Messages

1

Bugs prior to release 2.4 / Re: few visiual problems

« on: October 29, 2010, 09:38:36 pm »

Hello,

also in this version i think the SDL.dll bug from 2.3 has returned, after alt+tab or ctrl+esc the mouse get stuck in the corner and need to use the keyboard to end the task.

Each version of UFO:AI brings its own SDL.dll, so you have to replace the 2.4 with the SDL.dll from the 7z I provided in the well-known thread. Mattn did not put the SDL.dll from the thread to master, I think.

Regards

2

Newbie Coding / Re: (log) Actually got new build system to (partially) work on Windows 7

« on: October 25, 2010, 08:35:56 pm »

try make -r

Thank you, that worked fine. It's still 10 secs, but that seems to be because of my "old" hardware.

3

Newbie Coding / Re: (log) Actually got new build system to (partially) work on Windows 7

« on: October 23, 2010, 08:42:23 pm »

Hello,

just a small improvement - see attached patch. I also want to get rid of the first tasks, I want to implement a caching function for all tasks in Makefile and friends, which take a long time (file system searches, git check, etc).

Just to explain why I want to do this, please correct me, if I'm not right with that: I use Mingw/Msys to compile on Windows 7. I just type make in msys shell and this is the time it takes for doing nothing:

Code: [Select]

$ time { make; }
make: Nothing to be done for `all'.

real    0m29.291s
user    0m9.715s
sys     0m18.212s

I see there are some time consuming tasks such as find and I love to see that tasks are only done if I want to do it, for example after a make clean. How long does a build on Linux or other machines takes of there is nothing to do?

Best regards

4

Newbie Coding / Re: (log) Actually got new build system to (partially) work on Windows 7

« on: October 20, 2010, 04:18:22 am »

Hello,

I have a question to this: Is it possible (or generally a good idea) to put dependency task into its own make target to provide a faster build task? Each time I start a incremental build the dependency is recalculated even if I changed nothing on headers.

If it is an own task its up to the programmer to rebuild the dependencies during each build.

Just an opinion.

Regards

5

Sounds and Music / Re: Concept Music

« on: October 18, 2010, 08:17:54 pm »

Nice track. Sounds very relaxing.

What tools do you use to create that?

6

Newbie Coding / Re: (Probably) stupid question about the maps

« on: October 18, 2010, 04:20:56 pm »

Thank you very much, it keeps on compiling now with -quant 4. Sorry, that I don't read the page until end. I have to read the wiki more precisely.

Regards

7

Newbie Coding / Re: (Probably) stupid question about the maps

« on: October 18, 2010, 03:25:30 pm »

Now I tried to build the maps using 'make maps'. It took a very long time but finally results in an error (see attached) in a_hangar.map.

Could someone tell me, what I can do to make it work? Or provide the complete map bsp's somewhere?

Best regards

8

Artwork / Re: Solution - Rendering large city for animation video cut-scenes

« on: October 18, 2010, 02:34:00 pm »

Looks really great to me. Is there a tutorial about such things?

Regards

9

Newbie Coding / Re: (Probably) stupid question about the maps

« on: October 18, 2010, 01:01:33 pm »

Sorry for self-reply but I just want to attach of the refetch for map-get command, which I tried to perform because of the errors described above.

10

Newbie Coding / (Probably) stupid question about the maps

« on: October 18, 2010, 12:42:09 pm »

Hello,

maybe anyone could answer this question: I want to test some things in the new 2.4-dev and have many problems which could not be the case because anyone else would write "I have this, too", because they are essentially for the game.

I have some problems regardung the calculation of distances and points on the geoscape but they are not reproducable (it seems so). They occur many times on my PC but if I make a save and provide this to another guy, the problem seems to not exist. Also, if a reload my game the problem is gone away, but it will happen at any other timeframe later.

Is it possible, that I forgot something to build?

I explain, what I have done:

- Checkout from git
- Do some changes in the cl_input.c (for example)
- Compile & Link (build) the ufo.exe as well as all other targets which are build using "all"
- Start the new .exe => got problems

Has someone an explanation for that? I can only believe, that I got something wrong.

Best regards

PS: I mean for example this log entries:

Code: [Select]

2010/10/18 12:30:07ExecuteString: ''
2010/10/18 12:30:07ExecuteString: 'map day +ufocrash craft_crash_fighter'
2010/10/18 12:30:07Set server state to 1
2010/10/18 12:30:07------- Loading game.dll -------
2010/10/18 12:30:07not found at '/usr/local/lib'
2010/10/18 12:30:07Failed loading /usr/local/lib/game.dll: Das angegebene Modul wurde nicht gefunden.

