BOTF All in One installer 2.0.0 (Released 7-10-2020)

[Flocke]

That by some circumstances the game becomes unstable in later turns, is about normal with BotF. Usually it is some AI fleet order loosing it's target or another fleet related issue that can also be fixed by removing the ships.

I'd be interested to see if UE can identify and fix this with my improvements earlier this year, and if not, I'd be interested to take a look at the save as well. That a game becomes unplayable that late is a real pain, and when it is so reproducable we should be able to identify and fix the cause.

Another chance when you run into such issues btw is to rename and load the "last.sav" auto.sav backup of the turn before. That's specially helpful for multiplayer games, where auto.sav corruption happens more easily.

[Spocks-cuddly-tribble]

it crashes when I get deep in--maybe around turn 250 or so. But it isn't just a crash, it simply refuses to work with that saved game again no matter how many times I reboot.

Safest way is:

- Turning OFF the autosave feature (has more than one unsolved bug)

- manually usage of 3+ save slots, overwriting oldest of them each with current turn save

I know it sucks, but in event of corruption you lose 2 turns max (some bugs might corrupt current and the previous turn save).

[MAniX]

Hey guys - Happy New Year - I am a first time account but long time BOTF player (since release, the game has always been my nemesis to get running without crashes!).

I am currently running the 2.0 installer version with multiple mods using PlayOnLinux (under Wine), in Ubuntu 21.10. I have had success at getting the games installed. Having played these a lot under windows, linux and mac (via playonmac) I am versed in the usual crashes but I have a weird one I need to see if others have had (and any fixes?).

Issue:
Crash (total, sudden) when trying to access a system's resources page - game running in 1024x768 mode but other res affected

Symptoms:
Seems to affect only one or two systems, no real correlation to morale status or other indicators. No fleets nearby, no long system queue, happens both for subjugated and colonised systems. Using mouse right click access to resources, keyboard shortcuts or double click planets are all working on all non-affected systems but immediate crash when using any of those methods on affected system. Access to resource tab and all other systems will be possible. Once one system becomes 'inaccessible', others follow. I have noted that after 30-50 turns 'sometimes' the planet is accessible again but by then 2-3 more will not be. Not seen to affect homeworld.

Example:
Currently running under the v2.0 Delta Quad mod, normal difficulty, turn 90 noticed issue with subjugated Qo'nos, then with other owned planets. Has been replicated under Vanilla and RIF mods, irrespective of race chosen. Usually happens around 80-90 turns in.

Hardware:
Dell Inspiron 3511 16gb RAM, 13 11th gen, dual boot Win 11 (urgh), Ubuntu MATE 21.10, intel UHD integrated gfx 64mb, large 8gb swap file in place, game running in 1024x768 mode. Latest drivers and firmwares - opengl support included. the BOTF .ini has been changed by me to not display startup movie, otherwise is as per installed.

Logs:
Under WIne, botf debug log shows an odd behaviour prior to crash. On load, hashtabs around 13 (never more than 20, only occasionally 0), but when planet resource tab selected for affected system, leaps to 395 hastabs. It is almost as if it has a massive surge. Errors out to PlayOnLInux launch page with 'Error in Main'. This error has been replicated using both 2.0 installer i had (which is useful for vanilla in 800/600 on an old laptop of mine).

Solution:
None as yet - any ideas or help welcome. I have tried the following - reboot PC, reinstall game/PoL/wine, reconfigure wine, change wine from Windows 7 to Windows XP, save the file manually (no autosave). I have not been successful in getting the 2.0 installer to work on Windows 11 Home, and my only other machines are all Mac or Linux. Last known 'working' on a Mac about a year ago based on 1.0.1 installer. I could try the 1.0.1 installer but reluctant as older mods and more flaws - every attempt with 2.0 installer has resulted in above issues.

Any ideas guys? If anyone does still play this on PlayOnLinux / Wine, any config suggestions welcome, and anyone on Win11 Home, any feedback also interesting. And if this has been an issue for others - whatever OS - it might help me track the bug if you have had this and any solves you found along way.

[Flocke]

Happy New Year!

Well, I never tried to get it run on Linux or Mac, and I am still on Windows 8.1, but do you have any crash.log?
If not by the game, a crash dump might enlight us. In windows you find the application crash reports in the windows event viewer, with PlayOnLinux you debug the application instead to get a crash dump: https://www.youtube.com/watch?v=c10auEHA3JQ

Without, by the symptoms I can only guess that there is a graphic rendering issue, possibly related to the amount of build structures. I am surprised it doesn't work on windows 11 for you. What error do you get with it?

