PDA

View Full Version : Best automatic Galaxy/Planet/Stars plugin for 64 bit Lightwave?



AmigaNewTek
01-22-2011, 02:50 PM
Does anyone know any info about a plugin that generate Galaxy, Planets, Stars?

Thank you

nickdigital
01-22-2011, 03:00 PM
Liberty3D has training material for planets and space nebulas.
Not sure there's anything to automatically generate these type of models/sets.

For stars there's StarPro.
http://www.maasdigital.com/starpro/

Dexter2999
01-22-2011, 09:23 PM
I remember at one point Kat highly recommended starpro. ( I believe)

Shiny_Mike
01-22-2011, 09:55 PM
This is cool (and free) but only for 32bit modeler
http://www.lwplugindb.com/Plugin.aspx?id=a9cc383d

UltraViolet
01-22-2011, 11:43 PM
There is "Stars" plugin for Win32 ...

http://www.foundation3d.com/index.php?categoryid=38&p13_sectionid=82

Apparently nothing happened even though guy who developed it (or ported it, dunno exactly) said on Foudation3D forums he is gonna make 64 bit version, it would be fun to try it :D

jrandom
01-23-2011, 08:56 AM
I highly recommend StarPro (http://www.maasdigital.com/starpro/). Works for both 32 and 64 bit versions, and they make available a huge star database as well.

Here's a quick one-off (http://rainybrain.org/Images/Lightwave/StarPro_Example.jpg) I made using the default "small" star database and a basic bloom filter.

There are lots of options for controlling star brightness and also has a 3D-Stars mode where 1m == 1 light year so you can fly through the stars as well.

JohnMarchant
01-24-2011, 07:11 AM
Agreed StarPro is good, if a bit expensive. Points converted to polys with a good glow and bloom on them is as good if you dont knew and accurate representation of our galaxy.

As for whole planets and galaxies, ive seen a few models of them, try Foundation3D and some other places but there is no real plugin as such for them.

jrandom
01-24-2011, 10:39 AM
Points converted to polys with a good glow and bloom on them is as good if you dont knew and accurate representation of our galaxy.

What I always found tricky is that a random set of points doesn't actually look like real star arrangements -- stars form in clusters and groups which isn't accurately mimicked with a random starfield. This is very noticeable if you compare random starfields with actual telescope photos of our galactic neighborhood.

Actually, this brings up a good question: How would we generate a realistic-looking starfield that isn't just a set of random points?

Gumby22don
01-24-2011, 11:38 PM
overlay-ing sets of stars, possibly from real sample clusters seems to work well, and you can scale them up ala perlin noise to produce a larger galaxy maybe?

Don
have a great day

JohnMarchant
01-25-2011, 12:15 AM
What I always found tricky is that a random set of points doesn't actually look like real star arrangements -- stars form in clusters and groups which isn't accurately mimicked with a random starfield. This is very noticeable if you compare random starfields with actual telescope photos of our galactic neighborhood.

Actually, this brings up a good question: How would we generate a realistic-looking starfield that isn't just a set of random points?.

I dont imagine it could be that hard to write a script to do this, maybe have an option for clumping by percentage or something like that, sorry im not a script writer but there are a few around.

jrandom
01-25-2011, 12:26 AM
.

I dont imagine it could be that hard to write a script to do this, maybe have an option for clumping by percentage or something like that, sorry im not a script writer but there are a few around.

I'm a programmer by trade, just need to brush up on LScript (or just buckle down and learn LW's C++ API). I've certainly tried my hand at 2D starfields in the past, and while it's possible to get "clumps", I've yet to hit upon any kind of algorithm that distributes points in a star-like arrangement. It's a trickier problem than it looks on the surface.

I might have to dust this problem off and give it another go. It's been at least ten years since I last took a stab at it.

