torcs-devel Mailing List for TORCS - The Open Racing Car Simulator

Bernhard Kaindl wrote:
> On Thu, 27 Mar 2003, Eric Espie wrote:
> 
> 
>>>>- the xfree mouse cursor should be hidden during the race
>>>
>>>
>>>well, it's cool to see the cursor when you drive with the mouse...
>>>I'll see if I can do something for the other situations.
>>
>>Well GLUT does not seems to allow the mouse cursor back when it
>>was hidden... at least under Linux.
> 
> 
> Sounds like a bug in GLUT which I could look into.
> 
> BTW, one of the freeglut developers applied my Linux fixes
> for restoring the previous video mode on glutLeaveGameMode()
> so http://freeglut.sourceforge.net/dist/current.tgz contains
> my patch.

great !

BTW, I tried your SuSE package for 8.1 this afternoon at work
(caught, caught...) and it worked very well.
Based on your SRPM we rebuilded a version for the rc5 (I have put
them on <http://torcs.org/tmp>).
I'm only dispointed by the lack of Berniw robots... as they are
the best oponents on nearly all the tracks.


Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Gernot Galli wrote:
> On Thu, 2003-03-27 at 22:51, Eric Espie wrote:
> 
>>>>- the xfree mouse cursor should be hidden during the race
>>>
>>>
>>>well, it's cool to see the cursor when you drive with the mouse...
>>>I'll see if I can do something for the other situations.
>>
>>Well GLUT does not seems to allow the mouse cursor back when it
>>was hidden... at least under Linux.
>>It seems that we'll have to live with the cursor... sorry.
> 
> 
> I think a solution would be, if the position of the mouse cursor would
> be set to a screen corner and not to the center...

Oups, I forgot to move the mouse to the center *only* when
mouse control is activated...
I'll try other way to solve the mouse cursor presence...

> 
> Gernot
> 
> 
> 

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

On Thu, 27 Mar 2003, Eric Espie wrote:

> >> - the xfree mouse cursor should be hidden during the race
> >
> >
> > well, it's cool to see the cursor when you drive with the mouse...
> > I'll see if I can do something for the other situations.
>
> Well GLUT does not seems to allow the mouse cursor back when it
> was hidden... at least under Linux.

Sounds like a bug in GLUT which I could look into.

BTW, one of the freeglut developers applied my Linux fixes
for restoring the previous video mode on glutLeaveGameMode()
so http://freeglut.sourceforge.net/dist/current.tgz contains
my patch.

> It seems that we'll have to live with the cursor... sorry.

Have you checked if torcs would be able to call the needed X11
calls and bypass GLUT for this?

I haven't looked into this yet so I don't know what to pass to
GLUT in order to hide it, but if I now it, I could find the rest
by grep. "Use the source, Luke!" :-)

> >> - the setup_linux.sh file should not overwrite an existing configuration
> >>     in the users home directory
> >
> >
> > It's the old problem of the installation of new versions of products.
> > Should the config files overwritten with the new format or kept with
> > possible compatibility problems.
> > The best solution is to provide migration tools with the installation
> > application to convert the old parameters to the new format and keep the
> > user's preferences, but TORCS is not yet at this point.
> > So I decided to refresh the config files with the new versions, only
> > to avoid maintenance problems if new parameters are added in the config
> > files.
> > May be it would be cool for the users if the old config files are saved
> > before being overwriten ;-) I'll check that.
>
> A test has been added and the old config files are copied into .old
> files (verbosely at start).

That's cool, I don't know if there is a tool which could be used for this
already, but I think it should be possible to have a separate small xml
utitly which goes over the old file and checks if it finds the same xml
entry in the newly created file and copy this over in case he finds it
in the new file.

But I've seen XML only from far away so far, so I don't know if this
approach would work for TORCS.
Bernd

>> - the xfree mouse cursor should be hidden during the race
> 
> 
> well, it's cool to see the cursor when you drive with the mouse...
> I'll see if I can do something for the other situations.

Well GLUT does not seems to allow the mouse cursor back when it
was hidden... at least under Linux.
It seems that we'll have to live with the cursor... sorry.

