rec.autos.simulators

stupid licensing ruins everything

Mike Nun

stupid licensing ruins everything

by Mike Nun » Tue, 07 Aug 2001 23:21:48

Hi Mark,

Maybe I can add something to this discussion and help alleviate some
of your frustration. My own background is very different to yours, as
I have been the proud owner of a budget copy of GPL for all of two
weeks. I have sunk a fair bit of time into other "pseudo-sims" such as
GT1/2 on the PSX and Ferrari 355 Challenge on the Dreamcast, but until
recently didn't have a PC for ***. Now I do, so I'm taking the
plunge into the (IMO) much more interesting world of PC-based sims. I
also take my own road car (Subaru Impreza) to various circuits around
the UK several times a year for a bit of free lapping, so I do have
some RL track experience to compare the sims with, albeit not much in
terms of hours. Obviously with a car like that I have a few years of
experience at fast road driving, although the ragged edge is generally
out of bounds for the sake of safety. To give some idea of my pace -
or lack of - in GPL, I was relatively pleased to pull in a 1:33.4 at
Monza last night. I also had my first crack at the 'Ring, but I won't
say the time as it's just too embarrassing!

Ok, moving on to a few general points and observations:

PERCEPTION OF SPEED

I can remember the first time I saw in-car video shot by another
Subaru owner at a track day, and I was absolutely stunned at how
*slow* it looked. I could see the rev counter winding around to over
7000 in 1st and 2nd gear as he came out of the pit lane, but the
acceleration looked more suited to a trip down the high street for the
shopping. And this was in a Japanese import car with 280hp, weighing
less than 1300kg. The same applied to cornering speeds. It was a real
eye-opener to realise how the impression of speed in driving games was
being massaged.

SLIP ANGLES

Todd, regarding your comments on peak lateral force and slip angle, do
you think it's overdone even taking into account that the 67 cars were
running crossply tyres? I would definitely expect much higher slip
angles with these, and looking at the replays the car attitudes seem
to match my memories of film footage of GP cars from that era. (The
repeats! I'm not that old ;o) What do you think?

TRACK MODELLING

Looking at those comparison shots, it struck me how a tiny difference
in track modelling can make a huge difference to the nature of a
corner. Move a kerb by less than a foot and you can radically alter
out the whole cornering line, and crests, dips and cambers can have an
even more profound impact on the way you approach a corner. Compare
the GPL version of Fox Hole with RL. In GPL the crest around the
turn-in point seems more pronounced, as does the gradient change
between the approach and exit. The RL corner also looks sharper to my
eye. At Adenau, GPL has the approach drop into the corner much more
than it does in reality. These things will make a difference to entry
speed and the attitude and balance of the car through the corner and
on exit.

FEEDBACK

Mark, your comment about "occular disposition" is very interesting as
I believe it points to the heart of the issue with feedback. I feel
lucky as a sim racer, as I find very little need for physical feedback
even when driving a real car. I get most of my info from the changes
in cornering attitude, especially in pitch. These are pretty much the
only things I look for when trying to judge what I can get away with
either on braking or when feeding the power back in on exit. I have
done a fair bit of messing around with the spring and bar rates on my
real car, and at one particular stage the car was pretty wayward. I
didn't need to have a scare in order to work this out, even well
within limits it was obvious from the way that the car would go high
at the rear as the overly stiff rear springs transferred all the load
onto the outside rear wheel.

When it comes to issues such as single wheel lockups at the front or
rear, I can read what's going on from the yaw behaviour. For example,
if the rear has locked (rear brake bias too hot?) and you've got even
a fraction of steering lock dialed in, the tendency of the rear to
"buck" is detectable from the body movement alone. This isn't
translation, it feels very direct and natural to me. Even though I am
way down the learning curve compared to some around here, I can still
lock a wheel without flying off the track, and I can bring it back
from the great beyond at least 50% of the time in a broad variety of
situations. Most imporantly, the other 50% of the time I know what I
did wrong -- I'm not driving blind.

Another other thing I depend on strongly is the longer term feedback
that comes from learning what you can get away with, and what is
likely to stuff you in the barriers. I believe that it is possible to
build your own mental physics model, and develop a visualisation of
load transfer effects in various situations. Decisions such as whether
to trail brake on entry to a particular bend become much easier if you
can get a good feel for the interplay between pitching weight transfer
due to braking and the lateral grip limits (and weight transfer) at
your chosen cornering speed.

As my Monza time shows, I'm no expert, but for me GPL is an absolute
revelation, and I do not see myself spending time with any other
sim/racing title in the near future. It was clear from my first few
minutes experience that I was dealing with car behaviour that is very
close to RL, as percieved through the particular set of cues that I
pay attention to when I drive. But that's the clincher, and if you
focus on different aspects then that could easily explain why GPL
doesn't provide the deeply satisfying experience for you that it does
for me. Different people pay attention to different cues. For example
to some people the idea of doing without a FF controller would be
unthinkable, but I don't give a tinker's about that as long as I can
judge the load transfer effects from the car's attitude.

