Catalogs of Messier objects (V3a update)

[selden]

A zip file of catalogs of the 110 Messier objects is now available at http://www.lns.cornell.edu/~seb/celestia/messier-all.zip

It also includes a script which puts markers on them all.

See http://www.lns.cornell.edu/~seb/celestia/catalogs.html#3.7.5 for more information.

[selden]

Two of Messier's objects, M40 and M73, are small asterisms composed of several adjacent stars which are not included in Celestia. The 4 stars of M73 are not included in the expanded Tycho stars databases, either.

An STC file defining the locations of these 6 stars now is available at http://www.lns.cornell.edu/~seb/celestia/messier_stars.stc

This file requires Celestia v1.3.1pre4 or later since it includes the stars' names.

I've also added this file to the ZIP archive.

[selden]

http://www.lns.cornell.edu/~seb/celestia/messier_all_v2.zip (164KB)
is now available. It includes extremely low resolution pictures of the 16 objects listed in messier_other.dsc. Pictures of the other 94 Messier galaxies and clusters are not included because Celestia already generates images of them.

[added later]
v2.zip requires Celestia v1.3.1pre4 or later.

http://www.lns.cornell.edu/~seb/celestia/messier_all_v2a.zip (164KB) works with Celestia v1.3.0 or later.

http://www.lns.cornell.edu/~seb/celestia/messier_all_v3a.zip (833 KB) works with Celestia v1.3.0 or later. It has higher quality (q=90) images than the other two.



For more information, please see http://www.lns.cornell.edu/~seb/celestia/catalogs.html#3.7.5.2

For comparison, here are two screenshots of an object that is included in the v2 and v3a archives:

(as usual, these link to much larger images)

[clarification added later]
All of the billboard objects included in the v2 archive specify a wildcard texture file type. (Celestia v1.3.1pre4 or later is required for wildcards to be recognized.)

This means that you can replace, for example, the 512x512 quality=20 m1-5.jpg by a 4Kx4K lossless m1-5.dds with alpha channel. NOAO provides up to 4K TIFF images of most of the Messier objects. Their larger images have the same fields of view as the smaller ones used here. Links to all of them are included in the three archives.

The v2a and v3a archives explicitly specify jpg images, so they'll work with Celestia v1.3.0.

Note that the 16 512x512 images do require a total of about 16MB of graphics memory, so you'll need a card with at least 32MB in order to see them all at once.

[danielj]

It?s not working.I downloaded messier_all_V2,but the objects are not appearing.For example,the galaxy M102 is marked,but only the empty appears.How can I put the files in the correct places?

[selden]

Brief instructions are in the zip file in MESSIER_ALL_V2_README.TXT

If you followed those instructions but it didn't work, then please read the Web page http://www.lns.cornell.edu/~seb/celestia/addon-intro.html and work through its examples. That web page explains in great detail the method I use to organize my addons.

[Paolo]

Selden My compliments
Very beautiful. Add-On. .

I've seen that when the celestial grid is activated the billboard images of the nebulae covers the grid. Is it a bug or it is the default behaviour.

If I may give an advice, I think that should be better if the Nebula images are cleaned from the included stars. I made some experiments with a photo editor and the final result is better .

Bye - Paolo

[selden]

Paolo,

Nebula objects with opaque surface textures do block the "celestial grid". They also block view of Celestia's grey galaxy blobs.

If the surface texture includes an alpha transparency channel, then you can see the grid through the transparent regions. I used JPEG images, which can't have alpha channels, so they're opaque.

PNG or DDS textures with alpha channels would have made the addon much larger. One of my goals was to make the addon as tiny as possible.

There are several reasons why I left in the stars:

One of my intended uses of the Messier addon was to provide a visual comparison for someone who might be using Celestia as a "finder" for some of the Messier objects. This is rather difficult without the surrounding stars I suspect the images may be a little too small for that, though.

Also, some of the Messier objects are only a handful of stars. If you eliminate all the stars, there wouldn't have been anything left of them!

In addition, getting rid of the stars would have taken a lot longer. I've seen programs mentioned which will filter out the appropriate "point spread functions", but I haven't found any that are publicly available. Painting stars out by hand gets quite tedious after a while. I did it for the image of NGC 4622 that's on my Billboard web page.

And finally, I intentionally provided a "degraded" version of those 16 objects. There already are high quality models available for some of them. Hopefully someone will be inspired to provide high quality 3D versions of some of the others.

Most of the time I spent on this Addon was in making sure the orientations and locations were fairly accurate.

[dede]

Sorry Selden, but I did everything recommanded for the installation ( unzip file in the "extra" folder ) but I still can't see messier objects but just their names marks in Celestia ..... What's my problem ?

Thanks Dede

[selden]

dede,

Check to make sure that UnZip created the necessary folders and put the files into them. If it put all of the files into the extras folder, that might explain your problem.

The folders should be organized like this:

Code: Select all

extras
   |
  [-] messier_all_v2 (contains stc, dsc and html files)
        |
        -- models  (contains 16 3ds models)
        |
       [-] textures
              |
              -- medres (contains 16 .jpg images)


(There's also a folder used by the html file, but that's not used by Celestia)

Please read my Web page http://www.lns.cornell.edu/~seb/celestia/addon-intro.html That page explains one way addon directories can be organized. Most of mine use that design.

Does this help?

[Guest]

Thank you Selden for replying me, and for trying to solve my problem...
Your answer confirmed me the fact that everythings OK in my folders, even if it doesn't work anyway ! I really don't understand !
I read your explanations and I don't see what to do more ! I'm very sad , 'cause I really enjoy your awesome stuff !

Dede

[selden]

Dede,

If you're sure the files are in the correct places, then I don't know what might be wrong.