> 
>> - the setup_linux.sh file should not overwrite an existing configuration
>>     in the users home directory
> 
> 
> It's the old problem of the installation of new versions of products.
> Should the config files overwritten with the new format or kept with
> possible compatibility problems.
> The best solution is to provide migration tools with the installation
> application to convert the old parameters to the new format and keep the
> user's preferences, but TORCS is not yet at this point.
> So I decided to refresh the config files with the new versions, only
> to avoid maintenance problems if new parameters are added in the config
> files.
> May be it would be cool for the users if the old config files are saved
> before being overwriten ;-) I'll check that.

A test has been added and the old config files are copied into .old
files (verbosely at start).

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Gernot Galli wrote:
> Hi,
> 
> 
> the rc5 release is very cool - there are no crashes in the torcs menu
> anymore:)

thanks, I tried to fix all the problems you found.

>  
> But i have two things that should be fixed before the final release
> comes out:
>  
> - the xfree mouse cursor should be hidden during the race

well, it's cool to see the cursor when you drive with the mouse...
I'll see if I can do something for the other situations.

> - the setup_linux.sh file should not overwrite an existing configuration
> 	in the users home directory

It's the old problem of the installation of new versions of products.
Should the config files overwritten with the new format or kept with
possible compatibility problems.
The best solution is to provide migration tools with the installation
application to convert the old parameters to the new format and keep the
user's preferences, but TORCS is not yet at this point.
So I decided to refresh the config files with the new versions, only
to avoid maintenance problems if new parameters are added in the config
files.
May be it would be cool for the users if the old config files are saved
before being overwriten ;-) I'll check that.

> 
> Bye,
> Gernot
> 

Thanks,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi,

first, the resolution and color depth menu is nice;)
but i have noticed some strange behaviour. if i change the resolution
and color depth:
(Desktop resolution is 1280x1024, nvidia driver is 3141)
if i select 1024x768 an move the mouse - my screen is flickering
(sometimes) - i think this could be a problem with the nvidia driver (i
had no problem with rc4 and the 4191 nvidia driver)
The next thing happens if i change the resolution several times - the
menu becomes very slow - between clicking and reaction is a delay of one
or two seconds.
And finaly there is sometimes a second (blue) mouse cursor - which is
also very slow...

Gernot

Hi,


the rc5 release is very cool - there are no crashes in the torcs menu
anymore:)
 
But i have two things that should be fixed before the final release
comes out:
 
- the xfree mouse cursor should be hidden during the race
- the setup_linux.sh file should not overwrite an existing configuration
	in the users home directory

Bye,
Gernot

Hi,

Eugen Treise wrote:
> Hi all
> 
> The current release is very impressive. The developers did a very good 
> job. Thanks for this game.

Thanks :-) All the contributors will appreciate.

> 
> I had some problems with the display resolution in torcs.
> Sometimes, when i changed the resolution in game to 640x480 or 800x600 
> in Full Screen mode,
> the menu is displayed in wrong place, but the mouse acts like the menu 
> is where it should be.
> I dont know how to reproduce this.

It happens under Linux when the desktop is larger than the screen
resolution, when approching the mouse to the screen border, the desktop
scrolls (with KDE at least).

Another explanation could be that the game mode failled for that
resolution, and the "software" fullscreen mode is used.
This can happen when switching resolutions, sometimes the Nvidia drivers
don't allow the hardware acceleration for the new resolution, and if you
exit and restart all is OK.

> 
> When i change to Full Screen mode, no matter what resolution i set, it 
> is always displayed at 120Hz.
> My desktop is set at 100 Hz.

May be I'll add the freq selection in the screen configuration menu for
win32 (it seems that this is not useful in Linux)

> 
> In Full Screen mode the name of the application changes from torcs to GLUT.

Yes you are right, in GLUT game mode (the fullscreen mode used by TORCS)
no window is created, so the name is not set. Sorry for the
incovenience.

> 
> My system:
> Windows XP
> Geforce 2MX 64MB
> Standard res. 1024x768x16 @100Hz
> 
> 
> Eugen
> 
> 
> 

Thanks,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi all.

The current release is very impressive. The developers did a very good 
job. Thanks for this game.

I had some problems with the display resolution in torcs.
Sometimes, when i changed the resolution in game to 640x480 or 800x600 
in Full Screen mode,
the menu is displayed in wrong place, but the mouse acts like the menu 
is where it should be.
I dont know how to reproduce this.

When i change to Full Screen mode, no matter what resolution i set, it 
is always displayed at 120Hz.
My desktop is set at 100 Hz.

In Full Screen mode the name of the application changes from torcs to GLUT.