OTHER THINGS

Having been so glowing about GPL, I should at least say that I fully
sympathize with your comment about the "crapshoot" of staying on the
racing line. It certainly is hard, but I also agree with Jonny that it
may have more to do with the large drift angles (realistic or not) and
the sheer time that is spent sideways, rather than the physics model
per se. Also agree on the matter of controller setups, I guess I just
lucked up because linear works well for me.

The other game I purchased in the last fortnight was F1RC, which some
people around here seem to hold in high regard. Well, I could *almost*
get into it, but it seems very arcadey after GPL. Ok, the vastly
different cars make a direct comparison futile, but I just get a
"canned" feel about certain aspects of the handling. And don't even
mention the spins, which just made me reach for the EXIT button --
even GT does better than that!

Finally, can I also ask what this static***pit thing is all about,
and also what are CPR and MTBR?

Best regards,

-= mike =-

<much interesting stuff>

Gunnar Horrigm

stupid licensing ruins everything

by Gunnar Horrigm » Tue, 07 Aug 2001 23:38:21




> > > It's the shutter speed that makes motion blur, but anyone familiar
> > > with a camera knows that the 1/24 second speed needed for a perfect
> > > motion blur is still a too long exposure time.

> > sure, but who says the exposure time has to be 1/framerate?

> Nobody :-)

good :)

hmmmmmnoooo.  it's not that obvious.  you can have several rolls of
film seeing through the same lens.  doesn't immideately make much
sense, though. :)

--
Gunnar
    #31 SUCKS#015 Tupperware MC#002 DoD#0x1B DoDRT#003 DoD:CT#4,8 Kibo: 2
                 "Det er nok ingen ovn, men fartsm?leren v?r."

JM

stupid licensing ruins everything

by JM » Wed, 08 Aug 2001 02:45:16



> > Hmm, do wonder whether that _particular_ point isn't a factor of the
> > class of cars - I'd expect pretty much anything modern to feel more
> > precise than those skinny-tyred 1967 monsters.

> I agree with that, Jonny.

> Gerry

The tyres are certainly not skinny, and I'd take the 1967 GP cars precision
over a Ford Mondeo's any day.
I'm guessing you were comparing racing machines past and present, but people
don't have an automatic problem driving sloppy modern saloon cars and
getting them to go where they want.

cheers
John

Thom j

stupid licensing ruins everything

by Thom j » Wed, 08 Aug 2001 04:14:13

This is 'spot-on' Olav! I didn't know you knew this much
about photography aka video filming?? I learn something
everyday. See I knew there was a reason for me to wake
up each day! :)

| Very nice observation. :)
|
| It's the shutter speed that makes motion blur, but anyone familiar
| with a camera knows that the 1/24 second speed needed for a perfect
| motion blur is still a too long exposure time. Then again, maybe
| cinematic film needs longer exposure that the familiar 35 mm single
| picture camera ?
|
| The V5 got hardware motion blur, sadly, no game uses it, and none will
| :( Do the GeForce3 have this capability ?
|
| --
| Olav K. Malmin


| >
| > (*)  Actually it is often easy to see stutter when watching a
| >      movie. It seems to be most evident in slow pans filmed
| >      in bright light (motion blur hides stutter, maybe bright
| >      light scenes are filmed with shorter exposure times, thus
| >      reducing the motion blur?)

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.265 / Virus Database: 137 - Release Date: 7/18/2001

Gerry Aitke

stupid licensing ruins everything

by Gerry Aitke » Wed, 08 Aug 2001 05:50:13





> > > Hmm, do wonder whether that _particular_ point isn't a factor of the
> > > class of cars - I'd expect pretty much anything modern to feel more
> > > precise than those skinny-tyred 1967 monsters.

> > I agree with that, Jonny.

> > Gerry

> The tyres are certainly not skinny, and I'd take the 1967 GP cars precision
> over a Ford Mondeo's any day.
> I'm guessing you were comparing racing machines past and present, but people
> don't have an automatic problem driving sloppy modern saloon cars and
> getting them to go where they want.

I'd like to explore your logic a little more, but I fear I'd be wasting
my time ;)

Gerry

JM

stupid licensing ruins everything

by JM » Wed, 08 Aug 2001 06:16:26


yeah, up yours too charly.

John

J. Todd Wass

stupid licensing ruins everything

by J. Todd Wass » Wed, 08 Aug 2001 06:49:18

  Ah yes, very good point.  Hadn't thought of that.