However, just reading my web page really isn't enough. I'd suggest working through the examples one at a time so you understand where things have to be. Remember that Celestia really needs you to get everything precisely right. It doesn't (yet) provide error messages when you spell something wrong, for example, although unZip should have taken care of that for you.

If you can see the names and the marks turn on when you run the .CEL script, but the objects don't appear, then that means that Celestia cannot find either the models or the pictures, or that one or both are damaged.

Have you looked at the pictures in the directory Celestia\extras\messier_all_v2\textures\medres\ ? That's a quick test that the pictures are in the right place and undamaged.

[Guest]

Ok Selden, I'm going to cheack every files and textures and I'm going to look for corrupted files (even if, I never encontered any error message yet )....

Thanks again
Dede

[Paolo]

Thank you for the reply Selden

Paolo,

Nebula objects with opaque surface textures do block the "celestial grid". They also block view of Celestia's grey galaxy blobs.

If the surface texture includes an alpha transparency channel, then you can see the grid through the transparent regions. I used JPEG images, which can't have alpha channels, so they're opaque.

PNG or DDS textures with alpha channels would have made the addon much larger. One of my goals was to make the addon as tiny as possible.

Understood.

There are several reasons why I left in the stars:

One of my intended uses of the Messier addon was to provide a visual comparison for someone who might be using Celestia as a "finder" for some of the Messier objects. This is rather difficult without the surrounding stars I suspect the images may be a little too small for that, though.

Also, some of the Messier objects are only a handful of stars. If you eliminate all the stars, there wouldn't have been anything left of them!

In addition, getting rid of the stars would have taken a lot longer. I've seen programs mentioned which will filter out the appropriate "point spread functions", but I haven't found any that are publicly available. Painting stars out by hand gets quite tedious after a while. I did it for the image of NGC 4622 that's on my Billboard web page.

And finally, I intentionally provided a "degraded" version of those 16 objects. There already are high quality models available for some of them. Hopefully someone will be inspired to provide high quality 3D versions of some of the others.

Most of the time I spent on this Addon was in making sure the orientations and locations were fairly accurate.

At the end I agree with your considerations to leave the stars in the pictures

Bye - Paolo

[selden]

Dede,

I finally found why you can't see the pictures: I used wildcard image filetypes.
This new feature does not work with Celestia v1.3.0. It requires v1.3.1pre4 or later. Sorry.

I have created a new "v2a" zip file which specifies jpg images. It works with v1.3.0 on my system, at least, and hopefully will work on yours. This is the only change I've made, so people who are happy with v2 do not need this new version.

http://www.lns.cornell.edu/~seb/celestia/messier_all_v2a.zip (189KB, 7sep03)

Again, I'm sorry for the problems this caused..

[dede]

Thanks Selden, you finally found what's wrong when I decided to gaving up installing your addon, but.... hem, hermmmm..... Your link doesn't work for me

By the way, thanks again for your great effort.

Dede

[selden]

OOps

Sorry. Fixed.

(I left out the /celestia/ )

[dede]

OK, your "messier mod" works perfectly for me now !!!!

Many, many thanks Selden !
hep, don't be sorry for anything.

Dede

[selden]

After frustrating over it, I finally created an updated version of the archive with higher quality images: quality=90 instead of quality=20. They're all still 512x512, so all 16 images can be seen at the same time in cards with at least 32MB of graphics memory.

http://www.lns.cornell.edu/~seb/celestia/messier_all_v3a.zip (830MB, 26sep03)

See http://www.lns.cornell.edu/~seb/celestia/catalogs.html#3.7.5.2 for more information.

For comparison, here are two screenshots of an object that is included in the v2 and v3a archives:

(as usual, these link to much larger images)

If you look closely at the stars, significantly better resolution can be seen.

[mrzee]

Hi Selden,

I downloaded the v3a.zip (26sep03) and installed it sucessfully, looks great too.
Have one problem where the image does not display but is marked. If I move drag the display (using left click and drag) it sometimes appears and other times disappears. This only occurs with a few of the objects, m102 is an example of one that does this. If I use real time the object appears and disapears flashing quite regularly (new way to make a pulsar?...) If I pause, then it is either displayed or not depending on just when I hit the space bar!
Don't know if it's really got anything to do with the add-on but as its the only add-on I have installed I thought I'd let you know.
My system is win2k, tnt64 video card, 256M ram and latest Nvidia drivers.
Celestia v1.3.1pre9

Regards,

[selden]

Mr Z,

All of the models are identical, only the texture names that they specify are different. M102 is the last one loaded and is the most distant.

I can think of a few things that might be causing the problems you see.

1) Running out of graphics memory.

All of the images together total about 16MB when loaded. (512x512x4x16)
How much graphics memory does your TNT64 have?
According to the Web pages I've seen, that card can have 8, 16 or 32MB of memory. 32MB should be OK if you don't have a lot of other surface textures loaded. 16MB may not be enough to load all of the objects simultaneously. You can help this situation by disabling the Medres textures that come with Celestia.

2) Limitations in the graphics drivers.

The TNT card is relatively old and doesn't include as many hardware features as newer cards. I suspect more "software assist" code is being provided.

3) Problems in how Celestia draws Nebulas.

I have noticed that there seems to be a bug that causes only parts of surfaces to be drawn sometimes. I see this on my system at home (GF4 Ti4200) and at work (FX 5200). My impression is that it may be related to numeric errors in the calculation of what's visible near the corners of the screen. I'd expect that calculation to have more problems with more distant objects, although I see it happening with relatively nearby nebulas (M42 & M78) but I don't see it happening with M102 on my system at work, for some reason.


Sorry: I don't know if this helps at all.