rec.autos.simulators

As I said, I tried something along these lines in the GPL engine (we
have the ability in GPL's physics engine to capture more data than you
can shake a stick at - and a really big stick, at that).  I found it
very difficult to analyze and incorporate the data into my feedback
loop in real-time while driving the simulator.  However, I can
certainly envision display methodologies that would require less
conscious analysis, and so be more amenable to real-time use.  I didn't
spend a ton of time messing with it since this type of analysis tool
wouldn't be in keeping with the spirit of a simulation of late 60's
racing.  Mostly, I just wanted to be sure the required data gathering
and distribution mechanisms meshed well with GPL's simulation
architecture.  They do :-)

Randy

  Second question is if Nascar3 will use the same***pit
implementation as GPL.If you cant answer this I understand.

  Third question is if Dave Kaemmer worked on Nascar2.I dont see his
name in the credits and was wondering why not.

  Final question is about tire squeal.Is the way its implemeted in GPL
the best that can be done or can this actually be improved somehow for
more feel?I never tried GP2 since I could never find it but I heard
that it models the tire squeal differently.Just curious here:))
  Thanks
 Skeeter

Agreed. Heck, sometimes it's hard to catch a glimpse of your *tach* at
10/10ths. But, what I'm talking about is a display that would be used in
conjunction with the *replay view* for driving and setup analysis.

The display that I described, while sounding complicated, would be so
visually descriptive and intuitive that even a racing neophyte, with just a
minimum of instruction and contemplation, could comprehend the message being
delivered! It would make the vague concepts of vectors, traction, braking,
and cornering forces, limits of adhesion, weight transfer, wheel loading,
and friction circles visual and understandable!

How many times have we seen posts on RAS in which new GPL users have
expressed absolute astonishment over the fact that they can't even seem to
keep the car on the road? A visual display like I described would go a long
way toward their gaining an understanding of what's going on, and
substantially increase the probability that they become GPL fans. People
*want* to be able to drive these cars! Wouldn't this help make that happen,
insure your continued success, and our long-term enjoyment of your fine
products?

Imagine how effective such a display would be in giving a new driver an
understanding of why he can't, for example, dart around a slower driver
while accelerating away from the grid at full throttle, without taking a
wild ride across the grass and into the fencing. The more experienced driver
could work his way from corner to corner determining exactly where he still
has some speed left. The possibilities are virtually unlimited. Couple the
concentric circles with a couple of vertically oriented "progress bars"
displaying real time brake and throttle input and the driver/chassis tuner
has virtually all the analysis tools necessary to get around the track
faster and with better control!

If the data's there, please let us see it! A display like I described can be
built by periodically gathering just 12 bytes of data at each wheel (maximum
frictional force available, magnitude of the resultant force at the contact
patch, and direction of the resultant vector, at 4 bytes each assuming
single-precision floating point data). If file or memory size is an issue
give the driver the ability to turn data gathering on and off while driving.
With an option like that the driver could at least
analyze portions of the track that were particularly troublesome.

You may have read the recent review that criticized GPL because its physics
engine was *too* good. The review argued that because sim drivers must react
to only audible and visual stimuli, GPL is more difficult to drive than the
'67 cars because it lacks tactile feedback. As a former sports car racer
myself, I'd have to agree. An analysis tool like this would at least move us
toward being on equal footing. And again, I'm talking about something that
would be displayed, not while driving, but from the replay view. I can't see
how even the most fanatical F1 historian could object!

Great! If you could make the appropriate cross-process data available and
give the ability to display an applet on top of the replay view, and I'd
love to build it as a GPL add-on myself!

Look forward to hearing from you!

I think I can give you some idea of why the vertical line you've placed
on screen does what it does.  We model the driver's head as part of the
physics model.  It's connected to the chassis in such a way that it
moves relative to the chassis in response to the forces that the
chassis is undergoing.  The more force the chassis is under, the more
the driver's head moves.  When the chassis has lateral force applied to
it (e.g. when cornering), the driver's head will move laterally with
respect to the chassis.  When the chassis has longitudinal force
applied to it (e.g. while accelerating or braking), the head moves
longitudinally w.r.t. the chassis.  Since the in-cockpit camera is
located inside the driver's head, and since the head has moved w.r.t.
the chassis, the position of the chassis on-screen is different.

Once you're at the limit, turning the wheel farther doesn't generate
any additional cornering force.  Since the force on the chassis doesn't
change, the offset from the chassis to your vertical line doesn't
change.  (Actually, if you turn the wheel a lot farther, you'll
probably push the front tires well past their optimum slip angle,
causing less cornering force, causing the on-screen chassis to move
back towards center).

If I understand what you mean by 'model g-forces' correctly, it does.
When you accelerate, you'll see the***pit recede a bit into the
screen as the driver's head is pushed back away from it (it's most
apparent if you watch the position of the mirrors).  When you
decelerate, you'll see the***pit come towards you slightly as the
driver's head leans forwards.

I'd be surprised if it didn't use something very similar.

I don't recall whether Dave put much direct effort into N2.  Clearly,
though, N2 was derived from N1 and IndyCar 2, and owe much to the
creators of those products.  (I don't think my name is in the credits
for N2 either).

There's always room for improvement :-)  but we think the way we've got
it tuned works reasonably well.

