rec.autos.simulators

Also, as is often pointed out in this newsgroup, GPL is pretty much the high
point in sim racing physics, even though it came out years ago. Is there
simply no way to significantly improve physics modeling beyond GPL, or has
it essentially been perfected in mimicing what happens in the real world to
race cars ?  Somehow N4 and NR2002 do not seem to be much improved over GPL,
and in some respects actually feel inferior to GPL, IMHO. Are the additional
real world physical forces that are lacking in GPL's physics model simply
too trivial to waste CPU cycles on modeling, providing little or no
potential for improving the genre?

How about force feedback ?  Has everything that you would feel in a real
world steering wheel in a given situation been imitated ?    While asphalt
based sims like GPL seem to have sufficient force feedback effects, off-road
sims still seem to lack the upredictable feeling of small ruts and potholes
, or the way that deep sand areas can make steering a vehicle rather
different from mud or gravel or hard packed dirt. Rally Trophy is a very
good off-road sim, but I never had the impression that the road surface
irregularities were likely to cause me to crash while negotiating a dirt
straightaway at speed, while by comparison GPL sometimes leaves me feeling
one twitch away from being out of control even on smooth asphalt
straightaways at max speeds, as the car wanders a bit from side to side.
Shouldn't an off-road racer be that much more difficult to control at high
speeds coming down a bumpy dirt straightaway, putting you that much more on
the edge of being out of control ?
Having ridden motocross bikes down relatively smooth sand trails, I know
from experience that 60 MPH on sand is far more adrenaline inducing than 60
MPH down a freeway. Consequently, I would think that the force feedback
effects could be made to more effectively convey that feeling of lack of
control. I have noticed that in VIPER RACING the force feedback effects will
sometimes be so strong as to cause me to over compensate in trying to
control a fish-tailing situation, and actually cause me to make the car
fishtail even worse until I lose control entirely. I would like to see that
kind of an effect in more sims, under appropriate situations of course, and
perhaps even moreso in off-road sims.
Finally, each of us knows that there are numerous visual and other sensory
cues that are sorely missing from sim racing. Many of them can never truly
be incorporated into a home racing sim, for obvious reasons. However, are we
at a deadend with respect to possible innovative peripherals that might be
supported
in a games code ? Rumble seatpads akin to providing the rumble effects of
console joysticks, better 3D glasses support, HUD's , force feedback in
brake pedals and clutches, seats that tilt slightly in response to G-forces
giving you a hint of the forces direction and strength.
I am sure that there must be ways to make sims more immersive. Or have PC
racing sims pretty much gone the way of the light bulb, with not much room
for improvement, beyond maybe a prettier version of GPL.
Anybody have any ideas where PC racing Sims are headed over the next few
years given the practicalities of  the videogame business, or where you
would like to see them go ideally if developers actually cared what you
wanted or thought, instead of just about mass marketing appeal.

The demo of GP4 looks very impressive graphically.

F1 2001 can look excellent, but in order to keep a decent fps with a
full field on my P3 866 gft gts, I need to turn down so much detail
that is looks ***again...

The same goes for N2002. I -want- to see all the 43 cars in front of
me and behind me, but I have to give up on quite a lot of fancy
graphics to keep the fps acceptable.

NR2002 is imho way more advanced in the physical department than GPL.
I can't say why exactly, it just -feels- much more realistic. I'm sure
some people here can tell you exactly what has been improved in N2002
compared to GPL's physics :)

I agree with the rest of your excellent post.

Andre

Yes, and everything that there is to invent has already been invented, so
let's close the patent office too.

I feel we are at a high water mark until something revolutionary comes out
on the hardware side. Right now the improvements in hardware are not as
great as they use to be. Going from 1200 Mhz to 1400 Mhz is not the same as
going from 200 to 400 or even 400 to 600. Video cards are basically in the
same rut. Yes they can be amazing on paper but if you don't have software to
take advantage of it you don't gain a whole lot. I really saw this when I
made the change from a GF2 to a GF3. Yeah the GF3 was faster, but that's
about it. I'll be skipping the GF4 line because I don't see any great
improvements over the GF3 at this point.

I think there is currently three major problems in the sim-racing arena.

