Hard Light Productions Forums
Modding, Mission Design, and Coding => FS2 Open Coding - The Source Code Project (SCP) => Test Builds => Topic started by: taylor on June 02, 2005, 03:11:30 am
-
Changes from 20050530 build:
- all known OGG and streaming audio errors fixed for both OpenAL and DirectSound versions. Successfully tested with TBP 3.2 beta
- better error handling in OGG code to fail gracefully with multiple logical bitstreams
- proper Game_mode switching in the barracks to prevent single player campaign savefiles getting overwritten when initially in multi mode
- fix simulated speech in debrief where the full text would get repeated when you clicked on "Recommendations"
- better audio handling when game is minimized. It should now properly shut off sounds when minimized and pick them back up when you restore the game window. Some sound, such as on the mainhall, will have to finish playing before it actually stops but it will stop. Command briefings will pause audio and anis and continue them rather than restarting. Briefings should pause voice and briefing, not to continue until you restore window and finish with voice. If any of this is wrong or any screens still have audio (music, voice, sim speech, ambient sound) please report it.
- another attempt at the support ship docking crash. This fix may cause other problems so please verify that you don't have any docking issues with mods or support ships.
- error handling fix for jumpnodes which don't model_load() properly (WMCoolmon)
- fix spawn missiles locking on protected ships (WMCoolmon)
http://icculus.org/~taylor/fso/willrobinson/20050602-win32.rar (12meg, UPDATED with spawn missile fix)
http://icculus.org/~taylor/fso/willrobinson/20050602-win32r.rar (1.3meg, Release build only - DirectSound)
http://icculus.org/~taylor/fso/willrobinson/20050602-win32r-OAL.rar (1.3meg, Release build only - OpenAL)
That covers the main stuff. Report any problems. (immediately assumes this thread will be filled with redmenace posts ;))
While testing, please verify these bugs as fixed or not:
Spawn missiles lock on protected ships (http://mgo.maxgaming.net/mantis/bug_view_page.php?bug_id=0000432)
Slowdown in user interface sections with anis (http://mgo.maxgaming.net/mantis/bug_view_page.php?bug_id=0000419)
"Next" arrow not working (http://mgo.maxgaming.net/mantis/bug_view_page.php?bug_id=0000259)
Escape ignored in briefing (http://mgo.maxgaming.net/mantis/bug_view_page.php?bug_id=0000416)
-
Taylor, could we possibly see a FS2_open_R and anything neccessary to run that package alongside the proper total release? 11 Mb is murder on a 56ker
-
Originally posted by "me"
- fix spawn missiles locking on protected ships (WMCoolmon)
Crap! Looks like I forgot to update that file so this change is in fact missing from the build. I'm running off a new build set now and will update it shortly but anyone who already has the build will not have this change.
Originally posted by phatosealpha
Taylor, could we possibly see a FS2_open_R and anything neccessary to run that package alongside the proper total release? 11 Mb is murder on a 56ker
Sure, I'll add a link to that in a little while.
-
I found a couple of issues with Apocalypse.
-
I had a crash during The Roman's Blunder when I was turning to look at the fenris.
-
I went ahead and tried it again. Same result.
-
Take the spaces out of your filenames redmenace. It downloads correctly as "the" (by some miracle) but has no extension and so doesn't associate with a zip by default.
-
Here is the second one again
-
Not that I want to pressure you, but the fenris bug pretty much hampers my testing since many missions have fenris and Leviathans in them.
-
Originally posted by redmenace
Not that I want to pressure you, but the fenris bug pretty much hampers my testing since many missions have fenris and Leviathans in them.
I've tried to recreate this without success. Looks to just be driver related though and as I have an NVIDIA card it may not ever happen for me. There is likely something in the code triggering this but I don't know what that might be. What are your cmdline options? Also try deleting your cruiser01 and cruiser01x IBX files in case it's a data problem hitting some VBO issue. And for good measure I'll say "driver upgrade" too just in case but I don't know that it's actually a driver problem.
-
Well I will update the drivers and remove the IBX files from the cache.
-spec -glow -jpgtga -2d_poof -rlm -ship_choice_3d -targetinfo -orbradar -3dwarp -smart_shields -noparseerrors -decals -fps -stats -show_mem_usage -allslev -mod SCP
-
Wow, it was the drivers. I was using drivers to test d3d that worked with the shine maps.
-
Also here is an additional fs.log for you to look through concerning the bmpman error.
-
Apocalypse
I had killed the rest of the bombers and headed to the node and engaged my jump drive with 10 seconds on the clock.
Then this happened.
Assert: Viewer_obj
File: u:\src\cvs\fs2_open.testing\code\freespace2\freespace.cpp
Line: 1950
Call stack:
------------------------------------------------------------------
game_render_frame() game_frame() game_do_frame() game_do_state() gameseq_process_events() WinMainSub() WinMain() WinMainCRTStartup() kernel32.dll 7c816d4f()
------------------------------------------------------------------
This looks like something related to something I did, I remember setting Viewer_obj to NULL for free-camera stuff.
-
You might want to look into this one as well.
-
Something I've noticed with this and other more recent builds is absolutely terrible nebula performance. My fps are cut by more than half in some cases. I can use an older build like fs2_open_T-20050103.exe and have my fairly steady 85 fps and then use this latest one and get 30ish.
