GTR: You must be kidding me

> GTR, ... the > physics are far superior to GPL

I wouldn't make that claim. The cars in GTR are still too spin
prone, not as bad in version 1.0 (German version), but it's
still not quite right in version 1.5.

Getting good lap times with GTR seems too dependent on force
feedback. The force feedback is good, as it diminishes as you
approach the limits on the front tires, but with GPL, the force
feedback isn't as important, and you can get good lap times
without it (one joystick player has a GPLRank of -80).

It's difficult to drift the cars in GTR, but easy to do in
GPL and GTLegends.

Trying to induce understeer by pegging the front tires inwards doesn't
seem to work much at all in GTR, less than real life, while GPL
maybe a bit more than real life, but I've never driven a 67 F1 car
on the bias-ply tires with huge slip angles used in those days.

In my opinion, as a simulation game, GPL has better physics, because
you can run the cars closer to the limits and even exceed the limits
a bit and still maintain control.

Bottom line for me in a racing game, is being able to sense the limits
instead of memorizing them. GTR takes care of the steering inputs
with force feedback, but there's no real sense of what's going on
with throttle inputs.

"I wouldn't make that claim. The cars in GTR are still too spin
prone, not as bad in version 1.0 (German version), but it's
still not quite right in version 1.5."

My belief is that's a modern tires/setup issue. I don't have the
problem.

"with GPL, the force
feedback isn't as important, and you can get good lap times
without it (one joystick player has a GPLRank of -80). "

And that's definitely a GPL physics problem. Replays of those alien
laps show clearly the driver is slamming the steering side to side
rapidly. In a real car (or in GTR), you can't do that without breaking
one end or the other loose.

"It's difficult to drift the cars in GTR, but easy to do in
GPL and GTLegends."

Definitely real, not a problem. Modern tires and downforce.

"In my opinion, as a simulation game, GPL has better physics, because
you can run the cars closer to the limits and even exceed the limits
a bit and still maintain control."

Also realistic. Carroll Smith quote: "The one thing that's changed is
that the gap between I've got it and it's got me has narrowed
considerably.

If you like GPL's feel that's fine. But it isn't "better physics".

> "In my opinion, as a simulation game, GPL has better physics, because
> you can run the cars closer to the limits and even exceed the limits
> a bit and still maintain control."
>
> Also realistic. Carroll Smith quote: "The one thing that's changed is
> that the gap between I've got it and it's got me has narrowed
> considerably.
>
> If you like GPL's feel that's fine. But it isn't "better physics".

Probably based on the 1.0 version of GTR, a professional race driver
tried the game and made the comment that it was more realistic with
some of the assists turned on. I assume this was in reference to the
car being so spin prone.

I don't know if the problem is tire, differential, or momentum modeling
in GTR that is the issue, but the issue is that almost every low speed
spin results in the car facing backwards, as opposed to real life where
ending up sideways occurs just as much, if not more than backwards.

In real life, you can take off spinning both tires, and it's reasonably
easy to keep the car going forwards. Note that F1 cars have a much higher
power to weight ratio, have no traction control at the start of a race,
and it's not an issue for them. However, I haven't been able to duplicate
this with GTR, probably an issue with the GTR differential model letting
one tire spin and not both.

There's a Top Gear video where Tiff drives a 4 year old Ferrari formula
1 car (a Ferrari F50 is also tested in this save vided), and even in his
first time in the F1, he lost the rear and and recovered quite a few times
in the video. He was able to drift the car fairly well at low speeds, at
high speeds, the downforce is setup to make the cars more neutral (less
oversteer reaction), because of all the power it takes to keep the cars
moving at high speed with all the induced drag from the downforce.

