Magnitude limit bug in 1.4.1

Report bugs, bug fixes and workarounds here.
Topic author
LWK
Posts: 8
Joined: 09.12.2005
With us: 18 years 11 months
Location: Florence, Italy

Magnitude limit bug in 1.4.1

Post #1by LWK » 22.09.2006, 11:18

I noticed that varying the magnitude limit, while many stars fade gradually to black, as they should, some stars (and some bright ones!) at a certain point just disappear... An example is Cassiopeia: if you look at Cas from the solar system, and you vary the mag limit, you will notice that Cas 'beta' and Cas 'delta' will appear and disappear varying the mag lmit. They shouldn't, as they are two rather bright stars!
This happens also with automag turned on, it just varies the limit at which the stars disappear.

If I'm not the only one seeing this, then I presume this is a bug. Perhaps it's already been corrected in CVS, in that case let me know.

I'm using Celestia 1.4.1, winXP, ATI Radeon x800se.

Thanks for your time, regards,
LWK.

Toti
Developer
Posts: 338
Joined: 10.02.2004
With us: 20 years 9 months

Post #2by Toti » 22.09.2006, 11:49

This has been fixed for 1.5.0

Avatar
selden
Developer
Posts: 10192
Joined: 04.09.2002
With us: 22 years 2 months
Location: NY, USA

Post #3by selden » 22.09.2006, 11:55

Unfortunately, I'm not seeing the effect you describe on my system, either with v1.4.1final or with Celestia built from cvs early this morning.

What's your screen's color depth set to?
What screen resolution are you using?
What version of graphics drivers is installed?

(Don't say "most recent". Too often people think their systems are up to date but it turns out they aren't.)

System:
1GB 3.4GHz P4-550 WinXP Pro SP2
128MB Nvidia GF6600GT, Forceware v91.47
1600x1200 60Hz, highest (32bit)
Celestia v1.4.1f & from cvs

p.s. What kind of display are you using?
Mine is an Hitachi 812 CRT.
LCD displays are still very poor at displaying low-brightness objects and can have pixel-to-pixel variations in sensitivity.
Last edited by selden on 22.09.2006, 12:01, edited 2 times in total.
Selden

Topic author
LWK
Posts: 8
Joined: 09.12.2005
With us: 18 years 11 months
Location: Florence, Italy

Post #4by LWK » 22.09.2006, 11:56

System:
1GB RAM, 3.2GHz Pentium4 WinXP Pro SP2
128MB ATI Radeon X800SE
1024x768 75Hz, 32bit color depth
Celestia v1.4.1

In fact, my drivers aren't even the most recent..!
Let me see... Catalyst 6.1, that is:

Code: Select all

Driver Packaging Version   8.205-060104a-029602C-ATI   
CATALYST?® Version   06.1   
Provider   ATI Technologies Inc.   
2D Driver Version   6.14.10.6587   
2D Driver File Path   System\CurrentControlSet\Control\Video\{57A43E37-8434-4F37-8F99-452780612CAC}\0000   
Direct3D Version   6.14.10.0362   
OpenGL Version   6.14.10.5529   
CATALYST?® Control Center Version   1.2.2195.38647   
AIW/VIVO WDM Driver Version   6.14.10.6238   
AIW/VIVO WDM SP Driver Version   6.14.10.6238   


I'm using 32bit color depth, desktop resolution of 1024x768.
Perhaps in newer versions of the drivers the bug isn't showing. I didn't update the drivers because they work good as they are :-P
Besides, in every new version of the Catalyst, they corrected bugs for cards newer than mine (X1x00s), and nothing new was reported for X800 series (sure, I know that sometimes they fix a bug for >every< card...)

I'm using a Dell CRT, M783p to be exact.
But I'm absolutely sure this isn't a problem of brightness-contrast monitor calibration (I'm quite paranoid with that :roll: ). It's a rendering problem: I see this problem happening with stars much brighter than background stars.

Topic author
LWK
Posts: 8
Joined: 09.12.2005
With us: 18 years 11 months
Location: Florence, Italy

Post #5by LWK » 23.09.2006, 16:28

I updated today to Catalyst 6.9 drivers, and the problem persists. Any ideas?

This has been fixed for 1.5.0


You say so because it doesn't do that in your installation of 1.5.0, or because you know that there was this bug in 1.4.1 and in 1.5.0 is no more? That would be different...

Toti
Developer
Posts: 338
Joined: 10.02.2004
With us: 20 years 9 months

Post #6by Toti » 23.09.2006, 17:41

There was a bug in the octree code that caused this kind of misbehavior. I committed a fix to CVS some time ago.

Topic author
LWK
Posts: 8
Joined: 09.12.2005
With us: 18 years 11 months
Location: Florence, Italy

Post #7by LWK » 23.09.2006, 18:19

Well, in that case, I will wait until the 1.5.0 comes out, and will hope...
Thanks for the info!


Return to “Bugs”