My system:
Windows XP
Geforce 2MX 64MB
Standard res. 1024x768x16 @100Hz


Eugen

Hi All,

The version 1.2.0-rc5 of TORCS is available here:
<http://torcs.org/tmp/download.html>

The new features are:
- configuration/calibration screen for mouse control.
- some berniw robot enhancements.
- Win32 port working
- many rc4 bugs fixed (thanks to all the people reporting problems)

If there are problems report them as usualy, and if
all seems ok, then it will be the 1.2.0 public release.

Thanks to you all, your help contributes a lot in the project.
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

On Sat, 8 Mar 2003, torcs wrote:

> Hi,
>
> A new version...
>
> - fixed makefile issues about the setup_linux.sh script
The scrips works cool, it's really good (used some symlinks before)

> - fixed some "abandon race bugs"
> - warp the mouse at the center of the window if player present
>    and using mouse to drive.
Very good indeed.

> - X11 fullscreen 1st try from freeglut code.
>
> Missing:
>
> - mouse configuration screen.
> - keep the mouse pointer in the window in order to
>    avoid screen panning in low resolutions.

I got this working by linking against freeglut:
http://freeglut.sourceforge.net/dist/current.tgz

To compile it, run:
libtoolize --copy
aclocal
autoheader
automake -a
autoconf
./configure
make install

In Torcs RC4, replace -lglut with -lfreeglut-1.3 in Make-config and do
a make install

In this build there would be two codes doing video mode switching
but the do it the same way, so you'll only see the first.

> - color depth configuration (24bits now)

I use a Radeon Mobility M6 LY 16MB with XFree 4.3 and the included
DRI driver and I see it switching to such mode:

OK done for 640x480:24@160

But start the X Server with Depth 16, so it cannot switch to 24bpp,
but it looks like it feels confident with 640x480 in 16bpp.

So it looks like some X11 drivers or servers even do the switch
if the asked mode is compatible.

I would be interested if somebody has some issue with the fullscreen
mode,

Bernd

PS: It's not major, but if you want to remove the X11 vidmode stuff
from your torcs compile and rely on freeglut, you have to apply the
fixes below which should work for you also:

(sent the following to freeglut-developer also)

--- freeglut-1.3/freeglut_gamemode.c	2003/03/11 14:54:11	1.1
+++ freeglut-1.3/freeglut_gamemode.c	2003/03/11 15:01:55
@@ -137,6 +137,15 @@ void fghRestoreState( void )
                 displayModes[ i ]
             );

+	    /*
+	     * In case this will be the last X11 call we do before exit,
+	     * we've to flush the X11 output queue to be sure the command
+	     * is really brought onto it's way to the X server.
+	     * The application should not do this because it
+	     * would not be platform independent then.
+	     */
+	    XFlush(fgDisplay.Display);
+
             return;
         }
     }
@@ -200,6 +209,8 @@ GLboolean fghChangeDisplayMode( GLboolean haveToTest )
         if( fghCheckDisplayMode( displayModes[ i ]->hdisplay, displayModes[ i ]->vdisplay,
                                  fgState.GameModeDepth, fgState.GameModeRefresh ) )
         {
+	    if( haveToTest )
+		return( TRUE );
             /*
              * OKi, this is the display mode we have been looking for...
              */

Diffed against the current http://freeglut.sourceforge.net/dist/current.tgz.

