• Welcome to the new COTI server. We've moved the Citizens to a new server. Please let us know in the COTI Website issue forum if you find any problems.
  • We, the systems administration staff, apologize for this unexpected outage of the boards. We have resolved the root cause of the problem and there should be no further disruptions.

me being way too anal-retentive again

Pure Silicon has introduced a 2.5" 1 Terabyte solid state drive at 2009 CES (consumer Electronics Show). The drive is higher density than magnetic hard drives of that size. The 1TB Nitro SSD is the most compact SSD per gigabyte: 15.40GB per CC (cubic centimeter). Tech Level 7

Not commercial, yet, but there is a Stanford Research demonstrated concept electron quantum holography storage device. Multiple superimposed images of different wavelengths in the same hologram for approximately 3 Exabytes per 2 square inches using electron microscopes and a copper medium. Also from 2009.

Multiply ~1000X at each successive TL

I think you'll run into maximum possibly density awfully quick with such a generous multiplier.

And imo all this is pointless. I can see your desire for some guidelines shadowdragon but Traveller ship computers are not about data storage, at all. They are for all practical purposes limitless as far as any data you care to download or store in them. Consider the lowly Library program. At 1 slot it not only provides a whizbang interface and search program, but detailed information on everything in "the local stellar region", which I take as a subsector. Considerably more information than the collected encyclopedias of our single entire world. Imagine all the encyclopedias of Earth a dozen or so times over and you might be close. I generally also include the collected works of literature, art, and entertainment in MTU's Library program for the distraction of crew and passengers. And I still think there's room left in that 1 "slot" for all the personal logs and data the crew can ever imagine using :)
 
Last edited:
What's the max density?

The above holographic is the densest we've hit yet, and it's difficult to conceive getting much denser storage than that. You start to run into issues once you're involving scanning electron microscopes as your read/write interface. I certainly can't see getting anywhere near 3 to the 27th Exabytes on 2 sq inches from our current TL7 at TL15. Nor can I conceive of a purpose for that much information. Unless you want to map the entire galaxy to the atomic level ;)

Just a WAG, though I did see a figure for that somewhere, how many atoms in the galaxy...
 
I always thought of a ship’s computer as more of a fancy switch and timing device ... with more in common to an analogue telephone exchange than a modern desktop PC. IMTU loadable programs are hardwired ROM units in ruggedized containers (not dissimilar to old 80’s style Atari game cartridges). Thus the number of ‘CPU’ and storage rating represented the number of cartridge readers the ship computer has.

So the memory capacity (in bytes or multiples thereof) of a ship computer depends on the memory capacity of the program cartridges currently loaded. Most cartridges have no general purpose memory, but I suppose a library data cartridge might.

The question becomes what is the maximum possible memory capacity of a cartridge. Then, just as with modern USB sticks, you can have multiple cartridges all with the same form factor but different amounts of internal memory.
 
And imo all this is pointless. I can see your desire for some guidelines shadowdragon but Traveller ship computers are not about data storage, at all. They are for all practical purposes limitless as far as any data you care to download or store in them.

That is a point that seems valid, but I am not just talking about ships computers- I am also talking about the computers as designed through the computer design process. I just can not see a TL 9 "home" computer having unlimited storage for random data. I also cant see them not having any as it is not defined or even hinted at except for "XP" data storage. So lets look at this from a different tack- how much basic "data" is 1 xp of storage?

As far as the "library data" program is concerned, it has a benefit (+2 to gather info) far outweighing its "size" at 1 PP. Even normal skill based programs require 1 PP per point of skill rank. Most other programs are 2+ PP for a +2 to a specific skill or application of a skill. The Language Module is 5 PP for something that grants a character full fluency with 1 skill rank in the appropriate language skill. In light of this, I will likely be adjusting the Library Data program so that for each +1 to Gather Information checks it costs Cr1000 and 1 PP, similar to skill based programs.
 
okaaaaayyyyyy......... that's a start, but did you have a reason for your opinion or were you just throwing that out there and expecting me to have an apostrophe?:devil:

IIRC it has been bandied about that Traveller™ computer architecture is not what we use today, so I went with a decidedly low-ball option.

