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

----- BRIAN MARTIN Owner wrote:
> 
> David,
> 
> I have been using some of your wheel force changes with good 
> results.
> 
> I discovered that using this code:
> 
>                 tdble slipFactor = 0.36f;
>                 if (v < 0.1f)
>                         {
>                         //reduce drifting car on angled surface at very low velocities
>                         slipFactor = 2.0f;
>                         }
> 
>                 slipMagnitude = slipFactor + sqrt( MAX(MAX((v2), (vn*vn + wrl*wrl)), (vn*vn + (vt - wrl)*(vt - wrl))) );
> 
> prevents cars from drifting on the track when the car is stopped. This is
> especially noticeable on the ovals. The cars also seem to behave better when
> spinning out and coming to a stop.
>
Good.
>
One thing to note, is that my changes cause the tires to squeal
at low speeds. This is Not a problem with my code, it IS a
problem with the squeal code because the tires should not
squeal. 
> 
> I am using a hacked sim, but i think the results should be the same with simuv2.
>
I'm currently using my patch (wheel.cpp.2016jan14.diff) against
the original torcs-1.3.6.tar.bz2 source code, and default cars,
and reasonably simple user settings/calibrations.
> 
> Brian  
> 
> ----- Original Message -----
> From: "David Savinkoff"
> To: "David Savinkoff"
> Sent: Friday, January 8, 2016 3:08:23 PM
> Subject: Re: [Torcs-users] Patch for wheel spin velocity traction
> 
> ----- David Savinkoff wrote:
> > This patch removes legacy duct tape which was used as a
> > work-around for a previously nonexistent friction circle.
> > 
> > There is another piece of duct tape which impedes the car
> > from accelerating forward while allowing unimpeded sideways
> > acceleration when drifting. I'll remove that when I find it.
> > 
> > Note that RELAXATION2 is a low-pass Infinite Impulse Response
> > filter of the form: f(s) = as/(b+as) where s is in the s-domain,
> 
> correction:  f(s) = (a/s)/(b+a/s)
> 
> > 'a' is a spring-constant, and 'b' is a friction constant
> > mechanically acting on the rotation of the wheel. This will
> > certainly affect the physics of TORCS.
> > 
> > ----- David Savinkoff wrote:
> > > Hi Everybody,
> > > 
> > > This patch makes TORCS the best simulator wherever
> > > 'The Rubber meets the Road' is concerned. Try it.
> > > 
> > > ----- David Savinkoff wrote:
> > > > ----- David Savinkoff wrote:
> > > > > Hi,
> > > > > 
> > > > > In my last patch I had the following code:
> > > > > slip_magnitude = 0.000001f + sqrt( MAX(MAX((vn*vn + vt*vt)
> > > > > 
> > > > > Where (v2) should be equal to (vn*vn + vt*vt) but is not,
> > > > > because of lack of precision of the float data type.
> > > > > 
> > > > > Here I use (v2) for velocity squared because the car drives
> > > > > noticeably better, and (v2) is obviously more accurate.
> > > > > 
> > > > > slip_magnitude = 0.000001f + sqrt( MAX(MAX((v2)
> > > > > 
> > > > > BTW is there somewhere I can find recent TORCS source code
> > > > > changes so that I can test and experiment?
> > > > > 
> > > > > Sincerely,
> > > > > David Savinkoff
> > > > 
> > > > Hi,
> > > > 
> > > > Here is another patch for floating point improvements.
> > > > 
> > > > I used sqrtf() where it seemed to make a speed improvement
> > > > on my slow computer.
> > > > 
> > > > I used (double) to get more accuracy for CosA and SinA
> > > > 
> > > > It is obvious that (float) is not sufficient in some
> > > > circumstances, and that (double) is slower in others.
> > > > 
> > > > I hope these improvements in wheel.cpp help with making
> > > > other problems more characterizable.
> > > > 
> > > > Sincerely,
> > > > David Savinkoff
> > > > 
> > > > ps.
> > > > 
> > > > Speed Dreams should incorporate this patch too.
> > > > Note that this patch will give you deja vu a few times
> > > > per lap (even if you do motocross).
> > > 
> > 
> 
> 

Hi Brian,

I hope that a few improvements to the TORCS simulation
help with finding more improvements, so that TORCS
operates perfectly from first principles. Speed Dreams also.

