World Generation in 2300AD

[EvilDrGanymede]

Originally posted by Colin:
Honestly, I don't know yet, I have barely looked at that section. I suspect that it will still continue to generate large numbers of lifeless worlds. There may end up being an option to generate a life-bearing world, should the Director so decide.

Thanks,
Colin

The worlds in 2300AD always seemed more... interesting to me. And more varied than those in CT - they didn't all fit into some nice simple classification scheme. And it certainly seemed more realistic than CT too. I liked the fact that the gases in the atmosphere were determined by the molecular mass that the world could hold on to.

 

[aramis]

Originally posted by Colin:
IIRC, 2300's world generation system tended to generate relatively few habitable worlds, despite the large number of such in human space. T20 (and Traveller in general) has more of a bias towards generating habitable worlds.

2300's system avoids the "EVERY SYSTEM HAS A HABITABLE WORLD" that CT and T20 both seem to share. Fine for the 3I setting, with it's history of ancient manipulation...

All the systems in the human sphere that I've checked have checked with the world generation. That is, they are possible within the generation scheme to result in them. That the systems were hand-tweaked by the authors, I've no doubt. If you have a sufficiently massive world in the right orbit, life occurs in 2300... those two are the keys, tho...

Perhaps System and World gen should be outside the core for 2320? After all, stats were published for almost all mainworlds, plus several systems in total; use the descriptors, and put generation in a TA?

After all, either 2320 players will be using the setting for the setting... since the target audience is a D20/T20 baseline. (With a few crossovers... of course.) If they are buying 2320, they can fairly safely be assumed to not be buying it for purely mechanical reasons (it's a T20 supplement).

I know that If I buy it, it will be solely for the setting (I prefer 2300 to T20 mechanically), and don't need world generation to play in an established setting with all worlds aleady defined.

I also know that some will cry "Heresy!" or somesuch...

 

[EvilDrGanymede]

Originally posted by Aramis:
I know that If I buy it, it will be solely for the setting (I prefer 2300 to T20 mechanically), and don't need world generation to play in an established setting with all worlds aleady defined.

I also know that some will cry "Heresy!" or somesuch... [/QB]

Well, not 'heresy!', but I always considered the 2300AD world generation system as a very nice alternative option for generating worlds in one's own sci-fi backgrounds. And you could also obviously use it to make systems if PCs decided to go exploring beyond the existing explored Arms too.

 

[Colin]

I like the 2300 world generation ssystem as well. But it's kinda slow. Well, really slow. I don't think the system in 2320 will have that level of detail, though it is possible,. I have to see how I'm doing for space. There, will, however, be some sort of world-generation system.

 

[EvilDrGanymede]

Originally posted by Colin:
I like the 2300 world generation ssystem as well. But it's kinda slow. Well, really slow. I don't think the system in 2320 will have that level of detail, though it is possible,. I have to see how I'm doing for space. There, will, however, be some sort of world-generation system.

Sure, it might be slow, but when are people in a rush to make planets anyway? Yes, you could say 'oh, I need an NPC real fast so I need a rapid character generator', but planets aren't usually generated on the fly.

 

[Colin]

It also, I'm afraid, takes up a great deal of space, with all those charts and tables. Agreed, I'm of the mind that a ref should generate palnets before a game, but sometimes things happen, the players go off "thataway" and there you are...

We'll see hor it pans out. Like I said, though, space is an issue. All those colonies and nation descriptions come at a price.

 

[EvilDrGanymede]

Originally posted by Colin:
It also, I'm afraid, takes up a great deal of space, with all those charts and tables. Agreed, I'm of the mind that a ref should generate palnets before a game, but sometimes things happen, the players go off "thataway" and there you are...

The MMW retained table (that IIRC took up a whole page) can be reduced to a single small equation. Ditto with the surface temperature. The surface pressure table is also unecessarily detailed - it's not dependent solely on the gravity/mass of the planet. Going the Traveller route and saying certain types of atmosphere have a certain pressure range is probably more sensible IMO.

I'm sure it can be cut down considerably just by more judicious use of some equations and some reformatting. (I'd check, but I don't have my 2300AD Directors Guide here at all )

 

[aramis]

Originally posted by Colin:
It also, I'm afraid, takes up a great deal of space, with all those charts and tables. Agreed, I'm of the mind that a ref should generate palnets before a game, but sometimes things happen, the players go off "thataway" and there you are...

