermi Paradox'.
But yeah, there is something sexy about the bridge in Silent Running or the old Battlestar Galactica although it seemed they had a flat screen TV in Star Trek.
by theandrewbailey on 1/27/23, 1:35 PM
> Perhaps the most stunning discovery was the fact that most of the MiG-25’s avionics were based on vacuum tubes—not solid state electronics. This was considered woefully outdated for a top-of-the-line military jet in the 1970s, but the vintage system had its advantages. The vacuum tubes were more temperature-tolerant than modern avionics were, which allowed the MiG to fly without weighty environmental controls in the avionics bay. Plus, the tubes allowed for quick and easy maintenance at Russia’s primitive airfields, and the antiquated system would better withstand the circuit-frying power of an electromagnetic pulse created by a nuclear blast.
https://www.popularmechanics.com/military/aviation/a38083125...
by rolenthedeep on 1/27/23, 4:55 PM
The CRT itself is heavy, it's all glass and metal. Then you have to consider all the electronics required to drive it. Transformers and coils are very dense.
Whereas an LCD panel on its own is a very thin sheet of glass and some plastics. It's all driven digitally, so your entire control system is one fat chip.
Multiply this weight by the number of screens in a starship (hundreds? Thousands?) and you've got a problem. This extra mass significantly affects maneuverability and acceleration. You'd have to reduce mass elsewhere in the ship, most likely by reducing passenger count.
And good luck if you have to transport these things from a planetary surface. CRTs are inherently fragile and don't handle high acceleration well.
Magentic interference is also a likely concern. Every screen in the engine room would turn into a colorful mess any time the reactor is running. You'd have to invest even more mass into magnetic shielding.
All in all, I find this unlikely. Mass is too important in space travel. As other commenters mention, once you have the ability to produce ICs outside of earth, it's not that far of a leap to produce LCDs.
LCDs are simply better in every way that matters. Even refresh rate is arguably unimportant today, and will likely be solved entirely by the time we're thinking about interstellar travel.
by eqvinox on 1/27/23, 1:44 PM
Sure, but you may not have solved the problem of how to get the heat out of your spaceship. Since your only option is radiative heat transfer (no convection/conduction…), you're stuck with really fat radiators. So you really don't want to waste power into heat.
https://en.wikipedia.org/wiki/External_Active_Thermal_Contro...
by bee_rider on 1/27/23, 5:01 PM
If we suppose FTl travel is possible, 1 and 2 don’t really apply, the colonists can instead just get resupplies from Earth. In fact it will be very convenient, because FTL travel allows the traveler to go back in time, your spare monitor will arrive before you order it. The only downside is the sense of impending doom if you suddenly get a large shipment of splints and bandages from Earth.
3 seems unlikely to matter, humans are bad at quick precise movements compared to machines. By the time space colonization is possible, the (as RTS players would call it) “micro” at least of space combat will be handled by AIs. This seems independent of FTL, although we don’t know what the side effects of allowing FTL signal propagation inside a CPU could be. Really good branch prediction I guess, if we can just send a “which way did it go” bit back in time.
4 relies on point 0 I guess so we can toss that out.
5, if we’re moving humans we’ll have to come up with shielding anyway.
—-
If we don’t suppose FTL travel, then I think a colony will inherently need to include somehow the ability to make advanced computers.
There’s no practical way of returning resources from a space colony many light years from Earth. So, the colonists must be heading out there in their own self interest. Nobody is going to pay you to mine rocks on Alpha Centauri if it takes years to get anything back, you must have set out because you are interested in living there. So, the quality of life must not be much, much worse than Earth. Fairly up to date manufacturing will be table stakes for most colonists I think.
by tjpnz on 1/27/23, 2:51 PM
by pwg on 1/27/23, 2:37 PM
I.e., #1 -- ease of manufacture. A FTL species likely also has some technology similar to Star Trek's "replicators" -- and once said species is capable of assembling products by positioning individual raw atoms, difficulty of manufacture of a product onboard a space ship fades away and becomes little more than a raw supplies (you need the input raw materials) problem.
I.e., an "atomic" (as in placement of individual atoms) 3D printer allows for most anything to be assembled (provided one has the CAD files for the 3D printer and the required atomic materials).
by rocket_surgeron on 1/27/23, 2:31 PM
Any colony that can manufacture integrated circuits can easily manufacture LCD panels.
The ability to manufacture integrated circuits will likely be seen as important as the ability to manufacture oxygen in any colony, so there will probably be an LCD "printer" as part of their standard list of abilities.
This person made a simple passive LCD at home: https://www.youtube.com/watch?v=_zoeeR3geTA
The ability of simple automated machines stashed in a maintenance closet to churn out screens of arbitrary size as-needed is not far-fetched in a world of interstellar travel.
by simne on 1/28/23, 9:49 PM
Really, each beam is cloud of charged particles, and if two such clouds intersect, they will scatter, and will cause image distortion.
So must feed all image through 3-4 beams. And even with so low number, it is not easy and have imperfections.
More read next.
2. no. tft lcd, essentially, is large dram chip (each Thin Film Transistor have some memory effect, and in few tft designs, display cells are just memory cells, so cpu could access them via memory-like interface, or via system bus). In lcd, neighbor cells are so disconnected, that real hi-res displays work as 3-4 or even more logical displays in one physical, and this solves throughput problem (some monitors just used 3-4+ physical connectors).
3. no. This is limited by throughput and charges interaction.
4. unfortunately, no. Analog computers defeated by digital, because it is really hard and expensive, to make analog precision better than ~1/10000, but digital could stack few bytes, and even on old 8080 easy achieve 24bit (10^-6), or 32bit (10^-9) or even 64bit (10^-18).
5. partially agree. Small transistors more prone to single events, but large more prone for accumulated dose, because appear radiation matter move effects.
So real rad-hardened computers have few alu-s and circuit, which could disconnect non-working alu and connect spare.
ps it may sound non-politely, but really good read on magnetic tapes is Donald Knuth, and fortunately, their mathematics is very similar for hdd-s.
by somat on 1/27/23, 2:59 PM
Follow up question. Are there application where the CRT is preferred? And don't say to play old nintendo games, that does not count.
My dream setup would be to set up a computer system around one of those large monochrome SAGE raderscopes.
by cf100clunk on 1/27/23, 5:07 PM
by tinus_hn on 1/27/23, 5:49 PM
by rasz on 1/27/23, 5:13 PM
by warrenm on 1/27/23, 1:50 PM
Good on ya!