Big computers are no problem

[MThompson016]

It's been a long running complaint and joke about the size of computers in Traveller. As a younger fan, this is why I have no issues with the bulky computers aboard starships. It's largely an issue of power supplies, cooling, and how big a dTon actually is for volume.

https://travellersandbox.blogspot.com/2019/01/big-computers-no-problem.html

 

[mike wightman]

I agree with your assessment completely.

 

[Timerover51]

I just bought a 4 Terabyte hard drive that is about 3 inches square and an inch think for $109.99. Also, remember that the Ohio-class boats were designed for the government, which means that the computers were a minimum of 3 years out of date when first installed, due to the design and bidding process.

I still say that Traveller computers are far too big and expensive, as when Traveller was written, we were still using 80 column punch cards. That is what I was using in 1968-9.

A Windows PC with a 486 processor could be used to replace the computers used on some of the Soviet Union latest ships when it broke up.

 

[MThompson016]

I just bought a 4 Terabyte hard drive that is about 3 inches square and an inch think for $109.99. Also, remember that the Ohio-class boats were designed for the government, which means that the computers were a minimum of 3 years out of date when first installed, due to the design and bidding process.

I still say that Traveller computers are far too big and expensive, as when Traveller was written, we were still using 80 column punch cards. That is what I was using in 1968-9.

A Windows PC with a 486 processor could be used to replace the computers used on some of the Soviet Union latest ships when it broke up.

I'm still not convinced that computers will be that small. First, a lot of that volume in MCC was empty, to make repairs easier when pulling modules. Also, servos and system connections are going to take up volume.

Second, when I was onboard, a cabinet with 50 or so fibre-optic links in two redundant switches was installed in the Missile Compartment. It was about a sixth of the size of junction panel it replaced functionally, but it was still a sizable mount.

We knew everything was obsolete as hell (who uses SCSI and coaxial Ethernet still?), but it was (theoretically) reliable. We blew several power supplies in the upgrade by turning them on and off too frequently. Power and cooling requirements computers have are something often neglected in fiction. To be honest, most Traveller spacecraft must have some kind of amazingly awesome heat dispersal mechanism. Liquid helium cooling for high model number computers would take up that space!

 

[tjoneslo]

As someone who works with servers, I would say while the computers and technology get smaller, the space allocated to them doesn't get any smaller, it just increases in capability. And since the computer needs to monitor and keep running a fusion power plant, two drive systems, life support, a sensor suite, communications, internal security, and a set of weapons system.

I also recall the Space Shuttle was using an Apollo era computer as the main computer, a really old design. This was because it was hardened against radiation and both the hardware and software had been proven to work correctly.

I for one will not be running my ship on a 486 computer.

I do recall an article in White Dwarf in the early 80s talking about allocating space for computers. And noting that interfaces (i.e. terminals) which require a desk which takes a 1/2 dton of space. each. So that is one thing.

 

[AnotherDilbert]

As someone who works with servers, I would say while the computers and technology get smaller, the space allocated to them doesn't get any smaller, it just increases in capability.

Quite.

And since the need for cycles seems to grow more quickly than the capability of a single CPU, computers keeps growing in size.

E.g.:

Note that the power handling equipment in the foreground is part of the power supply to the "data centre" (≈ really big computer) in the background.


This is what is needed to keep your mobile terminal functioning smoothly...

 

[Timerover51]

How about we view it this way. I have a main and stand-by computer on the bridge, handling the control of the ship, and monitoring the engine room computers. I have a main and stand-by computer in the engine room, controlling the power plant, maneuver drive, and Jump drive, and being monitored and accepting commands from the bridge computer. I have three smaller computer handling life support, with the bridge computer monitoring them. I have a main and stand-by computer for ship security, which is being monitored by the bridge computer Overall, my numerous computers are equivalent in volume and cost to your single computer.

 

[mike wightman]

I don't think the computer is one stand alone device, rather it is the sum total of all of the computerised control systems of a spaceship or starship.

 

[aramis]

I just bought a 4 Terabyte hard drive that is about 3 inches square and an inch think for $109.99. Also, remember that the Ohio-class boats were designed for the government, which means that the computers were a minimum of 3 years out of date when first installed, due to the design and bidding process.

I still say that Traveller computers are far too big and expensive, as when Traveller was written, we were still using 80 column punch cards. That is what I was using in 1968-9.

A Windows PC with a 486 processor could be used to replace the computers used on some of the Soviet Union latest ships when it broke up.

The Corps of Engineers was still using punch-cards and punch-card slides, for various engineering purposes, into the late 1970's. The most recent asbuilt punchcards I maintained were in a 1980 collection within the National Archives system.

University of Alaska Anchorage had a functioning card-reader miniframe as recently as 1988 - It was gone by '92, but in the 80's, as a tween/teen, I made some spare change by carefully typing hand-written code into punch-cards for college students.

