[HellToupee]

whats with the eth being put on the end of words like downloaded

[ChronoReverse]

It's April 1st

[mrduckman]

And I wish I had another PC to try FS2 on linux

Rebuttal originating from the wisdom of ChronoReverse
It's April 1st

[Hirnfuhler]

Code: [Select]

WARNING: "medals.tbl(55):
Error: Required token = [$Wavefile Base:], found [$Wavefile 1: 1_badge_a.wav] .
" at parse/parselo.cpp:397
WARNING: "medals.tbl(56):
Error: Required token = [$Wavefile Base:], found [$Wavefile 2: 3_badge_a.wav] .
" at parse/parselo.cpp:397
WARNING: "medals.tbl(58):
Error: Required token = [$Wavefile Base:], found [$Promotion Text:] .
" at parse/parselo.cpp:397
WARNING: "medals.tbl(58):
Error: Required token = [$Wavefile Base:], found [XSTR("For having attained an out] .
" at parse/parselo.cpp:397
WARNING: "medals.tbl(62):
Error: Required token = [$Wavefile Base:], found [$end_multi_text] .
" at parse/parselo.cpp:397
WARNING: "Error: Unable to find required token [$Wavefile Base:]
" at parse/parselo.cpp:504
ERROR: "Error parsing 'medals.tbl'
Error code = 1.
" at stats/medals.cpp:346


This is error code prom latest fs2 installer.

Lst installer just shows splash and dies. And locks my mouse...

[mrduckman]

Perhaps you should try to compile it yourself.  Or perhaps is a bug in both the installer's binary and the code.

Let us know if it happens again.

[Hirnfuhler]

I tryed to compile. 2004* version - gets error while compile.

2003* doesn't work at all.

I tryed to build icculus CVS fs2 - works great, but it doesn't see data/ dir with WC - Saga....

[taylor]

What are the compile errors with the 2004* version?  You do need Mesa6 or NVIDIA drivers installed to not get OpenGL errors.

For the icculus.org version the directories are case sensitive and match the originals.  In other words data/ should be Data/, data/tables/ should be Data/Tables/, etc.  Basically just capitalize the first letter of each directory and you should be fine.  You could also try putting the data/ directory in a VP.  With the icculus.org version just type "make cfilearchiver" and it will make the utility to create VPs.  Remember the path to the cfilearchiver binary and then go to the game directory with data/ in it and type "/cfilearchiver mod_wcsaga data", where mod_wcsaga will be the name of the newly created VP and data will be the directory to create the VP from.  Just make sure before doing this that data/ doesn't contain any other VPs and you shouldn't have any problems.  Start the game as normal and it will use the mod data first.  If you want to return to normal FS2 then just move mod_wcsaga.vp to another directory.

[Hirnfuhler]

I have nvidia-53.28 drivers.

20043010 compile error:

Code: [Select]

In file included from globalincs/pstypes.h:331,
                 from anim/animplay.h:118,
                 from freespace2/freespace.cpp:901:
windows_stub/config.h:164: warning: malformed `#pragma pack'
windows_stub/config.h:187: warning: malformed `#pragma pack'
freespace2/freespace.cpp: In function `int game_hacked_data()':
freespace2/freespace.cpp:9547: implicit declaration of function `int ml_printf(...)'
make[1]: *** [freespace.o] Error 1
make: *** [all-recursive] Error 1
-DNDEBUG=1  -I. -I.    -fexceptions -I/c/dx8/include -Wno-missing-braces -Wno-multichar -Wno-deprecated -fsigned-char -DNO_DIRECT3D -DUSE_OPENGL -g -O2 -I/usr/include/SDL -D_REENTRANT -c -o freespace.o `test -f 'freespace2/freespace.cpp' || echo './'`freespace2/freespace.cpp
make[1]: Leaving directory `/home/anton/1/fs2_open/code'

[Hirnfuhler]

Well. Tryed to run cvs fs2 with Saga capitalized Data dir etc.

It writes to stderr lots of Parse mesgs and in the end writes this:
Assertion: (src/ship/ship.cpp:1627) hull_percentage_of_hits > 0.0f

FS2 shows splash, segfault and lock mouse.

[taylor]

Edit Makefile and uncomment the "#CFLAGS+=-DNDEBUG" line to get rid of most of the stderr messages.  Also run it with the "-w -g" options and it will run in a window and shouldn't lock the mouse.  I haven't checked but if WC - Saga requires FS2_Open then it's not going to work with the icculus.org version.

Back to the 2004* version:

I really need to get a new source release out since a lot of bugs have been fixed, including this one...

Did you configure with --disable-networking?  That will cause the problem.  You can either edit freespace.cpp and in the game_hacked_data() function put "#ifndef NO_NETWORKING" after the first line (int retval=0) and "#endif" before the last line (return retval) and it should fix the problem.  You could also just do a "make clean" and then ./configure again so that networking is enabled.

[Hirnfuhler]

Yes. i ./configured with --disable-networking.
Tryed to ifdef NO_NETWORKING.