We'll see hor it pans out. Like I said, though, space is an issue. All those colonies and nation descriptions come at a price.

if you're going to nut it anyway, why not go ahead and simply use the T20 gen? (Saves the space, eliminates the duplications...) the NSL will provide system locations and Stellar Types.

The problem is one of

Logically, Either you are
1) using T20's, and maintaining compatibility, but eviscerating the nature of systems in 2300 in favor of the CT/MT/T20 model...
2) using a published star list (so SysGen is a moot point)
3) Using 2300's, and maintaining compatibility with extant canon, but not with T20/CT/MT/TNE/T4)
4) doing a hybrid which takes up space and is not interchangeable for either one.
-Or-
5) trying to update to new astronomical data and invalidating both 2300 canonical worlds/systems and T20 mechanics.

And yes, simplification to formulae is a nice idea for saving space, but T20 did just the opposite...

 

[Pompe]

Regarding molecular masses and atmospheres:

I don't think that 2300AD's charts are a very good thing, not would an equation made from reducing them be better. Partially this is because the equations of MMW don't really fit with observations, partially this is because the atmosphere charts in general strike me as one of the flawed points of 2300AD which needs some fixing to allow for things like Venus and Titan.

Regarding equations instead of charts in order to save space:

I agree this is a quite valid solution.

However, if a problem of 2300AD's worldgen is that it is slow and already a bit too serious - something I don't necessary agree with - the question is if not equations could make it even slower to unprepared users as well as scare off people. Then again, that math isn't too bad.

2300AD's charts could be considerably scaled down, though, as a lot of the results are impossible, unlikely to be practical or silly. For worlds people actually land on only parts of the top third of the charts are likely to be used, as an example.

 

[Colin]

Nut it..? A colloquial expression that I'm not familiar with, I guess, but it doesn't sound good.

Here's my design goal for worldgen in 2320: I want to preserve the flavor of the 2300 worlds, first and foremost, and of the worldgen itself, if possible. At the end of the process, I want to be able to enter the wolrd information intot he T20 UWP as a shorthand for the world, which in no way replaces a detailed description, but provides useful information in chart form. This is especially useful for trade-based games, where Adventurers and Directors can use this information with the T20 trade charts.

Space is a consideration here, but I don't just want T20 worlds. T20 is biased in favor of generating habitable worlds, while 2300 was biased the other way. I do want to preserve that bias, and generate worlds compatible with 2300.

There's nothing in the T20 UWP that renders it unsuitable for 2300. But it is a shorthand, and for any detailed world there would also be a more in-depth description.

 

[Pompe]

Having a slightly dull afternoon, I decided to check out how common habitable worlds actually are in 2300AD's universe, i.e, how biased it is. So I went over the Chinese Arm.

As far as I can see, there are habitable worlds in 1/3 of the star systems on the Arm. It does thus seem as the setting is biased towards non-habitable worlds. But...

1: a fair number of these systems are off-the-main-roads, so to say, and the AG mentions specifically that "there are plenty of systems that have yet to be explored and colonized" in the Chinese Arm. So the 1/3 is a low number.

2: most of the fifty systems are red dwarf stars, and while habitable worlds are found around these too in the setting, the majority of the star systems with habitable worlds are F, G or K-type stars. Of these three star types, all of them on the Chinese Arm does have habitable worlds except for two, and those two systems are off the main roads and might well have habitable worlds - see #1 above.

Thus, my conclusion would be that 2300AD is heavily biased towards habitable worlds as long as the star is remotely sun-like. If the main star is a F, G or K-type main sequence or subgiant star, it is extremely likely there is a habitable world present. It might even be more common than in T20.

For an M-type red dwarf star it is considerably less likely (say 1/10 chance) but not impossible or unusual given the sheer number of them around.

 

[aramis]

Originally posted by Colin:
Nut it..? A colloquial expression that I'm not familiar with, I guess, but it doesn't sound good.

"Nut it" = Castrate it. as in, remove its testes, colloquially referred to as nuts.

