Please report problems with the MacOS X version here.
Re: same earth garbling texture again..
[quote="marcof"]
click for large image
as you can see, textures on earth garbled, sometimes it is af earth is only in blach&white, and only very rarely earth appears ok..
I haven't done extensive testing or anything, but I noticed that this problem was present when my screen resolution was set to millions of colors and it disappeared once it was set to thousands.
hope this helps...
click for large image
as you can see, textures on earth garbled, sometimes it is af earth is only in blach&white, and only very rarely earth appears ok..
I haven't done extensive testing or anything, but I noticed that this problem was present when my screen resolution was set to millions of colors and it disappeared once it was set to thousands.
hope this helps...
hankR wrote:I hope to have a new version of Celestia for Mac OS X based on the 1.3.0 code ready to go in a few days.
It will be incomplete, and there will be bugs, but it should be an improvement over the previous OS X version.
- Hank
Here I was really hoping that you'd have it out in time for my birthday. But, at this point, you've already missed it by two days...
-
- Site Admin
- Posts: 4211
- Joined: 28.01.2002
- With us: 22 years 9 months
- Location: Seattle, Washington, USA
Here I was really hoping that you'd have it out in time for my birthday. But, at this point, you've already missed it by two days...
Actually, I think that it was ready right on your birthday:
http://sourceforge.net/project/showfiles.php?group_id=21302
Thanks for your work on the MacOSX version, Hank!
--Chris
I love the new version
Thanks for the new version.
Anyone else seeing 'rainbow' patterns on their planets...I can't
figure out how to attach a picture, but when I look at the moon
or mars, they have a color spectrum superimposed over them.
I'm running on a 667 Mhz TiBook with 10.2.6, and 32 Mb video
ram on some flavor of ATI card (don't know exactly what...don't
know how to find out either).
Cheers
Anyone else seeing 'rainbow' patterns on their planets...I can't
figure out how to attach a picture, but when I look at the moon
or mars, they have a color spectrum superimposed over them.
I'm running on a 667 Mhz TiBook with 10.2.6, and 32 Mb video
ram on some flavor of ATI card (don't know exactly what...don't
know how to find out either).
Cheers
Re: I love the new version
[quote="bryanc"]Thanks for the new version.
Anyone else seeing 'rainbow' patterns on their planets...I can't
figure out how to attach a picture, but when I look at the moon
or mars, they have a color spectrum superimposed over them.
I do, and also on Mercury and Pluto. The Galilean moons look OK.
I'm running on a 667 Mhz TiBook with 10.2.6, and 32 Mb video
ram on some flavor of ATI card (don't know exactly what...don't
know how to find out either).
In the window "About this Mac" click on the "More Info..." button, this invokes the System profiler. Under the Devices and Volumes tab, there is information about the graphics card.
My TiBook is similar, but with only 16 Mb ATI card. I still found the bug with the Earth view; I see the surface only with the low resolution setting; it's garbled in the other settings. It seems more memory is required. I'll have to live with a lousy looking Earth
Anyone else seeing 'rainbow' patterns on their planets...I can't
figure out how to attach a picture, but when I look at the moon
or mars, they have a color spectrum superimposed over them.
I do, and also on Mercury and Pluto. The Galilean moons look OK.
I'm running on a 667 Mhz TiBook with 10.2.6, and 32 Mb video
ram on some flavor of ATI card (don't know exactly what...don't
know how to find out either).
In the window "About this Mac" click on the "More Info..." button, this invokes the System profiler. Under the Devices and Volumes tab, there is information about the graphics card.
My TiBook is similar, but with only 16 Mb ATI card. I still found the bug with the Earth view; I see the surface only with the low resolution setting; it's garbled in the other settings. It seems more memory is required. I'll have to live with a lousy looking Earth
-
- Posts: 312
- Joined: 04.03.2002
- With us: 22 years 8 months
new version
Holy crow, this new version is terrific.