New error appeared:

Code: [Select]


In file included from globalincs/pstypes.h:331,
                 from anim/animplay.h:118,
                 from freespace2/freespace.cpp:901:
windows_stub/config.h:164: warning: malformed `#pragma pack'
windows_stub/config.h:187: warning: malformed `#pragma pack'
freespace2/freespace.cpp: In function `int game_hacked_data()':
freespace2/freespace.cpp:9547: implicit declaration of function `int ml_printf(...)'
make[1]: *** [freespace.o] Error 1
make: *** [all-recursive] Error 1

[Hirnfuhler]

Jeez. Forgotten that i have to enable networking. Enabled. Great. Now it looks like that:

Code: [Select]

In file included from globalincs/pstypes.h:331,
                 from graphics/gropengl.cpp:589:
windows_stub/config.h:164: warning: malformed `#pragma pack'
windows_stub/config.h:187: warning: malformed `#pragma pack'
graphics/gropengl.cpp: In function `unsigned int opengl_create_vbo(unsigned int, void **)':
graphics/gropengl.cpp:4713: implicit declaration of function `int glGenBuffersARB(...)'
graphics/gropengl.cpp:4718: `GL_ARRAY_BUFFER_ARB' undeclared (first use this function)
graphics/gropengl.cpp:4718: (Each undeclared identifier is reported only once
graphics/gropengl.cpp:4718: for each function it appears in.)
graphics/gropengl.cpp:4718: implicit declaration of function `int glBindBufferARB(...)'
graphics/gropengl.cpp:4719: `GL_STATIC_DRAW_ARB' undeclared (first use this function)
graphics/gropengl.cpp:4719: implicit declaration of function `int glBufferDataARB(...)'
graphics/gropengl.cpp: In function `void gr_opengl_destroy_buffer(int)':
graphics/gropengl.cpp:4803: implicit declaration of function `int glDeleteBuffersARB(...)'
make[1]: *** [gropengl.o] Error 1
make: *** [all-recursive] Error 1


[mrduckman]

I've compiled the last Icculus' CVS build and tried. It works quite well.

What was odd is that I don't get the briefings voices played before the mission. I haven't checked if there are debriefings yet.
But the in-mission voices work well.

System specs:
Linux Kernel 2.6.5
processor      : Celeron (Mendocino)
cpu MHz         : 501.163
MemTotal:       252364 kB
GPU: NVidia TNT2 Pro (32Mb)
Lastest NVidia driver
SB AWE64 (without midi, I've to figure why the driver doesn't attach /dev/sequencer yet)

I will continue checking later today and see if I can get the joystick working...

I would like to ask if it would be possible to place the debug output to a file (debug.log, i.e.) instead of stdout.
There are quite a lot of debugging I'd like to do and a very short terminal buffer

[taylor]

Hirnfuhler: I've seen that problem before but I'm drawing a blank as to what I did to fix it.  I think it was just some wierd change to the gl headers in NVIDIA's newest drivers and I went back a version.  Download this and copy the two headers into /usr/include/GL.  If that isn't enough to get it going let me know.

mrduckman: Briefing/Debriefing vioces and music don't work in the Linux versions yet.  I had started on the code but moved on to other problems before getting very far.  Someday it will get done but I don't know when.  For the debug output you can either build a non-debug version (which I recomend) or just start it like this "./freespace2 &>debug.log" and that will direct all of the debug spew to debug.log.  If you need that debug spew then that's fine but debugging symbols are built into the binary and as long as you don't run "strip" on it debugging in gdb would work well.

[mrduckman]

I've to say you did a fine job anyway. The game runs quite smoothly. Too bad I couldn't get my MS Force Feedback to work on the 2.6.5.

[Hirnfuhler]

taylor: compiled fine (_endthread() was withaout arg, i gave it - don't remember what.. ).

Doesn't work.. I'm trying to start demo:

Code: [Select]


WARNING: "medals.tbl(55):
Error: Required token = [$Wavefile Base:], found [$Wavefile 1: 1_badge_a.wav] .
" at parse/parselo.cpp:397
WARNING: "medals.tbl(56):
Error: Required token = [$Wavefile Base:], found [$Wavefile 2: 3_badge_a.wav] .
" at parse/parselo.cpp:397
WARNING: "medals.tbl(58):
Error: Required token = [$Wavefile Base:], found [$Promotion Text:] .
" at parse/parselo.cpp:397
WARNING: "medals.tbl(58):
Error: Required token = [$Wavefile Base:], found [XSTR("For having attained an out] .
" at parse/parselo.cpp:397
WARNING: "medals.tbl(62):
Error: Required token = [$Wavefile Base:], found [$end_multi_text] .
" at parse/parselo.cpp:397
WARNING: "Error: Unable to find required token [$Wavefile Base:]
" at parse/parselo.cpp:504
ERROR: "Error parsing 'medals.tbl'
Error code = 1.
" at stats/medals.cpp:346


Same as from installer....

I've got a question: does FS2_Open work with demo?

[Christoph_Berg]

Just compiled the Linux version and I must it works really nice. You did a fine job. It's really good to see such a nice game on Linux.

One Question: I read before that the voices are not working in the briefings, but is it normal that music is not working throughout the whole game?

[mrduckman]

AFAIK, they've been taken "offline" from the game.
Voices and sound effects should be working well.

[taylor]

Briefing voices and music use the same code, they are both streaming.  That code isn't finished yet but is one of a handful of things remaing on my list.

A log of bug fixes are done.  We've got some command line help now ("./fs2 -h" would print out all available options) and more things are fixed then I can count.  I'm working as hard as I can on the remaining things to do and will put out a new 3.6 build shortly after it's released (on or close to May 9th).  I'll try and get as many todo list items done as possible before the new build is released.  The Linux code should go in CVS pretty soon after 3.6 is tagged.  After that everyone can keep up-to-date by typing "cvs up"

Things left to do are (in priority order): finish networking changes to be fully PPC compatible, streaming audio, code standalone server with GUI, make a GUI launcher similar to windows version that will also be OSX compatible, figure out why the installer seems to work for no one but me.

[mrduckman]

Uh? The installer? I used the installer at work and worked well.
Send me the last one (or its URL) to my address and I'll check it next thrusday night. I'm working late.. (Actually Wed, since 0 to 8)