I dont really have anything with which to compare a 1 cpu computer but maybe a wristwatch or something along those lines- what did you have in mind?


Just to clarify, I am working from the T20 design rules.

Skipping over the rest of this thread to toss my 0.02Cr into the pie...

I have a memory stick in my pocket right now that can store about 1,000,000 pages of text. It weighs less than my housekey, and takes up less than 1/1,000,000th of a Trav displacement ton.

I also own a 1 terabyte hard drive. 500,000,000 pages of text. I'm pretty sure you could easily fit 10,000 of them in a displacement ton.

So, with early 21st century technology, we're talking a trillion or so pages of text per displacement ton of data storage.

I'm willing to bet that in the next few thousand years, we'll be able to improve on that by an order of magnitude or 20....

In other words, Traveller computers should be able to store easily the million most important volumes of information about EACH world in the known universe.....
 
Skipping over the rest of this thread to toss my 0.02Cr into the pie...

I have a memory stick in my pocket right now that can store about 1,000,000 pages of text. It weighs less than my housekey, and takes up less than 1/1,000,000th of a Trav displacement ton.

I also own a 1 terabyte hard drive. 500,000,000 pages of text. I'm pretty sure you could easily fit 10,000 of them in a displacement ton.

So, with early 21st century technology, we're talking a trillion or so pages of text per displacement ton of data storage.

I'm willing to bet that in the next few thousand years, we'll be able to improve on that by an order of magnitude or 20....

In other words, Traveller computers should be able to store easily the million most important volumes of information about EACH world in the known universe.....

Skipping thru threads is a great way to miss information.:D

Again, you are talking about pure dedicated memory. This may well be true for a thumbdrive or a 1TB hard drive, but the traveller computers are not dedicated memory equipment. They are processor, program storage, active program memory, controls (sort of, there are rules for control panels that seem to have been ignored by everyone including hunter and the gang), display and whatnot included.

So, to reiterate, I am looking for a way to be somewhat realistic AND maintain some sort of game balance. And, just for the record, I completely disagree that a 10 cpu or even a 100 cpu computer can hold virtually unlimited amounts of data. I do not disagree that they can hold quite a lot of information and dedicated memory would hold even more. Keep in mind the traveller computer programs size in PP and what they can do (the computer can only give a 10% bonus, in most cases, on a chance to do something), and think realistically of what the same computer program in a real world computer would come out to in size (not to mention the fact that it would make many tasks almost guaranteed successes). This implies a less compact form of memory to me (among other things), especially since it has been bandied about on this board that Traveller computer tech is not real world computer tech.
 
Last edited:
Really mate, it is your call. Pick some base number and multiply it by the CPU squared, cubed or even raise it to a higher power. Done!

If your computer does not have capacity at least comparable to 10% of its tonnage as dedicated memory then people are just going to carry around the futuristic equivalents of thumb drives.

Your insistence on both realism and game balance is intriguing. What course are you trying to steer? If you want realism then some percentage of the computer tonnage, say 10% or even 1% will give the computer vast data storage capacities.

If the heart of the problem is that your players are trying to carry around enough memory storage to look up any plot clue rather than doing the investigating themselves, then the problem can not be solved by spelling out numbers of bytes of data. The problem is with them and how they want to play the game. No amount of "game balance" will fix it.

You have been given many answers that get to the heart of the problem and put forward suggestions of density of storage. You have even been given the crafty answer from Hemdian which gets around the problem completely - "none".

In the end, as in the beginning, it is your call. There are no rules that are going to help you.
 
Fair summary JustinInOz, but I wouldn't say there are no rules to help, I've just been to lazy to look them up :)

MT had some pretty hard numbers for various media storage with the peak being holographic crystal storage iirc. I also suspect that while the numbers seemed reasonable at the time they may not now.

