November Entry: Dont Panic !

[Redrobes]

For now this is a bit of a place holder (bucking the trend of gaussian distributed winning entries last month ). I know what I would like do but its a bit audacious and I am not even sure how to go about doing it.

I have some pics on my HD from Zaon Studios and on their old web site used to be this huge poster sized galactic map. Well its the most awesome space map ever. Its truly drool inducing. It looks like it was done on some high end app like Maya or something too. So I have some pretty high standard base line example to ref. Their current web site has some stuff on it like some great planet renders - it gives you an idea of the standard going on there. Does anyone here have the poster or the books ?

So I would like to try to do a sector map of a galactic region plotting the star systems only (planets being too small for this one) and their 3D space relation to each other with star names and paths from them all.

To do that by hand would be well tedious as I want hundreds of stars all named in the cluster. So what I would like to do is make a script or program to generate the positions randomly and then adjust them for some basic physics and try to allocate them types like red giants or red dwarfs etc.

The first problem I thought I would have would be names for all of these. What I did was look up some name generators and found a few. Some are algorithmic and some others had lists of peoples names in various languages like Swahili and Celtic etc. These weren't bad but that idea plus remembering that our current stars are named with Arabic stuff led me to download an Arabic dictionary. I munged them about, knocked out the diacritics, removed spaces and dumped all words less than 5 chars long and then randomized it. The names in there are pretty good and look like the Aldebaran, Canopus, Arcturus etc things so I think thats sorted now.

My next dilemma is what app to use as I want to make this map big so that its possible to read all the names. I could use ViewingDale but this would be a 3D map of the stars. I can either do it in 3D with Blender or LWave or artificially flatten the view to 2D and go with ViewingDale. The other issue is whether to go with vector or raster for this one. Ideally I would go with Blender but I don't know the app well enough yet to get all the stuff the way I would want it. Right now I am leaning towards getting the star program script to write an image magick script which when I run that will create the star systems and all the link up data and then import to VDale for finishing. The other idea I might try is to get the star script to write out a giant .obj file for the stars and load that into blender and render that and take the result into VDale for finishing. That would be better but more risky I think.

The other issue is that this could be a bit of a time sink and I am supposed to be doing other stuff as well.

[Redrobes]

I have decided to use an image of M81 from APOD as my base and probably unnamed galaxy. This will be just an icon where the map is a sector within that galaxy showing all the stars in that sector.

I want to have a galactic plane with the stars shown above and below with the angle and "height" above the galactic plane. i.e. cylindrical coords. Now with this being the 75th century etc do you think we would have decimalized by that time and have 100 degrees in a circle ? And although we have AU's in our solar system would another galaxy still use light years and parsecs ? or something else. In fact I don't even know if this map is one to be read by humans or not. Maybe its an alien translation to human map or vice versa.

Now I just have to re-watch "live organ transplants" to remind myself what scale we will be approximately at...

[Redrobes]

Been doing a little experimenting with sectors. It turns out that if we go with 100 angular segments and the bit to be mapped is on the edge then the space is very nearly cubic. As you get towards the center your angles have to increase obviously but in the center I would expect that you would change coord system to spherical since a) thats the arrangement b) the gravity there would be so immense that it would be hard to travel there and c) there would be zillions of stars there so its more of a cartographers problem than one of navigation.

So the coord system will be 100 angular units per circle and I will map 1 of them. Radially our milky way is 50,000 light years. The outer band in the image below is at 1/10th of the radius so for milky way thats 5000 light years.

List of closest stars..
http://en.wikipedia.org/wiki/List_of_nearest_stars
... means that we should have stars at about 15 light years apart or so. That means 5000/15 is 330 or so which is clearly too many to map (330x330x330)

(BTW that page has a diagram just like what I want to make but doing just a sector instead of a whole circle.

So as anyone who has star charts know they list only the brightest stars not all of them. What we would need would be the most relevant. This would mean those with inhabited planets, systems with planets possible to live on, or ones having a solid surface etc. You also need those which are big, bright and those which present a danger. We also need to mark on there outposts and non natural satellites.

So in the list of top 20 closest stars only 5 of them are of magnitude 5 or below but were still waaay off mapability.

