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

Hello,
I'm using torcs for a project, and I would like to know if it's possible to have
the force feedback while driving with a racer wheel. I just saw that it was
implemented by a student in a research project, maybe he gave the source to the
racer community.

Thank you in advance for your answer

David

On Sun, 18 Jun 2006 10:47:56 +0200, Kennet Eriksson  =

<eri...@ya...> wrote:

> First: Sorry for starting a new thread. I just cannot
> figure out how to answer to a post here. Please
> enlighten me.


I think if you just do reply-all with a client that does nothing more th=
an  =

add a "Re:" , the mail archive will be able to link it to the previous  =

post. Don't edit the title as you did in this case.

HTH

>  Do you think that it's reasonably accurate or "just

> for fun" ?

Hi everybody!
First: Sorry for starting a new thread. I just cannot
figure out how to answer to a post here. Please
enlighten me.

I cannot say how accurate it is. I found data on the
internet and modelled after that. A lot of stuff in
the xml file determines the cars behaviour. Can
somebody testdrive a real car and compare? I cannot
give a better answer, sorry.

I've got a sort of sweet tooth for this, I may look
into some other historic cars if no one objects.

I modified the xml file a little; the performance is
similar. The gearing a little different, the balance
and fueltank changed. The rear brake disks a little
smaller to lessen the locking of them.

Regards
Kennet

On Wed, 07 Jun 2006 18:55:18 +0200, Kennet Eriksson  =

<eri...@ya...> wrote:

> I experimented a little more with the Austin-Healey. I
> changed the gearing according to my previous post and
> then I tried to set up the torque to give the car
> somewhat similar speedresources to what I can read on
> the internet. The car accelerates a little too fast up
> to 100 km/h but I hope we can accept that. Given
> enough time i will also reach a top speed of a little
> under 190. I have driven it at 212 or so in this
> setting slipstreaming behind bernhist 4 on E-track 4.
> Feel free to comment this, if you have information
> that totally invalidiates this I'd love to hear it.
> Corrections of any kind are also welcome.
>
> I will stop pestering you about this now. I'm going
> cold turkey. I can do it! I will do it! Wish me luck.
>
> Regards
> Kennet
>

Hi, that's interesting.

I was wanting to do something involving breaking distances. What's your =
 =

impression of the modeling accuracy now that you have put some more  =

realistic data in for the Austin-Healey?

Do you think that it's reasonably accurate or "just for fun" ?

cheers.

  Hello , I was wondering if someone could 
     help me to make a Formula 1 track and Formula 1 cars on TORCS.
    I'm working on it and I would aprecciate your help very much.
    I'm Vicente from spain.
    
    The web of the project is:
    http://www4.uji.es/~al063357/tutorials/torcs/index.html
    
    I have read all information on the web:
    http://www.idiap.ch/~dimitrak/torcs/
    
    
    My intention is to make a guide that explain the proces of making a
    simple render with blender (a GNU free 3d rendering program).
    
    
    I have made a real track with trackeditor,
    I have searched the track map of Montmelo (Sanish F1-grand prix) on Google
    Earth. You can download the track on my web.
    
    Now I want to edit the track with blender in order to add objects and
    make a nice track. I mean, add trucks, signals of the distance of
    the curb, trees...
    
    
    The editing of the Formula 1 cars is OK, I can edit the car
    with blender in an easy way. This is explained on the F1 car section
    of the web page. You can download a car on my web.
    
    You can read more on the web of the project:
    
    http://www4.uji.es/~al063357/tutorials/torcs/index.html
    
    
    
    I was wondering is someone can explain how I can edit the track
    created with track editor with blender or another 3d program in order
    to improve the quality of the track.
    Or if is another way to create good tracks for torcs, please
    tell it to me.
    
    I'm very interested on make F1 cars and F1 tracks for torcs.
    
    Thanks to help me. Any comment is good to learn how to make
  a track for torcs with blender.
    
    
    By the way I have registerd on the
    http://torcs.sourceforge.net/user.php?op=new_user
    
    my nick name is: Vicente and vicentemarti
    The pasword wasn't sent to me.
    Can you tell it to the administrators of the
    http://torcs.sourceforge.net/ site to sent me the
    pasword?
    