The first is that the market is being over commercialized. We now have
companies only worried about the bottom line controlling the development of
games. These companies go after the mass markets to maximize profits. In the
past we had people making sims that were sim players and they wanted to do
they best they could. Now the corporate mentality has taken over. This is
why we see the "release it now patch it latter" way of doing things with new
titles. This mass market approach is also holding things back from the point
that these companies are afraid to put to much into a game for fear of
alienating customers by either making the games to hard to play or requiring
to much hardware to run. How many posts have we seen like this? "I have a
600 p3 and a V3 how come I can't max out the graphics on this POS game?"
GPL is a good example of what happens to a game that is declared too hard.
No matter how much we love it there is no denying that it was a finical
flop. This is also why we will never see a true damage model. You'd think
that people would just use the option to turn down the damage model, but
more on that latter. I also think this is why road course racing isn't as
popular as oval racing.

The second problem is what is there really left to add? The graphic engines
can really shine if there is enough hardware to let them. The physics models
are pretty darn good too. The two areas that need work IMO are AI and Force
feedback. Force feedback is really a subjective thing so I would just like
to see it being more open to adjustment. If you look at all the values to
edit in F12001 you should be able to really tweak it to your liking. There
just isn't anything left for that big leap between titles. How many times
have we heard that a new version should have just been a patch?  I
personally have been a fan of the expansion pack way of doing things. That
way you don't have to relearn the physics every new release, but you still
get new content like new tracks or chassis. Take NR2002 for example I really
hope they do the Craftsman truck series and Busch series add-ons, but why
stop there?  Why not take it farther to like ASA or ARCA? How many times
have we heard the cry for more short tracks? Why do we have to stick with
NASCAR? Give the short track crowd what they want with an ASA add-on. They
have the physics engine so all it would take is some new track models and a
new chassis. I think the fact that the basic game stayed the same from N1-N3
is why the series became so popular.  I'd love to see an expansion pack for
GPL with some new tracks and an update to OpenGL. I know there is a lot of
great stuff on the net. But it would be nice to have it all in one easy to
install setup instead of having 50 different add-ons to install in many
different ways.

The third problem is what I think is the worse one, and that is us. That's
right I said us. I truly believe we are our own worse enemy when it comes to
the sim vs. arcade titles.  We are not a very friendly group to outsiders.
When I say outsiders I don't mean newbies as much as people that think
differently. Since it's been debated so heavily lately I'll use the camera
view as an example. Its clear that anyone who doesn't use the***pit view
is looked down on as being an inferior driver. If someone posts a question
about say a Need for Speed title they get laughed at for playing an arcade
game. The worst part of it is that most of us can't let it go. We have to
start threads trying to blame Lara Croft for why we have chase views. Get
over it people. We need to make this a much friendlier place so that our
numbers will grow. If we quit chastising people for using arcade views or
moderate damage settings maybe they would work their way up to becoming full
blown sim racers. Then we could get the things changed in games to what we
want.  Right now I sure we scare off a lot more people than we invite in.
We must remember that we play these games for fun. I don't care how
realistic it is there is no fun in running a lap, smacking the wall, and
having to go to the pits. When trying to learn a new game. So if they run a
roof cam so they don't smack the wall who cares? How does it effect our
lives? If any thing I would like to think it makes our sim lives better. If
a roof cam helps them control their car and keep from hitting me in an
online race I'll hit f10/page down for them, and I'm sure as hell not gonna
force them to do something they are not comfortable with. If people were not
shamed so much for running in arcade mode maybe we would get a better damage
model for the sim side of a game. As it stands the game companies are too
afraid to make the damage too realistic because they don't want the "too
hard" label tied to their game.
  Another thing we do is illegal copies. Game software is not really all
that expensive. I'm sure most of us***away $50 bucks all the time with
out even thinking about it. If you don't believe me just look at what the
price for a copy of Microsoft Office XP is going for these days. Now talk
about software that is just a patch! At least with games we usually get more
improvements than a typical Office upgrade.  Hell taking a date to a movie
cost as much or more than most game titles these days. At the price of
cigarettes around here just take a few days off from that and you'd save the
price of a game. Quit altogether and in a year you could have a few games
and a new Geforce video card. Don't get me wrong we all love getting
something for nothing, but again we cut our own throats at the same time. We
all hate having to swap CDs to play games, but you're a complete idiot if
you think everyone that uses a no CD patch actually owns the game just
because the readme file says they have to. Like it or not these games are
made to make money, and if they don't make money no matter how much we like
them there won't be a squeal. It all comes down to priorities. If it's a
real priority to improve the sim-racing arena we must all do our part.

