Big Bug thread for 0.82.2/3 - Report bugs here - Page 4

Big Bug thread for 0.82.2/3 - Report bugs here

Discuss the source code and development of Spring Engine in general from a technical point of view. Patches go here too.

Moderator: Moderators

User avatar
hoijui
Former Engine Dev
Posts: 4344
Joined: 22 Sep 2007, 09:51

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by hoijui »

why jpeg?
png is the right thing for that (accuracy, detail). that is why we have png as default format for spring screen-shots.
User avatar
Jazcash
Posts: 5309
Joined: 08 Dec 2007, 17:39

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by Jazcash »

hoijui wrote:why jpeg?
png is the right thing for that (accuracy, detail). that is why we have png as default format for spring screen-shots.
Because it took me 1 minute and 24 seconds to load all the images.
Last edited by Jazcash on 20 Aug 2010, 23:28, edited 1 time in total.
User avatar
FLOZi
MC: Legacy & Spring 1944 Developer
Posts: 6240
Joined: 29 Apr 2005, 01:14

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by FLOZi »

Units using spheres rather than custom vols should just read the radius defined in the s3o, if someone broke that they are a tremendous idiot.
User avatar
Thor
NOTA Developer
Posts: 291
Joined: 05 Mar 2006, 10:26

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by Thor »

didn't think to check s3o's. They seem to be the same. So I guess it's just 3do.
User avatar
FLOZi
MC: Legacy & Spring 1944 Developer
Posts: 6240
Joined: 29 Apr 2005, 01:14

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by FLOZi »

Thor wrote:didn't think to check s3o's. They seem to be the same. So I guess it's just 3do.
I didn't consider 3dos (as its a silly outdated format noone has any reason to use :mrgreen: )

Guess something did change then; seems silly to change that behaviour after such a long time. Maybe it was part of the refactor of model loading.
Gouken
Posts: 35
Joined: 18 Aug 2010, 22:55

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by Gouken »

Another quite common bug since update - People respawn in a random place when game starts even though they clicked somewhere else

Example:
http://replays.adune.nl/?2709
Gouken
Posts: 35
Joined: 18 Aug 2010, 22:55

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by Gouken »

Another bug which isn't on your list is the fact that all replays are desynced, watching replays is pretty much broken so its quite serious

whole replay site is evidence

only happens since this update
Klopper
Posts: 146
Joined: 10 Jan 2007, 14:31

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by Klopper »

2 most likely ATI related bugs with graphics settings (using Radeon 4870 with Catalyst 10.7 Beta OpenGL ES 2.0 drivers):

-atioglxx.dll driver crash if you have dynamic clouds and high-res clouds activated AT THE SAME TIME (works fine if you have only one of the two settings activated, enabling both worked fine in last Spring version)

-if using bump-mapped water, using shorewaves ONLY works (fine, as in previous Spring version) on maps WITH water, but will cause the driver crash for maps WITHOUT water

Confirming most of the other bugs for me, especially pathfinding, replay desync, bright factories in idle builders widget, unit highlighting widgets turning great parts of the screen green and invisible bomber projectiles.
One more POSSIBLE thing (which i will have to test a bit to confirm or deny though): In several BA games i THINK i have noticed nano turrets (on roam) sometimes not properly executing the patrol command properly (like in previous Spring versions) and getting kinda freezed instead. Can someone confirm/deny that (could also have been lag or whatever)?
I would check replays but since they desync they currently can't be trusted...
User avatar
hoijui
Former Engine Dev
Posts: 4344
Joined: 22 Sep 2007, 09:51

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by hoijui »

Klopper, the problem you describe with nano turrets could be a combination of:
* nano turret not always turning correctly
* max particles reached -> it is not visible that nano turret does something.
User avatar
Jazcash
Posts: 5309
Joined: 08 Dec 2007, 17:39

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by Jazcash »

I managed to trim down the minimap bug a bit. The screen fadeflashes green whenever there are any nano particles showing on the map whilst a minimap widget is enabled. So anything such as restoring, reclaiming, building or anything that shows nanos particles would trigger this.
User avatar
CarRepairer
Cursed Zero-K Developer
Posts: 3359
Joined: 07 Nov 2007, 21:48

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by CarRepairer »

CarRepairer wrote:Projectile models are not visible anymore for weapons of the type 'cannon'. They were visible before the update.
Does this issue apply to my above quoted bug? I notice there is no mention of "cannon" in it, only missile.

http://springrts.com/mantis/view.php?id=2030
zerver
Spring Developer
Posts: 1358
Joined: 16 Dec 2006, 20:59

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by zerver »

CarRepairer wrote: Does this issue apply to my above quoted bug? I notice there is no mention of "cannon" in it, only missile.
http://springrts.com/mantis/view.php?id=2030
Fixed http://github.com/spring/spring/commit/ ... 9efc369883
User avatar
very_bad_soldier
Posts: 1397
Joined: 20 Feb 2007, 01:10

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by very_bad_soldier »