Here's my design goal for worldgen in 2320: I want to preserve the flavor of the 2300 worlds, first and foremost, and of the worldgen itself, if possible. At the end of the process, I want to be able to enter the wolrd information intot he T20 UWP as a shorthand for the world, which in no way replaces a detailed description, but provides useful information in chart form. This is especially useful for trade-based games, where Adventurers and Directors can use this information with the T20 trade charts.

Space is a consideration here, but I don't just want T20 worlds. T20 is biased in favor of generating habitable worlds, while 2300 was biased the other way. I do want to preserve that bias, and generate worlds compatible with 2300.

There's nothing in the T20 UWP that renders it unsuitable for 2300. But it is a shorthand, and for any detailed world there would also be a more in-depth description.

I don't care about the UWP's. Either way. UWPs and output formatting are irrelevant to the underlying mechanical issue.

What I'm asking, essentially, is "What changes to the way it works" make you feel it important to add a third system generation mechanic to the mix? What supposed gains will it provide. (UWP doesn't matter. it's not a mechanic issue itself.)

Are you going to mess with the underlying mathematics or processes of 2300 System generation? (Formula extraction from tabular data is not a mechanical change.)

Are you going to tweak the stellar data tables to fit modern data? Are you going to recconcile the new planetary observation data in to the system?


OT for llustrative purposes
It is like another key 2300 issue, not yet brought up... 2300 had a very different model for animal encounter tables (the actual critter design being irrelevant to the issue, as that should be t20 in any case). Are you goign to use T20's layout, 2300's layout, or something more hybridized, or some third style?

The 2300 encounter table, that d10 pyraid, would be a great thing to bring across to T20. It, too was a big "Look and Feel" difference from CT/MT/T20... and was a more realistic approach. (even if oversimplified.) No more and no less playable than Ct's 2d6 tables. But very different results for encounter odds and types from the "Classic" CT tables.

/OT

Really, the animal encounter issue is a trivial one to me, as I've 2300 and can make 2300 style tables with T20 mechanics for creature design. World Gen and System gen are VERY different, tho.

In CT, the size loosly affected atmosphere type/density, but type and density were linked tightly. T20 hasn't changed that.

In 2300, you had decisions to make... how much of a given gass was retained? How much was present to begin with?
(I do fault GDW staff for over-populating the 2300 universe with humans....)

 

[Colin]

Ahh, 'nut it' isn't good then?

To answer this fairly, I would have to go over the two design systems fairly carefully. I haven't ahd the opportunity to do so yet. Ideally, what I want to do is preserve the results of the 2300 world and system design. If I can do that from within the T20 system, then I will do so. If I have to use the 2300 system whole hog, I will do so. Whatever hgts me the best results, with reinveting the wheel. T20 provides a good, comprehensive system for world design. It doesn't quite fit the 2300 model, though. If a can make it fit with a little tweaking, that's what I'll do. If it would involve more than a little tweaking, though, I'll adapt the 2300 system. Final output will be raw data and the UWP, either way.

As for animal encounters, I really like the encounter pyramid, and will use it if possible. Animal design will use the T20 rules, perhaps with a few addtions.

As for changing the design process to take into account what we've learned in the last dozen or so years, well, if I can without invalidating anything that's come before (canon) then I'll look at it seriously. But the results have to be consistent with the 2300 AD universe as it is. I'm not going to change anything published to fit new data or new models of planetary formation.

That's it for now. As I've said, planet and system design isn't currently my priority. When it is, I'll have better answers to these questions.

Thanks.

Colin Dunn

 

[aramis]

Ah. THat does answer th question more fully. Thank-you Colin.

Go threateneing major changes and some people go nuts. Others geet defensive. I ask "Why?"

Integration of new real world astronomy is a hot issue, too. One which begs the question "Would it in fact be better to reinvent the wheel?" for both system generation and for the 2300 universe.

But that is a whole other issue.

 

[Colin]

Originally posted by Aramis:
<snip>

Integration of new real world astronomy is a hot issue, too. One which begs the question "Would it in fact be better to reinvent the wheel?" for both system generation and for the 2300 universe.

But that is a whole other issue. [/QB]

And one I'm not going to touch.