  >> I haven't driven a car via remote control through a camera mounted

  Good point there too.  Higher resolution requires higher frame rates.  Frame
rate is very important, but it still seems to me that after a certain point, it
can't be used as an excuse for not beating Gregor Huttu anymore :-)  I got a
1:30.xx lap at Monza after two or three hours running the lowest graphics
settings on a Pentium 60, at only 2-3 fps (default Lotus setup with low fuel).
Granted, it ended up taking about 3 minutes real time, so it was slow motion by
2x, but still....  At 12 or 13 fps I'd bet I could do better than that.

  It just seems odd to me that once someone's got 30+ fps they feel a need for
even more.  To me, it seems that regardless of resolution, you pretty much see
and react to everything there is to see and react to once you're around 25-30
fps.  When I'm looking at my braking zone, it doesn't move much in the screen,
when I switch and look at the apex, it doesn't move too much either.  At higher
rates, when looking at the edges, it does look smoother above 30 fps, but it
just doesn't seem to me that that would make much difference in driving if
you're looking where you should be looking.

  That's just me though, we're all different I guess :-)

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://www.racesimcentral.net/

J. Todd Wass

stupid licensing ruins everything

by J. Todd Wass » Wed, 08 Aug 2001 06:54:03

  There was a discussion on this awhile back, but I don't recall the outcome of
it.  I don't think the controller input is tied to framerate.  I ran GPL the
first time on a Pentium 60 at 2-3 fps.  Surely the 1:30.xx times I turned (only
a couple times ;-)) would be impossible if only 2 or 3 controller inputs were
being read each second.  It really didn't feel that this is what was happening.
 I was anticipating what the car was doing and making several adjustments each
frame, and it seemed to respond right, it just took 1/3 or 2/3's of a second to
see the results.

  I could be wrong though of course, but it really seemed this was the case.
Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://www.racesimcentral.net/

J. Todd Wass

stupid licensing ruins everything

by J. Todd Wass » Wed, 08 Aug 2001 06:58:18

  Yes, the blurring is something I hadn't considered.  That'd definately smooth
things out, good point.  What do you mean by depth of field?  Is this something
to do with the camera's focus and how objects closer or further from the focal
point appear slightly blurred?  If so, yes, that's another good point I hadn't
considered.

  >anyway; stuttering in computer games is often caused by double

  Yeah, I've seen this on the DX SDK demos.  You can play with the frame rate
and I was surprised to see that things do look smoother above 30fps.  Well
heck, maybe I've been talking out of my behind all along here :-)  

  You know a lot more about photography than I do.  Why can't they be compared
this way?

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://www.racesimcentral.net/

Asbj?rn Bj?rnst

stupid licensing ruins everything

by Asbj?rn Bj?rnst » Wed, 08 Aug 2001 07:34:49


I wonder, shouldnt the distance you sit from the monitor affect how
realistic a certain FOV feels? I tend to sit pretty close when I race,
I just can't sit back and drive. My theory is as follows (Bring out
your fixed width fonts):

Realistic: \       /
            [     ]
             \   /
              \ /
       Eye->   O

"Strange": \       /
            [     ]
             \   /
              \ /
               V

       Eye->   O

This would make a 180 degree FOV rather unusable if you're not extremely
nearsigthed. a 90 degree would require a distance of about 18 cm (I sit
about 25 cm. from the screen when I race.) etc.
--
  -asbjxrn

J. Todd Wass

stupid licensing ruins everything

by J. Todd Wass » Wed, 08 Aug 2001 10:33:44

  Your theory hits the nail exactly on the head :-)  There are really two
FOV's.  The monitor itself acts like a little projection screen, with perhaps a
90 degree field of view of its own.  I use OpenGL for 3-D graphics (not very
well yet, I must say), and the FOV is set by one little line of code.  If you
set it to 90 degrees, an object in our 3-D world that's 200 meters straight
into the screen and 200 meters to the left will appear along the left side of
the screen, because the angle between it and our "camera" is 45 degrees.  

   However, if your monitor is just a tiny thing off in the distance, that
object surely won't look like it's at a 45 degree angle to your real head.  But
as you bring the monitor closer and closer to your face (or increase its size),
the left or right edge gets closer to being truly at a 45 degree angle to your
real eyes and your sense of speed and life-sized animation/immersion improves.
This is the second FOV, the one you showed above, the angle between your
eyeballs and the edges of the screen.

   When the two FOV's are identical, you should see everything in actual size,
as true-to-life as you can get on a 2-D screen.  That's the key to really good
immersion and true sense of speed (minus 3-D glasses).  However, it's tough to
do as you pointed out.  For a 90 degree FOV on someone's monitor to look truly
correct, they could measure their screen width, then put their face at 1/2 that
distance to the screen.  For me, that'd be about 6 inches/ 15 cm away (you must
have a bigger monitor).  Way too close for comfort.  The other option is to
have a really big monitor further away.  Personally, even though the resolution
may not be good, I'd LOVE to try this with a giant screen television set :-)  I
bet the sense of speed would suddenly become awesome with absolutely no changes
to the sim itself.  

