The online racing simulator
Kunos Simulazioni - Assetto Corsa
(8882 posts, started )
Quote from Hyperactive :
The way inertias work in ac is that you just try to create a box that matches the dimensions the car. Then make the box longer if the cog is far away from the geometric center. For example a 4 metre long porsche would have inertia of something like x,y,4.5 meaning its yaw (and pitch) inertia is bigger than the car dimensions would produce. The jag length is 4.8metres. The inertia for the box is 5.2. I have not calculated the reverse of the 3 dimensional box kunos uses so I could use the inertias I've calculated for the jag. Maybe I should do that. But I don't think the problem is there. The tires have inertias and weight as well. Those are probably the heaviest parts in any ac apart from the race truck.

In real life you can have two cars both with mass CG right at the geometric center but with very different rotational inertias. Has to do with how far every unit of weight is from the mass CG. Still I think the box method is good. Do you have the bmw e92 box length?

Quote from Hyperactive :I'm thinking the dampers need more work. But also in the tires there could be something that makes the super wide tires handle too well. But I don't know what it is. The tires are not easier than the kunos tires. The xmu, load sensitiveness, camber gains. All those are very close to the 312t tires that I used as a base. Same make (dunlop), same construction (biasply) and same era. I've tried 90s slicks. The car handles even better. I tried the grpA tires from m3. Handles really really well.. On the first version of the car I made a mix of those 3.

When you tried 90s slicks and grpA where they of normal width or you made them super wide as well?

Quote from Hyperactive :I won't just start changing things randomly to make the car drive harder. I want to understand what is wrong. This is the best I can at the moment

Pretty sure you'll figure it out.
Quote from NotAnIllusion :Some time back I had the same issue with my own mod car, it turned out to be the tyres/unsprung weight, eventually I was able to tune it out.

If someone wants to have a go, https://www.mediafire.com/?uyz3mmimdcmzbx3

No need to point out that the model and sounds are borrowed

Now that was good fun....just took it around Brands Hatch, and loved it. Steering I think could do with being a little more responsive, to much turning circle??

You say you borrowed the model, I didn't know this Mod was out

Regards

Fordie
Quote from Fordman :
Steering I think could do with being a little more responsive, to much turning circle??

Turning circle is identical to the URD C6R, if that car is more responsive, it's likely to be a complex relation between geometries, dampers and tyres. I don't have an FFB wheel so it's impossible for me to tune. Someone else could try to tweak the various values of course, e.g. increase FFMULT or change STEER_RATIO and see what happens.

`; In car.ini[CONTROLS]FFMULT=1.20 ; Force Feedback power multiplayerSTEER_ASSIST=0.55 ; Variable steer assist, speed relativeSTEER_LOCK=320 ; Real car's steer lock from center to rightSTEER_RATIO=17.1 ; Steer ratioLINEAR_STEER_ROD_RATIO=0.00271`

Quote :You say you borrowed the model, I didn't know this Mod was out

The model was ripped from Forza by some Russians. Their "mod" is a payware scam: ripped model with URD C6R physics (literally, not changed at all). Mine is the model from that with Cobra sounds and 98% reworked physics, using real V8SC data where possible, tyres are also original.
Obviously can never be "released".

Lamborghini is coming!

Huracan GT3
Huracan LP620-2 Super Trofeo
Miura SV
Countach
Veneno
Asterion
Gallardo GT3
Gallardo LP 570-4 Superleggera

Bloody good news . Never been a fan of Lambrghini, but it may change
Like the Nissan GTR (I do not like it), when AC will provide them, I am sure I will be interested by those "4wd" beasts

Quote from kars19 :"Getting real close to a 1.0 version. (actually, it could be, but we'll likely find a thing or two, so I'm leaving myself a little head room until 1.0)"(c)Pierre-André Leclair
...
Gp circuit looks even better then KS version more trackside objects
..ok maybe not, but not far off

=> if Eric put this in LFS, I take a year off

Lamborghini is coming!