-
It is being looked at. It is isolate in D3D. Try running Open GL.
-
The Shards of death have returned.
-
Hm, I did wonder why my 6800GT was struggling with nebula missions so much but since you gents already know about it I'll shut up and let you get on with it. :)
-
Originally posted by redmenace
The Shards of death have returned.
For whatever ship or ships you thing is causing this, please send me the IBX files. I'd like to check them against what I have and see if I can figure out exactly what's wrong and if there is a way to guard against it. I am going to add a new IBX debug line to the log which will print out the checksum of the model and the IBX file. If I have to go so far as build a database of valid IBX file checksums then this will let me.
I still wish more people would distribute them. I've got custom MediaVPs with most of the hi-poly ships in them and everything has an IBX but since it's not official I don't publicly distribute those. I do have one volition_ibx.vp though which has IBX files every one of the original game models. I'll at least get that posted to the SCP download site this weekend and see if people use it. As far as I know, every IBX file in it is valid.
-
It might not be related to the IBX files. But here, I will post them and some more images.
http://fs2source.warpcore.org/temp/TheSixthWonder.zip
http://fs2source.warpcore.org/temp/cruiser01.zip
I am not sure which ibx files are for the manticore and triton.
-
Looks like something's screwed up with the radar on that Leviathan, because the primary hull looks ok most of the time. And the Triton is Freighter2t-01, don't know about the Manticore.
-
Freaky. The cruiser01.ibx is using the same model as my IBX file but yours has 6 more verts in it. :wtf: Same thing for cruiser01x.ibx. I deleted and regenerated my IBX files for those models about 4 times and always got the same results so my count should be correct. The problem must be something in find_first_index(). Either it's doing something wrong or is not accounting for something model related.
Of course I've been saying that for months so perhaps I should stop with the broken record imitation and do something about it... nevermind the bug is already assigned to Bobboau. :p :)
-
I thought bob said he doesn't have any bugs assigned to him?
-
Actually he said that all his bugs were driver and hardware related.
-
Bob?
-
Which bug report is it BTW?
-
http://mgo.maxgaming.net/mantis/bug_view_page.php?bug_id=0000412
Also related to: http://mgo.maxgaming.net/mantis/bug_view_page.php?bug_id=0000224 I believe. That bug is closed but it's not fixed. The IBX code just hides the fact that it will still hit that Assert(). It's been reported to me numerous times since but I didn't know how to properly fix it and didn't think anyone else was going to so I left it closed.
Through debugging (by myself and fizz) it looks like the verts can be NaN in some cases but I'm not sure if that's a model problem or a model reading problem. Since it's usually correct I assume that it's a model reading problem, or something isn't getting checked. There are near zero safety checks in the generation code to guard against processing crappy data.
-
Well then I will leave it in your collective hands. I will just have to continue to bug test. Although I am not sure that this shows up in D3D.
-
I have yet to see this in D3D.
-
I am posting an additional log concering the bmpman error.
-
Originally posted by redmenace
I have yet to see this in D3D.
Probably just a difference in the API and how it handles rendering the buffers. Doesn't change the fact that the code is wrong somehow though.
I am posting an additional log concering the bmpman error.
I haven't been able to recreate any of these so far. This is happening when you're going from one mission to the next right? The only similar error that I've seen is with the pause window popup in multiplayer but that's been fixed (not in CVS yet).
-
No this is during the mission.
-
Taylor, I got a chance test TBP 3.2 Beta today. I was playing Raider Wars 2.0 with OGG voice acting. I noticed that for Command Briefings, Mission Briefings, and Debriefings, the audio for each stage will repeat the first 3/4 of a second of the voicefile, then omit the last 3/4 of a second of it. Note, the 3/4 of a second figure is approximate.
-
The bmpman error also has occurred ub the breifing when I selected a enemy wing to see what they are. I think it might have been the render window that crashed it. But this does not occur during loading but only after playing a few mission in a row.
-
If there is anything you need to know just let me know.
And to clarify. The only time BMPMAN error occurs only after player a couple missions one after the other and then either selecting an icon in the briefing window(although not always) or in middle of a mission.
-
I still get my ANI bug taylor.
-
Originally posted by Trivial Psychic
Taylor, I got a chance test TBP 3.2 Beta today. I was playing Raider Wars 2.0 with OGG voice acting. I noticed that for Command Briefings, Mission Briefings, and Debriefings, the audio for each stage will repeat the first 3/4 of a second of the voicefile, then omit the last 3/4 of a second of it. Note, the 3/4 of a second figure is approximate.
One change to this, after some further testing with this build, I was incorrect in stating that Debriefings are effected by this. This has been Mantis'ed.
-
Originally posted by Trivial Psychic
One change to this, after some further testing with this build, I was incorrect in stating that Debriefings are effected by this. This has been Mantis'ed.
Ok. I'm working on this now though it doesn't seem to matter what build (DS or OAL) or where (de/briefing, cmdbriefing, etc.) the OGG gets played, it always repeats the first little bit. I put a re-encoded briefing voice on the mainhall music and it still repeats so it is specific to OGG streaming but not specific to where it's played from or what build it's on.
I'm going to try and get this fixed before I release a test build later tonight.
-
Happens to me ingame with some music I have converted to save space (I edited the table but didn't touch the sample size stuff so it could be that)