Here is a great link to a web site with nearest stars in a 5000ly redius - perfect....
http://www.atlasoftheuniverse.com/5000lys.html
... except that it looks like clouds.

So either we need to map a whole lot less than 1/100th of the circle or we have to be far out on the rim of the galaxy or the galaxy has to be less dense or less compact than the milky way.

Might have to fudge it a bit and pic something conveniently out on the rim to give just the right density of star systems with interesting enough properties to map.

This is always the problem with astro stuff. The numbers are always just too big. I wonder if Torstan is going to do a map... now that might get interesting.

[torstan]

Thinking about it I'm a particle rather than astro person so I'm actually almost as in the dark about this as the rest of you. However I do share an office with people working on the Sloan Digital Sky Survey - probably the largest space map ever made - so I might well tap them for ideas.

Just one note on your thoughts on units - light years are a sensible measure of distance as light is the only thing with a constant speed in the universe so measuring distances by the time that light takes to travel is a sensible choice. Obviously the concept of 'year' is the thing that would change.

[Redrobes]

I was reading somewhere today that there are 6 septasqillion or whatever stars and the size of the universe is such and such and that therefore one divided by the other means that you get an average of 70 stars per lightyear cubed which is obviously very high. Knowing also that we are in a galaxy and that the distance between galaxies is vast then that implies that the number of stars in the center of a galaxy must be just extreme - astronomical in fact. So I gather that the white ball around the center of galaxies is just packed with stars in a very small space.

You have mentioned before that in black holes the gravity is so great that your head and feet get ripped apart from gravity gradients. If there are that many suns packed close enough together then presumably that same effect is true - or lets be more real, the distance across a star ship capable of getting to the center of a galaxy must have a considerable gravity gradient across it.

My question is: where approximately on the swirling disk of a galaxy is the gradient so great that its unlikely that a tough starship hull could possibly withstand the gravity gradient ?

Thats what I was getting at with the change from polar to spherical coords near the middle because I doubt any ship could go there. Even if the tech was around to create an artificial gravity field I would imagine that would take energy to do it and the energy consumption would be more than that of the capability of the ship.

<snip>

[torstan]

I'm a UK person in the US so I'm just a fascinated observer. I'll let the actual Americans answer the US electoral questions

Interesting question about the gravity gradients. I'd be surprised if you get much in the way of serious tidal forces outside of the central black hole's immediate vicinity. There are a lot of stars, and so gravity will be much stronger - but in freefall - ie when you are out in space - its only the difference in gravity between two points that makes a difference rather than the absolute value of gravity at a point. It needs to be changing pretty fast for that to have a significant effect and I'd guess that only something as exotic as a black hole will have those sorts of gradients in its vicinity. I know that's awfully hand wavy. If I get a chance I'll have a look at getting you something more precise.

[RobA]

Shazam! Political OT posts moved to their own thread...

-Rob A>

[Redrobes]

Yes good call ! That diversion was getting a bit out of control. Back to the map !

[jfrazierjr]

Yes good call ! That diversion was getting a bit out of control. Back to the map !

Sorry for the thread derail. I was trying to figure out how to split the post up at the same time Rob completed it.

Anyway, I look forward to seeing how this works out. Did you say you were going to us Blender for this? Or is your knowledge level just not up to snuff there yet?

[Redrobes]

Did you say you were going to us Blender for this? Or is your knowledge level just not up to snuff there yet?

Actually I dont know at this stage. My thread is titled Dont Panic after Hitchhikers obviously but I have to say I am starting to panic !

I did some tests on blender and I can write out an obj file in perl but it doesnt handle text very well. ViewingDale handles text easily but its 2D. I also looked at POGL which is a perl based OpenGL interface and thats easy to use but... actually ill spare you details... but it looks like its not going to do more than screen sized unless I can figure some more stuff out on it - but that would be a good way of doing it. Then I thought that maybe perl writing image magick instructions would be good like the thumbs script is doing. That could draw all of these stars... I dunno still. Right now I am leaning to Perl generating a big ViewingDale file but with some extra stuff done in blender which I will composite into the scene. What would be ideal is if I can render stuff in blender and also have the transform for its camera to 2D so that I can do stuff in 2D at the same time and the coords would line up. Thats not easy tho.