Did you try re-download the installer for in case the download got corrupted? Or better, did you check the MD5 hash of your installer download, is it B721B61FC85C6F4730FE7B27F1E0C275 ?


@SCT In single player I don't remember any vanilla autosave issues, and with a quick forum search I don't find any specific. Can you give some pointers? THX

[Spocks-cuddly-tribble]

In single player I don't remember any vanilla autosave issues, and with a quick forum search I don't find any specific. Can you give some pointers? THX

It's been a long time (using BotF 1.0.2 CD-Rom version), so I can't be certain the issue persists in current multi-installer (or with current OS's):

Autosave = ON increased (seemingly) random freezes/crashes and savegame corruptions of regular save slots. My guess was the game had problems with the handing of certain simultaneous actions/orders/inputs (in this case the background task of overwriting last & auto sav in kombination with certain player inputs).

After turnig OFF autosave these issues disapeared for me.

Except at the specific moment when some background music tracks looped back to the starting point, some fancy actions could cause similar issues (IIRC).


EDIT OT wrt simultaneous actions/orders/inputs: Just moving the mouse increased scrolling speed of the arrow keys in main gal map.

[MAniX]

Happy New Year!

Well, I never tried to get it run on Linux or Mac, and I am still on Windows 8.1, but do you have any crash.log?
If not by the game, a crash dump might enlight us. In windows you find the application crash reports in the windows event viewer, with PlayOnLinux you debug the application instead to get a crash dump: https://www.youtube.com/watch?v=c10auEHA3JQ

Without, by the symptoms I can only guess that there is a graphic rendering issue, possibly related to the amount of build structures. I am surprised it doesn't work on windows 11 for you. What error do you get with it?

Did you try re-download the installer for in case the download got corrupted? Or better, did you check the MD5 hash of your installer download, is it B721B61FC85C6F4730FE7B27F1E0C275 ?

Happy New Year. Thanks for replying. I have verified the MD% hash and there is no corruption. I have replicated the error on multiple laptops running Ubuntu and PoL - I would equally suspect a graphics issue since the installers only allow game to run if I resolve the known 32bit driver issue for OpenGL by installing libgl1-mesa-glx:i386 [reference: OpenGL 32bit error PlayOnLinux]

I have a crash.log (that is created by game) and also have a debug log from PlayOnLinux (its a bit large, covers multiple restarts). The logs are fairly benign except for the hasttab references which spike to 395 and I am guessing break some buffer size or crash the main calls wthin the program as a result - and given the constant variable being that it only affects a specific system and the resource tab only (all other aspects of gameplay unaffected), leads me to think something within the specific build queue or mod is causing it.

On that point - I have re-run the game to replicate. Praxis is a system I colonised around turn 34 and at turn 95 has most structures built. Currently (i am Species 8471) Practice Faciltiy is in build queue and because of the issue preventing me access to system is showing 381 turns to complete build. This hints at the error since 'hastab' in the below logs seems to link to the queue builds across systems? I cannot clear the queue without access to system. This issue affects other mods, but yes it would seem to correlate to the build queue and specific elements (though, which ones I have not yet figured out).

Reference Windows 11 errors causing me issues, note I have not done much more than run the installer and it seemed to crash on video load. I didnt do any more investigation eg modify stbotf.ini or anything. It may indeed run if I tweak some registry edits etc. If the issues persist in Linux I may attempt a re-load but I will see what I can do as I really prefer Linux over Windows.

Crash logs excerpt:

Code: Select all

[12/28/21 20:05:11] - Running wine- --version (Working directory : /usr/share/playonlinux/python)
wine-5.0.3 (Ubuntu 5.0.3-3ubuntu2)

PlayOnLinux logfile
-------------------
Date: 12/28/21 20:05:11

> PlayOnLinux Version
  4.3.4
> uname -a
  Linux manix-Inspiron-15-3511 5.13.0-22-generic #22-Ubuntu SMP Fri Nov 5 13:21:36 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux
> lsb_release -a
  