    
    Tanks for help me again,
    
    Please I 
    
    
    ----------------------------------------------------------
    Vicente Marti
    http://www4.uji.es/~al063357/
    vte...@ho...
    ------------------------------------------------------------
 __________________________________________________
Correo Yahoo!
Espacio para todos tus mensajes, antivirus y antispam ¡gratis! 
Regístrate ya - http://correo.yahoo.es 

Hello,

I'm getting this error:
(...) configure: error: Can't find libm. (...)
What package am I missing?

		
___________________________________________________________ 
Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de

I experimented a little more with the Austin-Healey. I
changed the gearing according to my previous post and
then I tried to set up the torque to give the car
somewhat similar speedresources to what I can read on
the internet. The car accelerates a little too fast up
to 100 km/h but I hope we can accept that. Given
enough time i will also reach a top speed of a little
under 190. I have driven it at 212 or so in this
setting slipstreaming behind bernhist 4 on E-track 4.
Feel free to comment this, if you have information
that totally invalidiates this I'd love to hear it.
Corrections of any kind are also welcome.

I will stop pestering you about this now. I'm going
cold turkey. I can do it! I will do it! Wish me luck.

Regards
Kennet

hi,

i download torcs-1.2.4 and compile from sources

and all ok

when i run torcs it run ok but when i enter new race it closed with this 
error:

/usr/local/lib/torcs/torcs-bin: symbol lookup error: 
/usr/local/lib/torcs/modules/graphic/ssggraph.so: undefined symbol: gzopen

plz help me .

thanks .

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

Hi

> grscreen.h:81: error: extra qualification =91cGrScreen::=92 on member =91=
camDraw=92
> make[5]: *** [grmain.o] Fehler 1
> make[4]: *** [subdirs] Fehler 1
> make[3]: *** [subdirs] Fehler 1
> make[2]: *** [subdirs] Fehler 1
> make[1]: *** [subdirs] Fehler 1
> make: *** [restart] Fehler 2
>=20
>=20
> do you have an idea what i could do ... or what that mean

Change "void cGrScreen::camDraw(tSituation *s);"
to "void camDraw(tSituation *s);"

You use GCC 4.1 or higher, which is newer than TORCS 1.2.4, in CVS the pr=
oblem
has been fixed.

Bye, Bernhard.


--=20

Visit my homepage http://www.berniw.org
Official TORCS racing: The TORCS Racing Board, http://www.berniw.org/trb

Hi, I've experimented a little with the xml for the
Austin-Healey. It seems as the torque values for the
engine are too high. I suggest halving them for a
engine in the Mk III series, x0.45 for a MK II series
and, if you wish to experiment, try something like
x0.6 for a racing version.
The torque values goes up to 7000 rpm but the rpm
limiter is set at 6000, you may wish to set it to
7000.
http://www.supercars.net/cars/117.html have some
values for the gearing of the car, feel free to
experiment. (2.93:1, 2.05:1, 1.31:1, 1.00:1 perhaps
0.9:1 if you bought the optional overdrive :-) final
drive 4.22:1 or 3.91:1 depending on where you look...,
try something in between that gives the car a top
speed of < 190 for a MkII)
I believe the car is more realistic with these
modifications and of course I welcome corrections.

I'm not sure about how the licensing works with the
bundled xml file, but to be safe, do not ship with the
corrections. You may include this note though.
Comments?

Regards
Kennet

Hi all,

Christos is right, for all the model I imported (with the author=20
agreement of course) only the 3D and textures are non-free.
The xml files are free as indicated in the files, so you can
update the setups to provide better cars.

Bye,
Eric.

Christos Dimitrakakis a =E9crit :
> SpeedyChonChon wrote:
>> Hi
>>
>> The technical data and the 3D model are parts of a same "package" and
>> there is only one license for the whole "package". And we can't
>> published any modification of the "package" (data and/or 3d model), so
>> we must make a new complete "package".
>> Am I not right ?
>>
>> Speedy
>>
>=20
> Hm... I thought KC only gave us the 3d models and that the technical
> data .xml files were done by Eric.
>=20
> BTW, KC mentions the basic tech data in the readme.txt files, i.e. for
> the Austin:
>=20
> Size: 4 x 1.537 x 1.289 m
> Weight: 1250 kg
> Power: 148 ch @ 5250 RPM
> Torque: 22.8 mkg @ 3500 RPM
>=20
> These values seem to be twice higher than the values in the .xml :)
>=20
>=20
>=20

SpeedyChonChon wrote:
> Hi
> 
> The technical data and the 3D model are parts of a same "package" and
> there is only one license for the whole "package". And we can't
> published any modification of the "package" (data and/or 3d model), so
> we must make a new complete "package".
> Am I not right ?
> 
> Speedy
> 

Hm... I thought KC only gave us the 3d models and that the technical
data .xml files were done by Eric.

