Misc Replacement physics for AC DTM 1991 Car Pack by Tommy78 2

[AccAkut]

AccAkut submitted a new resource:

Replacement physics for AC DTM 1991 Car Pack by Tommy78 - Mustang GT, Opel Omega Evo and Audi Quattro V8

Tommy78's DTM 1991 car pack is a paymod you can aquire here:

https://tpongracz.com/shop/ac-dtm-1991-season-pass/

The models are really well done, lowish poly but pretty detailed, with 4 proper LODs which means they work well for big grids.
Only downsite (for the AC release of the cars, as he offers them also for RF2) are the physics, as all three cars run on the same suspension (DWB on all corners, very hard sprung)...

Read more about this resource...

 

[Billy Pilgrim]

Thanks a lot for doing and sharing these updates. Great!

I know these cars were way too fast to run against Kunos's 90s DTM cars. I see you've lowered engine power. So are your versions now BoPed to run with Kunos's 90s DTM cars?

 

[AccAkut]

Thanks a lot for doing and sharing these updates. Great!

I know these cars were way too fast to run against Kunos's 90s DTM cars. I see you've lowered engine power. So are your versions now BoPed to run with Kunos's 90s DTM cars?

somewhat, I let the AI run a few races around different circuits and after 20 laps around the twistier ones the field was pretty mixed, on high speed tracks with long straights these three will be top of the order tho (tracks like Sveg). But they all have more power than the E30 and 190E so that is expected.

I tried to get the wheel bhp to the official numbers, as I assume that was what they tested back in the day in homologation. No idea about the torque curves, for the Audi I took one of a N/A tuned VW KR engine (as the 3.6 was basically a doubled inline-4 Golf GTi KR) and went from there. Mustang I lowered the mid range as it was just too fast.

Aero changes also make an impact, as sold the cars all generate around 150 kg of downforce, even tho for example the real life Audi actually produced front end lift when it first hit the track. I lowered it for all of them to get them similar in downforce as the 190E, some 50kg max at top speed I think.

edit:

i think this indicates its not bad, 50 car mixed grid. The Mustangs crashed a lot and ended up at the back of the field

 

[Billy Pilgrim]

Yes, i think this indicates it's really very good Those results are great (and after six laps at Nords too).

Thanks a lot for this. I'm gonna enjoy it a lot.

I had bought the Tommy78 DTM cars hoping to run them in grids with Kunos's cars so was disappointed to find that wasn't possible, really.

You fixed it!

 

[AccAkut]

Just let this run, that Gran Turismo Motor Sport Land recreation (very small and twisty)

so here the brutes have no big advantage. The AWD of the Audi helps it recover after spins which helps with these extreme AI settings.

 

[Genex13]

I recently suspended these cars from my primary hard drive when I analyzed the physics in CM.
I'm glad you brought them back.

 

[Wilco88]

The balance is pretty good against the Kunos DTM cars. But ai can not drive the Mustang, they crash within a lap.

And in qualifing all these cars use the rain tires but in race they switch to the mediums..

Thank you very much for this mod!

 

[Genex13]

I too noticed that the mustang crashes a lot with the AI.

There is something weird going on with the front suspension in highspeed corners, or over curbs, where the outer front wheel will suddenly snap under the power and flip around in a weird angle, resulting in tremendous understeer. It's even possible to permanently break it that way with enough
force (see screenshot)

 

[Billy Pilgrim]

I can confirm what the above two posts say about the Mustang as AI. I also drove it and found it really strange in its cornering (that may be in the original version too, I don’t know yet as I hardly tried the Mustang before) there are points in the corner where the Mustang understeers massively.

The Audi seems ok except the cockpit judders up and down a lot in a way I have not seen with any other car in AC. I think this is a problem coming from this physics update.

This update is still a positive so far though as I can at least use the Tommy cars in the grid with the Kunos cars as they’re balanced now (although the Mustang may not make it round a lap)

I’m enjoying the Opel a lot.

 

[AccAkut]

Hmm, weird. I didn't notice that with the Mustang, it was the first of the cars I did tho I had issues with the limits set in the setup ini. Will have to check that.
About the Audi cockpit, do you mean the car judders or just the view shudders?

Funny that the Opel turned out best as I spend the least time on it AND did the biggest experiment with that trailing arm rear

 

[Billy Pilgrim]

Hmm, weird. I didn't notice that with the Mustang, it was the first of the cars I did tho I had issues with the limits set in the setup ini. Will have to check that.
About the Audi cockpit, do you mean the car judders or just the view shudders?

Funny that the Opel turned out best as I spend the least time on it AND did the biggest experiment with that trailing arm rear

It's not the view that judders, it's the cockpit. And it doesn't judder constantly. Its most obvious effect is when you look in the mirror, but the dashboard etc is judderring too.

I tried the cars out at Road Atlanta only. The Mustangs usually couldn't get round the first corner. And, as I said, when driving, they are rather strange to get round corners (i don't know if that's yor update or the original physics though).

Thanks a lot for working on this. You've already made an improvement as we've got a balanced grid now And the Opel is a nice drive.

 

[AccAkut]

It's not the view that judders, it's the cockpit. And it doesn't judder constantly. Its most obvious effect is when you look in the mirror, but the dashboard etc is judderring too.