Randy

Yes.  Clearly if the physics model is both detailed and accurate
enough, the ability to analyze telemetry gathered from it is something
to drool over.  And, yes, being able to visually depict the effects of
weight transfer and the concept of the traction circle would certainly
be enlightening for both the neophyte and the initiated.  (And, yes,
GPL's physics model is both detailed enough and accurate enough to
provide extremely useful telemetry.  For example, it was amazing to do
something as simple as graph individual wheel speeds against each other
(this was done with Excel, not inside GPL proper).  We could actually
see the two rear wheels spin up to the same speed as the throttle was
applied and the diff locked up, then see them diverge off throttle in a
corner as it opened up.  All very, very fascinating!)

Nothing comes without a cost.  GPL's physics engine is run at nearly
300Hz.  You can't do much of anything at that rate without gobbling up
serious resources (CPU and memory).  The telemetry gathering that we
have in GPL is turned off in the released executable (read: compiled
away - there's no way to turn it on, so don't go asking for a .ini
setting to enable it :-)

Randy

  Yes I defenitly notice the accel and braking effects in the
***pit.But its not actually showing the g-forces acting on the
drivers head.What I mean is that when braking the***pit moves back
to a certain extent dependent on the amount of brake travel you
apply.But when you start to downshift too fast,this causes more g's to
the driver as the gears are working like brakes,yet the***pit doenst
show it.Am I being to much of a perfectionist here??:)

I'll think about the 'neck stiffness'.  We can't make it too loose, or
small changes in direction would cause quite a swimming effect.  The
vertical line's an interesting idea.  I'll see what people think.

I notice a change in the longitudinal head position when I change gears
under heavy braking, but it is subtle.  I think we use a progressive
spring to model the driver's neck.  Once the head's already thrust
forwards by heavy braking, the small additional g loading caused by
downshifting isn't enough to cause a large change.

Randy

I remember talking about this thread about 2years ago about the difference
between Icr2 and Gp2 and why  Icr2 drivers normally had difficulties in Gp2
and vice-versa.

In Gp2, the tyre squeal is important. You clearly need it to get those great
laptimes.  But in Icr2, the less tyre squeal the better.  Actually tyre
squeal in Icr2 means that you will probably spin in not long.

This is why drivers normally liked one but hated the other.  Gp2 drivers for
them it was normal to hear tyre squeal, so each time they went into the
corner "a la Gp2" and spun.  Icr2 drivers, racing Gp2, had problems because
they back off when they hear tyre squeal, thus not pushing the car to it's
maximum.  I know Im like that ;-)   The same was for N2, who had a similar
tyre squeal system as Icr2.  The less squeal = the better laptimes on oval
tracks.

GPL for me is closer to the normal Papyrus mentality towards tyre squeal,
but since they are 1967 Formula 1 it doesn't mean anymore that tyre squeal =
upcoming spin.  The sound feedback is something I hope they will push to
it's limit in the future.  Racing simulators are visual and audio  virtual
reality, but I always felt all those companies lacked the audio feedback.

-= Fran?ois Mnard <ymenard/Nas-Frank>
-= NROS Nascar sanctioned Guide http://www.nros.com/
-= SimRacing Online http://www.simracing.com/
-= Official mentally retarded guy of r.a.s.
-= May the Downforce be with you...

"People think it must be fun to be a super genius, but they don't realise
how hard it is to put up with all the idiots in the world."

--
Pat Dotson
IMPACT Motorsports
http://www.impactmotorsports.com/pd.html

--
Pat Dotson
IMPACT Motorsports
http://www.impactmotorsports.com/pd.html

I think I understand what he's trying to say...  I find that when you
hear tire squealing in GP2, there's still room to push a little more,
but with ICR2 when you hear tire spealing it usually means you're
already on the limit and you shouldn't push any further.  The span
between the moment when the tires begin to squeal and when you
actually lose grip and spin out is, if I'm correct, a lot greater in
GP2 than it is in ICR2.

A. Renault

--
Pat Dotson
IMPACT Motorsports
http://www.impactmotorsports.com/pd.html

  Instead of making the neck stiffness more loose you could move the
vertical line instead.When making a right turn the***pit moves
towards the right of the stationary vertical line.When the***pit
shifts to its farthest point the stationary vertical line can then
start to move in the opposite direction of the turn indicating more
g's but not detracting from the feel of the***pit.This would work
wonders in cars that pull more g's than GPL's cars.

  Just some more ideas...How about also adding a horizontal line to
the vertical line that moves up and down the length of the vertical
line.It could be about half the length of the vertical line and be
used to show the balance of the chassis.For some reason I cant really
get comfortable with  using the engine rpm's to indicate if Im
accelerating,deccelerating or balancing the chassis.Between my braking

while throttling slightly to keep the car balanced all the while
trying to listen to my rpm's,I lose it somewhere.This horizontal line
right in my field of view could give me this feedback that I need.

 My final idea is about different tire squeal sounds for the fronts
and rears.I didnt wanna make the same mistake I did about the***pit
not  indicating g's under downshifting so I really sat down and tried
to listen to see if I heard a difference before posting this.To my
suprise I could swear that I hear a difference between front and rear
tire squeal.Am I right about this?If not then I think this could
really add to the sims in that itwould be great to be able to
differentiate the squeal sounds to know which set of tires is at their
edge.
 Thanks again for listening
Skeeter

On Thu, 25 Feb 1999 00:45:35 GMT, "Randy Cassidy"