Celestia 1.5.0 prerelease 6

[chris]

A new Celestia 1.5.0 prerelease package is now available for Windows. Phoenix has kindly volunteered to host the download on a site much faster than shatters.net:

http://www.purebytes.de/stuff/celestia/ ... .0pre6.exe

This is hopefully the very last prerelease before 1.5.0 final. There will be a change to the RA/Dec display; aside from that the only changes will be last minute bug fixes, but I'm optimistic that nothing new will turn up.

Here is the complete list of changes since 1.5.0pre5:

* New Bulgarian translation; other translations updated
* Updated stc parser to handle catalog numbers for OrbitBarycenter
* Added fade effect for constellation labels and lines when viewer leaves
vicinity of Solar System
* Added support for showing reference vectors: frame axes, body axes, velocity,
sun direction
* Implemented labels for markers
* Clamped simulation time to +/- two billion years
* Added hires textures to distribution
* New date/time formatting; date format setting added to config dialog
* Added child lists for stars
* Changed goto distance for stellar barycenters so that all child stars fit
in view
* Core: Default to the GLSL/OpenGL 2.0 render path if possible

Bug fixes:
* Fixed bug that caused solar system browser to come up empty if the observer
was close to a star without a solar system.
* Fixed incorrect shadow count assert
* Fixed label occlusion by irregular shaped objects
* Fixed star halo glitch at 1ly distance on Macs with some ATI GPUs
* Fixed bug that caused objects to show up black when they had a scattering
atmosphere, a mesh, and a texture with an alpha channel.
* Changed celx keyboard handler code for Lua 5.1; makes scripts that use old
* Switched to ARB_texture_cube_map from old EXT_texture_cube_map
style keyboard handling work again in new versions of Celestia.
* Fixed compiler warnings caused by unused parameters and numeric type
mismatches
* Fixed bug in star halo size calculation
* Fixed bug in calculating the position of objects with default reference
frames in orbit around objects with explicit orbit frames.
* Prevent rendering of planets as points in daytime skies of planets with
atmospheres.
* Fixed inappropriate Greek letter substitution in some object names
* Changed default rotation model to FixedRotation for objects without
periodic orbits.
* Changed the begin and end parameters for ScriptedOrbit and ScriptedRotation
to beginDate and endDate, because 'end' is a Lua reserved word
* Updated out of date descriptions in the tour guide

Data files:
* Updated Pluto-Charon orbits
* Updated Great Red Spot longitude
* Updated Nix and Hydra orbits (new data from Tholen et al)
* IAU naming changes: Erriapo -> Erriapus, new spelling of Hyrrokkin
* Added new IAU names for features on Mars, Io, Europa, Titan, Venus
* New IAU name for Saturnian ring feature: Roche division
* Updated crater sizes for Gaspra
* Applied brown dwarf textures to type M brown dwarfs RECONS 2 and LP 944-20
* New stars for extrasolar planets: XO-3, WASP-4, TW Hya
* New extrasolar planets: XO-3 b, WASP-4 b TW Hya b
* Added definition for RingSystemSectinos to celestia.cfg

Mac:
* Ignore bodies with empty names in solar system browser
* Fixed POSIX locales by setting LANG at startup
* Removed non-functional movie-capture menu (for now)
* Disabled point sprites for some ATI graphics cards; fixes star
glare rendering glitch at one light year distance.

KDE:
* Star type saved in config file
* Save texture resolution setting to config file
* Add texture resolution setting to config dialog

Windows:
* Removed unused Mark Features checkbox from Locations dialog
* Resized Locations dialog
* Automatically enable markers when 'Mark Object' is chosen in the
selection popup menu.
* Prevent fullscreen mode from crashing when graphics card or drivers
is changed in between program runs.

--Chris

[Chuft-Captain]

Has anyone else noticed broken celURL's in PRE6?
(EDIT: using URL's created in PRE5)

Chris, I've PM'd you with details.

[BobHegwood]

Many thanks, Chris, and MUCH thanks to Phoenix for the outstanding
download facility. That's the fastest I've ever gotten one of these
beauties.

[selden]

Has anyone else noticed broken celURL's in PRE6?

Chris, I've PM'd you with details.

Please don't be so vague.

Bugs should be reported using Tracker on SourceForge so everybody can see them.
https://sourceforge.net/tracker/?atid=1 ... unc=browse

I've been having some problems with high-precision (~ 1 meter) positioning, but they're hard to characterize.

[cartrite]

Has anyone else noticed broken celURL's in PRE6?

Chris, I've PM'd you with details.

The only cel url that was broken was created or copied from Celestia 1.3.2. Cel urls that I copied from version 1.5.0pre5 and 1.4.1 worked fine. These were simple url's though. I just selected different stars and typed "g". After I scrolled the mouse wheel to zoom in a little, I copied the url.
cartrite

[Adirondack]

... Cel urls that I copied from version ... 1.4.1 worked fine.

Really?
That is rather unlikely, see here!

Adirondack

[duds26]

A much faster site, sweet.