> wine --version
  wine-5.0.3 (Ubuntu 5.0.3-3ubuntu2)
> POL_WINEVERSION
  
> WINEPREFIX
  /home/manix/.PlayOnLinux//wineprefix/botf
> Distribution
  Ubuntu 21.10
> glxinfo \| grep rendering
  direct rendering: Yes
> glxinfo \| grep renderer
      GLX_MESA_query_renderer, GLX_MESA_swap_control, GLX_NV_float_buffer, 
    GLX_MESA_copy_sub_buffer, GLX_MESA_query_renderer, GLX_MESA_swap_control, 
Extended renderer info (GLX_MESA_query_renderer):
OpenGL renderer string: Mesa Intel(R) UHD Graphics (TGL GT2)
> OpenGL libs (Direct rendering testing)
  check_dd_x86 missing, test skipped
  check_dd_amd64 missing, test skipped

[12/28/21 20:05:12] - This is a 32bits prefix!
[12/28/21 20:05:12] - Running wine- cmd /c echo %ProgramFiles% (Working directory : /usr/share/playonlinux/python)
C:\Program Files
[12/28/21 20:05:29] - Running wine- /home/manix/Downloads/BOTF_All_in_One_2.0.0/BOTF_All_in_One_2.0.0.exe (Working directory : /)

C:\users\manix\Temp\is-U61H2.tmp\vanilla>BREAK OFF 
Can't recognise 'BREAK OFF' as an internal or external command, or batch script.

C:\users\manix\Temp\is-U61H2.tmp\vanilla>KIX32.EXE  mod.kix  

7-Zip (A) 9.04 beta  Copyright (c) 1999-2009 Igor Pavlov  2009-05-30

Scanning

Updating archive C:\botf\Vanilla\stbof.res
[REDACTED AS ALL SAME]
Compressing  wh_c.tga
Compressing  winlose.wdf
Compressing  yestxtop.tga
Compressing  yes_on.tga

Everything is Ok

C:\users\manix\Temp\is-U61H2.tmp\vanilla>if ERRORLEVEL 1 pause 

C:\users\manix\Temp\is-U61H2.tmp\vanilla>exit 

[REPEATS FOR EACH MOD INSTALLED - NO ERRORS]

Log file for last crash reads as follows:

Code: Select all

[01/01/22 21:24:16] - Running wine- trek.exe (Working directory : /home/manix/.PlayOnLinux/wineprefix/botf/drive_c/botf/DQM)
../src/intel/isl/isl.c:2095: FINISHME: ../src/intel/isl/isl.c:isl_surf_supports_ccs: CCS for 3D textures is disabled, but a workaround is available.
FOUND REGISTRY KEY: [SOFTWARE\MICROPROSE\STAR TREK: DQM]
hashtab has 13 items
hashtab has 13 items
hashtab has 13 items
hashtab has 13 items
hashtab has 13 items
hashtab has 13 items
hashtab has 0 items
hashtab has 0 items
hashtab has 0 items
hashtab has 395 items

and finally Crash.log for the specific mod has every crash listed and they are identical and as follows:

Code: Select all

-------------------------------------
ST:BOF Sat Jan  1 21:25:35 2022

Version Under Test: 72
File: guifile.c, Line: 131, funw.tga
Initialize State: 31
Player Empire: 2	Starting Seed: 1641068388
Galaxy shape: 0	Galaxy Size: 2

[thunderchero]

Example:
Currently running under the v2.0 Delta Quad mod, normal difficulty, turn 90 noticed issue with subjugated Qo'nos, then with other owned planets. Has been replicated under Vanilla and RIF mods, irrespective of race chosen. Usually happens around 80-90 turns in.

Delta Quad mod is released "as is" and from your crash log it is missing a file. (over 10 years and first report of issue)

to fix issue extract the fun.tga and rename funw.tga and re-add file to stbof.res with UE

[MAniX]

Example:
Currently running under the v2.0 Delta Quad mod, normal difficulty, turn 90 noticed issue with subjugated Qo'nos, then with other owned planets. Has been replicated under Vanilla and RIF mods, irrespective of race chosen. Usually happens around 80-90 turns in.

