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

Hi Joey

> I was built an ABS for my robot car. When I tested it, I found that the
> braking distance without abs is about 8 feet better than with ABS. Does
> torc account for static and kinnetic coefficient of friction? Cause, I

Yes. E.g. with car5-trb1 you have the best grip with 10% slip, slowly 
dropping then to 88% of the best grip. The drop-off etc. depends on the 
parametrization of the model.

> thought locking will transition the tire to kinetic coefficient of
> friction, thus increasing the braking distance.

So how does you ABS work, per wheel or in the robot with the brake? If 
you just operate the brake it could be that the brake-balance does not 
fit yet (maybe you lift the brake just because one wheel is a bit 
"locking", so your braking distance might be far from ideal).

Which car did you use, what are the test conditions (e.g. track, car, 
initial speed, etc.).

Kind regards

Bernhard

Hi

You have plenty of data to find it you, with the wheel spin velocity and 
the radius you can calculate the "speed" per wheel, with the car 
velocity and angular velocity you can find out the speed of the wheel 
position relative to the ground. With these you can calculate the slip 
per wheel. Or you can do it a bit more simple, like described in the 
robot tutorial:
http://www.berniw.org/tutorials/robot/torcs/robot/ch3/advanced.html

Kind regards

Bernhard

On 04/29/2015 02:26 AM, Joey Andres wrote:
> Thank you for the replies so far. You guys have been helpful. I have a
> new question, how do you detect if wheel is locking. I've looked at
> tWheelState->state, but I can't find any documentation abou it. tdble
> slipSide; and dble slipAccel; looks like somthing I should be intrested
> in, but then again, no parameters.
>
> I'm implementing my own ABS, and I'd like to know if a wheel is locking.
>
> Sincerely,
> Joey Andres
>
>
> ------------------------------------------------------------------------------
> One dashboard for servers and applications across Physical-Virtual-Cloud
> Widest out-of-the-box monitoring support with 50+ applications
> Performance metrics, stats and reports that give you Actionable Insights
> Deep dive visibility with transaction tracing using APM Insight.
> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
>
>
>
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users
>

Hello,

I was built an ABS for my robot car. When I tested it, I found that the
braking distance without abs is about 8 feet better than with ABS. Does
torc account for static and kinnetic coefficient of friction? Cause, I
thought locking will transition the tire to kinetic coefficient of
friction, thus increasing the braking distance.

Sincerely,
Joey Andres

Thank you for the replies so far. You guys have been helpful. I have a new
question, how do you detect if wheel is locking. I've looked at
tWheelState->state, but I can't find any documentation abou it. tdble
slipSide; and dble slipAccel; looks like somthing I should be intrested in,
but then again, no parameters.

I'm implementing my own ABS, and I'd like to know if a wheel is locking.

Sincerely,
Joey Andres

Hi

It is almost identical, differences:
- http://www.berniw.org/trb/forum/showthread.php?topicid=3421
- Some "char*" might be "const char*" now

Kind regards

Bernhard

On 04/17/2015 10:10 PM, Phanideep Gonthina wrote:
> Heyy David,
>
> Thank you so much for replying. I actually need Torcs for a project that
> I am doing and do need all the 'robot' features. I had found this
> tutorial online and was following it:
>
> http://www.berniw.org/aboutme/publications/torcs.pdf
>
> I will try 1.3.6 now following your instructions. Is the 'robot making'
> part similar in both the versions?
>
> Thanks again.
>
> -Phanideep.
>
> On Fri, Apr 17, 2015 at 3:37 PM, David Savinkoff <dsa...@te...
> <mailto:dsa...@te...>> wrote:
>
>     Hi,
>
>     Here is how I compile the latest TORCS in Linux.
>     TORCS is installed in your home directory without
>     root permissions (excellent for experimenting).
>
>     See the enclosed TorcsNotes.txt
>
>     TORCS-1.2.4 is old and requires several tarballs
>     (I have never compiled it and recommend torcs-1.3.6)
>
>
>
>
> ------------------------------------------------------------------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
>
>
>
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users
>

Hi Joey

See RtTrackSurfaceNormalL (function in Robottools).

