rec.autos.simulators

GPL physics model...

Steve Ferguso

GPL physics model...

by Steve Ferguso » Thu, 06 Jan 2000 04:00:00

Everyone of us (including me) has been e***d about Papyrus's multi-DOF,
288Hz mother of all physics models.  While all those rotating bodies are
nice, I would imagine the tire model is the real meat of the thing, and it
gets very little mention.  It would also seem to be the Achilles heel of a
good sim, as this will dictate the "feel" much moreso than all those
rotating flywheels, springs and dampers.  I have my doubts that Papyrus
actually could "just plug in a few numbers and model whatever car you
want", as we read when GPL first emerged.  If you read something like
"going Faster", or if you're a real ***, the SAE bible on racecar
engineering, you'll find that the tire modelling is a monster to get
right.  I would imagine that trying to capture the behaviour of a modern
CART slick would take almost as much programming and tweaking as the rigid
body model.

In the interview with the WSC developers, they imply that they have a lot
of work to do to get the tires right.  I'd be curious to know how much
work it is taking to get the NASCAR4 tire model working well.  Like I
said, I think we underestimate the work that is needed to re-adapt the GPL
model to fit a different racecar spec.

Stephen

Gregor Vebl

GPL physics model...

by Gregor Vebl » Thu, 06 Jan 2000 04:00:00

I guess more trouble than programming the tyre behaviour is actaully getting
the performance characteristics of the tyre, i.e. forces vs. slip angle/slip
ratio/camber, as the data is at best scarce (see Milliken's Race Car Vehicle
Dynamics for example). But this is in my opinion only a problem for true race
car engineers since they wnat to know exactly how a real car will behave
under setup changes. But in order to get only the true feel for the car, I
think that the available data plus some plausible educated guesses on the
missing parts should suffice.

I guess the reason why Papy does not give us a CART or any other modern open
wheel sim is mostly economical.

Dave Henri

GPL physics model...

by Dave Henri » Fri, 07 Jan 2000 04:00:00

  I pretty sure when ICR2 first shipped the programmers stated they
spent a great deal of time in the last few weeks getting the grip right
at the various tracks.  What that said to me at the time was they were
doing a-lot of trial and error until the AI ran lap times that matched
the current real world speeds.  I would hope we are beyond the trial and
error method of tire modeling.
dave henrie

> Everyone of us (including me) has been e***d about Papyrus's multi-DOF,
> 288Hz mother of all physics models.  While all those rotating bodies are
> nice, I would imagine the tire model is the real meat of the thing, and it
> gets very little mention.  It would also seem to be the Achilles heel of a
> good sim, as this will dictate the "feel" much moreso than all those
> rotating flywheels, springs and dampers.  I have my doubts that Papyrus
> actually could "just plug in a few numbers and model whatever car you
> want", as we read when GPL first emerged.  If you read something like
> "going Faster", or if you're a real ***, the SAE bible on racecar
> engineering, you'll find that the tire modelling is a monster to get
> right.  I would imagine that trying to capture the behaviour of a modern
> CART slick would take almost as much programming and tweaking as the rigid
> body model.

> In the interview with the WSC developers, they imply that they have a lot
> of work to do to get the tires right.  I'd be curious to know how much
> work it is taking to get the NASCAR4 tire model working well.  Like I
> said, I think we underestimate the work that is needed to re-adapt the GPL
> model to fit a different racecar spec.

> Stephen

Christer Andersso

GPL physics model...

by Christer Andersso » Sat, 08 Jan 2000 04:00:00

Interesting point... I would like to add the fact that all software has
bugs, even those that seem to work flawlesly... It's a pretty simple
mathematical excercise to calculate how long it would take to test all ways
a program can take... Forever, so there are of course bugs in the driving
model in GPL, both programmatical and mathematical (a minus where there
should be a plus, not noticeable in a F1 from -67, but perhaps in a NASCAR
from -98???). To punch in numbers in GPL physics model to simulate other
racing formats will be a very good way to find bugs in it. I've read some
rumours that N4 will have the GPL physics engine... Sounds good, but I'm
pretty sure they mean the improved GPL physics engine... I would love to
have that in GPL, a GPL 2 with better physics :o))).

/Chrille


>Everyone of us (including me) has been e***d about Papyrus's multi-DOF,
>288Hz mother of all physics models.  While all those rotating bodies are
>nice, I would imagine the tire model is the real meat of the thing, and it
>gets very little mention.  It would also seem to be the Achilles heel of a
>good sim, as this will dictate the "feel" much moreso than all those
>rotating flywheels, springs and dampers.  I have my doubts that Papyrus
>actually could "just plug in a few numbers and model whatever car you
>want", as we read when GPL first emerged.  If you read something like
>"going Faster", or if you're a real ***, the SAE bible on racecar
>engineering, you'll find that the tire modelling is a monster to get
>right.  I would imagine that trying to capture the behaviour of a modern
>CART slick would take almost as much programming and tweaking as the rigid
>body model.

>In the interview with the WSC developers, they imply that they have a lot
>of work to do to get the tires right.  I'd be curious to know how much
>work it is taking to get the NASCAR4 tire model working well.  Like I
>said, I think we underestimate the work that is needed to re-adapt the GPL
>model to fit a different racecar spec.

>Stephen

Dan Belch

GPL physics model...

by Dan Belch » Thu, 13 Jan 2000 04:00:00

Careful Chrille, if you talk bad about the physics of GPL in any way on this
board, you'll be charged with heresy.  <g>

Dan Belcher
Team Racing Unlimited
http://simcrashes.cjb.net

Christer Andersso

GPL physics model...

by Christer Andersso » Fri, 14 Jan 2000 04:00:00

I am. I'm posting in a thread very few will notice :o).

/Chrille



rec.autos.simulators is a usenet newsgroup formed in December, 1993. As this group was always unmoderated there may be some spam or off topic articles included. Some links do point back to racesimcentral.net as we could not validate the original address. Please report any pages that you believe warrant deletion from this archive (include the link in your email). RaceSimCentral.net is in no way responsible and does not endorse any of the content herein.