If you are having that "color spectrum" problem you can fix it by commenting out the bump map entries in solarsys.ssc. Just put a "# " before "BumpMap" everywhere it appears. You're right, it's the bump mapping that is broken.
I noticed that the specularity map is finally working for Earth's oceans on my ATI card. Nice!
Any ideas about the orbit display bug?
If you are having that "color spectrum" problem you can fix it by commenting out the bump map entries in solarsys.ssc. Just put a "# " before "BumpMap" everywhere it appears. You're right, it's the bump mapping that is broken.
I noticed that the specularity map is finally working for Earth's oceans on my ATI card. Nice!
Any ideas about the orbit display bug?
Where is the Mir ?
Hello Guys,
just downloaded the 1.2.5 Version for my iMac G4 17"... real cool Program but when i go to Mir it doesn't appear. Can Someone help me ?
____
Charly Weing?rtner
http://cweingaertner.themac.de
just downloaded the 1.2.5 Version for my iMac G4 17"... real cool Program but when i go to Mir it doesn't appear. Can Someone help me ?
____
Charly Weing?rtner
http://cweingaertner.themac.de
-
Topic authorHankR
Charly,
Why are you using 1.2.5? Try the new version, 1.3.0.
Re Mir:
Mir was launched on 20 Feb 1986 and deorbited on 21 Mar 2001. To see Mir, try setting your time to sometime in that interval. (Version 1.3.0 lets you set the time directly; with 1.2.5 you'll have to run the clock backwards.)
- Hank
Why are you using 1.2.5? Try the new version, 1.3.0.
Re Mir:
Mir was launched on 20 Feb 1986 and deorbited on 21 Mar 2001. To see Mir, try setting your time to sometime in that interval. (Version 1.3.0 lets you set the time directly; with 1.2.5 you'll have to run the clock backwards.)
- Hank
Hello,
I just d/led and installed Celestia 1.3.0 alpha for mac OS X. Great work, everyone! I was always waiting for such a nice app to hit the Mac OS X community... however:
1. I too experienced the terrible bug regarding the strange rendering of the Earth planet. It was working fine up to the momment when the earth started to render using textures from Jupiter! Then it scrambled a bit more and after a few minutes my mac hung with no KP screen but no GUI feedback whatsoever.
I noticed that the machine hung completely, since the fans which were working hard thanx to distributed.net client started to slow down.
There must be something terribly wrong with Earth rendering, since this bug only occurs when browsing Earth.
As a final note, I now cannot access the 3rd rendering path (the one after multitexture) even after a restart! Haven't tried to flush preferences, though... Is Earth rendering somehow different from the other planetary objects? If so, this could also explain such a behavior...
2. I also noticed a bug regarding the Orbital lines. They seem do draw slightly out of place, as well as a bit strange in general. Very noticeable at Jupiter in particular. Didn't read about this bug on the forum here, so I thought it would be nice to report...
My Mac: MDD DP867 / 512 RAM / GeForce 4MX
I still believe that there has to be something terribly wrong with Apple's OpenGL implementation and/or with drivers from ATi / nVidia, since even a bug in a program code sould by no means bring the whole system down! Untill they (Apple / ATi / nVidia) track the bug down (it's been around for since OS X hit the stores), I'm in the hope that you nice people developing Celestia will be able to fix it in your code...
I just d/led and installed Celestia 1.3.0 alpha for mac OS X. Great work, everyone! I was always waiting for such a nice app to hit the Mac OS X community... however:
1. I too experienced the terrible bug regarding the strange rendering of the Earth planet. It was working fine up to the momment when the earth started to render using textures from Jupiter! Then it scrambled a bit more and after a few minutes my mac hung with no KP screen but no GUI feedback whatsoever.
I noticed that the machine hung completely, since the fans which were working hard thanx to distributed.net client started to slow down.
There must be something terribly wrong with Earth rendering, since this bug only occurs when browsing Earth.