But please make it possible to install the 1.5.0 and 1.4.1 on the same system.
Couln't do it and
please make the default install folder Celestia 1.5.0 and the shortcuts name and program folder: Celestia 1.5.0.





This celURL thing is rather worrying.
Oh seen that celURL's from 1.5.0 will be compatable with future versions, that seems comforting enough.



-------------------------------
Don't forget to check development tools for updates before compiling Final Version of Celestia 1.5.0

[selden]

Duds,

You can as many different versions of Celestia installed at the same time as you want. You just can't do it blindly.

Rename the directory that your current Celestia installation uses e.g. from Celestia to Celestia_141.
Then run the installation program.
It'll recreate the Celestia folder containing the new version and associate the desktop icon and URLs with the new version.

Then you can move your addons at your convenience from the old directory to the new one.

You can still run the old version of Celestia by selecting Celestia_141/Celestia. It'll display all of the addons installed in its directory.

[duds26]

Okay

Too bad it isn't that easy, it isn't so difficult happily.

[BobHegwood]

Okay

Too bad it isn't that easy, it isn't so difficult happily.

Huh? It is VERY easy to do... I have three versions of Celestia
running on my PC right now, and I'm Brain-Dead.

[cartrite]

... Cel urls that I copied from version ... 1.4.1 worked fine.

Really?
That is rather unlikely, see here!

Adirondack

I see. I guess I did an invalid test because the cel url's were not complicated enough. I probably should have copied url's from a planet's surface. I'm not sure if this will work here but testcel-1.4.1
is copied while running 1.4.1 and see where it takes you in 1.5pre6. It should take you to kDra at around 14:48 UTC and the distance = .33832 au .
cartrite

[Chuft-Captain]

... Cel urls that I copied from version ... 1.4.1 worked fine.

Really?
That is rather unlikely, see here!

Adirondack

I see. I guess I did an invalid test because the cel url's were not complicated enough. I probably should have copied url's from a planet's surface. I'm not sure if this will work here but testcel-1.4.1
is copied while running 1.4.1 and see where it takes you in 1.5pre6. It should take you to kDra at around 14:48 UTC and the distance = .33832 au .
cartrite

Cartrite,

Just to clarify,

The problem I'm reporting is not about a difference between 1.4.1 and 1.5.0 URL's. That was a deliberate effect brought about by the re-positioning of the solar-system between 1.4.1 and 1.5.0pre1 a long time ago, and is something we have to live with.

The problem I'm reporting is when a URL created in 1.5.0pre5 is used in 1.5.0pre6, it is not correct. (It could be that this is only a problem for URL's related to objects using the new reference-frames and/or rotation models, which may be why you are not experiencing the same problem.

Chris is investigating.

[cartrite]

... Cel urls that I copied from version ... 1.4.1 worked fine.

Really?
That is rather unlikely, see here!

Adirondack

OK. Per my last post, I copied a url while running 1.4.1 from earth orbit and that was broken as expected. I also copied a url while running cvs version 10252007 and that took me to the same spot with cels associated to 1.5.0pre6 by clicking on the celurl. So if 1.5pre6 does break celurl's, it doesn't seem to break celurl's copied after cvs 10252007. That is the oldest cvs copy I have left on my system.

Chuft-Captain,
cross post, I just seen your post. OK.

cartrite

[Chuft-Captain]

No worries...

Perhaps if my first post had not been so brief, I might have saved you a lot of effort.

[ajtribick]

Comets are still using visual rather than bolometric magnitude to calculate when to draw the tail.

[chris]

Comets are still using visual rather than bolometric magnitude to calculate when to draw the tail.

This will be fixed in the next version.

--Chris

[Reiko]

Fixed bug that caused solar system browser to come up empty if the observer
was close to a star without a solar system.

Oh thank god! I thought I was losing my mind when I came across that bug.

[Adirondack]

* Automatically enable markers when 'Mark Object' is chosen in the selection popup menu.

Chris, that's fine, but ...

... when one chose 'Unmark Object' in the selection popup menu, markers are not automatically disabled.

Is this intended (though I think this behavior is unpleasant) or should I report this to the bug tracker?


Adirondack

[chris]

* Automatically enable markers when 'Mark Object' is chosen in the selection popup menu.

Chris, that's fine, but ...

... when one chose 'Unmark Object' in the selection popup menu, markers are not automatically disabled.

Is this intended (though I think this behavior is unpleasant) or should I report this to the bug tracker?

This behavior is intended. There may be many markers visible--unmark object is supposed to make just one of them disappear, not all of them.

--Chris

[bdm]

There is a duplicate star that needs to be removed.

* Gliese 661
* HIP 84140

These are the same star, as verified by the Nearby Stars Database (http://nstars.nau.edu/).

HIP 84140 is rendered properly; there are two stars orbiting each other (and very nice they look too in this version).
Gliese 661 is not rendered correctly; it appears to be one star in an orbit around nothing in particular.
The two systems are about 1/4 light year apart.
According to the Nearby Stars database, this binary star is located 6.39713 parsecs from Sol (20.865 light years). This is the distance that Gliese 661 has from Sol according to Celestia.