BTW, KC mentions the basic tech data in the readme.txt files, i.e. for
the Austin:

Size: 4 x 1.537 x 1.289 m
Weight: 1250 kg
Power: 148 ch @ 5250 RPM
Torque: 22.8 mkg @ 3500 RPM

These values seem to be twice higher than the values in the .xml :)



-- 
Christos Dimitrakakis
Homepage: http://www.idiap.ch/~dimitrak/main.html
Music: http://olethros.dmusic.com

Hi

The technical data and the 3D model are parts of a same "package" and=20
there is only one license for the whole "package". And we can't=20
published any modification of the "package" (data and/or 3d model), so=20
we must make a new complete "package".
Am I not right ?

Speedy


Christos Dimitrakakis a =E9crit :

>I don't think so.  In any case the technical data is independent of the
>3d model.
>
>
>
>SpeedyChonChon wrote:
> =20
>
>>Hi Kennet
>>
>>The problem of the historic cars is that most of them are non-free, so=20
>>only the author can modify them.
>>The only solution for us to correct the Austin-Healey is to make a 3D=20
>>model from zero and give it the technical datas we want.
>>I'm not familiar with licences but I think it is so.
>>Sorry.
>>

> I enjoy the game immensely
> but I am also very interested in history so accurate
> cars would enhance the experience.

That would be cool, but it is very difficult to get technical data for
most cars.  The hardest bit is to get all those physical quantities
(i.e. rotational inertia) which together determine how the car feels on
the road.  The motor can be easily adjusted as far as output horsepower
goes, but it is still a bit of a hit-and-miss affair to determine engine
braking coefficient and engine rotational inertia - and those things
influence a lot braking and acceleration.

-- 
Christos Dimitrakakis
Homepage: http://www.idiap.ch/~dimitrak/main.html
Music: http://olethros.dmusic.com

Hi Kennet

The problem of the historic cars is that most of them are non-free, so=20
only the author can modify them.
The only solution for us to correct the Austin-Healey is to make a 3D=20
model from zero and give it the technical datas we want.
I'm not familiar with licences but I think it is so.
Sorry.

Speedy


Kennet Eriksson a =E9crit :

>Driving at 247 km/h on the back straight(?) of E-track
>4 with the Austin-Healey knowing you will have
>difficulty braking at the end is quite interesting but
>I don't believe it to be historically accurate. You
>could buy the car with an electric overdrive so the
>fifth gear might not be too far from the truth but the
>most powerful version had 150 horsepower, would that
>be enough to scare the car up to 250? Maybe the engine
>is a bit too powerful in the gamecar?
>According to the always thrustworthy internet the car
>made 185/116 at most in a little less powerful
>versions. Maybe this car is modelled after some racing
>enhanced car? Some information about how the
>individual cars are modelled would be very
>interesting.
>Is nitpicking like this acceptable or should I just
>shut up and enjoying zapping along the tracks in my
>Austin-Healey? The game version is 1.2.4-1ubuntu7 on
>kubuntu and the extra cars from
>TORCS-1.2.4-data-cars-*. I enjoy the game immensely
>but I am also very interested in history so accurate
>cars would enhance the experience. Yes, most of the
>cars are well within reason as far as I can tell.
>Why do I drive this car? It is very interesting to
>drive it, when I try a modern car, say the 360-modena,
>I think; what a great car, and then I try some other
>car, it's just not interesting to drive. Driving a car
>that I have to treat gently is also, I believe, a good
>way to improve, something I need.
>
>Thank you for reading these ramblings,
>Regards
>Kennet
>
>
>
>_______________________________________________
>Torcs-users mailing list
>Tor...@li...
>https://lists.sourceforge.net/lists/listinfo/torcs-users
>
>
> =20
>