And what about us sim-heads?  The current physics models are pretty much as
sophisticated as engineering rigid body dynamics software.  Maybe they can
boost the frequency, but not much more.  There aren't many more DOFs which
need to be modelled, as I doubt you would noticed much difference going from
Papyrus's engine DOF model to one which models each piston, for example.
Improvements will come in the tire models (as each new SAE conference brings
new insight) and the aero models.  Non-rigid elements in the model would be
useful, but again I doubt most of us would notice the change in the feel.
Codemasters' "finite element analysis" damage model sounds a lot cooler than
it probably is (considering crash simultations need a supercomputer) but
it's still a step in the right direction.

Stephen

I am one of the great unwashed when it comes to sim games. I like the PC
version of F1 2001 best - mainly because the game has a sense of speed which
many games don't. I am not a decent driver but I still enjoy racing games -
particularily the sim/arcade variety. My personal wish list is rather
different from you '*** sim' drivers : I want more imersion - that
means better sound effects/animations. If I am Schumy and I'm racing at
Monza and I overtake into first place - I want to see the fans in the stands
stand up and cheer. If I take out Irvine - I want him to get out of his
***pit and give me a piece of his mind. F1 games are getting too samey - I
want a career mode where you start in a lowly team and you have to help
develop the car - basically take out the test car and determine the problems
with it. You would gain experience points for determining the correct
problems (like rpg games). If you reach a level of performance - you can
switch teams towards the top teams. I'd like a team tryout - where you are
told of what laptime you need to get to join a team - and you can basically
work your guts out getting the car to reach the set laptime. Really there is
so much more to F1 games that should be done. I'd also like a real
performance update - where you can race a season with the game updating
driver performances as the season progresses via the internet - ie for this
year Jaguar to be near the back of the pack. I want a historical gameplay
update - so perhaps a few days after a real gp - EA could do a replay of the
real race in the F1 2001 game and you can take over one of the drivers. So
if there was a a big pileup at the start as with the recent gp - you can be
Ralf Schumy and experience the crash  - and perhaps try things to avoid it.
It would be interesting to see the game ai as what would happen if Ralf
breaked earlier - would it have caused a bigger crash etc.  I want more fun
in my F1 games - I like the F1 intro that is at the start of every gp on
tv - its in the Sony F1 game - why not the other F1 games. I want a pit lane
sim aspect of the game where you can wander in the pit lane with all the
sounds/modelled real drivers*** around. I want a decent podium where
you can pour the champagne. I want a highlights mode where the game just
shows the best overtaking/crashes. F1 is such a cool life for a F1 driver -
lets have a F1 game with a rpg element.

Al, the unwashed F1 fan.  :)

I agree.  Part of this is the "common denominator" approach to
software development, so the game will run on as wide a customer base
as possible.  Also because at some point designers have to say "we
will develop game x to run on system y with these hardware specs".
Then 18 months later, when the game finally comes out, there have been
2 new generations of hardware.

I think 2 or 3 years down the pipe we will see some pretty amazing
stuff.  It may or may not happen to be in a racing game though...

I think part of the reason we see this so much is because FPS is the
only objective measurement of performance.  There is no way to
quantify eye candy, other than to say that a particular card supports
a particular instruction, and 99% of the folks buying cards have
ABSOLUTELY NO IDEA what trilinear filtering really is.

Lets say somebody somehow releases GAME A that takes advantage of all
the bells and whistles of the top of the line box, but can only muster
30 fps.  Somebody else releases GAME B with similar gameplay features
that can pull 45 fps on a mid-level box and still look "pretty good".
Which one is going to sell more copies that don't get returned by
pissed off consumers?  How much more is it going to cost to produce A
over B?  Would you pay $150 for a racing sim?  I might, you might, but
most people would just laugh at the idea.

There is much that could be done here, but hardware devices vary so
much from manufacturer to manufacturer, I think the best we could hope
for is probably for game designers and FF API developers to give us as
much customizability (is that a word?) as possible.

A natural extension of FF would be tactile transducers for seats,
pedals, etc.  There was a company that actually was/is/was_going_to
make hardware to do this, but I don't thing software support was ever
really forthcoming.  It's a shame this area hasn't received more
attention because this sort of effect would add so much immersion,
even moreso than graphical improvements.

Again, I am sure this really comes down to ROI on the software
development end.

This has next to nothing to do with the videogame business, but take a
look at

http://www.racesimcentral.net/

and

http://www.racesimcentral.net/

to check out the most realistic driving simulator in the world.  This
thing cost about $100 million dollars to build.  I don't know if this
includes software development costs or not.

The most intriguing thing to me is the 360 deg. FOV that is produced
from an array of 15 video projectors and a parabolic screen.
Suddenly, my home-theater/*** rig seems so inadequate...