Delta Quad mod is released "as is" and from your crash log it is missing a file. (over 10 years and first report of issue)

to fix issue extract the fun.tga and rename funw.tga and re-add file to stbof.res with UE

SOLVED
Absolute legend, thunderchero, thank you!! - Yes, on reload of game, Praxis system queue now seems editable without crash. I will test further in this game (no issues at turn 137 so far) and others to ensure no additional system faults and will also keep an eye out on the other mods to see if this error repeats out on anything or does anything that is an issue on Linux.
//EDIT: Game completed in 537 turns on a large galaxy, no issues, only one random crash which was isolated to a very large fleet movement (81 ship simultaneous attack). I will call this solved.

Not at all suprised in this error not being seen in 10 years - just my kind of luck. I have owned a copy of BOTF since 1999 (about a month after launch), but spent the rest of 1999 trying to get it to run consistently on a shaky intel 486DX2 processer running 66MHz overclocked to 90MHz and some 12mb mismatched RAM. My woes have never really changed since then - every few years I return to the game and have to battle to get it to work in new environments!

Solution in full for reference (based on Linux but may also be relevant on other OS's):
- using UE (under Linux), for anyone in future, download the UE from the forum and extract to a folder outside the PlayOnLInux folder
- run using

Code: Select all

java -jar UE.jar

- my above code is a Linux specific command line execution to run UE (MS windows can run as executable or with bat file at command line), if you need Java on Debian based systems (should be as standard) try

Code: Select all

sudo apt install default-jre

- you can verify java install before and after with

Code: Select all

java --version

- click Open in UE, and nav to stbof.res
- click Edit > Other > Files. Search and Extract fun.tga,
- rename extracted fun.tga to funw.tga in file explorer of choice,
- in UE, click Add, and save (compiles), on Exit, updates file and made changes to trek.exe.

Also, for future users of PlayOnLInux and specifically using BOTF under Ubuntu 20.04 and 21.10, I have noted in a prev post about the OpenGL 32bit drivers needed to resolve or error will be a fatal crash with screen resizing to resolution but not loading game (system will need force reboot in that case FYI). Movie on start can cause crashes so I also removed from stbof.ini to improve loading.

Resolve the known 32bit driver issue for OpenGL by installing libgl1-mesa-glx:i386 [reference: OpenGL 32bit error PlayOnLinux]

Hopefully some of the above will assist future resolutions and thanks to all for the welcome to the forum and assistance.

[MarkXXX]

That by some circumstances the game becomes unstable in later turns, is about normal with BotF. Usually it is some AI fleet order loosing it's target or another fleet related issue that can also be fixed by removing the ships.

I'd be interested to see if UE can identify and fix this with my improvements earlier this year, and if not, I'd be interested to take a look at the save as well. That a game becomes unplayable that late is a real pain, and when it is so reproducable we should be able to identify and fix the cause.

Another chance when you run into such issues btw is to rename and load the "last.sav" auto.sav backup of the turn before. That's specially helpful for multiplayer games, where auto.sav corruption happens more easily.

Hi, here's the save! I'd be interested in hearing any theories you might have after you look at it.

[MarkXXX]

when posting a reply, if you look below the submit button you should see a "attachment" tab. select that tab and upload file.

Thanks! Here's the save. Sorry I'm so late with this, work has been a bear...

[MarkXXX]

when posting a reply, if you look below the submit button you should see a "attachment" tab. select that tab and upload file.

Thanks! Here's the save. Sorry I'm so late with this, work has been a bear...

Whoops, let me try that again!

[MarkXXX]

it crashes when I get deep in--maybe around turn 250 or so. But it isn't just a crash, it simply refuses to work with that saved game again no matter how many times I reboot.

Safest way is:

- Turning OFF the autosave feature (has more than one unsolved bug)

- manually usage of 3+ save slots, overwriting oldest of them each with current turn save

I know it sucks, but in event of corruption you lose 2 turns max (some bugs might corrupt current and the previous turn save).

Thanks for the advice, I'll try this. It's funny, I finished a game a few days ago that went over 1000 turns, no problem. But then some other much shorter games just crash.

[thunderchero]

Thanks! Here's the save. Sorry I'm so late with this, work has been a bear...

Just tested your saved game went 3-4 turns without crash, did you do something during that turn?

running integrity check give no info to error.

[Mudpuppy630511]

What is the "Galaxy Mod" and what does it do? I've been considering adding it, but I'd like to know about it first before I do so. Thanks

[thunderchero]

What is the "Galaxy Mod" and what does it do? I've been considering adding it, but I'd like to know about it first before I do so. Thanks

best would be to check this post.

viewtopic.php?p=15819#p15819