As a final note, I now cannot access the 3rd rendering path (the one after multitexture) even after a restart! Haven't tried to flush preferences, though... Is Earth rendering somehow different from the other planetary objects? If so, this could also explain such a behavior...
2. I also noticed a bug regarding the Orbital lines. They seem do draw slightly out of place, as well as a bit strange in general. Very noticeable at Jupiter in particular. Didn't read about this bug on the forum here, so I thought it would be nice to report...
My Mac: MDD DP867 / 512 RAM / GeForce 4MX
I still believe that there has to be something terribly wrong with Apple's OpenGL implementation and/or with drivers from ATi / nVidia, since even a bug in a program code sould by no means bring the whole system down! Untill they (Apple / ATi / nVidia) track the bug down (it's been around for since OS X hit the stores), I'm in the hope that you nice people developing Celestia will be able to fix it in your code...
If you are a Greek Macintosh user, join Hellenic Macintosh Users Group: http://www.HelMUG.gr
-
Topic authorHankR
Thanar,
The problem with orbits will be fixed in the next release. (Matt McIrvin will be pleased to hear about this. He's been complaining about it for a long time.)
I don't know what the problem might be with the Earth texture. Are you using the standard textures, or have you installed any add-ons? Also, do you have the latest version (10.2.6) of Mac OS X? Apple has been gradually improving its OpenGL implementation.
- Hank
The problem with orbits will be fixed in the next release. (Matt McIrvin will be pleased to hear about this. He's been complaining about it for a long time.)
I don't know what the problem might be with the Earth texture. Are you using the standard textures, or have you installed any add-ons? Also, do you have the latest version (10.2.6) of Mac OS X? Apple has been gradually improving its OpenGL implementation.
- Hank
Yes, I have the latest versions of both OS X and Celestia. It is the same problem as the one many people were complaining about sometime ago in these forums...
The Earth texture turns random. I was playing the Demo of UT 2003 yesterday and when I fired up Celestia afterwards, I could see UT textures all over Earth.
I am almost certain that the routines used for earth are different compaired to these used to render the other objects, since this only happens with Earth. Sometimes clouds get funny, too...
As a bottom note, yes, I have the original textures. I don't think this is some kind of memory overload issue, since there is no disc activity when this happens, although I saw noumerous [Cache hits] in fs_usage when the Earth textures should be loaded. And that was after an hour of UTing! Seems like the method used in loading the textures for Earth somehow tricks the system in thinking that it's already there and serves them from the disc cashe somehow... dunno...
I just tried to duplicate the problem, but no! Whenever you want something like this to happen you get a perfectly working software! I tried pushing RAM (regular and video) to its limits, but Celestia works fine. However:
I launched it from the terminal (./Celestia) and got this output:
Loading texture: textures/hires/earth.png
Error opening texture file textures/hires/earth.png
Loading texture: textures/medres/earth.png
Loading texture: textures/hires/earthnight.jpg
Datafile::Open() - Couldn't open textures/hires/earthnight.jpg
Datafile::Read() - No file open, file of zero size, or no valid MemoryBuffer
CGBuffer :: Error loading JPEG texture file /Users/thanar/Library/Application Support/CelestiaResources/textures/hires/earthnight.jpg
Loading texture: textures/medres/earthnight.jpg
Loading texture: textures/hires/earth-clouds.png
Error opening texture file textures/hires/earth-clouds.png
Loading texture: textures/medres/earth-clouds.png
Loading texture: textures/hires/moon.jpg
Datafile::Open() - Couldn't open textures/hires/moon.jpg
Datafile::Read() - No file open, file of zero size, or no valid MemoryBuffer
CGBuffer :: Error loading JPEG texture file /Users/thanar/Library/Application Support/CelestiaResources/textures/hires/moon.jpg
Loading texture: textures/medres/moon.jpg
Loading bump map: textures/hires/moonbump1k.jpg
Datafile::Open() - Couldn't open textures/hires/moonbump1k.jpg
Datafile::Read() - No file open, file of zero size, or no valid MemoryBuffer
CGBuffer :: Error loading JPEG texture file /Users/thanar/Library/Application Support/CelestiaResources/textures/hires/moonbump1k.jpg
Loading bump map: textures/medres/moonbump1k.jpg
I really thing it would be better if the application was performing a check if these hires files exist before trying to access them. So many "onerr" routines can be sometimes dangerous...