Kind regards

Bernhard

On 04/23/2015 03:34 AM, Joey Andres wrote:
> I'm wondering how to get the road camber (angle of the road) in Torcs.
> I've been playing with the normal vector, but I'm somewhat lost in this.
>
>
> ------------------------------------------------------------------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
>
>
>
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users
>

Hi Joey

TORCS uses internally SI-Units, so tis is in m/s.

Kind regards

Bernhard

On 04/24/2015 05:30 PM, Joey Andres wrote:
> Like mentioned in the subject, what is the unit of tPublicCar.speed? I'm
> getting ~22units when I'm running 80kmh.
>
> Sincerely,
> Joey Andres
>
>
> ------------------------------------------------------------------------------
> One dashboard for servers and applications across Physical-Virtual-Cloud
> Widest out-of-the-box monitoring support with 50+ applications
> Performance metrics, stats and reports that give you Actionable Insights
> Deep dive visibility with transaction tracing using APM Insight.
> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
>
>
>
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users
>

Like mentioned in the subject, what is the unit of tPublicCar.speed? I'm
getting ~22units when I'm running 80kmh.

Sincerely,
Joey Andres

I'm wondering how to get the road camber (angle of the road) in Torcs. I've
been playing with the normal vector, but I'm somewhat lost in this.

Heyy David,

Thank you so much for replying. I actually need Torcs for a project that I
am doing and do need all the 'robot' features. I had found this tutorial
online and was following it:

http://www.berniw.org/aboutme/publications/torcs.pdf

I will try 1.3.6 now following your instructions. Is the 'robot making'
part similar in both the versions?

Thanks again.

-Phanideep.

On Fri, Apr 17, 2015 at 3:37 PM, David Savinkoff <dsa...@te...> wrote:

> Hi,
>
> Here is how I compile the latest TORCS in Linux.
> TORCS is installed in your home directory without
> root permissions (excellent for experimenting).
>
> See the enclosed TorcsNotes.txt
>
> TORCS-1.2.4 is old and requires several tarballs
> (I have never compiled it and recommend torcs-1.3.6)
>

Hi,

Here is how I compile the latest TORCS in Linux.
TORCS is installed in your home directory without
root permissions (excellent for experimenting). 

See the enclosed TorcsNotes.txt

TORCS-1.2.4 is old and requires several tarballs
(I have never compiled it and recommend torcs-1.3.6)

Hi,

1) I have finished the 'make install' step, gave permissions, and set the
environment for torcs-1.2.4 (as specified in the 'Torcs manual')
But when I execute 'torcs', it says:

"The program 'torcs' is currently not installed.  To run 'torcs' please ask
your administrator to install the package 'torcs'"

2) When I try 'make datainstall' it gives me the following error:

make[1]: *** data: No such file or directory.  Stop.
/usr/src/torcs/torcs-1.2.4/Make-default.mk:811: recipe for target
'installdatadirs' failed
make: *** [installdatadirs] Error 1


The problem is in the following lines of code of "Make-default'

installdatadirs:
    @if [ -n "${DATASUBDIRS}" ] ; \               // Line 811
    then R=`for I in ${DATASUBDIRS} ; \
    do echo $$I ;\
    done | sort -u` ; \
    RecurseDirs="$$R" ; \
    RecurseFlags="datainstall" ; \
    ${recursedirs} ; \
    fi


Kindly let me know what is to be done. I am unable to run "torcs".

Thank you.

-Phanideep

Hi,

1) I have finished the 'make install' step, gave permissions, and set the
environment for torcs-1.2.4 (as specified in the 'Torcs manual')
But when I execute 'torcs', it says:

"The program 'torcs' is currently not installed.  To run 'torcs' please ask
your administrator to install the package 'torcs'"

2) When I try 'make datainstall' it gives me the following error:

make[1]: *** data: No such file or directory.  Stop.
/usr/src/torcs/torcs-1.2.4/Make-default.mk:811: recipe for target
'installdatadirs' failed
make: *** [installdatadirs] Error 1


The problem is in the following lines of code of "Make-default'

installdatadirs:
    @if [ -n "${DATASUBDIRS}" ] ; \               // Line 811
    then R=`for I in ${DATASUBDIRS} ; \
    do echo $$I ;\
    done | sort -u` ; \
    RecurseDirs="$$R" ; \
    RecurseFlags="datainstall" ; \
    ${recursedirs} ; \
    fi


I have attached a screenshot for your reference.

