[REL] 1967 Endurance (Proto/Sports/GT) - modded Howston G4

The windshield thing is really an rF2 issue. S397 has to get inner windshield shading straight, what happens with it right now is that it basically works in reverse. It acts in a way it should if light sources would be from front when in fact light is being cast from behind. It happens with shadowign just the same, windshield shadows does nto appear when headign into the sun and drivign into the shadows, but rather when driving out of the sun and leavign the shadows...
 
Yes, that makes sense, it does behave as you say. Pity because it really causes a problem in certain situations. Although it is much more pronounced on this car than many others I drive in the same situation. Woochoo did suggest that something could be done but only if another edit were being carried out anyway. I.e not just to fix this. Which is fair enough.
 
Thanks for the feedback Benn and James,
Had never noticed the AI not replacing tyres. We did several 2/3/6/12hr accelerated time tests during development and there didn't seem to be any trouble with that. They sure stopped for fuel, and the cars tipped side to side suggesting they were taking tyres.

I've just spent a bit of time tweaking/checking the windscreen and have now put v251 up on the workshop.
It will be a little easier to deal with in the described situation, but also a little cleaner during daylight.
I hope it works out :)
 
Fantastic, thank you. I will have a go in a couple of days and report back but I'm sure the lighting issue will be much better now. I will also look at the tyre thing although I am sure that in my races they did change them as their stops were too long to have only taken on fuel.
 
woochoo said:
Thanks for the feedback Benn and James,
Had never noticed the AI not replacing tyres. We did several 2/3/6/12hr accelerated time tests during development and there didn't seem to be any trouble with that. They sure stopped for fuel, and the cars tipped side to side suggesting they were taking tyres.

I've just spent a bit of time tweaking/checking the windscreen and have now put v251 up on the workshop.
It will be a little easier to deal with in the described situation, but also a little cleaner during daylight.
I hope it works out :)
Click to expand...

@woochoo @davehenrie I agree, when they stop for fuel, they change tyres. But when you turn off fuel consumption (and I had to, because half of AI ran out of fuel before reaching pits, on various tracks), they don't make a decision to pit and change tyres. So they stay on track (at least 3h in my last test), using their tyres to the bone. And a long endurance race without pitstops is not so much fun.

Another solution could be that AI pit for fuel a bit earlier, when there are still few liters left in the tank.

Anyway, this is the 1% wrong with this mod. The other 99% is great, and you did fantastic job
 
What tracks did they run out of fuel at? I have done a couple of 40 lap stints at Le Mans 66 with them and they fuelled and changed tyres so if you want to try that I think you should not have a problem, but I agree, you need the pit stops working well in an endurance race.
 
Ok, did 30 laps again tonight, I found that the windscreen reflections are still quite drastic in dark sections of Le Mans, better than before and I can now see to get round the corners but only just, would still like to see these reduced a little further but that's just me and I genuinely appreciate what you have done. The AI were definitely pitting and changing tyres but I have fuel usage as normal. I lost quite a few AI in the race but not to running out of fuel but accidents. I do have AI mistakes enabled in the json file but the attrition rate seems high, will experiment with that. I will try to run a race tomorrow with fuel usage off and see if I can replicate the AI not changing tyres.
 
@woochoo @JamesB @davehenrie Guys, I got it. This is the ultimate result of maaaany tests.

When fuel usage is OFF, and tyres use is on, cars never pit to replace tyres, on any track.
When fuel usage is ON normal, cars change tyres when they stop for fuel.

But on tracks that have less pits spots than cars on track (Norisring, Longford, Brianza...), some AI have to do extra laps to wait their turn before coming in. That's where about 20% of AI run out of fuel at the first pit stops round.
When there are enough pit spots (Le Grand Circuit, Spa 2020 endurance,...) everything if fine with refueling.

English is not my first language, but I hope it's clear for everybody.
Conclusion: if you want to drive with tyres use only (no fuel usage, to avoid some extra DNF risk), AI never pit.

Cheers
 
Benn said:
@woochoo @JamesB @davehenrie Guys, I got it. This is the ultimate results of maaaany tests.

When fuel usage is OFF, and tyres use is on, cars never pit to replace tyres, on any track.
When fuel usage is ON normal, cars change tyres when they stop for fuel.

But on tracks that have less pits spots than cars on track (Norisring, Longford, Brianza...), some AI have to do extra laps to wait their turn before coming in. That's where about 20% of AI run out of fuel at the first pit stops round.
When there are enough pit spots (Le Grand Circuit, Spa 2020 endurance,...) everything if fine with refueling.

English is not my first language, but I hope it's clear for everybody.
Conclusion: if you want to drive with tyres use only (no fuel usage), AI never pit.

Cheers
Click to expand...
Thanks for your hard work. Your results also point out why we need to know PRIOR to loading a track just how many pit stalls are available.
 
Thanks for looking into it so precisely Benn.
I have entirely no clue about sorting that issue to get the cars to pit only for tyre wear. Perhaps it is something mantasisg has seen at some stage, or someone else. Same with cars waiting too long to pit while their boxes are occupied. No idea if they have ever worked their way around that or if it has always been an issue. And I can't promise either of those issues will be addressed soon (if possible), because we both have some other stuff on our plates right now. But again, thanks for working to tease out the issues and factors.
If we can't get to it i think we'd welcome anyone with an interest in tweaking the numbers. I assume it's not as easy as it sounds, but as i said, i don't have any experience with those factors
 
@Benn Good work, I will set of an AI only race later just to confirm that I see the same effects but what you have observed absolutely explains the behaviour even if there is not an obvious answer. At least knowing that, we can plan races accordingly.
 
I forget if this has been mentioned, in the suspension setup page, you cannot set separate slow rebound/fast rebound. To adjust the fast rebound, you have to click on the slow rebound and then the fast numbers mimic the slow. That was with the symmetrical link active, but that should only work for left and right side changes.
upload_2021-2-28_15-43-5.png
 
woochoo said:
I have entirely no clue about sorting that issue to get the cars to pit only for tyre wear.
Click to expand...
Found the problem. Let's examine, for example, HG4_275.tbc:

WearGrip1=(0.992,0.984,0.977,0.971,0.966,0.962,0.959,0.956)
WearGrip2=(0.954,0.952,0.949,0.945,0.936,0.913,0.85,0.765)
AIPitThreshold=0.7

I have absolutely no idea what the significance of having two different grip curves is, but the actual problem is that the AI is set to pit at a tyre grip level that the tyres are physically unable to reach. Setting it to something that's over 0.765 would fix the problem.
 
So it's something like an instruction to change tires, take fuel and repair damages that overrides any consideration the AI could take into account to decide to do this but not that?
 
You must log in or register to reply here.
Back
Top