Plus; any ideas on why fix_prebindings starts everytime Celestia is launched? (I cought it in my tests occupying arounf 150MB or RAM!):
10479 fix_prebin 0.0% 0:00.12 1 17 22 248K 864K 988K 14.1M
10478 Celestia 68.1% 0:10.20 1 60 182 13.1M 19.2M 22.0M 96.1M
The Earth texture turns random. I was playing the Demo of UT 2003 yesterday and when I fired up Celestia afterwards, I could see UT textures all over Earth.
I am almost certain that the routines used for earth are different compaired to these used to render the other objects, since this only happens with Earth. Sometimes clouds get funny, too...
As a bottom note, yes, I have the original textures. I don't think this is some kind of memory overload issue, since there is no disc activity when this happens, although I saw noumerous [Cache hits] in fs_usage when the Earth textures should be loaded. And that was after an hour of UTing! Seems like the method used in loading the textures for Earth somehow tricks the system in thinking that it's already there and serves them from the disc cashe somehow... dunno...
I just tried to duplicate the problem, but no! Whenever you want something like this to happen you get a perfectly working software! I tried pushing RAM (regular and video) to its limits, but Celestia works fine. However:
I launched it from the terminal (./Celestia) and got this output:
Loading texture: textures/hires/earth.png
Error opening texture file textures/hires/earth.png
Loading texture: textures/medres/earth.png
Loading texture: textures/hires/earthnight.jpg
Datafile::Open() - Couldn't open textures/hires/earthnight.jpg
Datafile::Read() - No file open, file of zero size, or no valid MemoryBuffer
CGBuffer :: Error loading JPEG texture file /Users/thanar/Library/Application Support/CelestiaResources/textures/hires/earthnight.jpg
Loading texture: textures/medres/earthnight.jpg
Loading texture: textures/hires/earth-clouds.png
Error opening texture file textures/hires/earth-clouds.png
Loading texture: textures/medres/earth-clouds.png
Loading texture: textures/hires/moon.jpg
Datafile::Open() - Couldn't open textures/hires/moon.jpg
Datafile::Read() - No file open, file of zero size, or no valid MemoryBuffer
CGBuffer :: Error loading JPEG texture file /Users/thanar/Library/Application Support/CelestiaResources/textures/hires/moon.jpg
Loading texture: textures/medres/moon.jpg
Loading bump map: textures/hires/moonbump1k.jpg
Datafile::Open() - Couldn't open textures/hires/moonbump1k.jpg
Datafile::Read() - No file open, file of zero size, or no valid MemoryBuffer
CGBuffer :: Error loading JPEG texture file /Users/thanar/Library/Application Support/CelestiaResources/textures/hires/moonbump1k.jpg
Loading bump map: textures/medres/moonbump1k.jpg
I really thing it would be better if the application was performing a check if these hires files exist before trying to access them. So many "onerr" routines can be sometimes dangerous...
Plus; any ideas on why fix_prebindings starts everytime Celestia is launched? (I cought it in my tests occupying arounf 150MB or RAM!):
10479 fix_prebin 0.0% 0:00.12 1 17 22 248K 864K 988K 14.1M
10478 Celestia 68.1% 0:10.20 1 60 182 13.1M 19.2M 22.0M 96.1M
As a side note, I have noticed that the screen "flickers" somehow (this is not hardware related) when a full-screen Celestia window is in the background and you are staring at a bright window in the foreground.