You probably already know about this crash, but just to be sure:
I get this one here quite regularly usually in the first minutes, seems to be related to aiInterfaceCallback_DataDirs_getWriteableDir().

http://pastebin.com/XFhhwNuj
User avatar
hoijui
Former Engine Dev
Posts: 4344
Joined: 22 Sep 2007, 09:51

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by hoijui »

i would say it is probably rather this:
[ 2526] Set water rendering mode to 1 (reflective)
[ 2629] Set Shadows to 1
[ 2752] Set water rendering mode to 3 (reflective&refractive)
[ 2864] Set water rendering mode to 4 (bumpmapped)
[ 2864] [BumpWater] water-shader compilation error: Vertex info
-----------
0(88) : error C1013: function "main" is already defined at 0(88)

Fragment info
-------------
0(120) : error C1038: declaration of "reftexcoord" conflicts with previous declaration at 0(120)
0(133) : error C1013: function "convertDepthToZ" is already defined at 0(133)
0(179) : error C1013: function "main" is already defined at 0(179)
[ 2864] Spring 0.82.3.0 (0.82.3) has crashed.
[ 2864] Exception: Access violation (0xc0000005)
[ 2864] Exception Address: 0x14a40538
as the crash happened in the same frame in which you switched to bumpmapped.

aiInterfaceCallback_DataDirs_getWriteableDir is just the last exported symbol of spring.exe, so it just shows a lot of/most stuff after that as an offset to that symbol (my guess). it surely is not involved an the crash.
User avatar
very_bad_soldier
Posts: 1397
Joined: 20 Feb 2007, 01:10

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by very_bad_soldier »

Ok thanks, I will try to disable FPSManager widget, that should stop that water switching.
I get that shader error nearly every game but I usually get away with it without crashing.

EDIT:
Regarding the function addresses: So the stack is corrupted?
User avatar
hoijui
Former Engine Dev
Posts: 4344
Joined: 22 Sep 2007, 09:51

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by hoijui »

no, does not mean the stack is corrupted. it just makes no sense that addr2line does use this symbol in translation. do not know the details why it gets used.
Last edited by hoijui on 01 Sep 2010, 08:50, edited 1 time in total.
User avatar
Beherith
Posts: 5145
Joined: 26 Oct 2007, 16:21

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by Beherith »

A really really wierd bug: dragon teeth randomly disappear.

Edit: Correction, they are still visible from far view (billboards) but not visible up close. (from spectator mode)
User avatar
very_bad_soldier
Posts: 1397
Joined: 20 Feb 2007, 01:10

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by very_bad_soldier »

very_bad_soldier wrote:You probably already know about this crash, but just to be sure:
I get this one here quite regularly usually in the first minutes, seems to be related to aiInterfaceCallback_DataDirs_getWriteableDir().

http://pastebin.com/XFhhwNuj
I still get this one in 0.82.5 even without switching the water renderer.
User avatar
SpliFF
Posts: 1224
Joined: 28 Jul 2008, 06:51

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by SpliFF »

Been fighting a lot with MinGW builds. Seems pretty flakey in general.

I'm trying now with MinGW 4.5.0 (from the installer) and I get a strange error, not sure what to make of it

Code: Select all

g++.exe: CreateProcess: No such file or directory
process_begin: CreateProcess(C:\Temp\make2784-1.bat, C:\Temp\make2784-1.bat, ...
) failed.
make (e=5): Access is denied.
mingw32-make[2]: *** [spring-multithreaded.exe] Error 5
mingw32-make[1]: *** [rts/builds/GML/CMakeFiles/spring-multithreaded.dir/all] Er
ror 2
process_begin: CreateProcess(D:\Apps\CMake\bin\cmake.exe, D:\Apps\CMake\bin\cmak
e.exe -E cmake_progress_report D:\SpringTest\debug\CMakeFiles, ...) failed.
make (e=5): Access is denied.
mingw32-make[2]: *** [rts/builds/HL/CMakeFiles/spring-headless.dir/__/__/Game/UI
/KeyBindings.cpp.obj] Error 5
mingw32-make[2]: *** Waiting for unfinished jobs....
[ 25%] mingw32-make[2]: *** [rts/builds/default/CMakeFiles/spring.dir/__/__/Game
/UnsyncedGameCommands.cpp.obj] Error 1
mingw32-make[2]: *** Waiting for unfinished jobs....
I've turned off parallel building to see if the processes are fighting for the same files and I'm deactivating the "Program Guard" feature of Online Armour but if anyone knows why this might be happening let me know.

What I do know is the disk isn't full and the temp directory is writable.
imbaczek
Posts: 3629
Joined: 22 Aug 2006, 16:19

Re: Big Bug thread for 0.82.2/3 - Report bugs here

Post by imbaczek »

bogus createprocess failures may mean several things, most often:
- command line too long (had that problem last time i tried to rebuild spring)
- path too long (there are some workarounds for that IIRC, but I don't know whether they work with mingw)
Post Reply

Return to “Engine”