Thanks for your input. I hope I've answered your questions, at least for now. I'll have more information on what I'm doing with world generation once I get to that section of the book.

Colin

 

[Colin]

Just started working on the world generation section for the 2320 rules. I think I'm going to stick closely to the original 2300 rules, while providing a method to convert those results into the standard T20 UWP. 2300AD is a little table-happy, so I will look at reducing that if possible. But I'm just starting that section now.

<<typos corrected>>

 

[EvilDrGanymede]

Originally posted by Colin:
Just started working on the world generation section fo the 2320 rules. I think I'm going to stick closely tot he original 2300 rules, while providing a method to convert those results into the standard T20 UWP. 2300AD is a little table-happy, so I will look at reducing that if possible. But I'm just starting that section now.

One thing that bugged me about 2300AD, if I remember it correctly, was that it had a huge full page table that basically said that atmospheric pressure was pretty much exactly the same as surface gravity. While the two are related, that kind of direct 1:1 link is nonsensical - it depends more on the mass of gas in the atmosphere, the density of the gas and the gravity. So if you're gonna get rid of any tables, I suggest that's the first one you should be looking to remove

 

[mandelkubb]

Originally posted by Colin:
2300AD is a little table-happy, so I will look at reducing that if possible.

The "Life Zone" table could easily be replaced by the formula which is already provided.

The orbit calculation may be in need of some work. The cold stars (M-class) will have a hard time having any planets within their life zones. maybe it should be possible to have orbits closer to the star than 0.1 AU, perhaps dependent on the mass of the stars (high-mass stars have high mass because they have absorbed all matter close to themselves).

I like the hard-science part of world generation in 2300, and I'm happy to see that you are working on a "modernized" version of the game.

 

[Tanuki]

Originally posted by mandelkubb:
The orbit calculation may be in need of some work. The cold stars (M-class) will have a hard time having any planets within their life zones. maybe it should be possible to have orbits closer to the star than 0.1 AU, perhaps dependent on the mass of the stars (high-mass stars have high mass because they have absorbed all matter close to themselves).

Well, the non-captured parts of the moon systems of Jupiter, Saturn, etc., look like solar systems in miniature, and ALL of those moons are within .1 AU of their primaries.

The only limitation on planet placement should be the Roche limit, and/or the distance within which stellar insolation would vaporize a planet. Given that, once you've placed the innermost planet the rest should more or less follow Bode's rule until you get to the outer system where things can get a little sloppier.

Mass of the primary should play a role in how far out that outermost stuff is likely to be. For innermost, add another column to the lifezone tables showing the Roche/vaporized limit (whichever is greater) and allow the planetary system to begin at any point beyond.

I need to stick my copy of 2300AD in my backpack and bring it to work. I'll start going over the system generation pages and give you my recommendations.

 

[aramis]

Originally posted by Tanuki:
</font><blockquote>quote:</font><hr />Originally posted by mandelkubb:
The orbit calculation may be in need of some work. The cold stars (M-class) will have a hard time having any planets within their life zones. maybe it should be possible to have orbits closer to the star than 0.1 AU, perhaps dependent on the mass of the stars (high-mass stars have high mass because they have absorbed all matter close to themselves).

Well, the non-captured parts of the moon systems of Jupiter, Saturn, etc., look like solar systems in miniature, and ALL of those moons are within .1 AU of their primaries.

The only limitation on planet placement should be the Roche limit, and/or the distance within which stellar insolation would vaporize a planet. Given that, once you've placed the innermost planet the rest should more or less follow Bode's rule until you get to the outer system where things can get a little sloppier.

</font>[/QUOTE]Actually, there should be one other concern: Game Playability.

Seriously, the Titus-Bode relationship is easy enough to use, but the problem becomes "In what units do you generate A and B? (A and B are orbital distances for 1st and second body...)

If you have different scales for different sized stars, where do you place those breaks?

How granular should the orbital data be for ballancing reality and play ballance? 0.1 AU? 0.05 AU? 0.01 AU?

for some external comparison: Starfire used a roll 1d10 2 times, rerolling if they are within two of each other, and read as LM, to generate slar systems. (1 lm ~ 0.1 AU) It used imcrements of 0.5 LS for "tactically significant moons".