Pyramid #3/37: Tech and Toys II

[Tyneras]

Originally Posted by LoneWolf23k

Well, I was wondering how to rebuild Phasers and Disruptors for Prime Directive using stats closer to Ultra-Tech. This might work.

Depending on your campaign assumptions, it can work wonderfully, it worked out great for me. But I have discovered that if you get 3 trekkies to watch an episode, you'll get 4 interpretations on what all that technobabble actually means.

[tg_ambro]

This issue looks awesome. Too bad I'm not currently running a future campaign would I could easily subject . . . I mean introduce my players to these new goodies.

[PseudoFenton]

Originally Posted by Kromm

Just be sure not to push the red button without first asking what it does.

I hate warriors, too narrow-minded. I'll tell you what I do like though: a killer, a dyed-in-the-wool killer. Cold blooded, clean, methodical and thorough. Now a real killer, when he picked up the ZF-1, would've immediately asked about the little red button on the bottom of the gun...

[seasong]

I think Starmaker, Starbreaker is my favorite (pure MacGuffin bacon flavoring!), but these two have just become essential to my space opera settings:

Originally Posted by Kromm & Rev. Pee Kitty

• David Pulver, the man who came up with all of those lasers, masers, and bobasers (wait, sorry, wrong game), gives us a behind-the-scenes look at the Blaster and Laser Design system that gave him his starting numbers for GURPS Ultra-Tech. If you want a damaging beam weapon that isn't in the book, it's no longer a problem – just roll your own!
  
• Let's say your vehicle is damaged and you don't have time to fix the whole thing. Maybe you only need to get the thrusters up, or the teleporters. That's when you reach for Mr. Fixit, Matt Riggsby's simple rules for partial repairs to vehicular systems and other large toys. Now you can get power to your lasers without fixing the control panel or safeties – and you just know that's going to be fun!

Plus there's that computer article written by some hack ;-)

[Trachmyr]

The Computer Article is very useful for creating small computers and systems... but it seems to significantly Nerf Large systems, the possibility of AI (using the AI IQ limits currently in Ultra-Tech), and the possibility of high Complexity systems as a whole in futuristic settings.

In fact, unless you also use the Fast/Genius options from ultra-tech, Complexity 6 isn't even possible at TL8 except for billion dollar Server Farms. This is quite different than what's presented in High-tech and Ultra-Tech.

[seasong]

Originally Posted by Trachmyr

The Computer Article is very useful for creating small computers and systems... but it seems to significantly Nerf Large systems, the possibility of AI (using the AI IQ limits currently in Ultra-Tech), and the possibility of high Complexity systems as a whole in futuristic settings.

First, thank you for commenting! I had to avoid software, because it ballooned into too-many-thousands of words.

For AI, as a cinematic option: 5+(IQx0.2) for a shadow mind emulation (Ultra-Tech, p. 220) seems like a good starting place. If you decide that every molecule is important, 6+(IQx0.2); if even the tiniest quantum fluctuation is important, 7+(IQx0.2). Volitional AI reduces this by -1; non-volitional by -2; and dedicated by -3.

With that said, it is very likely that AI (with any IQ comprehensible to humans!) will only require a Complexity 7 computer. Possibly less, if our brains are inefficient at producing intelligence.

In fact, unless you also use the Fast/Genius options from ultra-tech, Complexity 6 isn't even possible at TL8 except for billion dollar Server Farms. This is quite different than what's presented in High-tech and Ultra-Tech.

Yes, it is quite different. Keep in mind that if you combine the rules from Ultra-Tech and High-Tech, then an IQ 10 human's mind could run on a TL8 supercomputer.

In the real world, one of IBM's BlueGene/P computers ("Dawn") can just manage to run a cat's brain at a 1:100 ratio - i.e., each second of brain activity requires 100 seconds of simulation. (The current world-record holder for computer speed could theoretically manage a 1:33 ratio.)

A cat's brain has approximately 1/100th as many neurons as a human brain. To run a real-time human brain in simulation would require 3,000x as much power as the current best supercomputers in the world . . . assuming that increasing the size by 100x did not substantially increase the complexity of the simulation.

My goal was plausible and consistent computers. So I strove for compatibility with the Basic Set, but I had to throw out High-Tech and Ultra-Tech.

[lexington]

Originally Posted by seasong

My goal was plausible and consistent computers. So I strove for compatibility with the Basic Set, but I had to throw out High-Tech and Ultra-Tech.

Do you consider Fast or Genius to be compatible with your system?

[seasong]

Originally Posted by lexington

Do you consider Fast or Genius to be compatible with your system?