Huracan GT3
Huracan LP620-2 Super Trofeo
Miura SV
Countach
Veneno
Asterion
Gallardo GT3
Gallardo LP 570-4 Superleggera

oh my
Hyperactive I think your lateral/longitudinal grip ratio is too high. I tried to lower the DY0 and the car felt more like it should IMO. It also helped with the wandering while going straight.
Quote from PhilS13 :Do you have the bmw e92 box length?

nope

Quote from PhilS13 :When you tried 90s slicks and grpA where they of normal width or you made them super wide as well?

Wide. 345 front, 365 rear! (yolo) Maybe I should try just making the tires skinny to see what it does to the car.

Quote from NotAnIllusion :Anywho, the main issue I have with is that it seems to "wonder" quite a bit on straights and sometimes in corner exits. In the latter case I've straightened up the steering a second or two ago, have no oversteer, but the rear keeps rotating a bit which causes a loose feeling in the steering. In the former case it's a matter of driving in what I believe to be a fairly straight line, but find that it wants to go left or right.

Maybe this is a wheel issue because for me I don't really notice any wandering. Some people are saying it does it but personally I only seem to notice it when I drive with keyboard (testing stuff) or with a wheel braking into a corner where it does get a bit squirrelly. But with a keyboard I notice the car has a tendency to keep its rotation. Like if you make a turn on the straight the car kinda keeps going to that direction. But I don't notice it with wheel (fanatec 911 tbo s).

Quote from NotAnIllusion :That issue could easily have something to do with using mouse steering although it's the only car it happens with. I think there was a bit of that in one of the early Cobra releases too, now that I think of it.

Hmm, maybe I should tripple check the steering arms. Does the problem go away with more toe-in? Sinbad also mentioned light feeling in the center.

Quote from NotAnIllusion :Some time back I had the same issue with my own mod car, it turned out to be the tyres/unsprung weight, eventually I was able to tune it out.

If someone wants to have a go, https://www.mediafire.com/?uyz3mmimdcmzbx3

No need to point out that the model and sounds are borrowed

The model and sounds are borrowed.

I'll take a look. What was wrong? Too little unsprung weight, tires..?
Quote from PhilS13 :Hyperactive I think your lateral/longitudinal grip ratio is too high. I tried to lower the DY0 and the car felt more like it should IMO. It also helped with the wandering while going straight.

Those peak values come from my only source of info about the tires. Historix mod. Comparable tires (3.0 csl) which raced same times (in the same races even: http://touringcarracing.net/Pages/p%201977%20Zandvoort.html). The csl in rf had something like 1.5lat, 1.3long.
Quote from NotAnIllusion :Obviously can never be "released".

That is what I was afraid you was going to say

I will thought "for testing purposes only ) have a look at your ini file
Quote from Hyperactive :
Quote from PhilS13 :Hyperactive I think your lateral/longitudinal grip ratio is too high. I tried to lower the DY0 and the car felt more like it should IMO. It also helped with the wandering while going straight.

Those peak values come from my only source of info about the tires. Historix mod. Comparable tires (3.0 csl) which raced same times (in the same races even: http://touringcarracing.net/Pages/p%201977%20Zandvoort.html). The csl in rf had something like 1.5lat, 1.3long.

Was that type of ratio common in Rf in general ? Cause it might just be a specific thing that worked well on the Rf physics. I have never seen higher than 1.08 ratio in AC.
Quote from Hyperactive :I'll take a look. What was wrong? Too little unsprung weight, tires..?

I really couldn't say, because I never kept any sort of a log of what I changed. Any combination of the DX/Y values, flex, xmu, friction limit angles, relaxation length, hub masses.. Sorry, I can't be any more specific. Besides, the Jag may not even have a problem as such. It could be down to individual players' wheel settings and such. Various people with the same wheel seem to have very different opinions on some of the cars in AC :-\.