2010/10/18 12:30:07not found at 'C:\Users\Maik\AppData\Roaming\UFOAI/2.4-dev/base'
2010/10/18 12:30:07Failed loading C:\Users\Maik\AppData\Roaming\UFOAI/2.4-dev/base/game.dll: Das angegebene Modul wurde nicht gefunden.

2010/10/18 12:30:07found at './base'
2010/10/18 12:30:07==== InitGame ====
2010/10/18 12:30:07SpawnServer: +ufocrash
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_ufo
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_crash_fighter
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_drop
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_drop_firebird
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_ufo
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_crash_fighter
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_drop
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_drop_firebird
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_ufo
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_crash_fighter
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_drop
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_drop_firebird
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_ufo
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_crash_fighter
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_drop
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_drop_firebird
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_ufo
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_crash_fighter
2010/10/18 12:30:07SV_GetCvarToken: cvar replacement: rm_drop
2010/10/18 12:30:07SV_ParseAssembly: cvar replacement value: +craft_drop_firebird
2010/10/18 12:30:07numTiles: 23
2010/10/18 12:30:07numAssemblies: 5
2010/10/18 12:30:07Use assembly: 'craft_crash_fighter'
2010/10/18 12:30:07tiles: -ufocrash/uc_ +craft_drop_firebird +craft_crash_fighter +d01 +h07 +h01 +h05 +h01 +h02 +h02
2010/10/18 12:30:07pos: 8 -8 0 -24 0 0 16 8 0 -24 -16 0 -24 16 0 0 -16 0 8 -16 0 0 0 0 24 -16 0
2010/10/18 12:30:07tiles: 9
2010/10/18 12:30:07CM_LoadMap: "-ufocrash/uc_ +craft_drop_firebird +craft_crash_fighter +d01 +h07 +h01 +h05 +h01 +h02 +h02" "8 -8 0 -24 0 0 16 8 0 -24 -16 0 -24 16 0 0 -16 0 8 -16 0 0 0 0 24 -16 0"
2010/10/18 12:30:07CM_LoadMap: token: +craft_drop_firebird
2010/10/18 12:30:07CM_AddMapTile: ufocrash/uc_craft_drop_firebird at 8,-8,0
2010/10/18 12:30:07********************
2010/10/18 12:30:07ERROR: Couldn't load maps/ufocrash/uc_craft_drop_firebird.bsp
2010/10/18 12:30:07********************
2010/10/18 12:30:07Shutdown server: Server crashed.
2010/10/18 12:30:07Set server state to 4
2010/10/18 12:30:07Shutdown the game thread
2010/10/18 12:30:07==== ShutdownGame ====
2010/10/18 12:30:07Free inventory slots in game on shutdown: 0
2010/10/18 12:30:07Unload the game library
2010/10/18 12:30:07ExecuteString: 'music_change geoscape'
2010/10/18 12:30:07music change to karlmacklin_geoscape (from Crystan-Geosphere04)
2010/10/18 12:30:07ExecuteString: ''
2010/10/18 12:30:07Mission is cat2_interest32_0, 2
2010/10/18 12:30:07ExecuteString: '-leftmouse 200 206105'
2010/10/18 12:30:07ExecuteString: '-select 200 206105'
2010/10/18 12:30:07Mission is cat2_interest32_0, 2
2010/10/18 12:30:07Mission is cat2_interest32_0, 2
2010/10/18 12:30:07Mission is cat2_interest32_0, 2
2010/10/18 12:30:07Mission is cat2_interest32_0, 2
2010/10/18 12:30:07Mission is cat2_interest32_0, 2
2010/10/18 12:30:07Mission is cat2_interest32_0, 2
2010/10/18 12:30:08Mission is cat2_interest32_0, 2
2010/10/18 12:30:08Mission is cat2_interest32_0, 2
2010/10/18 12:30:08Mission is cat2_interest32_0, 2
2010/10/18 12:30:08Mission is cat2_interest32_0, 2
Here the game keeps up on geoscape, I can reproduce this all of the time.

11

Coding / Re: Question to the Linux users

« on: October 17, 2010, 11:58:26 am »

You are right, it has nothing to do with it. It was just some q&a thinking. You told already, that it works, to load linux savegames in windows.

I thought, it could be the case, that there is a problem, because if I load my own windows savegame, I do not have such entries in log.

Anyways, topic closed.

12

Coding / Re: Question to the Linux users

« on: October 16, 2010, 12:28:55 pm »

So you think its basically a 2.3 => 2.4 problem?

Finally, I ask me, why its not possible to load 2.3 in 2.4-dev. What do you think about an importing method?