Both are highly cinematic even in the books they were published in. For a supers, Mission: Impossible, space opera, or similar setting, however, there's no reason to skimp. They're not going to break the game balance - just my poor realism-loving heart ;-)

The Advanced option (Pyramid 3/37, p. 20) is a less cinematic version of those two options, and is about as good as you can realistically expect an advanced computer to be without time travel.

[Trachmyr]

Originally Posted by seasong

Both are highly cinematic even in the books they were published in. For a supers, Mission: Impossible, space opera, or similar setting, however, there's no reason to skimp. They're not going to break the game balance - just my poor realism-loving heart ;-)

The Advanced option (Pyramid 3/37, p. 20) is a less cinematic version of those two options, and is about as good as you can realistically expect an advanced computer to be without time travel.

Thanks for your response first off!

I think the thing that bugged me most was the AI rules interaction... using the formulas in UT, to get an IQ 20 super-AI (which isn't exactly super-human intellect, but at the border-line), it would require a TL12 Planetary Brain! Using the numbers you provided earlier, a IQ20 SAI could run on a desktop at TL12, which I find much more acceptable.

I tend to lean towards somewhat cinematic/hopeful future tech, so with that in mind I think I'll be doing the following... tell me if you see any major issues with the system you designed other than these modifications being very optimistic.

• Smartcell Computers: Complexity 2 at TL10.25.

• Mainframe: Complexity 4 at TL7.75 then Normal progression. Complexity 9 at TL11.

• Macroframe: Complexity 4 at TL7.5, Complexity 5 at TL8.25, Complexity 6 at TL9, Complexity 7 at TL9.5, Complexity 8 at TL10, Complexity 9 at TL10.5, Complexity 10 at TL11.

• Megacomputer: Complexity 7 at TL9, Complexity 8 at TL9.25, Complexity 9 at TL9.75, Complexity 10 at TL10.25, Complexity 11 at TL11.

• Skyscrapper Computer: Complexity 10 at TL9.5, Complexity 12 at TL11

• Living City: Complexity 12 at TL10.5, Complexity 13 at TL11

• Planetary Brain and larger are all +2 Complexity

• Cutting-Edge Option (TL8.25): Like Advanced, but the Computer is a full 0.25 TL better for all purposes, including Complexity. Cannot be combined with Advanced, Genius, Old or Obsolete. 20x Cost.

• Genius Option (TL9): Like Fast, but the Computer is a full 0.5TL better for all purposes. AT TL 11, this gives a flat +1 to complexity. 500x Cost. Cannot be combined with Advanced, Cutting-Edge, Old, Obsolete or Dedicated. 500x Cost.

• Causality Option (TL11^): Functions as a Quantum Computer but with +1 Complexity. 2000x Cost and 2x Weight. Cannot be combined with Quantum.

This gives +1 Complexity to large computers and +2 to huge ones by TL11... with Genius and Causality options, it allows for a maximum complexity of 20. Does that surpass Optimistic and go fully into superscience? (Other than the Causality Option which is Superscience)


And I have two questions for you as well, concerning the Blood Swarm option for Viral Computers. First, would they be able to use a dedicated Direct Neural Interface, and if so, at what additional cost? Second, would it be possible for SmartCell Computers to use this option, seeing how "Smartcells are the largest computer that can be placed in a nanobot", and if so, at what additional cost?

Edit: Ok, one more question... Does it seem feasible to run an AI on a dedicated computer? (as in the Dedicated option). My assumption is that it could only run one specific AI, but would be possible.

Thanks again for the wonderful article and your responses to my queries!

[lexington]

Originally Posted by Trachmyr

Thanks for your response first off!

I think the thing that bugged me most was the AI rules interaction... using the formulas in UT, to get an IQ 20 super-AI (which isn't exactly super-human intellect, but at the border-line), it would require a TL12 Planetary Brain! Using the numbers you provided earlier, a IQ20 SAI could run on a desktop at TL12, which I find much more acceptable.

Even leaving aside stat normalization issues that IQ20 AI is going to be vastly "smarter" than a human with IQ20 would be. A middle of the road C12 computer can in the back of its mind run 1 million games of Crysis at max settings (assuming its a game meant for TL9 computers) at once, that's what its dreams and random fleeting thoughts are like. C12 is enough to run 40 million iterations if it really wanted to and that ignores any increased capacity. Skill enhancing programs, language files, and databases are where the real apparent brilliance is going to come from.

A C12 computer with IQ20 will be utterly god-like to a human being. More like the main computer of the Enterprise than an awesome desktop.