rec.autos.simulators

Yeah, I think it's difficult to judge speed too.  Therein lies part of my
problem with tracks like Monaco and Rouen.  The hairpin turns are an adventure.
 For some reason though, I don't have the same problem at La Source.  Not sure
why...

Eldred
--
Dale Earnhardt, Sr. R.I.P. 1951-2001
Homepage - http://www.racesimcentral.net/~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.

--

"Racing! - Science for the action minded."

mark

--

"Racing! - Science for the action minded."

mark

--

"Racing! - Science for the action minded."

mark

Gerry

Lazy, sloppy, over confident, insensitive, egotistical.... stupid.

--

"Racing! - Science for the action minded."

mark

Critical inspection is a big part of my craft. I also think it is a big part
of a racers craft. Sim or not. Sometimes it's very hard to pinpoint a
problem. Once noticed, it can be even harder to find a cause and even harder
still to find a cure. At my job, we A/B constantly. A/B is a simple
comparison. A is a product that we know to be 'good' and B is our product.
We play them both at the same time then monitor A, then B... etc paying
careful attention to the differences. That practice in itself will reveal so
much. Finding the differences is as much a craft as anything else and
learning to be objective, one of the hardest accomplishments. Especially of
one's own work. We also A/B on an array of what we hope to represent the
spectrum of possible consumer equipment. A practice I'm pretty sure computer
software and hardware engineers use.

Now, we know, in this industry, that no one person is going to go into a
room by themselves and come out a month later with quality product.
Subjectivity being the main reason here. My tastes will preclude diversity
in my work. A certain grayness will manifest simply from my chemical nature.
Personality, ability, talent. My work will usually lay within my known
boundaries because inspiration rarely comes from within and there is no one
there with different aspects to refute my ideas. But even one other person
in the room, with all their nature and achievement to that point, there own
angle.... will see things that I can not possibly see. Imagine five
different people even. And that is where the discussions start.

So if I go to the audio programmer and the physical programmer and I say, "I
can't tell if the brakes are on or off." They may have some spare time to
explain to me how they've modeled these things. They may not. But it's up to
me to insist to the producer that my perception is valid. The producer must
then prioritize and ask anyone involved to put it on there list if it's
important. Some days you go, "Oh, this 6 should be a 7." and everything
falls in place. Other days you rely on faith that the physics guy respects
your opinion and will get back to you. Either way, when a group of people
are in a room together, and really looking into something, things get done.
It's not like a news group. Drawings and eyebrows and jokes and sound
effects and wild hand, arm and full body gestures. Abstract word
associations. Incomplete sentences. One person completing another's is
usually when you can tell when you've been awake long enough to get down to
the nitty gritty. Brainstorming is definitely gets the best results with no
small amount of sleep deprivation. :-) But face to face, no social ladder or
authority or ranking when troubleshooting.

I would expect the GPL team worked very much like that. Except for the A/B
part. It's very obvious they stayed in house for this thing. N4 drives that
point home. Anyway, if I were producing, I'd also be simming so I could beta
myself. Even just being me, I knew the brakes and perspective were goofy 8
months before Papy released the game. I just didn't know who to call.

2 parts to go....
--

"Racing! - Science for the action minded."

mark

Tires

Tires

What are the tires doing? For me, that's ninety nine percent of driving at
the limit.

Tires - So whattayou gonna simulate in your sim?

--

"Racing! - Science for the action minded."

mark

David G Fisher

Jan.
=---
"Pay attention when I'm talking to you boy!" -Foghorn Leghorn.

Grip levels are a pretty huge part of the car modelling to get wrong.
Especially since so much of the feedback you would get in real life is
missing. I can't enjoy any sim for very long if the grip modelling seems too
far off the mark one way or the other.

Remember you said recently that you found you could re-create in F1RC what
you saw the driver doing on TV, but not in GP3? I can do the same in F1RC
and that, along with the knowledge I'm driving on almost perfect replicas of
the tracks tells me a lot about the driving model. Back end breaking loose,
wheel action, using the same amount of curb, etc., on the same sections of
tracks in F1RC as I see the real life drivers do in TV means a lot when
judging the accuracy of the sim. Driver view, or trackside camera view
replays look exactly like what I see on TV. That isn't the only way of
judging a sim's driving model obviously, but it is important.

From the very beginning, replays I saw from experienced, talented GPL
drivers did not look like what I saw in real life footage, and from the GPL
AI. The racing line seemed irrelevent. People were criss-crossing all over
it, and were drifting for way too long. Something seemed way off IMO, and
was the main reason why I didn't stick with it longer.

David G Fisher

I thought you would have more decorum than the old "everybody hates
me"-routine. I also thought we'd built up more of an understanding since our
earlier run-ins than that.