On Tue, 11 Mar 2003, Bernhard Kaindl wrote:
>
> Hi all!
>
> I've tried the GameMode of the current freeglut CVS on current
> Linux XFree86-4.3 with TORCS(torcs.sf.net) and noticed two problems:
>
> - glutGameModeGet(GLUT_GAME_MODE_POSSIBLE) calls
>   int fghChangeDisplayMode( int haveToTest ) with
>   haveToTest == TRUE, but the XFree86 branch of
>   fghChangeDisplayMode() does not check the
>   flag and always changes to the mode if it
>   was possible. Simple to fix in the loop
>   when a matching mode was found.
>
>   This may not be a problem in some(most?) cases but
>   if a call to glutEnterGameMode() follows
>   glutGameModeGet(GLUT_GAME_MODE_POSSIBLE),
>   then it calls fghRememberState() which uses
>   XF86VidModeGetModeLine() to get the current
>   modeline from XFree, but as
>   glutGameModeGet(GLUT_GAME_MODE_POSSIBLE)
>   already switched the mode, fghRememberState
>   gets the already witched video mode from
>   the X server and remembers this mode and not
>   the initial video mode. glutLeaveGameMode()
>   tries to switch to the GameMode videomode
>   and not to the initial video modes because
>   the videomode saved was the GameMode videomode.
>
> - glutLeaveGameMode() calls fghRestoreState() which
>   uses XF86VidModeSwitchToMode() in the XFree86 branch
>   but it looks like that XFree86 does not execute the
>   VidModeSwitch immediately(maybe also because the X
>   protocol is asyncronous by default) and you have
>   to call anoter X11 call afterwards to ensure that
>   the X server switches to the mode. So a call to
>   XFlush(display) may be done afterwards to flush the
>   output buffer in case the application does no calls
>   to XPending, XNextEvent or XWindowEvent afterwards
>   which would flush the buffer also.
>
>   In my case I put an call to XF86VidModeGetModeLine()
>   after the XF86VidModeSwitchToMode() to check the mode
>   switch which seemed to execute the mode switch on the
>   server, but an eXpert told me that he have heard about
>   such things with these functions and a final call to
>   XFlush() should be done after calling the switch mode
>   functions when the application does not do other X calls
>   like create window or so immediately afterwards.
>
> I could send you simple patches to fix these two issues,
> and I wanted to send you this description beforehand to
> check for any feedback or concerns you may have.
>
> After fixing these problems TORCS works fine with the
> current freeglut-1.3 CVS.

Hi,

A new version...

- fixed makefile issues about the setup_linux.sh script
- fixed some "abandon race bugs"
- warp the mouse at the center of the window if player present
   and using mouse to drive.
- X11 fullscreen 1st try from freeglut code.

Missing:

- mouse configuration screen.
- keep the mouse pointer in the window in order to
   avoid screen panning in low resolutions.
- color depth configuration (24bits now)

I'm out tonight so don't expect quick answers before tomorrow ;-)

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Bernhard Kaindl wrote:
> On Fri, 7 Mar 2003, Eric Espie wrote:
> 
>>Bernhard Kaindl wrote:
>>
>>>I've started investigations why GlutGameMode which seems to work
>>>in Windows as I guess did not work out for Linux so far. Maybe it's
>>>not implemented in Mesa GLUT.(I've asked somebody who might know)
>>
>>Yes there are no code in GlutGameMode for Linux, only Win32 :-(
> 
> 
> I see it, what a schame.
> 
> I've checked the current CVS of freeglut (the tarball is outdated).
> 
> In freeglut-1.3/freeglut_gamemode.c there is an Implementation for
> XFree86 with the XF86VidMode Extension using XF86VidModeSwitchToMode()
> 
Good, I'll try to have it work for torcs.

Thanks,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

On Fri, 7 Mar 2003, Eric Espie wrote:
> Bernhard Kaindl wrote:
> >
> > I've started investigations why GlutGameMode which seems to work
> > in Windows as I guess did not work out for Linux so far. Maybe it's
> > not implemented in Mesa GLUT.(I've asked somebody who might know)
>
> Yes there are no code in GlutGameMode for Linux, only Win32 :-(

I see it, what a schame.

I've checked the current CVS of freeglut (the tarball is outdated).

In freeglut-1.3/freeglut_gamemode.c there is an Implementation for
XFree86 with the XF86VidMode Extension using XF86VidModeSwitchToMode()

> If someone knows how to resize the X11 screen... independently from
> GLUT/SDL...

Yes, the functions used in freeglut_gamemode.c should be ok to use
in other programs.

The license just says that you have to keep the copyright notice but
this should be all.

A fresh implemtation which just swites to a fixed mode if it is
available using the XF86VidMode function also looks doable.

I guess SDL is using the same functions.

> It's a bit exaggerated to require SDL only for full screen, no ?

Yes, it was just the only place which I knew.

> I don't know if TORCS works with FreeGLUT, but
> the GlutGameMode is checked if it answers that modes are available
> it is used, else GlutFullScreen is used.

Sounds ok, maybe I can give FreeGLUT or the GlutGameMode code from
it a try.
Bernd

Bernhard Kaindl wrote:
> On Fri, 7 Mar 2003, Eric Espie wrote:
> 
>>This version of TORCS should be compiled with PLIB 1.6.0
>>where the lens flare are present. Earlier versions of
>>PLIB did not contained the lens flare code.
>>The lens flares are visible also on Linux.
> 
> 
> I think an autoconf test for checking the plib version would
> be good to have which tells people got the plib-1.6.0 from
> plib.sf.net.