I'm fully in the camp of CT computers are not what most people today think of as computers (or is that the other way around?). CT computers include a workstation (or several - 1 per model number in MTU) and access space, subsystems, and so on. They are not a modern desktop, laptop, netbook, or anything nearly so mundane. Even the CT Hand Computer is far more capable and advanced from anything most people today think of as a computer. The ship's computer is excessively hardened, with multiple redundancy (just look at the combat damage rules). It'll run reliably 24/7/365 without EVER crashing (short of being hit by a starship weapon strike, and even then it is more likely to keep working than not, until hit several times). Try that with your desktop running any current OS, even without being hit by a starship weapon. In short, it's not just a computer, it's a Computer. A name meaning something similar to what we think of as a computer, but also much much more.

If pressed for an answer to the OP question, I'd borrow from the above noted electron holography storage and off the cuff go with something like this for CT...
:alpha: A Memory blade is a standard TL9 data storage device built to interface with a ship's computer. It takes up 1 slot in the computer and costs MCr 0.1 to purchase. Built to the standard size of 10cm x 50cm x50cm it houses a scanning electron read/write head and the copper storage media. Data is stored as multiple superimposed images of different wavelengths in the same hologram for approximately 243 Exabytes total write once information. This should be enough for anyone*.
* ;)

Some Exabyte examples from Wiki:


  • All the words ever spoken by humans would require about 5 Exabytes. Some debate about if that is as a text file or audio.

  • The sum of all human-produced information (including all audio, video recordings and text/books) is about 12 Exabytes of data. Presumably this is just information of lasting value (though I could be wrong) and does not include blogs, vacation video, and pictures of ones children and grandchildren that must be shared with everyone ;) I also hope Reality TV is never considered for inclusion ;)

  • One years worth (2002) of telephone calls worldwide on both landlines and mobile phones would require about 17.3 Exabytes if stored in digital form.

Again it seems to me this pretty much defines the Library program, but this option is for a blank slate to record your own data on, while the Library one comes loaded with info (hence the extra MCr0.2), but probably allows some storage for personal notes and updates, maybe 10%? Call it 24 Exabytes? Maybe more depending on the particular subsector of Library data.

I imagine a Type S would have one or two of these for recording survey and incidental data, perfect for those Detached Duty Scouts as well. And it's sort of how I imagined the X-Boats worked.

The IN probably has a few on it's ships for various uses as well. Mission recordings and such. And at least one for CrewNet storage needs. Which may get swapped out from time to time as it fills and crew leave and mark their personal data for deletion (soft deletion of course, the IN is going to have banks of blades going back millenia of course - I picture huge vaults deep underground in Depot systems... )

Yeah, that should be enough data storage, even for Traveller PCs ;) If not they can always buy another blade...
 
Last edited:
I realize it is ultimately my call, I was looking for suggestions and for opinions of my original thoughts on the subject. The focus on ship computers is part of why I am keeping the thread running, since ship computers are a small portion of the available computers I see no reason to form a ruling based solely on them, especially since you cannot make a model/1 ship computer (TL5) with TL5 computer equipment (electromechanical). Yes I know they are completely different, but since a ship computer is significantly smaller, then EVERYONE would be swiping them from ships to use in their personal lives rather than making the substantially larger and clunkier ones from the design sequences. There would be a thriving black market in ship computers. In addition, many replies are focusing on dedicated memory, which is also not what I was asking about. Dedicated memory will of course be much higher per vl than the computers themselves.

Ok so what I will be going with (which I am sure most of you will have some sort of reason for calling me an idiot) is:

Base Memory (Data) of computers: ~100 pages of text (my definition from above) x 10^(TL-5) per point of cpu. I will work out the math later for various sizes of computers to see how it flows.