It's really quite complex the way the values interact with each other. When I was tweaking my tyres I found that anything more negative than -0.02 for the DX/Y1 values were undrivable, yet I've seen tyres that work fine at 3x that. Crazy.
Quote from PhilS13 :Hyperactive I think your lateral/longitudinal grip ratio is too high. I tried to lower the DY0 and the car felt more like it should IMO. It also helped with the wandering while going straight.

Hmm, I think I'll send a pm to aris to see if he responds. As far as I know he was one of the guys doing tires and physics for the gtr2 power&glory mod so he should have good bit of info about the tires these suckers have.

I'll check that reducing dy0 but again I just don't want to it blindly. So I need to figure out if it works why it works.

Quote from PhilS13 :Was that type of ratio common in Rf in general ? Cause it might just be a specific thing that worked well on the Rf physics. I have never seen higher than 1.08 ratio in AC.

I did check some rf mods. And nope. I don't see such pattern.

Quote from NotAnIllusion :I really couldn't say, because I never kept any sort of a log of what I changed. Any combination of the DX/Y values, flex, xmu, friction limit angles, relaxation length, hub masses.. Sorry, I can't be any more specific. Besides, the Jag may not even have a problem as such. It could be down to individual players' wheel settings and such. Various people with the same wheel seem to have very different opinions on some of the cars in AC :-\.

It's really quite complex the way the values interact with each other. When I was tweaking my tyres I found that anything more negative than -0.02 for the DX/Y1 values were undrivable, yet I've seen tyres that work fine at 3x that. Crazy.

It is odd because on my car I can pretty much change the dx1/dy1/xmu values in a major way without causing big difference to the car handling. What I have found is that my tires seem to be (or were at one point) somewhat camber sensitive.
To me it feels like there's very little or no caster angle- is that set independently or simply a result of the locations of suspension attachment points?
Quote from sinbad :To me it feels like there's very little or no caster angle- is that set independently or simply a result of the locations of suspension attachment points?

It's calculated with the attachment points.

open Steam\SteamApps\common\assettocorsa\sdk\dev\ksSusEditor\knSusEditor.exe and load any suspensions.ini file into it

caster is 4 degrees on the jaguar iirc (actually 3 degs)
Meh. I just drove around vallelunga and noticed the wheel holes show through. For some reason ac doesn't show the bits behind the wheel but shows what's behind the car...
Quote from JackDaMaster :

Lamborghini is coming!

Huracan GT3
Huracan LP620-2 Super Trofeo
Miura SV
Countach
Veneno
Asterion
Gallardo GT3
Gallardo LP 570-4 Superleggera

oh my

Where is the Diablo?
Quote from kars19 :
Quote from JackDaMaster :

Lamborghini is coming!

Huracan GT3
Huracan LP620-2 Super Trofeo
Miura SV
Countach
Veneno
Asterion
Gallardo GT3
Gallardo LP 570-4 Superleggera

oh my

Where is the Diablo?

lambo diablo SE30 Jota would be a nice car. Apart from the countach are those all awd? (I don't know much about modern lambos)
Quote from Hyperactive :
Quote from kars19 :
Quote from JackDaMaster :

Lamborghini is coming!

Huracan GT3
Huracan LP620-2 Super Trofeo
Miura SV
Countach
Veneno
Asterion
Gallardo GT3
Gallardo LP 570-4 Superleggera

oh my

Where is the Diablo?

lambo diablo SE30 Jota would be a nice car. Apart from the countach are those all awd? (I don't know much about modern lambos)

only Aventador, Veneno, Asterion and Gallardo LP 570-4 Superleggera
Great to hear there will be proper cars in the game soon.
Quote from Fordman :I watched some V8 video's on U tube

Watch some Symmons Plains, they have > 360 degrees of lock, but not much more: https://www.youtube.com/watch?v=atD9oXfGmCQ&t=18s and https://www.youtube.com/watch?v=fmqVn1WR0JI&t=40s

edit: "but not much more".

