Cars (DATA REPLACEMENT) Nissan Skyline R34 GT-R V-spec Improved Physics by Arch

[Kyuubeey]

A-LSD doesn't actually have any effect during braking: unless that's what you mean. It's almost completely open on coast and the only lock it has is internal friction.

I tested for quite a bit, trying to decide between using A-LSD or ditching it, and A-LSD is an integral part in getting the V-spec to handle how it should. It quite noticeably reduces power-oversteer and turns it into a more gradual rotation.

 

[Kyuubeey]

Kyuubeey updated R34 GT-R V-spec Scratchmade Improved Physics by Arch with a new update entry:

Minor physics improvements, fixed tire shaking bug

13.03.2019 Version 1.35

- Tire changes
- Changed grip dropoff and minimum level
- Removed shader patch tire raytracing until fixed
- Suspension changes
- Changed S3 version rebound valving
- Removed HICAS from S3 version: for now just disabled
- Electronics changes
- Adjusted ABS

Read the rest of this update entry...

 

[JackCY]

Kyuubeey updated R34 GT-R V-spec Scratchmade Improved Physics by Arch with a new update entry:

Minor physics improvements, fixed tire shaking bug



Read the rest of this update entry...

Trying it now on the "factory" version.

A decent lap with S1 around the Nords (8:01.1). I was too cautios around 5-6 minutes marks, basically in a crash avoidance mode, so lost some time there.

7:35 is stock Kunos R34 on Nord. Tourist.

M7 R34 on Nord. with blankets: https://acstuff.ru/s/zCzQ
7:57.216

It could go faster if it was stable on turn in after brakes and over bumps.

Here is what it does:
Open rear diff coast + damper balance + toe change with ride height make it oversteery on brakes especially without left foot braking and giving it gas, over bumps rear loses traction spontaneously, locks rear inner wheel on braking, it loves to spin on it's front. It's "good" for oversteering it in and trying to drift...
Lower max brake force and adjusting toes to combat their undesirable change with ride height helps a little but it's still a far cry from a nice ride since the issue source is elsewhere to be fixed.

Tyres do grip more than stock Kunos tyres, about 5km/h if not more when cornering. But they also behave a little differently when they are losing traction and start to slide.

Car is unstable at the limit and unsuited for track use.

 

[Kyuubeey]

I mean, the car does more or less the time it should (Real alleged time is 7:54, but conditions and setup unknown. I think the sim car would be a little faster if we match conditions.), and I've matched up video footage on Suzuka quite well. I don't know if your criticism are legitimate or just your opinions or what: but you are definitely driving in a way which inherently unsettles the car. When I made the R33 V-spec, I could damn near put in the exact same inputs over the entire Nordschleife lap and it almost matched up.

 

[JackCY]

Well do a 7:55 smoothly with strong brakes, I want to see. Or at least sub 8 with any brakes.
I know the car isn't setup for me and likely neither for track use, on a limit the car is murderous. Wants to spin around after brakes and snap around over bumps. It's set that way right now. Is the original car supposed to drive this way, maybe but I doubt it and no one who values their life would drive it on a limit on Nordschleife set that way.

 

[Kyuubeey]

It is in my attention the ABS balance isn't quite right. On the 33 it was fine, but they changed it around for 34, and especially for late-model 34 when stability was increased. So I'll have to take a look. The suspension, it just is like that. V-spec dampers are too stiff. Stronger spring with Nismo dampers will feel much softer than stock spring with V-spec dampers.

 

[Kyuubeey]

Hey, thanks for the feedback, especially @JackCY. I'm taking another pass at some details of the car. Just fine tuning now to get the entry balance correct. I wouldn't have noticed it myself without feedback as I'm so used to the car now.

 

[Kyuubeey]

Because @JackCY owns an R34, I would be open to a further discussion. I never saw this brake-oversteer as a particular issue: more-so an inherent feature of the open differential and overall suspension balance. Actually there are some scenarios where more brake oversteer is clearly needed.