Kindly let me know what is to be done. I am unable to run "torcs".

Thank you.

-Phanideep

---------- Forwarded message ----------
From: Phanideep Gonthina <pha...@gm...>
Date: Tue, Apr 14, 2015 at 6:34 PM
Subject: Re: [Torcs-users] Torcs run error
To: be...@bl...


Dear Sir,

1) I have finished the 'make install' step, gave permissions, and set the
environment for torcs-1.2.4 (as specified in your guide)
But when I execute 'torcs', it says:

"The program 'torcs' is currently not installed.  To run 'torcs' please ask
your administrator to install the package 'torcs'"

2) When I try 'make datainstall' it gives me the following error:

make[1]: *** data: No such file or directory.  Stop.
/usr/src/torcs/torcs-1.2.4/Make-default.mk:811: recipe for target
'installdatadirs' failed
make: *** [installdatadirs] Error 1


The problem is in the following lines of code of "Make-default'

installdatadirs:
    @if [ -n "${DATASUBDIRS}" ] ; \               // Line 811
    then R=`for I in ${DATASUBDIRS} ; \
    do echo $$I ;\
    done | sort -u` ; \
    RecurseDirs="$$R" ; \
    RecurseFlags="datainstall" ; \
    ${recursedirs} ; \
    fi


I have attached a screenshot for your reference.

Kindly let me know what is to be done. I am unable to run "torcs".

Thank you.

-Phanideep

Hi

It depends where you have installed it, the command would be "torcs". If 
you have no path set to the installation location you have to give the 
full path, e.g. on my system I start it with "~/torcs_bin/bin/torcs" 
(this will not be the case for your system, just as example).

Btw. did you also run "make datainstall"?

Kind regards

Bernhard

On 04/13/2015 10:36 AM, Phanideep wrote:
> Dear Sir,
>
> I have finished the 'make install' step, gave permissions, and set the environment for torcs-1.2.4 (as specified in your guide)
> But when I execute 'torcs', it says:
>
> "The program 'torcs' is currently not installed.  To run 'torcs' please ask your administrator to install the package 'torcs'"
>
> How do I run Torcs?
>
> -Phanideep.
>
>
>
>
> ------------------------------------------------------------------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
>
>
>
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users
>

Hi,

I found the shocking truth that I independently derived
a formula for tire friction. The formula is:

sx = (vt - wrl) / (MAX(fabs(vt) , fabs(wrl)) )

Please read page 2 of this authoritive paper:
http://bsesrv214.bse.vt.edu/Hop/Papers/Tire-Road%20Friction%20Coefficient.pdf

TORCS and Speed Dreams need fixin'