"I don't know if the problem is tire, differential, or momentum
modeling
in GTR that is the issue, but the issue is that almost every low speed
spin results in the car facing backwards, as opposed to real life where
ending up sideways occurs just as much, if not more than backwards. "

If you want controlled slides in GTR here's a way to do it (probably
not the ultimate, but good ebough for demonstration purposes). Use
Dunlops (the others will heat up and lose too much traction. Springs,
shocks, and bars reasonably balanced. Max neg camber on front, small
negative camber on rear. Hard compounds. Diff max power, min coast.
Have a ball. You'll wind up sideways as well as backwards. Of course,
it's not a fast setup. An underappreciated quality of excellent
drivers is that they can drive fast setups that punish mistakes _right
now_. It's especially true in NASCAR. One big reason the guys at the
back of the pack are slow is that they can't drive a faster setup.

"In real life, you can take off spinning both tires, and it's
reasonably
easy to keep the car going forwards. "

You can make real pretty twin tire marks with the above setup.

"There's a Top Gear video where Tiff drives a 4 year old Ferrari
formula
1 car (a Ferrari F50 is also tested in this save vided), and even in
his
first time in the F1, he lost the rear and and recovered quite a few
times
in the video. He was able to drift the car fairly well at low
speeds..."

Tiff's lurid slides on Top Gear are a running theme (or a running
joke?). It's especially a hoot when he's supposed to be driving fast
on a track. Sideways ain't fast, just entertaining TV. I would
imagine they mess around with setup, tire pressures, etc. to put on a
show.

If you enjoy GPL, fine. But to say it has better physics than GTR is
silly. The relative responses to setup changes and control inputs are
concrete proof. There's also logic. Do you not think the guys who
did GTR are smart? Do you think they didn't bother to analyze GPLs
physics for strengths and weaknesses and incorporate the former and
reduce the latter? Do you think they didn't use increased CPU/video
card processing power and technical information from GT race teams to
do it better?

enlightened us with:
> It worked on mysystem, with excellent framerates, but, after a month
> of rFactor, the GTR demo left me unimpressed; I felt like a spectator.
> Could be the dark, sunset settiing. It lacks RF's visceral,
> you-are-there feel. However, I've heard GTL ups the ante- if you like
> the old cars.
Ahem, if you have enough machine for it... I love the damn game, bus
SHeesh...

Everything I have read and or found turned off, p4 2.4ghz machine (early via
board) geforce6600gt, I get sub 20 fps with the 20 car fields (dips to 1
when scoreboard is shown some races and not the next...) BTW I am talking
about now, when doing the offline CUP challenges. rFactor, I get over 50
fps stock settings upon install with almost as many cars, and nascar and GPL
are everything turned up, 1024x768 no aa, I get over 60. I just dont get
why it is such a gfx hog, nor why I cannot tune it down enough to make it
drivable?

> Do you not think the guys who
> did GTR are smart? Do you think they didn't bother to analyze GPLs
> physics for strengths and weaknesses and incorporate the former and
> reduce the latter? Do you think they didn't use increased CPU/video
> card processing power and technical information from GT race teams to
> do it better?

In spite of all the development that went into GTR, the developers
admitted that the German version 1.0 was too spin prone and addressed
this in the later versions.

My issue is not so much a physics issue, but a feedback issue. It's
difficult to push the the throttle inputs to the limits on any simulation,
because there are no forces to feel. The question is should the simulation
compensate for the lack of feel by making the cars more forgiving?
Ok, the assists are one way of doing this, but it's not quite the same.

Regarding the physics of GTR, is why doesn't induced understeer (pegging
the steering inwards to wash out the front end) work? Real racers
use some amount of induced understeer in mostly two situations: under
braking on a car that would otherwise be spin prone; on very high speed
turns with non-downforce cars because generally the throttle is pegged
just to maintain high speed against aerodynamic drag.

In comparing GPL to GTR, I submit the video below as an example, a Greg Stewart
hot lap at Rouen. You see induced understeer, counter steering and a lot
of steering corrections made at a rate that seems to match reality. With
GTR and other games, the steering inputs are done at a lower rate, mostly
because the feedback just isn't there for those subtle but quick
corrections. The force feedback in GTR and many other games is almost like
a steering assist, which dampens out some of the subtle steering corrections
a player needs to make. The physics in GTR may be better, but which game
GPL or GTR, gives the player a more realistic experience?