I have taken the PLIB version test in the configure.ac of flightgear...
now PLIB must be 1.6.0 ir order to pass the configure step.

Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Bernhard Kaindl wrote:
> Hi,
>   in the rc3 annoucement, Eric said about fullscreen under Linux:
> 
> 
>>- The fullscreen option is back under Linux, so you'll have to
>>   adjust your screen resolution to the window size if you are
>>   using the fullscreen mode until I find a way to change the
>>   resolution from inside the program.
> 
> 
> I've started investigations why GlutGameMode which seems to work
> in Windows as I guess did not work out for Linux so far. Maybe it's
> not implemented in Mesa GLUT.(I've asked somebody who might know)

Yes there are no code in GlutGameMode for Linux, only Win32 :-(

> 
> I know that SDL is a possible alternative to allow a switch to
> a e.g. 640x480 fullscreen game mode, but I don't know what impact
> converting a GLUT program to an SDL/GLUT program has. I only know
> that the gl-117 flight sim (gl-117.sf.net) can be compiled either
> with GLUT only or with SDL and at least has suhc Game mode then.

If someone knows how to resize the X11 screen... independently from
GLUT/SDL...
It's a bit exaggerated to require SDL only for full screen, no ?

> 
> On http://www.arianne.info/, I've also read about FreeGLUT, a
> compleley free reimplementation for GLUT: freeglut.sf.net
> (note the tarballs are outdated, CVS copiles for me)

When I looked in the code few monthes ago and FreeGLUT did not support
the GlutGameMode under Linux.

[...]
> I installed the current freeglut CVS now, and now just
> need a small test program to check it. I should be able
> to do it myself with the Manuals I've found, however if
> somebody has a test program or have tried freeglut, please
> tell me, thanks!

I don't know if TORCS works with FreeGLUT, but
the GlutGameMode is checked if it answers that modes are available
it is used, else GlutFullScreen is used.

> 
> Bernd
> 
> ---------- Forwarded message ----------
> From: Eric Espie <eri...@fr...>
> To: Bernhard Kaindl <ber...@gm...>, Bernhard Kaindl <bk...@su...>
> Date: Tue, 04 Mar 2003 21:26:22 +0100
> Subject: Re: Hardware Requirements for TORCS?
> 
> Bernhard Kaindl wrote:
> 
> 
>>BTW, I don't get the fullscreen selectable on my installs,
>>I guess it could be only the GlutGameMode() which rejects
>>it, do you have any hint how I can look to enable it?
>>
>>Some other Game (gl-117) I know uses SDL with GLut for
>>fullscreen/mode switching on Linux, maybe this can be
>>done for torcs also, but I'm just adding data, not that
>>I think it's very easy, maybe though, I don't know.
>>
>>Thanks, Bernd
>>
>>
>>
> 
> I did not find yet how to have the fullscreen working
> correctly under Linux... I'll try to check the SDL code
> to see if I can hack some code from it.

I have tried to find in SDL sources where the fullscreen
code was, but I got rapidly lost...

> 
> Eric.

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi,
  in the rc3 annoucement, Eric said about fullscreen under Linux:

> - The fullscreen option is back under Linux, so you'll have to
>    adjust your screen resolution to the window size if you are
>    using the fullscreen mode until I find a way to change the
>    resolution from inside the program.

I've started investigations why GlutGameMode which seems to work
in Windows as I guess did not work out for Linux so far. Maybe it's
not implemented in Mesa GLUT.(I've asked somebody who might know)

I know that SDL is a possible alternative to allow a switch to
a e.g. 640x480 fullscreen game mode, but I don't know what impact
converting a GLUT program to an SDL/GLUT program has. I only know
that the gl-117 flight sim (gl-117.sf.net) can be compiled either
with GLUT only or with SDL and at least has suhc Game mode then.

On http://www.arianne.info/, I've also read about FreeGLUT, a
compleley free reimplementation for GLUT: freeglut.sf.net
(note the tarballs are outdated, CVS copiles for me)

The programmer of the Arianne 3D Clients says about it:

> I created a new client using FREEGLUT and PLIB + PUI,
> and thats prolly what we gonna end up with,
> it works great right now...
>
> SDL just doesnt cut it, its too bloated and too slow.
> I had massive speed updates just from exchanging SDL
> with FREEGLUT.
>
> Skyflash

I've also read on the freeglut webpage that glutGameMode
works on Linux with it, so I'm about to check it out.