To experience this yourself, just open Celestia, maximize its window and open a new Finder window, maximizing it as well. You will then (bring the brightness of your monitor down) this flickering I am talking about. It is a very mild change in the brightness. Veeery mild. It hit me while I was staring at my monitor and thought... what the heck... is it brocken? I noticed it on my CRT display. Maybe it doesn't show up on a TFT, since the crystals in TFTs may be less responsive to so small brightness changes...
I guess it has to do with Quartz Extreme, since minimizing the Celestia window cancels the effect! I just don't get it... Does Celestia draws the screen using just OpenGL, or does QE gets somewhere in the way as well? Would it be possible to add a Full-screen option in Celestia so as the only libraries working will be OpenGL's?
To experience this yourself, just open Celestia, maximize its window and open a new Finder window, maximizing it as well. You will then (bring the brightness of your monitor down) this flickering I am talking about. It is a very mild change in the brightness. Veeery mild. It hit me while I was staring at my monitor and thought... what the heck... is it brocken? I noticed it on my CRT display. Maybe it doesn't show up on a TFT, since the crystals in TFTs may be less responsive to so small brightness changes...
I guess it has to do with Quartz Extreme, since minimizing the Celestia window cancels the effect! I just don't get it... Does Celestia draws the screen using just OpenGL, or does QE gets somewhere in the way as well? Would it be possible to add a Full-screen option in Celestia so as the only libraries working will be OpenGL's?
I don't get it... Since Celestia first froze my Mac after the corrupt Earth texture issue appeared for the first time, every time I try to access the "nVidia combiners" render method via CNTR-v, the application crashes... Here's the basic Console output:
Date/Time: 2003-06-20 20:24:46 +0300
OS Version: 10.2.6 (Build 6L60)
Host: Thanars-Computer.local.
Command: Celestia
PID: 481
Exception: EXC_BAD_ACCESS (0x0001)
Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000034
Thread 0 Crashed:
#0 0x0003e654 in _ZN7Texture7getNameEv
#1 0x3feff9d6 in 0x3feff9d6
#2 0x000317e4 in _Z20SetupCombinersSmoothR7TextureS0_5Colorb
#3 0x0005f5c4 in _Z16renderSmoothMeshRK9GLContextR7Texture7Vector3IfE10QuaternionIfE5ColorfRK7Frustumb
#4 0x0006065c in _Z22renderSphere_CombinersRK10RenderInfoRK7FrustumRK9GLContext
#5 0x00063fa4 in _ZN8Renderer12renderObjectE6Point3IfEfd10QuaternionIfEff7Vector3IfE5ColorRNS_16RenderPropertiesE
#6 0x00065234 in _ZN8Renderer12renderPlanetERK4Body6Point3IfE7Vector3IfEffd10QuaternionIfEff
#7 0x0005e848 in _ZN8Renderer6renderERK8ObserverRK8UniversefRK9Selection
#8 0x0000ea20 in _ZN12CelestiaCore4drawEv
#9 0x00048e00 in -[CelestiaOpenGLView drawRect:]
#10 0x930809ec in -[NSView _drawRect:clip:]
#11 0x93096a34 in -[NSView _recursiveDisplayAllDirtyWithLockFocus:visRect:]
#12 0x930761e8 in -[NSView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#13 0x930765d0 in -[NSView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#14 0x930765d0 in -[NSView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#15 0x930c8224 in -[NSFrameView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#16 0x930b2c14 in -[NSThemeFrame _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#17 0x930963f4 in -[NSView _displayRectIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:]
#18 0x930a5778 in -[NSView displayIfNeeded]
#19 0x930b5608 in -[NSWindow displayIfNeeded]
#20 0x930a03b0 in _handleWindowNeedsDisplay
#21 0x90148464 in __CFRunLoopDoObservers
#22 0x901487d4 in __CFRunLoopRun
#23 0x90180f58 in CFRunLoopRunSpecific
#24 0x969a3b70 in RunCurrentEventLoopInMode
#25 0x969b3b00 in ReceiveNextEventCommon
#26 0x969dabbc in BlockUntilNextEventMatchingListInMode
#27 0x9308dedc in _DPSNextEvent
#28 0x930a0158 in -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:]
#29 0x930b1d88 in -[NSApplication run]
#30 0x9315fc58 in NSApplicationMain
#31 0x00002ac4 in _start (crt.c:267)
#32 0x00002944 in start
The strange thing is that, if I go to the Sun, the renderers switch fluently. I can then return to Earth and it works fine! However, as said, once I try to rotate the renderers modes when looking to any planet or moon, the application crashes! It works fine on stars and comets... Erased the preferences, but no luck!