Sincerely,
David Savinkoff


----- David Savinkoff wrote:
> ----- Bernhard Wymann wrote:
> > Hi David
> > 
> > It makes in my opinion no sense in respect to the tire model, the slip 
> > definition looks consistent as is for me
> 
> I cannot make an argument here because I am not absolutely certain how
> the tire model interacts with 'sx = (vt - wrl) / fabs(vt)'
> 
> >
> > (in your analysis there is a little glitch, vt is not the car velocity,
> > it is the wheel specific velocity, all wheels have usually at least
> > slightly different velocities).
> >
> 
> I may have not have expressed myself properly when I said 'car velocity'
> I envisioned a single wheel in contact with the road where 'vt' is the
> velocity in the direction the wheel is pointed... and where 'wrl' is the
> wheel spin velocity (can only be in the direction the wheel is pointed).
> 
> >
> > The proposed change would cause a discontinuity which would usually 
> > lower the grip.
> >
> 
> 'sx = (vt - wrl) / ( 0.000001f + MAX(fabs(vt) , fabs(wrl)) )' has an
> inflection point where the wheel is actuated to slip by
> deceleration, or slip by acceleration.
> The discontinuity is caused by the 'act' of braking or accelerating.
> 
> I believe 'sx' may need three different functions to include:
> 1) braking
> 2) accelerating
> 3) wheels rotating backwards to the direction of travel.
> The formula for 'sx' is my best so far, and it seems to work ok.
> 
> I visualize vt and wrl as two parallel lines showing their respective
> lengths. Where vt - wrl is the length part that differs. eg.
> _________________________
> _____________________________________
> 
> The formula for 'sx' suggests that one wants to determine a 'normalized'
> slip value given vt, wrl, and vt-wrl. Note that the slip happens in
> several different situations.
> 
> > I think a better solution for drifting is to adjust the parameters of 
> > the tire model
> 
> I just tried different tire pressures (other than the default 40 psi).
> I will comment on this list when I come across interesting things.
> 
> ***
> Please actually try my latest patch traction2.diff on different cars
> and tracks. Drive aggressively (off-track also) or ride with Olethros 6.
> ***
> 
> > , differential and suspension. If you don't mind to edit 
> > the XML you can adjust as well the engine braking coefficient (instead 
> > of adapting your accelerator control in your mind;-) ).
> > 
> > Other opinions?
> >
> 
> Others on the list... Your opinions please.
> 
> Sincerely,
> David Savinkoff
> 
> 
> ------------------------------------------------------------------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users

Dear Sir,

I have finished the 'make install' step, gave permissions, and set the environment for torcs-1.2.4 (as specified in your guide)
But when I execute 'torcs', it says:

"The program 'torcs' is currently not installed.  To run 'torcs' please ask your administrator to install the package 'torcs'"

How do I run Torcs?

-Phanideep.

Hi,

I am getting the following errors in the 'make install' step for Torcs
(screenshot attached):

/usr/src/torcs/torcs-1.2.4/Make-default.mk:405: recipe for target
'torcs-bin' failed
make[2]: *** [torcs-bin] Error 1
make[2]: Leaving directory '/usr/src/torcs/torcs-1.2.4/src/linux'
/usr/src/torcs/torcs-1.2.4/Make-default.mk:791: recipe for target
'installdirs' failed
make[1]: *** [installdirs] Error 1
make[1]: Leaving directory '/usr/src/torcs/torcs-1.2.4/src'
/usr/src/torcs/torcs-1.2.4/Make-default.mk:791: recipe for target
'installdirs' failed
make: *** [installdirs] Error 1


I am following the steps exactly as instructed in the 'Torcs Manual
Installation and Robot tutorial - Bernhard Wymann' guide.

Could you help?

Thank you.

-Phanideep

----- BRIAN MARTIN Owner wrote:
> David,
> 
> I tried a couple of your patches and found i liked the original code
> better. Of course, this is completely subjective.
> 
> Tire pressure, the last time i looked, is not implemented in simuv2.
> 
> Changes to the car physics file will get you far more profound changes

Most of these changes are too profound for me.

> in car handling than small sim hacks, especially  changes to mu, clift, 
> wings and brake partitioning. I have attached a sample physics file that works
> very well for me in terms of drifting and just overall fun factor.
> 
> Brian 
>

Thank you very much. I will be trying this file out.
The tire pressure can be changed by changing a value in wheel.cpp
and re-compiling. It does make a difference.

For a few years I changed parameters in the .xml files. Eventually
I decided to look at the code. I have taken into account that
 TORCS car parameters are important.

Some of my concerns are:
- I could not find satisfactory performance by changing the parameters.
- I found that by using 'the golden ratio' in making some empirical
  formulas for TORCS parameters, that I could get a sweet spot.
  (this may show that TORCS has a chaotic function)
- I have been concerned that the operating points for TORCS are
  narrower than in nature.
- I have noticed that the road has abrupt traction differences on either
  side of the racing line that can cause the car to wander similarly to
  loose steering linkage.

