FPS has nothing to do with the physics. There's lots of examples of sims
that have a physics engine which updates independantly of the frame rate. I
believe the ISI engine games update the physics at up to 400 times per
second.
FPS has nothing to do with the physics. There's lots of examples of sims
that have a physics engine which updates independantly of the frame rate. I
believe the ISI engine games update the physics at up to 400 times per
second.
>> High frequency bumps? BTW, the fps don't matter when you're talking
>> physics...
>> Cheers!
>> Remco
> Eh? Are you saying that simulating high frequency bumps is nothing to
> do with physics? They certainly would be in my world coz they're going
> to affect how the car behaves. And I thought hfb could not be
> reproduced in an effective enough way because, in the time it takes to
> jump from 1 frame to the next, the car would have travelled too far
> along the track in order to allow for hfb's to be simulated correctly.
> Hence more fps required.
Alex.
Hey they at least gave us a rainbow from time to time at Dover <g>
--
-- Fran?ois Mnard <ymenard>
-- This announcement is brought to you by the Shimago-Dominguez
Corporation - helping America into the New World...
Btw, what's happened with him? I guess he's gone back living on his island
with all the ca$h he made cowing a 15 year old game engine.
--
-- Fran?ois Mnard <ymenard>
-- This announcement is brought to you by the Shimago-Dominguez
Corporation - helping America into the New World...
> >> On Sun, 21 Nov 2004 15:10:48 -0000, Peter
> >> >That watchamacallit to do with being able to emulate the bumps on the
> >> >road. Yeah, I forget the name and I'm pressed for time to look it
> >> >up. <BG> Unfortunately, we'll need sims that run at 1000fps I'd
> >> >guess. :)
> >> High frequency bumps? BTW, the fps don't matter when you're talking
> >> physics...
> >> Cheers!
> >> Remco
> > Eh? Are you saying that simulating high frequency bumps is nothing to
> > do with physics? They certainly would be in my world coz they're going
> > to affect how the car behaves. And I thought hfb could not be
> > reproduced in an effective enough way because, in the time it takes to
> > jump from 1 frame to the next, the car would have travelled too far
> > along the track in order to allow for hfb's to be simulated correctly.
> > Hence more fps required.
> He said that simulating HFB has nothing to do with fps. FPS reflects how
> often the image is drawn on the screen. When the time to draw the next
> frame arrives, renderer takes the current state from the physics engine
> (which may either operate on higher frequency or use continuous model, or
> combination of both).
> Alex.
[TRACKVARS]
RoadDryGrip=1.00
RoadWetGrip=0.80
RoadmetalGrip=0.80
RoadDustGrip=0.90
RoadBumpAmp=0.010
RoadBumpLen=13.0
RumbleDryGrip=0.95
RumbleWetGrip=0.60
RumbleBumpAmp=0.015
RumbleBumpLen=8.0
MiscBumpAmp=0.08
MiscBumpLen=4.0
There are also other stanzas in the file that modify these HF variables as
well as add more for washboard bumps, road cracks, cement patches, dusty
berms, grass, gravel and rumblestrips.
SB
>> >> On Sun, 21 Nov 2004 15:10:48 -0000, Peter
>> >> >That watchamacallit to do with being able to emulate the bumps on the
>> >> >road. Yeah, I forget the name and I'm pressed for time to look it
>> >> >up. <BG> Unfortunately, we'll need sims that run at 1000fps I'd
>> >> >guess. :)
>> >> High frequency bumps? BTW, the fps don't matter when you're talking
>> >> physics...
>> >> Cheers!
>> >> Remco
>> > Eh? Are you saying that simulating high frequency bumps is nothing to
>> > do with physics? They certainly would be in my world coz they're going
>> > to affect how the car behaves. And I thought hfb could not be
>> > reproduced in an effective enough way because, in the time it takes to
>> > jump from 1 frame to the next, the car would have travelled too far
>> > along the track in order to allow for hfb's to be simulated correctly.
>> > Hence more fps required.
>> He said that simulating HFB has nothing to do with fps. FPS reflects how
>> often the image is drawn on the screen. When the time to draw the next
>> frame arrives, renderer takes the current state from the physics engine
>> (which may either operate on higher frequency or use continuous model, or
>> combination of both).
>> Alex.
>Cheers to those who put me straight. So are HFB's being simulated in
>some sims? I remembered an old discussion here on ras which, after some
>google groups research, may have been to do with GPL and some time ago,
>that said that pc's weren't fast enough to simulate them. Has that
>changed now?
>--
>Pete Ives
>Remove All_stRESS before sending me an email
If your graphics are running at 30 fps, then the car is moving 10 times, then
the graphics are drawn, etc.. So when talking about high frequency bumps it's
not frame rate dependent at all. Instead, it's about the physics engine
sampling frequency (the 300, not the 30).
How far can the car go in 1/300th of a second? If there's a bump smaller than
that then the car will drive right through it without seeing it. In practice
the car will sort of go through a bump that's quite a bit longer than that, I'd
think at least 4 times the size.
So the smallest bump you'd probably feel consistently at 100mph would be:
With:
1/300 = 0.0033333.... * 4 = .01333333... seconds of travel
100mph = 146.67 ft/sec
146.67 ft/sec * 0.0133333... = 1.95 feet
And that would be a bit inconsistent too, sort of the limit in a way. I.e., if
there was a bump 2 feet long and you drove "through" it at 100mph you would
definitely hit it every time, but you'd hit it a little differently on each lap
probably. I.e., a wheel would hit the bump 4 times with as you passed, but you
might hit the peak on one and not another pass.
At 200mph of course you'd be at about 4 feet minimum wavelength then. Seems
it's worse than I thought. Granted, if one sat to think about this a bit, this
sort of standard limitation could undoubtedly be overcome :-)
Todd Wasson
Racing Software
http://PerformanceSimulations.com
http://performancesimulations.com/scnshot4.htm
Yep. I remember someone from Papy saying how much more accurate the
physics engine in GPL was due to the cpu's of the day allowing 2XXhz (two
hundred and something) running of the physics engine! :)
<snip<
Oh Yea? But what would happen if one of the crew members git's sent
a'packin for the weekend? Where would he watch the race from? City 17?
Mike
-Larry