Celestia Forums

I would like to start an ongoing thread on the Celestia Explorer script. I think this script is great, but it has some limitations. I would like for there to be a place for people to post their thoughts on the script, ways that they have found to workaround limitations, and features that they would like to see in the script. And possibly, if Mr. Harris is agreeable, to form a group to work on developing the script.

I though that I would start out by giving my ways of navigating to objects outside of sensor range in order of difficulty:

1. Just use the "Go To Object" function in Celestia.
2. Use the "Select Object" function, Open the Sensor panel in Celestia Explorer and use the "home" key to approach the object until it registers on the sensors.
3. As above, but use the "Center" function to center the object and "fly" towards it using Celestia Explorer until it registers on sensors.

This is a great idea and I'll lend my support as I have time. I have some questions:

Is anybody having trouble downloading the manual?

I downloaded the script back in April. I'm curious as to why nobody had any questions until August.

Is anybody interested in adding to the interstellar cloud database?

Is anybody interested in the dockable object concept? There is quite a bit of nuance to this idea that could be explored.

I'm guessing that some of the "limitations" are actually my design decisions intended to create some sort of psuedo realism. Perhaps this could be a selectable option.

Henry

You should include a link to the Explorer script in this thread, too.

There are some bugs in the V1.0 documentation that's in the Zip file of the Explorer script on the Motherlode as of August 18, 2006. Is there a more recent version somewhere?

To read it, I'm using Microsoft Word Viewer 97 and OpenOffice 2.0.2. (WordPad doesn't show any of its figures at all.)

Suggestion: provide the document as HTML. That would allow the reader to specify font and page sizes that might be more appropriate for their screen and visual acuity. This also would avoid some of the bugs in Word Viewer (WV) and OpenOffice (OO), but surely would have different ones. (CSS has different bugs in different browsers.)

1. Table 2, page 5: column 2 wraps around into column 1. (bug in WV)

2. Table 3, pages 5&6: title is "orphaned"
i.e. the title is on page 5 but the body of the table is on page 6. (OO, OK in WV -- which shows the table a half page later, in the middle of page 6)

3. WV complains that the image does not exist for "Figure 2 - Celestia Explorer Coordinate System" in Section 2, on page 8. OO garbles its text and shows an otherwise empty box with a tilted Y in it. (An image of the galaxy with an appropriate graticule might work better.)

4. Bug in WV: The page numbers wrap around. The word "Page" is at the right margin, but the number is at the left margin. OO is OK.

p.s. a very small fraction of people ever provide feedback about anything. There have been thousands of downloads of the Web pages I've written about Celestia, for example, but I think fewer than a dozen have ever asked questions or reported problems with them.

Some operational comments:

0. I've tried the script with Celestia as built from CVS and with v1.4.1 final. Both have the same problems with the numeric keys. (see #3 below)

1. When I double click on the script's icon when Celestia isn't running, it starts Celestia, but it leaves me out in the middle of nowhere, at Celestia's default startup location. (Celestia coordinates [0,0,0], I think.) Apparently the ISS has to be within some reasonable distance for the initial automatic docking to work. It'd be nice if there were a way to tell the script "start from scratch at the ISS" and/or "start where I left off last time".

2. When I start Celestia, let it take me to Earth, and then start the script, it takes me to the ISS but doesn't reflect that in the console display. It'd be nice if ISS were shown as the Target. The "Docked" message blinks off and on and is off more than it's on so I didn't notice it initially.

Parenthetical comment: Since the next release of Celestia casts "eclipse" shadows on models (v1.4.1 doesn't), often the ISS is pitch black, orbiting on the night side of the Earth. As a result, it isn't always obvious that the script has taken you there. It sure would be nice if the ISS had some running lights I finally realized I should turn on Celestia's "ambient light".

3. The manual says to press "9" to undock. It isn't obvious if it means the 9 at the top of the keyboard or the 9 on the numeric keypad. It's be nice if that were clarified. Either way, it doesn't actully seem to undock me

9 above the keyboard centers pluto on the screen. 0 centers the sun.

9 on the numeric keypad rotates the viewpoint (command Roll Right)
0 does nothing.

Apparently Celestia's own "numeric pad" mode needs to be disabled and apparently the script doesn't do that.

*sigh*

I need to go fix breakfast, so I'll stop now.

Sorry, I thougt that I had mentioned the location in my first post. Here is the URL for the script http://www.celestiamotherlode.net/creators/henry_harris/CelestiaExplorer.zip

There was also a recent post about the about the keyboard conflicts with Celestia http://www.shatters.net/forum/viewtopic.php?t=9894

I have had the same issue with starting Celestia from the script, but I find it a postitive as it eliminates any problems with undocking. I often had Celestia post an error message and stop the script while trying to undock.