>
> ----- Original Message -----
> From: "David Savinkoff"
> To
> Cc: "torcs-users" <tor...@li...>
> Sent: Wednesday, April 8, 2015 4:05:20 PM
> Subject: Re: [Torcs-users] Patch for wheel spin velocity traction
> 
> ----- Bernhard Wymann wrote:
> > Hi David
> > 
> > It makes in my opinion no sense in respect to the tire model, the slip 
> > definition looks consistent as is for me
> 
> I cannot make an argument here because I am not absolutely certain how
> the tire model interacts with 'sx = (vt - wrl) / fabs(vt)'
> 
> >
> > (in your analysis there is a little glitch, vt is not the car velocity,
> > it is the wheel specific velocity, all wheels have usually at least
> > slightly different velocities).
> >
> 
> I may have not have expressed myself properly when I said 'car velocity'
> I envisioned a single wheel in contact with the road where 'vt' is the
> velocity in the direction the wheel is pointed... and where 'wrl' is the
> wheel spin velocity (can only be in the direction the wheel is pointed).
> 
> >
> > The proposed change would cause a discontinuity which would usually 
> > lower the grip.
> >
> 
> 'sx = (vt - wrl) / ( 0.000001f + MAX(fabs(vt) , fabs(wrl)) )' has an
> inflection point where the wheel is actuated to slip by
> deceleration, or slip by acceleration.
> The discontinuity is caused by the 'act' of braking or accelerating.
> 
> I believe 'sx' may need three different functions to include:
> 1) braking
> 2) accelerating
> 3) wheels rotating backwards to the direction of travel.
> The formula for 'sx' is my best so far, and it seems to work ok.
> 
> I visualize vt and wrl as two parallel lines showing their respective
> lengths. Where vt - wrl is the length part that differs. eg.
> _________________________
> _____________________________________
> 
> The formula for 'sx' suggests that one wants to determine a 'normalized'
> slip value given vt, wrl, and vt-wrl. Note that the slip happens in
> several different situations.
> 
> > I think a better solution for drifting is to adjust the parameters of 
> > the tire model
> 
> I just tried different tire pressures (other than the default 40 psi).
> I will comment on this list when I come across interesting things.
> 
> ***
> Please actually try my latest patch traction2.diff on different cars
> and tracks. Drive aggressively (off-track also) or ride with Olethros 6.
> ***
> 
> > , differential and suspension. If you don't mind to edit 
> > the XML you can adjust as well the engine braking coefficient (instead 
> > of adapting your accelerator control in your mind;-) ).
> > 
> > Other opinions?
> >
> 
> Others on the list... Your opinions please.
> 
> Sincerely,
> David Savinkoff
> 

----- Bernhard Wymann wrote:
> Hi David
> 
> It makes in my opinion no sense in respect to the tire model, the slip 
> definition looks consistent as is for me

I cannot make an argument here because I am not absolutely certain how
the tire model interacts with 'sx = (vt - wrl) / fabs(vt)'

>
> (in your analysis there is a little glitch, vt is not the car velocity,
> it is the wheel specific velocity, all wheels have usually at least
> slightly different velocities).
>

I may have not have expressed myself properly when I said 'car velocity'
I envisioned a single wheel in contact with the road where 'vt' is the
velocity in the direction the wheel is pointed... and where 'wrl' is the
wheel spin velocity (can only be in the direction the wheel is pointed).

>
> The proposed change would cause a discontinuity which would usually 
> lower the grip.
>

'sx = (vt - wrl) / ( 0.000001f + MAX(fabs(vt) , fabs(wrl)) )' has an
inflection point where the wheel is actuated to slip by
deceleration, or slip by acceleration.
The discontinuity is caused by the 'act' of braking or accelerating.

I believe 'sx' may need three different functions to include:
1) braking
2) accelerating
3) wheels rotating backwards to the direction of travel.
The formula for 'sx' is my best so far, and it seems to work ok.

I visualize vt and wrl as two parallel lines showing their respective
lengths. Where vt - wrl is the length part that differs. eg.
_________________________
_____________________________________

The formula for 'sx' suggests that one wants to determine a 'normalized'
slip value given vt, wrl, and vt-wrl. Note that the slip happens in
several different situations.