Date/Time: 2003-06-20 20:24:46 +0300
OS Version: 10.2.6 (Build 6L60)
Host: Thanars-Computer.local.
Command: Celestia
PID: 481
Exception: EXC_BAD_ACCESS (0x0001)
Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000034
Thread 0 Crashed:
#0 0x0003e654 in _ZN7Texture7getNameEv
#1 0x3feff9d6 in 0x3feff9d6
#2 0x000317e4 in _Z20SetupCombinersSmoothR7TextureS0_5Colorb
#3 0x0005f5c4 in _Z16renderSmoothMeshRK9GLContextR7Texture7Vector3IfE10QuaternionIfE5ColorfRK7Frustumb
#4 0x0006065c in _Z22renderSphere_CombinersRK10RenderInfoRK7FrustumRK9GLContext
#5 0x00063fa4 in _ZN8Renderer12renderObjectE6Point3IfEfd10QuaternionIfEff7Vector3IfE5ColorRNS_16RenderPropertiesE
#6 0x00065234 in _ZN8Renderer12renderPlanetERK4Body6Point3IfE7Vector3IfEffd10QuaternionIfEff
#7 0x0005e848 in _ZN8Renderer6renderERK8ObserverRK8UniversefRK9Selection
#8 0x0000ea20 in _ZN12CelestiaCore4drawEv
#9 0x00048e00 in -[CelestiaOpenGLView drawRect:]
#10 0x930809ec in -[NSView _drawRect:clip:]
#11 0x93096a34 in -[NSView _recursiveDisplayAllDirtyWithLockFocus:visRect:]
#12 0x930761e8 in -[NSView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#13 0x930765d0 in -[NSView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#14 0x930765d0 in -[NSView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#15 0x930c8224 in -[NSFrameView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#16 0x930b2c14 in -[NSThemeFrame _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
#17 0x930963f4 in -[NSView _displayRectIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:]
#18 0x930a5778 in -[NSView displayIfNeeded]
#19 0x930b5608 in -[NSWindow displayIfNeeded]
#20 0x930a03b0 in _handleWindowNeedsDisplay
#21 0x90148464 in __CFRunLoopDoObservers
#22 0x901487d4 in __CFRunLoopRun
#23 0x90180f58 in CFRunLoopRunSpecific
#24 0x969a3b70 in RunCurrentEventLoopInMode
#25 0x969b3b00 in ReceiveNextEventCommon
#26 0x969dabbc in BlockUntilNextEventMatchingListInMode
#27 0x9308dedc in _DPSNextEvent
#28 0x930a0158 in -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:]
#29 0x930b1d88 in -[NSApplication run]
#30 0x9315fc58 in NSApplicationMain
#31 0x00002ac4 in _start (crt.c:267)
#32 0x00002944 in start
The strange thing is that, if I go to the Sun, the renderers switch fluently. I can then return to Earth and it works fine! However, as said, once I try to rotate the renderers modes when looking to any planet or moon, the application crashes! It works fine on stars and comets... Erased the preferences, but no luck!
HankR wrote:Looks to me like you results - system freezes (and similar reports by others) may be due to problems with the ATI drivers. Hopefully Apple will fix these eventually.
Well, here's an idea...
Since we've all been having problems with OpenGL core service one way or another since OS X came around, why don't you great people e-mail Celestia to Apple themselves and tell them to fix those bugs using it? Maybe they'll listen to you...