JohnMarchant
01-25-2011, 12:31 AM
I'm a programmer by trade, just need to brush up on LScript (or just buckle down and learn LW's C++ API). I've certainly tried my hand at 2D starfields in the past, and while it's possible to get "clumps", I've yet to hit upon any kind of algorithm that distributes points in a star-like arrangement. It's a trickier problem than it looks on the surface.

I might have to dust this problem off and give it another go. It's been at least ten years since I last took a stab at it.

Well we are half way there already there are starfield scripts around i believe and also random points scripts, find a non compiled LS and see what makes it tick, then only need to find out how to add clumping of the points.

If you do not know LScript at all and are a hardcore member then maybe it will be more of an advantage to learn C++ for Core then LScript which will eventually die, but not for a few years yet.

Rayek
01-25-2011, 01:18 AM
This article is informative as well - if you use Photoshop for 2d starfields, that is.
http://gallery.artofgregmartin.com/tuts_arts/making_a_star_field.html

I included an action that will guide anyone through the steps (not mine). Still needs some manual 'artistry', nonetheless.

oliversimonnet
01-25-2011, 05:46 AM
This article is informative as well - if you use Photoshop for 2d starfields, that is.
http://gallery.artofgregmartin.com/tuts_arts/making_a_star_field.html

I included an action that will guide anyone through the steps (not mine). Still needs some manual 'artistry', nonetheless.

Just thought id say: Thanks for that link :)

safetyman
01-25-2011, 06:00 AM
Just use the "God" plugin. Heh heh. :devil:

Sorry about that.

I have a plugin on my 'puter called "imitate galaxy", but I don't think I've ever used it. It's from someone named Takahiko Date, back in 2001. The text file is all jibberish, but I did find a web link in there that might be useful...

http://www.vfxengine.com/dtt/

DISCLAIMER: If the above website is questionable, invalid, etc. don't blame me. I can't check it right now, 'cause I'm at work. :D

JohnMarchant
01-25-2011, 06:27 AM
This article is informative as well - if you use Photoshop for 2d starfields, that is.
http://gallery.artofgregmartin.com/tuts_arts/making_a_star_field.html

I included an action that will guide anyone through the steps (not mine). Still needs some manual 'artistry', nonetheless.

Nice indeed

Here is a link to imitate galaxy


http://www.lwplugindb.com/Plugin.aspx?id=a9cc383d

prometheus
01-25-2011, 08:46 AM
What I always found tricky is that a random set of points doesn't actually look like real star arrangements -- stars form in clusters and groups which isn't accurately mimicked with a random starfield. This is very noticeable if you compare random starfields with actual telescope photos of our galactic neighborhood.

Actually, this brings up a good question: How would we generate a realistic-looking starfield that isn't just a set of random points?

You could try and use a partigon emitter, and use an image map or procedural textures on the birth rate, that could give some clusters if set up right, use the save transformed object command to save out the particles to points.
You should probably add the particles to a sphere firstly thou.

Michael

AmigaNewTek
03-03-2011, 08:00 AM
Thank you for all the replies.

@safetyman
"I have a plugin on my 'puter called "imitate galaxy", but I don't think I've ever used it. It's from someone named Takahiko Date, back in 2001."
can you create a 64 bit version?

StartPro is great for stars

I'm working on alternative solutions. I will post pictures as soon as i will get decent results.

prometheus
03-03-2011, 08:23 AM
star clusters should be easy to either pre build by creating simple points or spray a couple of clusters, or simply use and place several smaller clusters of particle emitters in a tight cluster within the huge particle emitter cluster.

Add some tiny hypervoxels in sprite mode..
Make sure to have some depth to the huge particle field to get some variance on luminosity, or use procedural on the luminosity channel.

use the convert to partigons tool if you would like to use glow or motion blur.

Michael

jeric_synergy
04-26-2011, 11:02 AM
I think hypervoxels would be good, especially to add some variation to the 'stars' sizes and magnitude. Simple points, tmk, don't give you enough control.

Although, with a nodal approach, who knows what's possible?? (not me.) It may be possible to somehow use nodes to affect semi-random variations on apparent size of individual points.