My post is related to problems with the position display. Nothing to do with the starting position.
I think the best way to solve the problem you describe is still manually creating the starting position and orientation for the reversed track.
Yeah, I think if you want real reversed starting positions that make sense, then you'd have to manually define them. I kind of like starting on the gates, though, just backward. Try Monza for example... you don't see the starting gate, but that's okay, you're in the start area. I'd rather be in the start area and not see the gate than out of the start area and see the gate, I guess.
Quote:I think we need one more change in BEZIER::Attach(), that is to reverse the sequence of the 4 corner points.
Alright, everything seems to be working now (R1566).
I still think this is a bit incomplete...I tried racing on Monza in reverse, with two opponents. The AI cars are placed in front of me (which doesn't give me much of a chance), and as I mentioned before all the cars start out in front of the starting line.
I think we should define a second set of starting points for reverse tracks as Joe suggests...this is the only way to fix the problems I describe above, and also keep things like having a race start on a curve still possible.
Another idea which might work, would be to take the whole set of car starting point vectors, and mirror them on the other side of the starting line, using the starting line as an axis sort of. This wouldn't work for every shape of track, but it would probably do for us for now...especially since the most cars starting at once will be 4.
R1570 has starting positions reversed, so you start ahead of the AI. I think this is Good Enough For Now.
That's definitely better. I still wish the cars were moved back behind the starting line...is it really hard to do this? I guess I can live with this for the next release. People will probably report it as a bug though.
I'd rather start in the start area instead of turn 1. I think it makes more sense to start on a start marker than in some random part of the road facing the start area. If you want to make an option to start in turn 1, you're welcome to. :-)
What do you mean by starting in turn 1? This is what I mean...go from this (forward):
Code:
: : : :
| | \ \
| | / /
| |_____________________/ /
| * * ] /
\ * * ] /
-------------------------
To this (reverse):
Code:
: : : :
| | \ \
| | / /
| |_____________________/ /
| ] * * /
\ ] * * /
-------------------------
Sorry for the crappy drawings. This is sort of what most of our tracks look like on the starting line area. the ] ] marks represent the starting line, and the *s are the cars' starting positions. I
don't think they should start in turn one (which I'm guessing would look like this)
Code:
: : : :
| | \ \
| | / /
| |_____________________/ /
| ] * * /
\ ] * * /
-------------------------
Okay, I was exaggerating, but you know what I mean. Look at Monaco... you'd start downhill on a bend... c'mon.
One problem with the current implementation is that the race starts at a point before the start/finish line. So the cars need to run an almost full lap before the lap counter is triggered to start the 1st lap. It's like we are having a rolling start instead of standing start.
Well Joe, how about a compromise. Leave the default way of reversing start positions as it is, but allow for custom reverse starting points so that we can start them behind the starting line if we want to. This would fix the problem rookie1 mentions which is what kinda bugs me about this also.
Another thing about reverse tracks: The finish line seems moved back about one road patch. So when I race in reverse, it doesn't end the race when I cross the finish line, but about 10 meters before I get to it.
All right, all right, I give. If we have defined reverse points, then let's have it use those.
thelusiv: the problem you mentioned is because it looks for you to drive onto the first lap sequence patch to trigger the lap. In reverse, you drive onto the first lap sequence patch from the other end, so it's in front of the finish line... I should be able to fix this by moving the first lap sequence patch back 1 patch during the reversal process.
I think I fixed this in R1578, please confirm.
The finish line seems to be in the right place now.
Very good fast work. My compliment
pwp71