  The fact that people have a 180 degree field of view is sort of irrevelant to
car sim racing, imo, unless you insisted on having a monitor that completely
covered your entire field of view so you couldn't see your
furniture/girlfriends on either side :-)  If your monitor's/sim's "programmed"
FOV is the same as your true FOV measured from your eyes to the edges of the
screen (as you showed in your illustration), it'd look just like you're peering
through a window into our 3-D world, everything's life sized and looks like
it's moving at you at exactly the right speed.  The question then is (for me
anyway), how big a window do you need to see through, and how close are you
willing to sit to the screen?

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

J. Todd Wass

stupid licensing ruins everything

by J. Todd Wass » Wed, 08 Aug 2001 11:14:10

   The only thing that I've actually been able to roughly measure is the slip
ratios (percent slip) at the tires when the throttle is applied.  These were
around 20+% in GPL when traction control is on (default when using the
keyboard).  The tires seem to not do a whole lot at lower slip ratios IMO.  I'm
not a tire engineer though, so....maybe old crossply tires were really this
flexible/balloon-like.  I'm looking at some data for one particular Nascar tire
now, and it shows peak force coming in at about 8% slip ratio, although the
force really doesn't start to drop much until around 20%, but by the time you
hit this, the tire will probably go up in smoke.  This is for a modern tire,
however, and as you pointed out, it's probably a lot stiffer than the '67 tires
were.  My impression is still to answer "yes" to your question, but I'd rather
see more video first before really trying to argue the point.  Especially since
I've never heard of crossply tyres before :-)  

  When looking at the GPL intro video, they just don't seem quite as sloppy as
the sim's tires do, but to be fair, you've seen more video than I have, and I
don't have any test data to back any of this up, so I'd be inclined to take
your word for it.  My friend's father raced F-1 around this time frame (had an
accident in '71 I think that ended his career, hit the pit wall, splitting his
car down the middle and shattering his leg, ouch!)).  Unfortunately, he passed
away about 9-10 years ago.  Otherwise, we could just ask him :-)

  In fact, come to think of it, I'm working on a sim project for radio
controlled cars (supposed to be as accurate as possible)  The head engineer
told me the little ***s peak at roughly 20 degrees slip angle, and I would
never guess that from looking at the tires themselves, as they're rather wide
and very low profile.  This is probably roughly the same peak that the GPL
tires use.

  My point earlier was really that sloppy/unresponsive/unpredictable tire
behavior can really be traced to the shape of the slip ratio/slip angle vs.
force curves, something that most sim drivers don't seem to be aware of.  GPL's
are certainly looser than anything else out there, but that doesn't necessarily
mean it's not close to right.  If the RC car tires do it maybe those big '67
crossplys did too.

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://www.racesimcentral.net/

J. Todd Wass

stupid licensing ruins everything

by J. Todd Wass » Wed, 08 Aug 2001 11:24:18

  Ah yes, of course...   Thought it was some new fangled sim doo-dad I missed
:-)  lol

  Agreed.  120 degrees looks like you're going fast as the devil, but the road
gets pretty skinny.  Hmmm...  Is there an FOV for girlfriends??  ;-)
Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

Eldre

stupid licensing ruins everything

by Eldre » Wed, 08 Aug 2001 11:52:44


Maybe that's why the people who have the larger monitors are usually better at
the driving games?  I always thought it was *resolution*, but never even
considered FOV.  I ran on a 15" monitor for the longest time, and only
recently(about a year or so?) borrowed a 17" from work.  I can't afford a 19"
or 21" yet... :(

Eldred
--
Dale Earnhardt, Sr. R.I.P. 1951-2001
Homepage - http://www.umich.edu/~epickett
F1 hcp. +16.36...Monster +366.59...

Never argue with an idiot.  He brings you down to his level, then beats you
with experience...
Remove SPAM-OFF to reply.

J. Todd Wass

stupid licensing ruins everything

by J. Todd Wass » Wed, 08 Aug 2001 12:34:46

  Probably right.  I really hadn't given any of this much thought until this
thread started, but yes, a bigger monitor should make going fast easier than
going to higher res would.  Heck, I know where the corner is already at
640x480, even if it's a little fuzzy, I just don't hit it where I want all the
time because it's hard to guage what the car is doing on that little screen.  I
suppose the bigger you go, the easier it would be to eyeball your
speed/attitude changes, since you're now a little closer to seeing things "out
the window" as they would look in reality.  Should be much easier to drive.
Did you notice a difference when going to the 17" monitor?  

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com


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.