I installed the current freeglut CVS now, and now just
need a small test program to check it. I should be able
to do it myself with the Manuals I've found, however if
somebody has a test program or have tried freeglut, please
tell me, thanks!

Bernd

---------- Forwarded message ----------
From: Eric Espie <eri...@fr...>
To: Bernhard Kaindl <ber...@gm...>, Bernhard Kaindl <bk...@su...>
Date: Tue, 04 Mar 2003 21:26:22 +0100
Subject: Re: Hardware Requirements for TORCS?

Bernhard Kaindl wrote:

> BTW, I don't get the fullscreen selectable on my installs,
> I guess it could be only the GlutGameMode() which rejects
> it, do you have any hint how I can look to enable it?
>
> Some other Game (gl-117) I know uses SDL with GLut for
> fullscreen/mode switching on Linux, maybe this can be
> done for torcs also, but I'm just adding data, not that
> I think it's very easy, maybe though, I don't know.
>
> Thanks, Bernd
>
>
>
I did not find yet how to have the fullscreen working
correctly under Linux... I'll try to check the SDL code
to see if I can hack some code from it.

Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

On Fri, 7 Mar 2003, Eric Espie wrote:
>
> This version of TORCS should be compiled with PLIB 1.6.0
> where the lens flare are present. Earlier versions of
> PLIB did not contained the lens flare code.
> The lens flares are visible also on Linux.

I think an autoconf test for checking the plib version would
be good to have which tells people got the plib-1.6.0 from
plib.sf.net.

I don't have this test ready, but an example patch which could
be used for older plib versions to allow compile on these, loosing
lens flares of course with older versions:

--- src/modules/graphic/ssggraph/grscene.cpp
+++ src/modules/graphic/ssggraph/grscene.cpp
@@ -408,6 +408,7 @@
     glEnable(GL_LIGHT0);
     glEnable(GL_DEPTH_TEST);

+#ifdef _SSGA_TYPE_LENSFLARE
     if (!sun) {
 	ssgaLensFlare      *sun_obj      = NULL ;
 	sun_obj  = new ssgaLensFlare () ;
@@ -416,6 +417,10 @@
 	sun      -> addKid          ( sun_obj  ) ;
 	TheScene    -> addKid(sun) ;
     }
+#else
+#warning For lens flares you should use plib-1.6.0 or higher
+#warning you can get it from plib.sourceforge.net/
+#endif

     /* GUIONS GL_TRUE */
     glLightModeli(GL_LIGHT_MODEL_TWO_SIDE,GL_FALSE);

A similar check might be used in the configure script,
and either warn about missing lens flares and possible
other compile problems with ssga functions or aborts
if the check fails.

Bernd

Hi,
   which version of plib do you use?

Maybe I've seen this or another plib error some time ago
as I was using plib-1.4.2. It declares something different
with const.

ssgaLensFlare is provided by plib-1.6.0, the current stable
release of plib. If you want to see sun/lens flares, just
get it from plib.sf.net, compile and install it.

BTW, for installing self-compiled packages there is the
great checkinstall preload utitily which builds you a
package of what the install installs so that you can
track it and uninstall it(rpm and slackware supported)
It's listed on freshmeat.net, just type it in the search.
http://checkinstall.izto.org

Bernd

On Fri, 7 Mar 2003, Chris Bennett wrote:

> Hi!
>
> Just thought I'd raise the question,
>
> In src/modules/graphic/ssggraph/grscene.cpp, line 427 :-  is that block of
> code meant to be windows only?
>
> Output from compiler:
>
> g++ -I/home/zortar/devel/torcs/torcs/torcs/export/include  -ansi -Wall
> -Wstrict-prototypes -g -DDEBUG -DDEBUG_OUT -D_SVID_SOURCE -D_BSD_SOURCE -DSHM
> -c grscene.cpp
> grscene.cpp: In function `int grInitScene()':
> grscene.cpp:427: `ssgaLensFlare' undeclared (first use this function)
> grscene.cpp:427: (Each undeclared identifier is reported only once for each
>    function it appears in.)
> grscene.cpp:427: `sun_obj' undeclared (first use this function)
> grscene.cpp:428: parse error before `(' token
> make[5]: *** [grscene.o] Error 1
> ....
>
> On closer look, the ssgaLensFlare class is defined in
> src/windows/include/plib/ssgaLensFlare.h, and so I thought the code in
> grscene.cpp was probably only for windows binaries...
>
> I don't know the torcs source that well so I'm not confidently saying this
> patch is appropriate/correct... All I know is torcs cvs compiles with this
> patch applied :)
>
> Regards,
>
> Chris Bennett