> I think a better solution for drifting is to adjust the parameters of 
> the tire model

I just tried different tire pressures (other than the default 40 psi).
I will comment on this list when I come across interesting things.

***
Please actually try my latest patch traction2.diff on different cars
and tracks. Drive aggressively (off-track also) or ride with Olethros 6.
***

> , differential and suspension. If you don't mind to edit 
> the XML you can adjust as well the engine braking coefficient (instead 
> of adapting your accelerator control in your mind;-) ).
> 
> Other opinions?
>

Others on the list... Your opinions please.

Sincerely,
David Savinkoff

Hi David

It makes in my opinion no sense in respect to the tire model, the slip 
definition looks consistent as is for me (in your analysis there is a 
little glitch, vt is not the car velocity, it is the wheel specific 
velocity, all wheels have usually at least slightly different velocities).

The proposed change would cause a discontinuity which would usually 
lower the grip.

I think a better solution for drifting is to adjust the parameters of 
the tire model , differential and suspension. If you don't mind to edit 
the XML you can adjust as well the engine braking coefficient (instead 
of adapting your accelerator control in your mind;-) ).

Other opinions?

Kind regards

Bernhard

On 03/27/2015 07:29 PM, David Savinkoff wrote:
> Hi Bernhard,
>
> Please apply this patch; which makes for better maneuvering while
> drifting and overtaking in corners.
>
> What I found interesting with the line of code:
> sx = (vt - wrl) / fabs(vt);
>
> Is that:
> sx = (car_velocity - wheel_spin_velocity = slip_velocity) / car_velocity;
> Has no [units] and is 'with respect to' car_velocity.
> This works well with braking because the wheel_spin_velocity is less
> than the car_velocity.
>
> When wheel_spin_velocity is faster than car_velocity, it appears that this
> unitless value works better with respect to wheel_spin_velocity.
>
> Thus:
> sx = (vt - wrl) / ( MAX(fabs(vt) , fabs(wrl)) );
>
> Sincerely,
> David Savinkoff
>
>
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming The Go Parallel Website, sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for all
> things parallel software development, from weekly thought leadership blogs to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
>
>
>
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users
>

Hi

You can subdivide the segments, example:
- You have a pit staring right at the start with length 15m
- You have a start line with 0.5 m length in track direction
- Your initial side banner is 4 meters

Solution:
- surface "texture 1", wall with banner, length e.g. 4m, texture 
type=continuous, texture link with previous=yes, texture start on 
boundary=yes
- surface "texture 2," wall, length e.g. 15m, , texture type=continuous, 
texture link with previous=yes, texture start on boundary=yes

- Segment 1, length 0.5m, surface="texture 1"
- Segment 2, length 3.5m, surface="texture 1"
- Segment 3, length 11m, surface="texture 2"

Alternative (just differences to the above):
- Two equally sized textures (so you can take the same image and just 
paint on one the letters)
- Both textures are of same length, e.g. 15m
- "texture 2" has texture start on boundary=no (so the texture 
coordinates are not reset to 0 on the texture change)

Regarding the bumps, sudden changes: Did you try to increase the 
resolution of such critical segments, I usually adapt the "profil 
steps"/"profile steps length" property?

Regarding the grandstand, I would create this by revolving too 
(important: from a line), did you use ac3d? IIRC you can set various 
options and the center point, eventually you need to edit the object 
afterwards. Remember to use just single sided smooth shaded surfaces.

To distribute objects around a center you can use the "Replicator" in 
ac3d IIRC.

If you have trouble with a certain model part feel free to post it, then 
people can help you to fix the issues.

