Celestia Forums

I have put together an asteroid add-on pack containing simulation data for 40 (real) asteroids. About half of them are displayed using the models the University of Helsinki generated from lightcurve data. The rest are just ellipsoids. Some of them have satellites too. Of course, all the asteroids are tinted with colors from the SMASS spectroscopic survey. I'm quite pleased with how it turned out. Pictures:


Download it from the motherlode or here:
http://www.wronkiewicz.net/moreasteroids.zip
Please let me know if you try it out and run into problems. The pictures were taken on Mac OS, and I noticed that the colors were washed out on the Windows version. I don't know if it's because of a bug in Celestia or if I did something wrong with the cmod files.

Matt

Tx Matt!

I don't think your color problem comes from the cmod conversion, but perhaps it's related to this thread...

http://shatters.net/forum/viewtopic.php?t=10032

Matt, your spheroids are a bit crude, they could gain a few poly to eliminate the hard edges...

Some of the cmods need a remapping because in certain case the pole artifact could be less visible if placed elsewhere. Also a few ones need a bit more smoothing to eliminate a few poly artifacts...

In all case, Tx again for the nice work!

Argh, another question: in your pack there is a asteroidcolors.ssc; aren't they the old values?

Thank you very much for the feedback.

in your pack there is a asteroidcolors.ssc; aren't they the old values?

I am certain that they are the most recent colors. In the old compilation, some asteroids had blue levels higher than the red values. In the new compilation, all the asteroids are redder than they are blue.

your spheroids are a bit crude, they could gain a few poly to eliminate the hard edges...

The polygon count of the models is reduced to 800 from the original 1600 to improve rendering performance. I didn't find the differences to be very visible though. Also, some of the asteroids are very angular. Are you sure you're seeing edges? Are you seeing them in the texture mapping, the outline of the shape, or the shading? Any asteroids in particular? Also if you could take a look at the original renderings here:
http://www.rni.helsinki.fi/~mjk/thirty.pdf
and here:
http://www.rni.helsinki.fi/~mjk/IcarPIII.pdf
it might help to see the original generated shapes.

Some of the cmods need a remapping because in certain case the pole artifact could be less visible if placed elsewhere.

I agree. I wrote a custom uv mapper to paste the texture to the asteroid shape. It handles the seam along 0 longitude fine, but it's still having trouble with the poles. I'll see if I can figure out a way to make it work. Maybe it would help if I added an extra vertex at each pole.

I don't think your color problem comes from the cmod conversion, but perhaps it's related to this thread...

I do expect the colors to be a little darker on the Mac because of the different gamma setting. I don't think that is the problem I am seeing though. Actually, the Mac does have the same problem, but it is only visible when the ambient light is turned up. Here are some pictures of the problem:



Notice how the side of the rock lit by the sun changes color as the ambient light level goes up. Windows has this problem even when the ambient light level is 0. I am thinking it's either a problem with the material settings, which are absent in my cmod files, or it's a bug Celestia exhibits when dealing with color blending.

Matt

wronkiew wrote:The polygon count of the models is reduced to 800 from the original 1600 to improve rendering performance.

Oops. That should be vertex count, not polygon count.

Matt

wronkiew wrote:

your spheroids are a bit crude, they could gain a few poly to eliminate the hard edges...

The polygon count of the models is reduced to 800 from the original 1600 to improve rendering performance. I didn't find the differences to be very visible though. Also, some of the asteroids are very angular. Are you sure you're seeing edges? Are you seeing them in the texture mapping, the outline of the shape, or the shading? Any asteroids in particular? Also if you could take a look at the original renderings here:
http://www.rni.helsinki.fi/~mjk/thirty.pdf
and here:
http://www.rni.helsinki.fi/~mjk/IcarPIII.pdf
it might help to see the original generated shapes.

Oups, I'm not talking here about your models but about the ellipsoids.cms
which are a bit crude on the outline shape, unfortunatly as it's a displacement of a default sphere, I suppose there is nothing to do...

Now a few cmods need a few more degrees of smoothing to eliminate some poly artifact...

On those papers, they talk about almost 100 bodies; do you plan to add more to your catalogue?

ElChristou wrote:On those papers, they talk about almost 100 bodies; do you plan to add more to your catalogue?

They only posted model files for 20 or so of them, so I could add more if I can get them to give me more files, or if I can figure out their shape generating algorithms and create model files on my own. I have some ideas for other Celestia projects that I want to tackle before adding more asteroids. Of course, if someone else wants to create asteroid models, I have a handy tool for CMODifying them.

Matt

Many thanks to ElChristou for his feedback on my asteroid expansion pack. I have updated the add-on with fixes for the model files and some documentation improvements. Get it here:
http://www.wronkiewicz.net/moreasteroids.zip

Matt

You are welcome Matt! Tx again for the nice work!

wow another reallly cool picture where do you guys get these great graphic renderings?