I also have the same problem with orbits, just as I am about to start the orbit the script crashes with an error message.

At the moment I am dealing the the problems by simply not undocking or orbiting, thinking that this was a bug with my computer. Are other people having these problems?

If this is a common problem then the bug must be in the script.

One of my reasons for starting this thread was to take the burden of bug fixes and development off of Mr. Harris. One of the first projects that I wanted to take on was a rewrite of the manual. Now that Mr. Harris has given his OK to other people working on the script, are there any other people interested in working of a new version of the manual. I have already seen the suggestions made in this thread, does anyone else have suggestionsof things that they would like to see. Screenshots of the various panels are a must, IMHO.

I think that it is a neat idea for a group of people to modify the Celestia Expl. script. I recently downloaded the script and have had similar experience with unable to undock or orbit. There are certainly many limitations, but this idea has extreme potential. Unfortunately, I am not well versed in programming or script writing, thus I can contribute only by offering ideas:

1) The top speed of the shuttle being 0.1ly seems to by slightly too slow for very far stars, planets.
2) I first found the UI a little bit unsettling, since unless you knew that you were running Explorer, it just looked normal, I expected to be able to at least tell that I am in a ship.
3) The common problems of unable to undock/orbit. Eventually working into the script of being to dock at other objects beside the ISS.

sigh, neat idea, just needs work. If a newer version of the manual/CelesX comes up, could someone please notify me of it. That would be great

Dante - The best way I know to get that "In my ship" experience is to use the Celestia-Ed version of the program. It includes an overlay function of the front window of a shuttle. You can find it at :

http://www.celestiamotherlode.net/catalog/educational.php

Just remember, the HUD for Celestia Explorer take up a portion of the screen.

Also the Celestia-Ed version is only available for windows at the moment, but developers are working on the Mac and Linux codes.

At the moment the best this thread can offer is workarounds for problems but that is going to change.

Quick question. It has been suggested that the manual be in HTML format for easier viewing. What do you think of making it a PDF document?

First of all, thanks for all the comments on Celestia Explorer. They are appreciated. I'm currently in a situation which allows me to devote a little time to this, not a whole lot, but that should improve with time.

The comments on the documentation suggest to me that the rtf standard is not a standard at all, or perhaps my application (NISUS Writer Express) does not do a good job implementing the standard. I think the suggestion to covert it to HTML is a good one. I?€™ll work on that as I have time.

I wrote Explorer on a Mac so that might suggest the reason for some of the key assignment problems.

I tested the dock/undock feature again and it works for me. However, having some time since I've looked at this again made me think that it's easy to get confused as to how this works. As I explained in the manual, the order of the commands is significant, but I?€™m still convinced given the limitations of the interface, the effort expended in learning how to command should be worth it. But I promise to think about this some more. This may be a Mac/PC problem. The ?€?9?€

Can you give a more detailed description of how the orbit function works when you redo the manual.

Whenever I try to go into orbit around a object, I get the TTT countdown as I approach, but just when I should enter orbit the program always crashes.

The message is that the program tried to calculate using a nil value.

Orbiting an object is accomplished using a celestia:createcelscript() function. Can you give me some more detail on your crash?

What type of orbit are you attempting? equatorial,polar, inclined?

What type of object are you attempting an orbit? My code determines the parameters of the orbit using several methods to determine mass of the object, depending on the type of object.

I don't mind explaining things like this in the manual -- I'm a scientist so I love it -- but not everyone will appreciate it. Perhaps this should be in an appendix?

Henry

I've completed coverting the manual to HTML and I'll be able to publish a link to it as soon as I've figured out how to ftp the file to my Earthlink web space.

I was unable to reproduce the "undock" problem. It's pretty simple. Just hit the tab key to get the UNDOCK menu item. Select UNDOCK using the number keys at the top of the keyboard and you should be backing away from the ISS at that point. Can someone give me some more detail on this problem?

I've discovered there is a bug in the code that handles proximity to an object. This bug is probably responsible for the computer crash when trying to orbit. This code is fairly complicated because it has to handle a lot states and types of objects.

Henry

Would it be possible to add a function that would allow you to target the currently selected object in Celestia. In your recent post on the "dolly" function you had the script get its selection from Celestia. Could you do something similar with Celestia Explorer using a keypress such as :

"C" target currently selected object

If not in the main release, would it be possible for you to post instructions on how to manually edit the .celx file to add this function?