Chris Bennett wrote:
> Hi!
> 
> Just thought I'd raise the question,
> 
> In src/modules/graphic/ssggraph/grscene.cpp, line 427 :-  is that block of 
> code meant to be windows only?
> 
> Output from compiler: 
> 
> g++ -I/home/zortar/devel/torcs/torcs/torcs/export/include  -ansi -Wall 
> -Wstrict-prototypes -g -DDEBUG -DDEBUG_OUT -D_SVID_SOURCE -D_BSD_SOURCE -DSHM  
> -c grscene.cpp
> grscene.cpp: In function `int grInitScene()':
> grscene.cpp:427: `ssgaLensFlare' undeclared (first use this function)
> grscene.cpp:427: (Each undeclared identifier is reported only once for each
>    function it appears in.)
> grscene.cpp:427: `sun_obj' undeclared (first use this function)
> grscene.cpp:428: parse error before `(' token
> make[5]: *** [grscene.o] Error 1
> ....
> 
> On closer look, the ssgaLensFlare class is defined in 
> src/windows/include/plib/ssgaLensFlare.h, and so I thought the code in 
> grscene.cpp was probably only for windows binaries...
> 
> I don't know the torcs source that well so I'm not confidently saying this 
> patch is appropriate/correct... All I know is torcs cvs compiles with this 
> patch applied :)
> 
> Regards,
> 
> Chris Bennett

This version of TORCS should be compiled with PLIB 1.6.0
where the lens flare are present. Earlier versions of
PLIB did not contained the lens flare code.
The lens flares are visible also on Linux.

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi!

Just thought I'd raise the question,

In src/modules/graphic/ssggraph/grscene.cpp, line 427 :-  is that block of 
code meant to be windows only?

Output from compiler: 

g++ -I/home/zortar/devel/torcs/torcs/torcs/export/include  -ansi -Wall 
-Wstrict-prototypes -g -DDEBUG -DDEBUG_OUT -D_SVID_SOURCE -D_BSD_SOURCE -DSHM  
-c grscene.cpp
grscene.cpp: In function `int grInitScene()':
grscene.cpp:427: `ssgaLensFlare' undeclared (first use this function)
grscene.cpp:427: (Each undeclared identifier is reported only once for each
   function it appears in.)
grscene.cpp:427: `sun_obj' undeclared (first use this function)
grscene.cpp:428: parse error before `(' token
make[5]: *** [grscene.o] Error 1
....

On closer look, the ssgaLensFlare class is defined in 
src/windows/include/plib/ssgaLensFlare.h, and so I thought the code in 
grscene.cpp was probably only for windows binaries...

I don't know the torcs source that well so I'm not confidently saying this 
patch is appropriate/correct... All I know is torcs cvs compiles with this 
patch applied :)

Regards,

Chris Bennett

G.Galli wrote:
> Hi,
> 
> the last two days i have tried to build&run torcs rc3 and some strange
> problems apears on my debian box. 
> first i made a normal rc3 debian package with the debian config-files
> from my rc2 packages.  Torcs runs - with some small problems (i couldn't
> add a human driver as non-root). after that i made a new package (to

May be the access rights on the $HOME/.torcs/drivers/human/human.xml
file were not correctly set by the torcs shell.

> sove some other small problems withe my package) - with some
> modifications to my debian configfiles - but now torcs won't start
> anymore.i get the following error:
> 
> /usr/local/bin/torcs: line 47: 26098 Segmentation fault      ./torcs -l
> $LOCAL_CONF

Bad news :-(

> 
> I think this is not a problem of my debian config-files (i tried to
> build/install torcs without making a debian package - but this error
> apears also)
> 
> Hmmm, any suggestions?

you can try to recompile with debuging mode:

$ make clean
$ ./configure --enable-debug
$ make && make install

then go under the target torcs directory and run

$ gdb ./torcs
(gdb) r -l $HOME/.torcs
...
<reproduce the crash>
...
(gdb) bt

and send the results (all the gdb session if possible),
this way I'll have more chances to spot the problem.