I'm currently pondering over why TORCS seems to have
an excessive force component that acts in opposition to
the car moving forward (doesn't affect sideways) when 
the car is drifting with acceleration (possibly with braking
too). This problem makes the car spin-out, and also affects
donuts. I think there is some ad-hoc code somewhere,
or maybe excessive wind-resistance at slow speeds, or
some other vector component problem, or an errant
wheel-hop algorithm (where is this code?).

I will be testing your code over the next while. It's good
to debug the non-computer aspects of TORCS.

Hi,
This patch will probably be the last I make for
wheel.cpp because I've meticulously examined and
tested the code here repeatedly with results that have
converged to perfection.

Please apply this patch.

----- David Savinkoff wrote:
> This patch removes legacy duct tape which was used as a
> work-around for a previously nonexistent friction circle.
> 
> There is another piece of duct tape which impedes the car
> from accelerating forward while allowing unimpeded sideways
> acceleration when drifting. I'll remove that when I find it.
>
This problem seems to be diminished as TORCS is made more
correct.
> 
> Note that RELAXATION2 is a low-pass Infinite Impulse Response
> filter of the form: f(s) = (1/as) / (b+1/as) where s is in the s-domain,
> 'a' is a spring-constant, and 'b' is a friction constant
> mechanically acting on the rotation of the wheel. This will
> certainly affect the physics of TORCS.
> 
> ----- David Savinkoff wrote:
> > Hi Everybody,
> > 
> > This patch makes TORCS the best simulator wherever
> > 'The Rubber meets the Road' is concerned. Try it.
> > 
> > ----- David Savinkoff wrote:
> > > ----- David Savinkoff wrote:
> > > > Hi,
> > > > 
> > > > In my last patch I had the following code:
> > > > slip_magnitude = 0.3f + sqrt( MAX(MAX((vn*vn + vt*vt)
> > > > 
> > > > Where (v2) should be equal to (vn*vn + vt*vt) but is not,
> > > > because of lack of precision of the float data type.
> > > > 
> > > > Here I use (v2) for velocity squared because the car drives
> > > > noticeably better, and (v2) is obviously more accurate.
> > > > 
> > > > slip_magnitude = 0.3f + sqrt( MAX(MAX((v2)
> > > > 
> > > > BTW is there somewhere I can find recent TORCS source code
> > > > changes so that I can test and experiment?
> > > > 
> > > > Sincerely,
> > > > David Savinkoff
> > > 
> > > Hi,
> > > 
> > > Here is another patch for floating point improvements.
> > > 
> > > I used sqrtf() where it seemed to make a speed improvement
> > > on my slow computer.
> > > 
> > > I used (double) to get more accuracy for CosA and SinA
> > > 
> > > It is obvious that (float) is not sufficient in some
> > > circumstances, and that (double) is slower in others.
> > > 
> > > I hope these improvements in wheel.cpp help with making
> > > other problems more characterizable.
> > > 
> > > Sincerely,
> > > David Savinkoff
> > > 
> > > ps.
> > > 
> > > Speed Dreams should incorporate this patch too.
> > > Note that this patch will give you deja vu a few times
> > > per lap.
> > 
> 

----- David Savinkoff wrote:
> This patch removes legacy duct tape which was used as a
> work-around for a previously nonexistent friction circle.
> 
> There is another piece of duct tape which impedes the car
> from accelerating forward while allowing unimpeded sideways
> acceleration when drifting. I'll remove that when I find it.
> 
> Note that RELAXATION2 is a low-pass Infinite Impulse Response
> filter of the form: f(s) = as/(b+as) where s is in the s-domain,

correction:  f(s) = (a/s)/(b+a/s)

