rec.autos.simulators

Slot

James Larson

James Larson
Omaha, Nebraska, USA

What N3.ini or NL.ini settings are you using for FF?

James

James Larson
Omaha, Nebraska, USA

Slot

My interpretation of the GPL FF parameters is as follows:

The description of force_feedback_damping seems to be missing
the important part, i.e. that you should use it to tell GPL how
much inherent damping you have in your force feedback wheel.
With a perfect force feedback wheel, sending a 10% force command
should generate a force that feels like one tenth of the
maximum force. But all FF wheels have internal friction
making it necessary to send e.g. a 20% force command to generate
a force that feels like one tenth of the maximum. IMO, the damping
value controls how much extra force GPL should generate for
weak forces to compensate for this. A higher damping value means
that you have a wheel with high internal friction that needs a
larger compensation for weak forces.

All this means that you increase the damping value until the weak
forces show up as expected, but _NOT_ to the point where forces
that should be weak (e.g. grass rumble) are getting stronger than
forces that should be strong (e.g. wheel loading). I find it
easier to start with a rather high damping value and then decrease
it until the weak forces are realistically weak but still there
to feel. (I would not use 'drive into the wall' as a test for
strong forces since it is very uncertain if GPL models that
case correctly, e.g. forces applied to the wheel at any other
place than the contact patch between the wheel and the track.)

The description of max_steering_torque is correct. What it means
that if you set it to 250, a force computed by the dynamics engine
to be 250 whatever unit (foot pounds?) will result in sending
a maximum force command to the wheel (i.e. 10000). A force computed
to be 125 will result in a 50% force command etc. Any force
computed to be stronger than 250 will of course be clipped to
a 100% force. If you look at it the other way around and think
about what happens with the force computed in a given situation
it becomes a bit easier to grasp. Say the force is computed to
be 100 foot pounds in real life. If your max torq setting is
200, a 50% force will be generated. If your max torq setting
is 400, a 25% force will be generated and so on. Hence the
'increasing this value will reduce the force'.

The force_feedback_latency description is also correct, but few
seem to understand that prediction of a value microseconds ahead
in time isn't going to make much difference :-)  Even my value
of 25 milliseconds is borderline to being meaningless.

Here are my current Ferrari wheel setups:

Windows Game Controller:
Overall Device Gain:    100%  (maximum)
Spring Gain:              0%  (anything but zero seems to trigger bugs)
Damper Gain:           0-10%
Default Spring:          off

Core.ini:
[ Joy ]
allow_force_feedback = 1
force_feedback_damping = 80
force_feedback_latency = 0.025
max_steering_torque = 200-250

      _
Mats Lofkvist

I agree with you Mats on the effects of his settings. Good to hear that I'm not
alone to think like this on GTX_SlotCar's settings. I haven't experimented
enough to get myself an opinon on the meaning of the parameters though.
Just wanted to share my opinion.
For the settings I'm using, check:: home.swipnet.se/roger3
/Roger

Sounds like magic. But I was trying to make weak forces stronger. It works
as far as road feel. It makes it a lot easier to feel when you start losing
grip.

Not according to the GPL text file (1.1 I believe) or any other game I know
of.  The control panel damping adds viscosity to the wheel. In the game
settings, it dampens the spikes (according to the GPL text).

Interesting. I don't see how you came to this, and I certainly wouldn't
agree,  but it is interesting.

I think it's pretty certain that it does. It even models the chain link
fence differently than the armco.

It's Newton-inches, I believe.

Well, is it 125 or 250 or what. What is it's reference point?

The GPL text file states that if you go too low on Torque, it will tend to
clamp the forces. That means compress them so it's hard to distinguish one
from another. On damping it states to use a value from 0 to several hundred.
Actually, my settings aren't that radical. I like a light wheel, but have
always said if you want more loading to use 300 for torque and 600 for
damping, for example. This torque figure is certainly within the bounds of
the common GPL theory and settings, and the damping of 600 is certainly
between 0 and several hundred as described in the GPL text. The values that
I personally use have less wheel loading, which I've already said I prefer,
and I can feel the traction better because of it. Also, I still believe that
whatever setting you use that you're happy with, is the right setting for
you.

[snip]

Ok

No, it only means that forces computed to be _larger_ than what translates
to a 100% force command sent to the wheel will also result in a 100% force
command being sent. Maybe 'clip' is a better word than 'clamp'.

E.g: force commands sent to the wheel as a function of the computed
force and the settings. The damping=0 columns I'm quite sure about,
the non zero damping values (with question marks) are my unconfirmed
theory.

computed     torq=100     torq=200    torq=200    torq=200
force        damp=0       damp=0      damp=50     damp=100
0            0%           0%          0%  ?       0%  ?
50           50%          25%         40% ???     50% ???
100          100%         50%         60% ???     67% ???
150          100%         75%         80% ???     83% ???
200          100%         100%        100% ?      100% ?
250          100%         100%        100% ?      100% ?
300          100%         100%        100% ?      100% ?

Of course. Much of what I have written is my opinions only, but regarding
your theory about the damping and torque settings being backwards
I do believe you are wrong :-)

      _
Mats Lofkvist