Celestia Forums

I had from Martin (Cham) the possibility to test his very interesting Sgr A* (galactic core) addon, whose thread with images is here:
http://www.shatters.net/forum/viewtopic.php?f=2&t=13213&st=0&sk=t&sd=a&start=15

But I found two strange problems, i.e.:

1- when I'm in the "Galactic center" (I can see the stars around the black hole, still invisible) and try to go to the "Sgr A*:Central Black Hole", when I reach about 1.01 ly distance from star system barycentre, the system crashes;
2- even when I’m sufficiently far away, seeing only the stars, I cannot show their orbits (obviously with ON orbits, both with Celestia "View options" window and the LUA EDU TOOLS 1.5, as you can see in the image.
And this happens at every renderpath setting.

I checked with the various renderpath settings, basic, vertex, OGL2, with same crashing result.
Then I completely removed the LUA EDU TOOLS, with same result, and finally added the latest LUA EDU TOOLS 1.2beta2.
No variations.
FYI, I have the latest 180.48 whql Nvidia drivers and the latest Cartrite’s 4556 SVN installed.
Any idea on the reasons of this?
Is this a real bug or only a problem on my PC?
Thank you for your help.
Bye

Andrea

I'll need to try the add-on myself in order to investigate... If I can reproduce the crash on my own computer, I can almost certainly fix the problem very quickly.

--Chris

I'm expecting that this bug is related to the fact that I placed the black hole model around a barycenter, while all the stars are moving around the barycenter. Personally, I'm using a SVN built made a month ago and don't experience the bug. However, I was experiencing a crashing bug when no stars were present, while only the black hole was there : Clicking the black hole sphere crashed Celestia. Placing the stars solved the issue on my machine. This bug is apparently related to the illumination of the black hole model.

chris wrote:I'll need to try the add-on myself in order to investigate... If I can reproduce the crash on my own computer, I can almost certainly fix the problem very quickly.--Chris

Chris, you can find Martin's SgrA address in your PM.
Bye and thank you very much

Andrea

ANDREA wrote:

chris wrote:I'll need to try the add-on myself in order to investigate... If I can reproduce the crash on my own computer, I can almost certainly fix the problem very quickly.--Chris

Chris, you can find Martin's SgrA address in your PM.
Bye and thank you very much

I fixed an off-by-one error in an assert in the lighting code--a less-than sign should have been less-than-or-equal. This one-character fix is revision 4560 in SVN.

The disappearing orbits is caused by Celestia's restriction that star systems be less than one light year in size. I want to address that, but I think it will have to wait until after 1.6.0. I'll investigate further and see if it's possible remove this limit without significant modifications to the code.

Martin: this is a GREAT add-on!

--Chris

chris wrote:I fixed an off-by-one error in an assert in the lighting code--a less-than sign should have been less-than-or-equal. This one-character fix is revision 4560 in SVN.
The disappearing orbits is caused by Celestia's restriction that star systems be less than one light year in size. I want to address that, but I think it will have to wait until after 1.6.0. I'll investigate further and see if it's possible remove this limit without significant modifications to the code.

This is very nice, Chris, thanks a lot for your help and the speed in solving the problem.
Very appreciated, indeed!
I hope that the orbits problem may be solved easily like this, otherwise, while waiting for 1.7, we'll try to make changes in SgrA* stc ands ssc files, just to see it in full glory.
Crossing fingers!

chris wrote:Martin: this is a GREAT add-on! --Chris

Oh yes, and very helpful for educational purposes.
Cannot wait for Cartrite's SVN 4560.
Bye

Andrea

Both sites now been updated to revision 4560.

cartrite wrote:Both sites now been updated to revision 4560.

Steve, I checked your page immediately after sending my previous message, and there was the #4559.
Who are you, Superman?
Thanks a lot, you are very very kind, as usual.
Bye

Andrea

Chris, Martin and Steve, I'm very happy to inform you that everything works flawlessly, as you can see in the first image.
Moreover, now that I can see them, I found that orbits of all those stars can be seen, paying some attention to both distances and orientation, as shown in the second image.
Very well done to all of you, a zillion of thanks from me and my students.
Bye

Andrea

ANDREA wrote:

cartrite wrote:Both sites now been updated to revision 4560.

Steve, I checked your page immediately after sending my previous message, and there was the #4559.
Who are you, Superman?
Thanks a lot, you are very very kind, as usual.
Bye

Andrea

I actually rebuilt this when I saw the email at sourceforge saying the code was updated with a bug fix before I even saw this thread. I was checking often because Chris asked that a patch be tested for Linux versions and I had some trouble with the original patch. So you must have visited the site just as I was about to update it. Glad that this fix worked well.
cartrite

cartrite wrote: Glad that this fix worked well. cartrite

Thanks a lot, Steve, for your incredible promptness in SVN upgrading, very appreciated.
Bye

Andrea