> 
> 
> Gernot
> 

Yesterday I have fixed some crashes about the sound shutdown
between the races and some crashes that occured when the races
were stopped and then relaunched.
I fixed also some OpenGl state management conflicting with
PLIB, so there should be less visual problems.
The version has been checked in the CVS repository.

I'll do an RC4 when the modifications about the player controls
(mouse specificaly) will be more usable and configurable.
it is just a matter of days.

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

Hi,

the last two days i have tried to build&run torcs rc3 and some strange
problems apears on my debian box. 
first i made a normal rc3 debian package with the debian config-files
from my rc2 packages.  Torcs runs - with some small problems (i couldn't
add a human driver as non-root). after that i made a new package (to
sove some other small problems withe my package) - with some
modifications to my debian configfiles - but now torcs won't start
anymore.i get the following error:

/usr/local/bin/torcs: line 47: 26098 Segmentation fault      ./torcs -l
$LOCAL_CONF

I think this is not a problem of my debian config-files (i tried to
build/install torcs without making a debian package - but this error
apears also)

Hmmm, any suggestions?


Gernot

Hi all,

The TORCS 1.2.0-rc3 is available here: <http://torcs.org/tmp>

The main evolutions:
--------------------

- The configure script can be launched with the options:
      --prefix=   for all the data to be in <prefix>/games/torcs
                  (default /usr/local)
      --bindir=   for the torcs shell (the one to use to run torcs)
                  (default <prefix>/bin)
      --libdir=   for the shared libs (default <prefix>/lib)

- The configuration data and results are stored in the $HOME/.torcs
   directory created by the torcs shell.
   the torcs shell now creates the configuration directories under
   $HOME/.torcs and copy as necessary the config files from
   <prefix>/games/torcs to $HOME/.torcs
   note to the developpers, the old $HOME/.torcs file is now
   $HOME/.torcs.rc and if you want to compile from within the source
   tree (somewhere else than the root) you'll have to define
   the env variable MAKE_DEFAULT=<torcs sources>/Make-default.mk

- The torcs binary (located in <prefix>/games/torcs) accept now
   a parameter "-l <local-dir>" used to specify the corresponding
   $HOME/.torcs
   Calling the binary without parameters gives the previous behaviour
   (the config files used are in <prefix>/games/torcs)

- The cars and tracks packages now contains only useful data to run the
   game, all the data used to edit the cars/tracks are not more in the
   packages.

- Trademarks were removed from some car skins.

- The fullscreen option is back under Linux, so you'll have to
   adjust your screen resolution to the window size if you are
   using the fullscreen mode until I find a way to change the
   resolution from inside the program.

I'll work on the mouse control and on the problems seen those
days to enhance the stability.

Bye,
Eric.
-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=

G.Galli wrote:
> On Tue, 2003-03-04 at 19:16, Bernhard Wymann wrote:
> 
> 
> 
>>at my installation i couldn't reproduce it... did you also hit escape?
>>
>>Single Player -> Quick Race -> New Race -> Hit Escape -> Abandon Race -> ...
>>
> 
> Yes, I hit the escape key.
> 
> 
>>my system:
>>- SuSE 7.3 (glibc 2.2.4, gcc 2.95.3).
> 
> 
> my system:
> 
> debian/unstable (glibc 2.2.1, gcc 3.2.3)

I'm using gcc 3.2 (.0 i guess)
TORCS is not using the glibc.
may be the problem can be the compiler version, I have to do some
tests on that.

> 
> 
>>the only special thing on my installation is that the user owns 
>>/usr/local/games/torcs, perhaps it doesn't work for you because of the 
>>permissions (torcs writes some files in /usr/local/games/torcs)?
>>
> 
> The crashes also happens if i start torcs as root, so i think that the
> permissons for the torcs directory are not the reason. 
> (But it would be great if torcs would save changes in the
> homedirectory;)
> 
> Gernot
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Etnus, makers of TotalView, The debugger 
> for complex code. Debugging C/C++ programs can leave you feeling lost and 
> disoriented. TotalView can help you find your way. Available on major UNIX 
> and Linux platforms. Try it free. www.etnus.com
> _______________________________________________
> Torcs-devel mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-devel
> 
> 


-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
                        TORCS
             The Open Racing Car Simulator
                  http://torcs.org
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=