Anyhow, I wasn't being sarcastic (otherwise I would have "flagged" the
sentence more clearly). I was being serious. You have claimed to have tested
(and to still have opportunities to test) software in the beta stage. I'm
just the guy who buys software and sits in amazement at some of the stuff in
there. So if either of us is in a position to know how some of that makes it
"in there" it is you, or am I mistaken?

Take GP3 (pre-2000 add-on) for example... when a real F1 driver cuts a
chicane there's no hand of god holding him down afterward, nor does the FIA
jamm his engine management resulting in loss of power. So what kind of black
flag penalty was Geoff simulating there? GPx players have been miffed about
this *feature* since the Amiga days.... yet it had been present in every
incarnation since (even in the add-on during qualifying... I mean, how hard
can disallowing that particular laptime possibly be??) How can any
betatester not have noticed *that*??

To take the experience closer to home, there are those in this newsgroup who
claim you tested RC2000. If this is the case, how come the word "chicane"
made it into the pacenotes? I can understand it would be very hard to vary
the "reading speed" according to the lie of the land as it is done in real
life, but to throw in this stop-gap without any indication of the severity
of the turns or the width of the road (narrows/opens) is nothing short of
staggering. If I'd ever done that to Johan I'm sure he would have had me
dragged along by the service van on the way home.

Obviously one can argue the simmer would practise the stages more than the
real life rally driver, but if the driver is going to go from memory as in
circuit racing, why bother with pacenotes at all?

True. However, you have also pointed out the AI is terrible at avoiding
wrecks. Do you have any indication this was picked up in-house and it was
just lack of time which prevented a fix before release?

Also true, yet, and I'm 100% positive on this, at the time there was also a
strong sentiment of "this is miles ahead of anything at the moment, let's
not nit pick" and "there is always the option not to exploit it". I should
know as I was eventually swayed by this argument on the issue.

Jan.
=---
"Pay attention when I'm talking to you boy!" -Foghorn Leghorn.

I knew some of the people who did the physics modelling for Vanishing
Point (on the Dreamcast IIRC).  Apparently they ended up using
approximately double the grip / tyre stiffness (I don't know details),
since an 'accurate' model left the car feeling like it was on ice.

Jonny

It did seem to be sarcastic in some way. My reply was not serious though.
Just thought I'd fill in the blank before someone else did (like my buddy
Thom).

I think I read where Crammond doesn't take feedback from testers. They are
probably only used to find bugs after teh game is completed. He obviously
likes it that way, or something in the program makes it difficult to do it
the right way (like F1RC).

They had Derek Ringer and Clare Mole do the pace notes, and about 17 other
real life drivers from the various teams, including Martin Rowe (BRC champ).
I'm sure they noticed what you did. My only guess to explain this has would
be that it probably has something to do with the track editor. I think that
the editor automatically added the correct pace note as a particualr section
of track (type of turn or straight) is added to a stage. To go through all
500 miles (I think that's right) of track and individually describe each
section perfectly would of been a lot of added work. The sim was a pretty
massive undertaking as it was.

Well, for me it's not nit picking at all. Just the opposite. I wouldn't play
F1RC if the grip was as far off as it seems to me it is in GPL. I used to
say that GPL would be simply ridiculous to play if it had rain racing.
Someone else said it's a shame that there wasn't some way to adjust the grip
since it's release, and I agree.

Ok sorry 'bout that, anyway...

That is one way these things happen, of course, ignoring feedback.

Like I said, I can understand that. Trouble is this method is bound to throw
up some accuracy/timing errors. Corners which, on the basis of radius would
equate to a certain number on the severity scale usually can either be taken
faster or should be taken slower according to certain factors (camber /
bumps / width). Similarly, in the case of a much slower corner closely
following a fast one, a pure "geographical" note might be called too late to
slow enough for the 2nd corner. Both types of error are indeed quite common
in RC2000 (not a lot of *really* bad ones though), leaving the player having
to remember where the notes can be trusted and where not. An activity which
is not exactly part of a rally driver's job. Ideally the player should be
able to change/amend these base notes, insert warning calls and such. An
idea for a sequal perhaps.

Ok, fair enough.

Well, two things:

1) 1967 is a long time ago. Footage contradicting GPL is pretty thin on the
ground nowadays. In F1RC's case any car attidude inaccuracies would be
flaunted in our faces every forthnight as <sarcasm>350 million people cannot
help to watch some part the races</sarcasm>.

2) it is possible to change GPL, not in a nice, user friendly way, but it is
possible. I know for a fact one person edited in his very own 4 cylinder, 5
speed, 10500rpm F3 engine (based on real life data of a 1972 Ford unit
IIRC). I've seen a replay of it in action. This same person claimed it was
possible to alter the simulated shape/weight/drag (as opposed to the drawn
shape) of the cars. I assume a person armed with that knowledge could find
the data for the tyres too if he/she knew what those might look like.
However the person in question didn't want to take it further, much less
release it due to the negative reaction the GPL Engine manager received.

Jan.
=---
"Pay attention when I'm talking to you boy!" -Foghorn Leghorn.