The USAF was still using mainframes into the 1990's, as well. (My dad's office at 11AF DE) switched from a Wang mainframe about 1991... Even as he was receiving a "fast" unix-desktop which took 10-20 hours to calculate air-route environmental impact...

Also, note that Vacuum tube computers are, in fact, far more resilient to both power fluctuation and radiation... which is why they're being looked at again.

https://www.sciencemag.org/news/2012/05/return-vacuum-tube

I read an article in the 1990's about soviet-era mini-tube computers. the tubes were essentially 2D; thicknesses of fractions of an inch.

 

[Timerover51]

The Corps of Engineers was still using punch-cards and punch-card slides, for various engineering purposes, into the late 1970's. The most recent asbuilt punchcards I maintained were in a 1980 collection within the National Archives system.

University of Alaska Anchorage had a functioning card-reader miniframe as recently as 1988 - It was gone by '92, but in the 80's, as a tween/teen, I made some spare change by carefully typing hand-written code into punch-cards for college students.

The USAF was still using mainframes into the 1990's, as well. (My dad's office at 11AF DE) switched from a Wang mainframe about 1991... Even as he was receiving a "fast" unix-desktop which took 10-20 hours to calculate air-route environmental impact...

Also, note that Vacuum tube computers are, in fact, far more resilient to both power fluctuation and radiation... which is why they're being looked at again.

https://www.sciencemag.org/news/2012/05/return-vacuum-tube

I read an article in the 1990's about soviet-era mini-tube computers. the tubes were essentially 2D; thicknesses of fractions of an inch.

I understand that vacuum tube computers are much better at handling radiation, and that the Mig-25 Foxbat's electronics were all vacuum tube. That is all I will say about it.

 

[wellis]

Do you think computers should still weigh less and take up less space aboard starships though, even if they have vacuum tubes in them, thanks to advances in manufacturing and technology?

 

[aramis]

I understand that vacuum tube computers are much better at handling radiation, and that the Mig-25 Foxbat's electronics were all vacuum tube. That is all I will say about it.

I'm aware of that bit of tech, but don't recall the source. And not just the MiG-25... the Soviets figured upon a tac-nuke war, and kept planes in the inventory that wouldn't be grounded because of it.

It was publicly available data in the late 1990's, IIRC. Heck, it's mentioned as such on wikipedia...

Oh, and it should be MiG, not Mig - it's for Mikoyan-Gurevich

 

[coliver988]

I don't think the computer is one stand alone device, rather it is the sum total of all of the computerised control systems of a spaceship or starship.

I have to agree with the original author & Mike. We're not just talking about the computer per se, but enough backup and resiliency to maintain reliability under extreme conditions, calculate jump parameters of which we've no idea how complicated that can be (as complicated as the plot requires ). It is also access to the system, ergonomic controls, and a lot of supporting gear.

And if you think Traveller computers are big - check out the ones in Star Trek. You could almost fit a Scout Courier into the one on the Enterprise (it is several decks and has to have a counter weight to balance it).

Sure - a 4TB drive can fit in the palm of your hand. A minor variance in power and there it goes. For decades I've actually assumed virtually unlimited memory (back in the day I had an everything watch that preceded cell phones, but had complete holographic systems. Basically a very fancy Dick Tracy watch. This was mid-80s, when I was, in fact, using punch cards for assembly programming on an IBM 360. Last class to do so, but I do remember the JCL (job control language) cards that had to be loaded before your program got loaded)

Somewhere in the original article I think the author also postulated the computer was actually a series of computers for redundancy. I like that idea. Not sure I'd trust the jump to a PC (well, it will calculate it, but it will take about 12 years to get the calculation to about 100 diameters out. If you are okay with being in the same system we can have that for you in a few days....)

 

[whartung]

The last mainframe I was involved in replacing was for a warehouse company in 1992-93.

Punch cards, room filling monster. The cut off date was very important because that was the day support would be stopped -- and support would not be coming back.

It was replaced with a pizza box Sun workstation and a 25 port serial concentrator.

They were very excited.

I've also replaced more than my share of refrigerator/washer/dryer combo sized computers with small towers and tiny drive array.

I've also dragged around 1/4 rack sized "small" computers crammed in to a cargo hold of a commuter plane.

I have no problem with Traveller computer sizes because their computing problems are not my computer problems.

 

[aramis]

To give an idea about "commercial applications"...

An elementary school I worked at about 2002-2008, built in 1998 or so...

30 classrooms (not all used as such), library, gym, music room, computer lab, 5 special use classrooms, 6 individual offices, 5 room main office. Main classrooms all served 5 computers on 8 available ports, plus 1 or 2 phones sharing ports. A 35 machine lab, 7 computers in the library proper, 5 extra computers in each of the 4 kindergarden rooms, 5× 1000-B-T RJ 45 to wi-fi routers, plus a pair of 40 machine mobile labs. 6 laser printers, 1 color laser printer. Roughly 500 students.