I tried the cars out at Road Atlanta only. The Mustangs usually couldn't get round the first corner. And, as I said, when driving, they are rather strange to get round corners (i don't know if that's yor update or the original physics though).

Thanks a lot for working on this. You've already made an improvement as we've got a balanced grid now And the Opel is a nice drive.

I want to investigate but I'm at work now, and not at home the whole weekend Really the last thing I did on the Mustang without testing it myself on track was that I lowered the ride height via ROD_LENGTH= as the car sat very high. And I ran like Targa Florio and a full 4 lap stint on Nords with it before.

 

[Billy Pilgrim]

lowered the ride height via ROD_LENGTH=

I don't really know but that could well be the problem with the Mustang. I have noticed the AI in the Mustang finds ascending corners very tricky (but I'm making suppositions here too a bit)

 

[AccAkut]

I too noticed that the mustang crashes a lot with the AI.

There is something weird going on with the front suspension in highspeed corners, or over curbs, where the outer front wheel will suddenly snap under the power and flip around in a weird angle, resulting in tremendous understeer. It's even possible to permanently break it that way with enough
force (see screenshot)

View attachment 409854

ok, found a minute... the wishbone folding 100% upwards and then inverting IS an issue. Tried quick fix by adjusting bumpstops and ride height, worked out for that corner at Mantorp so far .. makes no sense to dunk the front into the tarmac anyway

also see what you guys mean with judder in the Audi... wtf

 

[Wilco88]

This seems to be the issue as said above, fast corners with altitude change does seem to produce it.

ROD_LENGTH=0.075 fixes the problem. I tried 0.07 and lower but that didn't work. So I am inclined to think 0.075 might not be enough to fix it in every situation. Will test some more and report back here.

Edit 2: 0.075 seems enough to fix it even on Algarve and Bathurst.

 

[AccAkut]

View attachment 410113
This seems to be the issue as said above, fast corners with altitude change does seem to produce it.

ROD_LENGTH=0.075 fixes the problem. I tried 0.07 and lower but that didn't work. So I am inclined to think 0.075 might not be enough to fix it in every situation. Will test some more and report back here.

Edit 2: 0.075 seems enough to fix it even on Algarve and Bathurst.

Yeah changing the rod length "fixes" it but the main problem seems to be the combination of too wide bumpstop range and the wishbone being much too narrow, it just folds upwards with enough force. AC has no checks for intersection of suspension parts

I got very helpful documentation from Stereo and will try making new front suspension myself for it. I'm just not at home this weekend.

Would be cool if any of you could try driving the Audi with varied spring and damper settings, I wonder if somehow managed to match a condition that induces oscillation.

 

[Kyuubeey]

I really suggest to just use DWB to make trailing arms. Check out my Mini or 106 Rallye physics for an EX=0 FWD trailing arm, and E30 physics for an EX=0 RWD semi-trailing arm. IIRC in the 106 I made the arm longer to avoid popping out of place on big compressions, but it's not a big issue.

EX=2 has some serious issues with antilift if you make a trailing arm the easy way, but I'm yet to figure out the correct way to circumvent that.

The problem with strut is that the principle is just completely different. How camber, IC, antisquat etc. are formulated. It's easier and more correct to do it with DWB where you can more freely control everything.

Also, the toe is unlikely to be adjustable at all, as the tie rod is completely longitudinal. Lengthening or shortening it shouldn't really be able to change the toe, at least it never did for the geos I've made.

For the Mustang front, I doubt the arm is that short IRL looking at them. In vanilla physics you should be pulling the SAI along strut top to LBJT, not along the damper tube. CSP lets you separate them for correct IC by separating the steering axis (and FFB?) from strut*. There you can pull the strut along its own axis and place the LBJT correctly for SAI. I would bet the inner tie rod point is too far as well.

*Oops, I wrote it how it is in vanilla without thinking.

 

[Billy Pilgrim]

ROD_LENGTH=

How do i change this? I guess it's in the Data.acd file but when I open that up, it's just nonsense symbols.

EDIT: think i found it, using CM's UNpack data. It seems to be in suspensions.ini

That right?

EDIT 2: but I can't seem to save it. Any tips on this?

 

[Wilco88]

That is right, it is in suspensions.ini, after you changed it you have to save and go back to CM and "Pack Data".
Keep in mind this is just a short term fix, it works perfectly when you let AI use te car. The car is drivable but does not work as intended.

 

[Billy Pilgrim]

Ok, I managed to save rod length =.075

Thanks

The rod length change seemed to improve things for the poor Mustang. But I saw the Mustang AI off track, broken down and looking sorry for itself on one occasion so we’re not all the way there yet. Anyway, it was working mostly fine.

Re balancing the grid
I have done a couple of races in championship mode (so qualy + race) with all these cars (Tommy’s + Kunos’s + Ford Sierra 90s DTM). The AI level was at 100% and opponent strength at 100%, one race at Suzuka and one at Oulton Park. The Audi won both races by quite a bit, Mustang came second in both and Sierra came 3rd. Opel came further down.

I read in another thread (@ales100i was saying this, I think) that, on track in the 90s, the Audi was indeed the most powerful but its brakes were not good enough.