GREAT NEWS, I just commited the modif of code for rain parameters in track !
So, in graphic menu :
- put 3000 at visibility%
- put 12000 to sky dome
- put 100 to rain density
take e-track-6, ruudskogen or wheel-2 for test.
For the rain the friction parameter is 0.6 by default, maybe modified ?
Look the code and test please
From: Annick et Jean-Philippe <jpmeuret@fr...> - 2009-03-17 21:52:54
Hi, Xavier, and all.
> GREAT NEWS, I just commited the modif of code for rain parameters in track !
> So, in graphic menu :
> - put 3000 at visibility%
> - put 12000 to sky dome
> - put 100 to rain density
> take e-track-6, ruudskogen or wheel-2 for test.
> For the rain the friction parameter is 0.6 by default, maybe modified ?
Not checked the code for the moment, but it's a nice beginning !
Some comments and ideas to imporove :
- to my mind, the settings you added are good for testing, but not very
* sky dome : what does this setting means ?
* rain density (= likelyhood ?) : for me the user should rather have
the choice of the weather on a "level" basis : sunny weather,
sunny but some clouds, more clouds, heavy black clouds, fog, light rain,
normal rain, full rain (the fact that this level is random, as currently
implemented, or user-choosen should be a setting for me,
and I think the user should not have to set a number for that,
but use a "combo-box" like control)
- to me these settings should be completely independent from the choosen track,
that is : it could apply to any track with no added track settings,
apart from the friction coefficient ... and may be even this could be
simply computed from the dry one and the rain level
(ex: dry coef * 80 % for light rain, 50 % for normal rain,
20% for heavy rain, for a very rough model)
- the user should also be able to choose the hour of the day for the race,
or let Torcs-NG decide on a random basis (as for the weather)
- finally, the water projections are not as realistic as I would like,
as they don't look "continuous" (it seems the car is alternatively rolling
on wet road and dry road : projected water, then no projected water,
then again ...)
Nice job, Xavier. I think you've got the main part of this first attempt,
go on !
Then, for a truly realistic simulation, some more or less hard work to complete:
- water on the windshield,
- really visible falling rain
- water pudles on the track, on a random basis, or even better on a geometric
basis, with associated dynamic friction coefficient management (up to 0
when too much water !)
- no more dry tire sound in the turns when water on the track
(yes, I know they also do that in american movies but ... so bad ;-)