Kind regards

Bernhard

On 03/11/2015 09:36 AM, stockroom wrote:
>
> Well, the building construction seems to be progressing, however slowly.
> Anyway, there's still something in this 3D design that causes problems
> or some sort of limitations. Perhaps they could be handled in this same
> thread if there are answers and solutions.
>
> First, the addition of text (objects) to certain surfaces, most
> frequently the track barrier or pit wall. (see example image,
> http://stcreativedesigns.fhero.net/img/torcs-sfl00.png) This primarily
> deals with placing a track logo at the start/finish line, and for oval
> tracks, all the wall markings (like "turn 4", track name and possible
> wall ads).
>
> Now, before any suggestion of simply putting them in some wall texture
> and editing the track.xml where needed, it doesn't always work. For the
> start/finish line, the tracks here have two 0.5-meter segments at the
> start/finish line, so the track logo just won't fit there. And those
> segments (first and last) can't be lengthened because that would cause
> problems with the pit stalls (being cut short or pushed too far ahead on
> either side of the line), and obviously with the robots' already erratic
> pitting. So an external object for that track logo, placed right at the
> start/finish line, touching the pit wall, seems to be the only
> conceivable option. But as that example image shows, the text is totally
> blurred.
>
> I tried editing the wall itself in the .ac file, hoping to get a decent
> rectangle there, cutting it away and adding the logo to that new object,
> but it failed completely. There were so many lines from the triangles on
> that wall that the logo texture didn't appear like it should have
> (nothing to do with "texture repeat", which was 1 like usually). An
> attempt at moving the vertices of that object, to create a "clean"
> rectangle for the logo texture made no difference.
>
> On oval tracks, the banking is the obstacle. As I've experienced
> numerous times before, short track segments (like something needed for a
> "turn 4" wall texture) where banking is changing always cause stupid
> bumps in the track surface and the nearby wall, to the detriment of
> racing. It doesn't matter if you use "banking start/end" or "z end". Any
> "profil steps length" manipulations I tried always failed too. So the
> lesson learned was not to have such short track segments in turns with
> the changing banking. And as those "turn 4" types of markings stand
> exactly where the banking is changing (rising or falling), putting them
> in the wall texture and generating the track will result in those
> annoying bumps. So they should be some kind of external objects, like
> earlier with the start/finish line.
>
> Now the question is, can the graphical issues be solved? Clearly, that
> kind of blurred black mess can't be left in the final version. In that
> object, I used a simple one-sided rectangle with the track logo (on a
> transparent background) as its selected texture. The normal was pointing
> in the right direction. An alternative, the same logo texture with a
> background color, close to the wall's color, looked equally awful. I
> don't know what else is relevant.
>
> I suppose the start/finish line logo could be scrapped and placed
> somewhere else on the track, but those oval track markings (commonplace
> on real-life ovals too) can hardly be moved around. I hope there's some
> solution left.
>
> -------------------
> The second graphical obstacle deals with curving structures outside the
> track, which may only be added after track generation (very difficult to
> predict their position beforehand). Those include walls and fences of
> different kind (mainly as an additional barrier between the track and
> buildings/grandstands, imaginary roads and paths etc.) and the
> grandstands themselves. The curving grandstand would mean a part of it
> connecting two rectangular grandstands, like in a turn. That can be
> avoided on fictional road courses but would be absolutely essential to
> most oval tracks, instead of gaps between grandstands in turns or on the
> curving front stretch.
>
> I wonder if it's possible to create a curving structure that would
> function without graphical problems/ugliness and would neatly join two
> rectangular structures. I tried creating one such grandstand (see
> example image, http://stcreativedesigns.fhero.net/img/torcs-grst00.png)
> with lots of short (straight) parts connected at an angle, and the end
> result is pretty terrible. There is obviously overlap of objects and
> surfaces, so possibly another graphical mess awaiting. "Revolve" seems
> to be useless too: that usually generates gigantic objects that always
> have to be resized, not to mention the inevitable flickering of 6473
> surfaces.
>
> If this is way too complicated for my limited skills, I'll have to
> settle for something simpler. The 3D learning curve is much more like a
> steep slope, and I feel dizzy already here at the bottom. Good advice is
> again appreciated.
>
> Sincerely,
> stockroom staff
> ST Creative Designs
> http://stcreativedesigns.fhero.net/
>
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming The Go Parallel Website, sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for all
> things parallel software development, from weekly thought leadership blogs to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
> _______________________________________________
> Torcs-users mailing list
> Tor...@li...
> https://lists.sourceforge.net/lists/listinfo/torcs-users
>

Hi All,

This patch is a new standalone experiment, similar to the previous.