Driving at 247 km/h on the back straight(?) of E-track
4 with the Austin-Healey knowing you will have
difficulty braking at the end is quite interesting but
I don't believe it to be historically accurate. You
could buy the car with an electric overdrive so the
fifth gear might not be too far from the truth but the
most powerful version had 150 horsepower, would that
be enough to scare the car up to 250? Maybe the engine
is a bit too powerful in the gamecar?
According to the always thrustworthy internet the car
made 185/116 at most in a little less powerful
versions. Maybe this car is modelled after some racing
enhanced car? Some information about how the
individual cars are modelled would be very
interesting.
Is nitpicking like this acceptable or should I just
shut up and enjoying zapping along the tracks in my
Austin-Healey? The game version is 1.2.4-1ubuntu7 on
kubuntu and the extra cars from
TORCS-1.2.4-data-cars-*. I enjoy the game immensely
but I am also very interested in history so accurate
cars would enhance the experience. Yes, most of the
cars are well within reason as far as I can tell.
Why do I drive this car? It is very interesting to
drive it, when I try a modern car, say the 360-modena,
I think; what a great car, and then I try some other
car, it's just not interesting to drive. Driving a car
that I have to treat gently is also, I believe, a good
way to improve, something I need.

Thank you for reading these ramblings,
Regards
Kennet

hallo everyone,

i got a error at compiling:

grscreen.h:81: error: extra qualification =91cGrScreen::=92 on member =91=
camDraw=92
make[5]: *** [grmain.o] Fehler 1
make[4]: *** [subdirs] Fehler 1
make[3]: *** [subdirs] Fehler 1
make[2]: *** [subdirs] Fehler 1
make[1]: *** [subdirs] Fehler 1
make: *** [restart] Fehler 2


do you have an idea what i could do ... or what that mean

thx

Hi Jim

> Is This effective hardware accelleration  ?

Never. My old GeForce3 gives me more than 2500 FPS. You can simply check it this
way:
glxinfo | grep render

Direct rendering must be "Yes", and the renderer String should represent the
message of your driver. My string is:

direct rendering: Yes
OpenGL renderer string: GeForce3/AGP/SSE

> Is that -DRI message important ?

It depends on the driver: NVIDIA has AFAIK its own hack to talk to the hardware,
but for the ATI driver it might be important. The best places to get help are:
- The ATI driver documentation supplied with the driver, there is for sure a
trouble shooting section (I guess).
- Forums for your distro, you are for sure not the only one which has trouble
setting up 3D.

Bye, have fun,

Bernhard.

-- 

Visit my homepage http://www.berniw.org
Official TORCS racing: The TORCS Racing Board, http://www.berniw.org/trb

many moons ago, I succeeded in getting Torcs running
from source,  on my ATI laptop, running Fedora Core 4,
01:00.0 VGA compatible controller: ATI Technologies Inc RV350 [Mobility 
Radeon 9600 M10]
, running with ATIs  fglrx package of that (unknown) vintage.

With fglrx working, I was seeing frame-rates around 30-sec,
without it, I was getting an unusable 0.3 FPS

Kernel upgrades etc broke this somehere along the line,
and the whole effort to stay current collapsed.

Ive currently got Torcs installed from FC extras rpms,
and it consumes 100% cpu just sitting on the entry screen.
It gets 0.5 FPS when 'driving'

Very recently, I tried ATIs installer, which apparently does
a better integration with existing linux installation, but I havent
been able to discern a difference yet.  It  installed into
/lib/modules/fglrx, which implies its OS independent.
But with this in, I still only get 0.5 FPS

glxgears does run, giving these numbers for small/large screens

[jimc@harpo dev]$ glxgears
Xlib:  extension "XFree86-DRI" missing on display ":0.0".
1207 frames in 5.0 seconds = 241.400 FPS
1440 frames in 5.0 seconds = 288.000 FPS
1440 frames in 5.0 seconds = 288.000 FPS
1440 frames in 5.0 seconds = 288.000 FPS
1200 frames in 5.0 seconds = 240.000 FPS
600 frames in 7.0 seconds = 85.714 FPS
120 frames in 9.0 seconds = 13.333 FPS
120 frames in 10.0 seconds = 12.000 FPS
240 frames in 6.0 seconds = 40.000 FPS
1320 frames in 5.0 seconds = 264.000 FPS
1440 frames in 5.0 seconds = 288.000 FPS
1440 frames in 5.0 seconds = 288.000 FPS


The big dip in hte middle is where I went full-screen.

Is This effective hardware accelleration  ?
What should I be seeing, via what tools ?

Is that -DRI message important ?
Im running FC-x's standard  xorg server, not xfree86.

I have a steering wheel now, :-D
so Id love to get back to a fast, working, cvs-version of torcs.
Sorting my 3D seems a necessary 1st step.

tia
-jimc