Re: [nbos] [AS] Suggestions for AS 2.1"David Loman" Wed Oct 3rd, 2007
As a programmer, I can comment on a few of your statements Mike. All orbits
are simple vectors undergoing a continual change. Computers are really good
nowadays at calculating, multiplying, adding, etc vectors. I have written
simple scripts (aka not compiled code) that can push several million vector
calcs per second. Compiled code would easily double that. Thus, the
resultant vector of any given celestial body will be the result of the
'gravity' vectors of all the surrounding bodies. Ultimately will this be
accurrate enough to be a "universe simulator"... probably not, but for what
Sam is describing, definetly!
On the visual side of the house, DirectX, OpenGL and all the graphics cards
on the market can handle amazing amounts of visual data... However, the
average graphics application programmer takes lots of short cuts in order to
get their product out the door quicker and this results in a less than
Extrapolating this over thousands of stellar systems SHOULD make your brain
shutdown! On that exact note, it would be futile for an application to
attempt to display all that data on the screen at one time. each planet
would be soooooooooooo tiny that it would measure less that a pixel on your
screen... so whats the point? Now, all the celestial body positioning
calculations would still be happening in the background, so should you
'zoom' in on a system that you wish to view, all the bodies would be in
their right place. And again, if you 'zoom' in on a body with moons, they
too should all be in the right place.
There is even room for pre-emptive calculations, aka, where will everything
be in, 1 year, 10 years, 100 years, etc.
Tons of possibilities and AS2 has the stage set for some pretty cool
On 10/3/07, Mike Oliver <mike-oliver-at-blueyonder.co.uk> wrote:
> I think you may be expecting a bit much of the program. Its main purpose
> is to provide users with a means of generating a galactic sector and its
> entire contents. I believe the graphic displays are intended as a means of
> showing players the general layout of a star system and the planetary
> systems within it. I view of the distances involved, I think trying to show
> the star system together with all the planetary and satellite bodies within
> it would prodice such a visual muddle that it would defeat the object.
> I can't imagine the complexity of a computer programme that attempts to
> replicate, visually, all the orbital perturbations of a complete star
> system. Then if I try to extrapolate that to cover the thousands of stellar
> systems within a galactic sector, my brain shuts down.
> I may have misunderstood you or Ed my chime in and tell us he's planning
> to incorporate it all into V3 of AS.
> It's been a long time since I tried out Celestia and I'm not familiar with
> ChView2 but I have run several SF campaigns with AS2 as my only tool and I
> haven't had any problems with convincing my players that what they see is a
> reasonable representation of the section of space they are in. Generally,
> they are impressed with what AS2 shows them.
> However, having said all that, some of the things you are suggesting would
> be nice……
> -----Original Message-----
> *From:* nbossoftware-bounces-at-nbos.com [mailto:
> nbossoftware-bounces-at-nbos.com] *On Behalf Of *Sam Orton
> *Sent:* 02 October 2007 17:11
> *To:* nbossoftware-at-nbos.com
> *Subject:* Re: [nbos] [AS] Suggestions for AS 2.1
> Hmmm.... I think I may be taken as asking for more than I actually am. If
> you go to a star, you can pull up a system display. It shows planetary
> orbits, but not all the moons. To get satellite orbits you can go to any
> parent body (such as a gas giant) and pull it up there, and it shows planet
> and satellites, but not the parent star or other planets. There's probably a
> basic underlying assumption (a valid one I think) that for example, the
> current location of Mars isn't going to make a measurable difference in
> conditions on Callisto. The effects of the immediate parent body, Jupiter,
> will override all other factors.
> Okay, so if you click on a multiple star system, allow it to bring up a
> single level system display, like those at planetary and satellite level.
> True, depending on the spread of the stellar system, a GM may want to take
> into account that the orbits of outer planets of a binary may be captured or
> slung by the other star. But a picky GM will be taking that into account
> anyway, and one who isn't picky is unlikely to be bothered by the logical
> inconsistencies. The only tough part I see about having such a display is
> having it orbit around system barycenter rather than a given star. If one or
> more of the children is a multiple itself, display the orbit of that
> subsystem's barycenter. Taking it to 2 levels so you can show contents of a
> child multiple star system would be good, but that's also probably a pretty
> fundamental change to how the system display works, too.
> I dunno, maybe I'm just showing how ignorant I am of modern programming
> principles. It just seems to me that the system display code is already
> there, you just need a means for multiple star systems to call on it in the
> same way single star systems do. Assuming you can find a way to tag system
> barycenter as center instead of an object, the rest of the commands to call
> up a system display should be the same throughout. The only place I see it
> getting really weird is in close multiples, where objects are so deep in
> each others' gravity wells that tidal forces are trying to shred everything
> in the system.
> I'm trying to see this from all sides. Yes, gamers and GMs can be like
> spoiled children, wanting everything. To be fair, they're creating a
> universe, in order for it to be complete they *have to* have EVERYTHING. And
> yes, programmers have no problem trying to give it to them, provided that
> the nights they stay up figuring out this cool new feature actually
> translates into money for the bills at some point. Can't say as I blame them
> a bit. And yes, I fully sympathize with "Oh I could write it, sure.... but
> your machine couldn't run it."
> I'm just thinking that the less likely a GM is to try to run AS *and*
> Celestia AND ChView2 all at once, because AS takes care of all of it, the
> better AS will sell.
> "I do not feel obliged to believe that the same God who endowed us with
> sense, reason and intellect intended us to forego their use." - Galileo
> Be a better Globetrotter. Get better travel answers
> someone who knows.
> Yahoo! Answers - Check it out.
> Nbossoftware mailing list
Nbossoftware mailing list