Has anyone a 2.4 Linux save game to add on this thread, please?

13

Coding / Question to the Linux users

« on: October 14, 2010, 09:50:54 pm »

Hello,

I tried to load a Linux Savegame in a Windows version. I got this log entries:

Code: [Select]

2010/10/14 21:40:27XML Written to buffer (2743397 Bytes)
2010/10/14 21:40:27ExecuteString: ''
2010/10/14 21:40:27ExecuteString: 'map day +ufocrash craft_crash_bomber'
2010/10/14 21:40:27Set server state to 1
2010/10/14 21:40:27------- Loading game.dll -------
2010/10/14 21:40:27not found at '/usr/local/lib'
2010/10/14 21:40:27Failed loading /usr/local/lib/game.dll: Das angegebene Modul wurde nicht gefunden.

2010/10/14 21:40:27not found at 'C:\Users\Maik\AppData\Roaming\UFOAI/2.4-dev/base'
2010/10/14 21:40:27Failed loading C:\Users\Maik\AppData\Roaming\UFOAI/2.4-dev/base/game.dll: Das angegebene Modul wurde nicht gefunden.

2010/10/14 21:40:27found at './base'
2010/10/14 21:40:27==== InitGame ====
2010/10/14 21:40:27SpawnServer: +ufocrash
Because it is a Savegame from 2.3 I will not think much about it. But: Has anyone a savegame from Linux for me to test this in 2.4-dev? I want to know, whether it is possible or not to load Linux saves in Windows. If this not possible, I will write a feature request or fix this by myself. I think it should be possible to provide a good debugging opportunity.

Best regards

14

Bugs prior to release 2.4 / Re: translating executable addresses to file:line in backtrace - addr2line

« on: October 14, 2010, 08:36:42 pm »

Here is an initial draft version of the backtrace in windows for an unhandled exception (segfault, e.g.).

Please apply patch in ascending order (0001, 0002)

Tested by adding following lines somewhere in code:

Code: [Select]

int *i = NULL;
*i = 4;
This caused the following backtrace in ufoconsole.log:

Code: [Select]

2010/10/14 20:19:48
----- network initialization -------
2010/10/14 20:19:48libcurl/7.21.1 zlib/1.2.3 initialized.
2010/10/14 20:19:48
------ server initialization -------
2010/10/14 20:19:48mapcycle add: '+africa small' type '1on1'
2010/10/14 20:19:48mapcycle add: '+frozen small' type '1on1'
2010/10/14 20:19:48mapcycle add: '+italy default' type '1on1'
2010/10/14 20:19:48mapcycle add: 'farm' type '1on1'
2010/10/14 20:19:48mapcycle add: 'estate' type '1on1'
2010/10/14 20:19:48mapcycle add: '+japan default' type '1on1'
2010/10/14 20:19:48mapcycle add: 'tower' type '1on1'
2010/10/14 20:19:48added 7 maps to the mapcycle
2010/10/14 20:19:48Stack trace:

0x41064f : E:\cpp\ufoai\ufo.exe : e:\cpp\ufoai/src/client/cl_main.c (1260) : in function (CL_Init)
0x4dc28a : E:\cpp\ufoai\ufo.exe : e:\cpp\ufoai/src/common/common.c (1105) : in function (Qcommon_Init)
0x53b433 : E:\cpp\ufoai\ufo.exe : e:\cpp\ufoai/src/ports/windows/win_main.c (149) : in function (WinMain@16)
0x53c676 : E:\cpp\ufoai\ufo.exe : main.c
0x4010db : E:\cpp\ufoai\ufo.exe : crt1.c
0x401158 : E:\cpp\ufoai\ufo.exe : crt1.c
Failed to init bfd from (C:\Windows\system32\kernel32.dll)
0x76d61194 : C:\Windows\system32\kernel32.dll : BaseThreadInitThunk
Failed to init bfd from (C:\Windows\SYSTEM32\ntdll.dll)
0x773ab495 : C:\Windows\SYSTEM32\ntdll.dll : RtlInitializeExceptionChain
Failed to init bfd from (C:\Windows\SYSTEM32\ntdll.dll)
0x773ab468 : C:\Windows\SYSTEM32\ntdll.dll : RtlInitializeExceptionChain

May it is helpful.

Regards

15

Bugs prior to release 2.4 / Re: translating executable addresses to file:line in backtrace - addr2line

« on: October 12, 2010, 11:12:18 pm »

on windows is no backtrace anyway

Well, I found this: http://code.google.com/p/backtrace-mingw/

Maybe an option for the debug build?