I think the best chances we have of getting major improvements in the
areas we are talking about here will come from some kind of
open-source community or third party development.  A commercial
venture just can't afford to do be this ambitious and cutting edge.
Even if they charged huge sums of money for the finished product.

Aside from technological issues, there is still a HUGE way to go in
the gameplay department - fodder for another discussion I think.

Cheers,
Gerald

<snip>
<snip>

Right. I have a P4-1.5GHz with a GeForce2 MX-400. That's pretty old hat
by now, yet I haven't had any trouble with any game yet (I've no idea
about NR2002, as I haven't played it). Not so very long ago, by the
time so many months had elapsed I'd be starting to struggle.

<snip>

Mmmm. The problem is that costs are going up and up, which means that
only sure-fire hits are getting made in the first place. Geoff
Crammond, for all we love to moan about GP2.75, has done some amazing
things, and Revs (the BBC Micro game) was one of the greatest -
possibly the first real simulator on a home commputer (and how many F3
games have we seen since...?). That game would have had no chance of
being made in today's climate.

<snip>

...in the US. GP3 sold far more than any NASCAR game in the UK, but of
course the market isn't so big. (On a similar note, in 10 years time,
when far more Indians have PCs, there will be a lot more cricket games
around.)

<snip>

FF I'll pass over, but AI... yes indeed. I think around 1997-8 a lot of
game developers got so e***d by the net that they neglected the
single-player game, and AI suffered (there are of course honourable
exceptions like Thief and Half-Life). Even on well-done games, some
things could be improved - if only the AI could slipstream in GPL, it
would improve offline racing enormously.

<snip>

You're dead right, and most of us are at fault. I suppose the title of
the group doesn't help - if it were something like
comp.sys.ibm.pc.games.driving thigns might be different. What we ought
to do is move from a sim/arcade split to a good/bad split, so that
something like Powerslide, which is a fun blast, would be treated more
kindly than Prost GP, which is a disaster.

<snip>

I can be smug here, as I don't do this. At all. All these "try before
you buy" excuses don't wash if you'ree living - as most piraters do -
close to a shop like EB that has a decent returns policy. (In fact, if
the mail-order firms started doing it, the excuse would vanish
altogether.)

<snip>

Actually, I think people make far too much fuss about this. It takes
about 15 seconds - who really can't spare that?

--
"After all, a mere thousand yards... such a harmless little knoll,
really" - Raymond Mays on Shelsley Walsh.

The GPL Scrapyard: Back again! http://www.racesimcentral.net/~gplscrapyard

   Can you and your friends join together for an online endurance event
where you all share the same car?  When we can have 40+ cars with 80 to 160
drivers running around Le Sarthe for 24 hours, THEN....THEN you might say
technology is topping out...
dave henrie

Did it reach it in 1980?  1985?  1990?  1995?  1998?  2000?

No.

Has every single detail of the handling of a racecar been simulated?

No.

Has every single environmental detail been implemented?

No.

Has Artificial Intelligence been perfected?

No.

So I'd say, overall, no.

But really, if you want higher visual quality, you ultimately need
faster fill rates and higher polygon counts rather than gimmicky
effects, which is what NVIDIA is attempting to provide.  I want to be
able to run Nascar 2002 at 1600x1200x32 with 4x FSAA and full details
using D3D at 100fps with a full field of cars at Watkins Glen.  I'm
not confident that there is a card capable of doing that yet.

<snip force feedback bit>

I hate force feedback and find it distracting and nothing at all like
a real car (considering NASCAR uses power steering, I wonder how much
you actually feel through the wheel anyway) so I can't add to what
you've said.

read more »

--
-- Fran?ois Mnard <ymenard>
-- http://www.ymenard.8m.com/
-- This announcement is brought to you by the Shimato Dominguez
Corporation - helping America into the New World...

Oooh.  Now that'd be nice...!

<snip>
I've heard there is something like this under development, a kind of cross
between a real-time Virtual Spectator and F1 2001 where you can race with
the real F1 drivers. It doesn't sound impossible technically, just get the
telemetry data from the cars and then feed it to the other PC drivers the
same way as multiplayer racing. I don't think you can take out the real
drivers though :-)

Tony

When I posted this I got the message:
Outlook Express could not post your message.  Subject 'Re: Have we reached
the end of the road with sim racing development ???', Account:
'news.freeserve.net Tony', Server: 'news.freeserve.net', Protocol: NNTP,
Server Response: '441 Article posted in the future', Port: 119, Secure(SSL):
No, Server Error: 441, Error Number: 0x800CCCA9

"Article posted in the future"? Spooky...