Dedicated memory: I will most likely go with the rules for "xp data storage" for a guideline as to basic size and cost as well, maybe just a straight conversion of ~100 "pages" of text being ~1xp. Good enough if I need it and not so complex it will take much time away from the game.

Thanks for the assistance - have a day.

Dan:

Interesting concept there, fairly workable I suppose if a little more powerful than I would have at TL9. But if such a small item can hold so much information, how does one explain the limits on "xp data storage" for AI's? One blade should be more than enough for any and all xp storage.

This is what I am talking about. I am trying to keep consistent with the rest of the T20 rules, and while they are based on CT there are substantial differences as well. While I understand T20 is slightly less popular than T4 or RTT, it is what I am using for now. So again, thank you for your assistance all, but I'm going Sinatra on you.......
 
Last edited:
I have to agree with Far-Trader here. Much of the comparison is irrelevant. I would think that a starship computer (the main one at least) would be heavily biased in favor of specialized operating systems and programs for a specific purpose like making jumps rather than being some super "PC" general purpose computer. That is, the ship's computer is more like the avionics and computers on modern aircraft that serve a specific purpose and have specially written software.
For what reason would anyone want a ship's computer to be all-encompassing in ability? Wouldn't it make more sense that the ship's computer be for running the ship's systems and doing things the crew needed to do with the ship while secondary and tertiary computers like portable PC's do mundane stuff like let you play games and send e-mail?
Basically, the StarTrek model is counter-intuitive particularly on a military vessel. Putting all system control into a single central computer makes it too vulnerable to failure.
 
*snip*
Basically, the StarTrek model is counter-intuitive particularly on a military vessel. Putting all system control into a single central computer makes it too vulnerable to failure.

Again- I am not just talking about starship computers, I am talking about all computers. Neither are Traveller computers ST, SW or RW computers (personally I think they are their own special blend of handwavium and tubes). But to answer you directly:

The fact is, all system controls are already in one place- the ships computer. Ain't no other computers listed aboard any ship design I have seen so far. Not to mention- why put in extra computers when there is space in the one already there? No self respecting bureaucrat or businessman would in anyway deviate from the maxim "max results / minimum cost". It could cost money to do so.

The fact that the T20 errata replaces the "free cpu" column with "PP Max / Total" of the ship computers, identical to the master 1-9 computers from the design sequences? I leave it to you to ignore that obvious (and deliberate) link between the two.

So what I am doing is just defining it a little better (?) for me and my (potential) players. In my opinion, all of the design sequences in T20 leave big freaking holes. I personally don't like them. OTOH, I don't really require anyone else to dislike them. If you choose to ignore them, feel free. I don't want them there so I try to fill them with some sort of rational glop. It is the way I am.

I was just asking for advice and opinions- I got them and I made my decision.

Thanks for the input.:)
 
I'm fully in the camp of CT computers are not what most people today think of as computers (or is that the other way around?). CT computers include a workstation (or several - 1 per model number in MTU) and access space, subsystems, and so on. They are not a modern desktop, laptop, netbook, or anything nearly so mundane. Even the CT Hand Computer is far more capable and advanced from anything most people today think of as a computer. The ship's computer is excessively hardened, with multiple redundancy (just look at the combat damage rules). It'll run reliably 24/7/365 without EVER crashing (short of being hit by a starship weapon strike, and even then it is more likely to keep working than not, until hit several times). Try that with your desktop running any current OS, even without being hit by a starship weapon. In short, it's not just a computer, it's a Computer. A name meaning something similar to what we think of as a computer, but also much much more.

This really captures the orders of magnitude difference between what we use today vs. a couple thousand years from now.
 
Here are a few of my rambling thoughts about rationalizing how Traveller computers relate to real computers. Though in the end, I mostly just go with the "it's scifi, not reality" approach.