Let's say that a V8SC has
- 24 degrees of steering angle
- 400 degrees of rotation (so a bit over 180 per side)
Steer ratio would be 400/24 = 16.7

By using the below method to get the rod length, I get 0.00261.
Quote :; Enter AC in dev app mode enabled
; Open SUSPENSIONS app
; turn your steering wheel by 90 degrees
; check the actual steer ratio value
; modify LINEAR_STEER_ROD_RATIO value until ingame steer ratio and car.ini STEER_RATIO values are similar

So the final values for control would be as below, feels alright.
`STEER_ASSIST=0 ; Variable steer assist, speed relativeSTEER_LOCK=200 ; Real car's steer lock from center to rightSTEER_RATIO=16.7 ; Steer ratioLINEAR_STEER_ROD_RATIO=0.00261`

A bigger problem is that as far as I can tell, real V8SCs run a minimum of ~-14 degrees of caster. The current geometry only gives -6.4. Perhaps that's why it lacks feel that upping the FFMULT can't fix.
I want a clean post for this so no edit. I was told that the iRacing V8SCs have a steering ratio of 13.56. Assuming 400 degrees of steering wheel rotation, 400 / 13.56 = 29.5 degrees of steering angle.

- All of the AC console INIReader errors fixed.
- -15 degrees of caster.
- 400 degrees of steering lock (lock to lock, 200 per side).
- 29.5 degrees of steering angle.
- 13.56 steering ratio.
- LINEAR_STEER_ROD_RATIO is 0.00296.
- FFMULT is set to 1.5 from because I suspect that the increased caster will help with feel.

https://www.mediafire.com/?we1uzp11u7uf363

Lamborghini is coming!

Huracan GT3
Huracan LP620-2 Super Trofeo
Miura SV
Countach
Veneno
Asterion
Gallardo GT3
Gallardo LP 570-4 Superleggera

in reddit I also saw a mention of BMW M235i Racing
I worked on the steering issue a bit more, changed some values. Now it feels better for me, of course take such comments from me with a tablespoonful of salt considering my steering input device.

suspensions.ini changes:
`WBCAR_STEER= 0.390, -0.180, 0.122 ; Steering rod car side attach pointWBTYRE_STEER= 0.108, -0.193, 0.149 ; Steering rod tyre side attach point`

suspensions.ini original:
`WBCAR_STEER=0.390, -0.180, -0.122 ; Steering rod car side attach pointWBTYRE_STEER=0.108, -0.193, -0.149 ; Steering rod tyre side attach point`

---

car.ini changes:
`STEER_RATIO=12.5 ; Steer ratioLINEAR_STEER_ROD_RATIO=0.00248 ; Because of AC complex suspension geometry, you need to calculate manually the steer rod ratio`

car.ini original:
`STEER_RATIO=-12.5 ; Steer ratioLINEAR_STEER_ROD_RATIO=0.00216 ; Because of AC complex suspension geometry, you need to calculate manually the steer rod ratio`

In summary, changed the steering rod attachment points to invert the steering ratio to be positive and adjusted the steering rod ratio using the method in the file comment.

Out of curiosity, where did you get the dimensions of the car? A modern V8SC's height is quoted to be
Quote :Overall height (mm): 1225
(source: http://jackdanielsracing.nissa ... ategorised&Itemid=101)

and just by looking at images, the Jag doesn't look taller.

I noticed that sometimes the car gets into very severe oscillations that takes ages and ages to settle, after hitting kerbs. I reduced the car's height by 0.2 to be close to the height of the road-going XJ12C which I saw quoted to be 1372mm, and these huge oscallations reduced noticeably while retaining the boat-like characteristics intact.

Also found that some of the vagueness in the handling was down to the setup as well, I used the one attached (remove .txt).
Attached files

Kunos Simulazioni - Assetto Corsa
(8882 posts, started )

© 2002-2021 Live for Speed - Scawen Roberts, Eric Bailey, Victor van Vlaardingen
All timestamps on this website are displayed in your local timezone
FGED GREDG RDFGDR GSFDG