> 'a' is a spring-constant, and 'b' is a friction constant
> mechanically acting on the rotation of the wheel. This will
> certainly affect the physics of TORCS.
> 
> ----- David Savinkoff wrote:
> > Hi Everybody,
> > 
> > This patch makes TORCS the best simulator wherever
> > 'The Rubber meets the Road' is concerned. Try it.
> > 
> > ----- David Savinkoff wrote:
> > > ----- David Savinkoff wrote:
> > > > Hi,
> > > > 
> > > > In my last patch I had the following code:
> > > > slip_magnitude = 0.000001f + sqrt( MAX(MAX((vn*vn + vt*vt)
> > > > 
> > > > Where (v2) should be equal to (vn*vn + vt*vt) but is not,
> > > > because of lack of precision of the float data type.
> > > > 
> > > > Here I use (v2) for velocity squared because the car drives
> > > > noticeably better, and (v2) is obviously more accurate.
> > > > 
> > > > slip_magnitude = 0.000001f + sqrt( MAX(MAX((v2)
> > > > 
> > > > BTW is there somewhere I can find recent TORCS source code
> > > > changes so that I can test and experiment?
> > > > 
> > > > Sincerely,
> > > > David Savinkoff
> > > 
> > > Hi,
> > > 
> > > Here is another patch for floating point improvements.
> > > 
> > > I used sqrtf() where it seemed to make a speed improvement
> > > on my slow computer.
> > > 
> > > I used (double) to get more accuracy for CosA and SinA
> > > 
> > > It is obvious that (float) is not sufficient in some
> > > circumstances, and that (double) is slower in others.
> > > 
> > > I hope these improvements in wheel.cpp help with making
> > > other problems more characterizable.
> > > 
> > > Sincerely,
> > > David Savinkoff
> > > 
> > > ps.
> > > 
> > > Speed Dreams should incorporate this patch too.
> > > Note that this patch will give you deja vu a few times
> > > per lap (even if you do motocross).
> > 
> 

This patch removes legacy duct tape which was used as a
work-around for a previously nonexistent friction circle.

There is another piece of duct tape which impedes the car
from accelerating forward while allowing unimpeded sideways
acceleration when drifting. I'll remove that when I find it.

Note that RELAXATION2 is a low-pass Infinite Impulse Response
filter of the form: f(s) = as/(b+as) where s is in the s-domain,
'a' is a spring-constant, and 'b' is a friction constant
mechanically acting on the rotation of the wheel. This will
certainly affect the physics of TORCS.

----- David Savinkoff wrote:
> Hi Everybody,
> 
> This patch makes TORCS the best simulator wherever
> 'The Rubber meets the Road' is concerned. Try it.
> 
> ----- David Savinkoff wrote:
> > ----- David Savinkoff wrote:
> > > Hi,
> > > 
> > > In my last patch I had the following code:
> > > slip_magnitude = 0.000001f + sqrt( MAX(MAX((vn*vn + vt*vt)
> > > 
> > > Where (v2) should be equal to (vn*vn + vt*vt) but is not,
> > > because of lack of precision of the float data type.
> > > 
> > > Here I use (v2) for velocity squared because the car drives
> > > noticeably better, and (v2) is obviously more accurate.
> > > 
> > > slip_magnitude = 0.000001f + sqrt( MAX(MAX((v2)
> > > 
> > > BTW is there somewhere I can find recent TORCS source code
> > > changes so that I can test and experiment?
> > > 
> > > Sincerely,
> > > David Savinkoff
> > 
> > Hi,
> > 
> > Here is another patch for floating point improvements.
> > 
> > I used sqrtf() where it seemed to make a speed improvement
> > on my slow computer.
> > 
> > I used (double) to get more accuracy for CosA and SinA
> > 
> > It is obvious that (float) is not sufficient in some
> > circumstances, and that (double) is slower in others.
> > 
> > I hope these improvements in wheel.cpp help with making
> > other problems more characterizable.
> > 
> > Sincerely,
> > David Savinkoff
> > 
> > ps.
> > 
> > Speed Dreams should incorporate this patch too.
> > Note that this patch will give you deja vu a few times
> > per lap (even if you do motocross).
>