IMTU (which is MgT flavored), I assume that a Starship computer has effectively unlimited "long term memory", and the Rating is more about what it can keep accessible at any one time. I make the possibly naive assumption that computers will continue to have short- and long-term memories as distinct pools. :) There are practical limits to long-term memory size, but frankly I tend to hand-wave that part. (Such as, "You can't store Regina's whole Internet in your computer." and "Your passengers have access to a few hundred movies during the trip, but you may want to try to buy some more recent ones; those have been in circulation for quite a while....")

As an aside, I also consider "Library Data" to be both a database and a web-browser of sorts. You can often get more detailed Library Data about a world you are currently visiting and often about neighboring worlds, at least if you are visiting a TL-8+ world or one with a Class-C+ starport. If the Starport provides the data, you will probably have to pay for it. A Scout base will sell the best quality data, and will give it to Scouts for free. Otherwise, the local government determines access restrictions and fees (as well as content :D).

A model/1 is equivalent to a 70's or early 80's supercomputer in performance (that is, pretty much a modern desktop machine) but as a starship computer, it has multiple redundant systems and various fail-safes built in. Naturally, it's not just "a laptop plugged into the Ship's USB jack", but more like a rack-mount server farm. Since in MgT computers don't take up tonnage, I imagine each bridge has a dedicated space for the "computer rack" with access to the data feeds of all ship systems, and most likely the entire bridge's construction is modular to make repairs easier.

I view model/2 as being "the most reliable and performant computer we could imagine with current computer architecture". It should be just beyond what we can reliably manufacture today, but not into any "alternative" techs, like quantum or molecular.

From my experiences with dedicated hardware, in the end general purpose hardware always catches up and surpasses it. But to save money and time, specialization is very useful. Hence, "bis" - you don't *need* quantum computer technology to compute a Jump/2 if you build a special purpose J-Drive interface module, but a real Model/2 computer will be more useful, if you can get one.

Model/3 appears around the same time as synaptic computers are available, so presumably they incorporate some of that technology. Higher models probably follow from integration with quantum and other technologies. This is "beyond the Singularity" (in the strict sense of the term, not the "apocalyptic" sense used by Kurzweilites), so it's hard to guess what they will actually work like.

I toyed with the idea that Traveller computers are really NOTHING like modern computers, using some sort of handwave like "Transition to Jump Space causes an EMP, so silicon is actually a dead-end technology, so Starship computers use XXX processors, which have some odd properties." In the end, I was too lazy to follow up on that too much. :)
 
A model/1 is equivalent to a 70's or early 80's supercomputer in performance (that is, pretty much a modern desktop machine) but as a starship computer,

Interesting but, that wouldn't even cut it for the computers used to fly a B-2 bomber, much less a Jump capable Star ship. (see Far Traders post below for a better feel)
 
You won't mind if we just carry on though...

Not at all- just feel that being told there is no reason to bother was getting old .

Again, considering the last few posts, let me reiterate, I am looking more towards the computers as created from the T20 Computer Design sequence than ship computers (although for MTY they are basically the same- YTUWV). The whole idea was merely to give me a baseline for consistency. I also wanted something easy I can just plug into Excel if I need the specifics.

This really captures the orders of magnitude difference between what we use today vs. a couple thousand years from now.


The problem with that though is that with the limiting factors of Tech Levels, the number of years is irrelevant, it is the available technology as defined by Tech Level that is important.

*snip*
I toyed with the idea that Traveller computers are really NOTHING like modern computers, using some sort of handwave like "Transition to Jump Space causes an EMP, so silicon is actually a dead-end technology, so Starship computers use XXX processors, which have some odd properties." In the end, I was too lazy to follow up on that too much.

I quite like that, I think. Consider it stolen- I will post something after I toy with it for a while.
 
The problem with that though is that with the limiting factors of Tech Levels, the number of years is irrelevant, it is the available technology as defined by Tech Level that is important.

I agree but, when written we were early in our TL. Closer would a supercomputer today (think Lawrence Livermore Labs supercomputer). That wouldn't violate TL and would be more believable.
 
Back
Top