Personally, I'm decent, but not expert at either game. My GPL rank is -10,
negative on all track, with an 8:15 at Nurburg Ring my relative best.
My best lap time with GTR 1.4 at Spa is a 2:14.7 using a momo racing wheel
and pedals, and 2:15.4 when using twin joysticks (no force feedback) and
assists on.

Dual view video (top is on board, bottom is chase) of Greg Stewart at
Rouen:

http://jeffareid.net/gpl/gplrngs.wmv

"It's difficult to push the the throttle inputs to the limits on any
simulation,
because there are no forces to feel. The question is should the
simulation
compensate for the lack of feel by making the cars more forgiving? "

You may have isolated our differences there. I think GPL is more
forgiving, but also has some truly bizarre behavior with regard to
things like setups and (too) rapid control inputs. I'll take accurate
physics instead and cope with the fact it's a computer simulation.

I have dial up so a 30M file is out of the question, but I've seen
plenty of GPL laps with "rapid inputs". They wouldn't work in a real
car, and they don't work in GTR, which is why they're not used, not
because of lack of feedback. Also, in a real race car, countersteering
is a desperation move to save a car. It's dog slow on pavement. But
it's used extensively in GPL hot laps I've seen.

> I have dial up so a 30M file is out of the question, but I've seen
> plenty of GPL laps with "rapid inputs". They wouldn't work in a real
> car, and they don't work in GTR, which is why they're not used, not
> because of lack of feedback. Also, in a real race car, countersteering
> is a desperation move to save a car. It's dog slow on pavement. But
> it's used extensively in GPL hot laps I've seen.

It depends on the car. Here's a link to a maniac in a Porsche at
Nordschliefe (old Nurburg Ring), with inputs just as rapid and severe
if not more as what I've seen in GPL, plenty of countersteering too.
Maybe it's just the nature of rear engine low downforce cars.

http://www.green-flag.com/pages/welc...&id2=videomenu

> It depends on the car. Here's a link to a maniac in a Porsche at
> Nordschliefe (old Nurburg Ring), with inputs just as rapid and severe
> if not more as what I've seen in GPL, plenty of countersteering too.
> Maybe it's just the nature of rear engine low downforce cars.
>
> http://www.green-flag.com/pages/welc...&id2=videomenu
>

Thats 'Sideways Stephan' and he does it just for show on the RUF videos

Fast porsche laps at the ring look slower as they're tidy
eg:
http://vanadium.prg.dtu.dk/~pmc/ande...09_GT3_745.WMV

--

Robin

> > It depends on the car. Here's a link to a maniac in a Porsche at
> > Nordschliefe (old Nurburg Ring), with inputs just as rapid and severe
> > if not more as what I've seen in GPL, plenty of countersteering too.
> > Maybe it's just the nature of rear engine low downforce cars.
> >
> > http://www.green-flag.com/pages/welc...&id2=videomenu
> >
>
> Thats 'Sideways Stephan' and he does it just for show on the RUF videos
>
> Fast porsche laps at the ring look slower as they're tidy
> eg:
> http://vanadium.prg.dtu.dk/~pmc/ande...09_GT3_745.WMV
>
It doesn't depend on the car that much. Sideways is never
fast. If the tires are at that slip angle, they ain't developing max
cornering force, for sure. Seems like you've been looking at some bad
examples; Tiff, "Sideways Steven".

My final thoughts (I'm done here, but I'll read responses).
If the GTR cars on radial slicks with downforce _weren't_ a lot less
forgiving than GPL cars on narrow bias treaded tires with no downforce,
_that_ would be highly unrealistic. The setups (basically go-carts)
and driving style that leads to the fastest laps on GPL are not
realistic ones. That is not true for GTR.