The switch room was 3x6m, with 2 banks of switches , double sided, and massive throughput for datastream (I am aware that 50+ simultaneous 480p streams was possible... because I've done it.) Also, one rack was closed circuit coax boosting hubs for the CCTV. Each classroom had roughly 1 cubic meter of data/telecom, as did the specual use rooms. Battery backup was included on the racks for the telecom multiplexer, but NOT for the data system.

So, roughly (35 classroom)+(5 for the lab)+(5 for the two mobile labs)+(5 printers)+(4 for office, music, and library). cubic meters of devices, 36 cubic meters of data and telecom switching, and a few dozen kilometers of wiring... individual 100-B-T lines to each port. 8 ports per classroom. Ends of the halls tend to be about 50m, total, from the switch room

So probably 60 cubic meters overall of computing & telecom, not counting seating space for the computers. Since then, i am aware they added IP-driven smartboards - 1 per classroom, 0.3 m each...

Adding in the video equipment .5 cubic meters per room for televisions, on closed circuit with DVD & VHS, in every room but the gym and MPR... another 20 cubic meters of Video Equipment, plus the wiring for that (low impedance coax)...

Standard classrooms are about 10x12 meters in that building, and ceilings throughout are dropped to 3m, with 1m to 2m above the drop ceiling.

So, that's a lot of AV/TC/Data network. the building is roughly 1100 to 1200 Displacement tons, with 6 tons of AV/Electronics, not counting operator space.

That's without a central miniframe.

 

[Timerover51]

It looks like I am odd man out, as usual. However, I would like to point out that the IBM 1401 that I learned on in 1968-69 occupied a room of roughly 25 by 25 feet with its various components. It had less computing power and storage than the Mac IISI that I bought in November of 1990, which had the largest hard drive we could get at 80 MegaBytes, and cost $3200, without monitor or printer. My cell phone has 500 times the storage and more computing power than the first Mac that I owned, along with a tiny keyboard that drives me nuts. The iPad that I bought for my wife for Christmas has even more power, and a nice size keyboard, which folds up into a case of 9 inches by 9 inches and less than an inch thick.

I have seen computers shrink to a degree that makes some of the earlier 1990s science fiction seem hopelessly out of date. Remember as well that the initial computations for Hohman Transfer Orbits were done using a slide rule in the early 1920s. The Atomic Bomb was designed with slide rules. I will continue to hold the view that the Traveller computers are far too large and expensive, and will be making them smaller in my version of the Cepheus Engine Rules. They will also be cheaper. I will be catching the players with operating costs rather than capital costs to soak up their credits/dollars/stellars.

 

[BlackBat242]

Remember as well that the initial computations for Hohman Transfer Orbits were done using a slide rule in the early 1920s. The Atomic Bomb was designed with slide rules.

And how many days/weeks with how many people operating slide rules did it take to calculate those things.

CT has a single Navigator calculating a Jump in less than an hour, just himself & the computer.

 

[kilemall]

I was unhappy with the CT computer rules, when I was playing with them, in the 1980s when I was working on said mainframes.


Moore's Law was already clear and the TL progression annoyed me no end.


So I had rules for virtual machines (yes that wasn't invented recently, first trick of IBM before a proper OS actually), exponential power increases per TL, rules for analysis complexity and time it took, etc. etc.


The complexity rules from Gurps Cyberpunk and Space Opera were helpful too.



Nowadays I don't worry about it. Especially given the damage sinks computers are in CT ship combat proper, the computers are clearly overengineered and/or spread all over rather then one room/cabinet, and tying them to the sensors helps cover the costs and justification for the extreme differences in commercial starship vs. military/research capability.


Nowadays I spend a lot more time with EM bands then modeling computing.

 

[whartung]

My cell phone has 500 times the storage and more computing power than the first Mac that I owned, along with a tiny keyboard that drives me nuts.

Yet it does less.

There's more to computing than raw CPU power.

My iPhone is a marvel. From its fluid gestures, on board image correction, the "AI Chip" doing 600 BILLION operations per second just so it can recognize that it's me trying to turn it on.

My phone made me a "movie" from the photos stored on it. It was a montage of stills and video of my cats. Remarkable since my photos are not part of the cloud, I don't store any of it online. So, I know that my phone didn't just go an upload my photos to Apple where acre feet of servers plodded on it to make me a kitty cat movie. The phone figured all that out on it's lonesome during downtime while just sitting on my desk.

But, from a utility point of view, it doesn't do as much as my desktop does. With it's large screen, big keyboard, the drive array, etc. I'm certainly not typing this post on it.

So, computing utility is not necessarily tied to the size of the microchip doing the work. There's all sorts of other stuff potentially involved.

 

[Timerover51]

As I am more than a bit tired of getting piled on, I will not post in this thread again. I have expressed my views and gotten hammered for it.

Edit Note: What all of you who are arguing for large computers are saying is that after 5000 more years of development, the computer of the Far Future will resemble those of the late 1960s, 1970s, and 1980s in size, bulkiness, and cost.