I have some basis for it as well:

For example here at 4:02, and various other sections, pay mind to the attitude of the car and the fashion and amount of brake application:

Or 4:20 where a mere lift of the throttle fully unsettles the rear. Or 5:20 where a mere tap of the brake unsettles the rear fully. Moreover none of my testers ever reported any instability on brakes, on R32 or R33. My R33 tester also owned an A-LSD, so it was open on coast.

Also here in this video, at 21:25 is something that stands out to me.

Notice the need to countersteer after braking, in a stock R33 V-spec. The BeeR car in the front probably has a rollbar and/or brake bias issue, but that is not to be used as reference.

22:10 as well. This appears excessive, no? But there it is.

Due to factors like this, I never really considered it an issue, mainly because the inputs matched up well. It's very well possible I'm driving around it and not paying attention to the issue. It does feel like there is too much rear looseness on brake application, if I would consider how I would set up the balance. So I'm keeping an open mind. I tried a lot of things and nothing worked: so do try your own if you wish and post the results.

PS: The R33 has almost identical suspension to the R34, and identical brakes. So I believe it's a sound reference. And the 33 is considered more stable than the 34, with more understeer...

EDIT:

Why not, have some more. Although too unclear to use as reference.

At 7:12 onward. Also 7:50 onward. Also 8:15 onward. The only place it seems where R34 doesn't oversteer on brakes is low speed, high front-slip scenarios.

 

[Kyuubeey]

Here, a comparison video. It's my first lap in a long time, and of course I'm not anywhere as good as Gan-san, so it's hard to say anymore what is a car or driver difference. I think both. Also I paused at one point to sync them up because A: He drives that section way better and B: He has a bit more boost than I do. The throttle oversteer is 100% my lead foot, for sure. Just compare the throttle inputs.

From the looks of this, the car is only very slightly too imbalanced, in some specific braking situations. Perhaps the master brake pressure is 1 or 2% too rearward, and it's 77% instead like on the Sumitomos and my 75% is incorrect for Brembos. Perhaps.

And like I said, R33 and 34 share the same brake system and suspension, barring dampers. Brakes also are a different color, different structural part in a few areas etc. but same performance.

 

[Kyuubeey]

Quite a strange situation I find myself in. I tested more and I agree there is a lack of controlled understeer when coasting. The old R33's indeed did perform it correctly: stable with little to no oversteer when coasting, unless in specific conditions. R34 slides regardless of anything.

It is not AWD related, nor is it tire, engine compression, diff or suspension related. So I really do wonder what it is. It happens at low speed as well, so it cannot be an aero imbalance: besides, they are correct. My current state is something resembling but I'll get it fixed soon I believe. Thanks for the feedback as always.

 

[Kyuubeey]

Well, I tried everything. It's most likely a function of the lack of a locking differential on coast, the track ratio, wheelbase/track ratio, CoGH etc. combined. There's literally nothing you can do and still remain inside the data parameters to fix the issue: if it is one. Again, no testers ever reported anything like this as an issue, among A-LSD owners.

A thought that came to mind is that perhaps it is an unnatural jacking force from either axle, but I can't confirm nor fix that. Nor should there be any present.

The reason I don't remember R33's doing this is because A: I drive around it and B: I had ported the buggy Shader Patch physics to the R33's: which do indeed mess up the wheel loads, making it into a more of a FWD CoG: which greatly overloads the front tires, I believe. Using that model will have other issues sadly: namely said wheel loads.

Now, if someone wants to post some video/anecdote proving that A-LSD GT-R's do indeed stick obediently to the road when coasting at the grip limit despite a naturally oversteer bias suspension and open differential, then we will have something concrete to work off. Up to now it has only been opinions, mine included. The only people who agree with the behavior are owners of A-LSD GT-Rs.

EDIT: I did also try things such as porting S-chassis front suspension geometry into the GT-R, changing the front roll center and camber curves, changing the rear bumpsteer curve, (The rear has a bit too much camber gain so it should already have excess grip...), jacking the front ARB way up while making the rear very low etc:

If this is a bug, it is a very devious one. But the video evidence suggests otherwise. *shrug*

EDIT: I'll still play with the rear upper link. ML sadly has serious bugs which prevent it's use, but as it is the roll center and camber gain in the rear is definitely not 100% right. I have some play-room to adjust the upper link, so I'll put it wherever provides me a closer video comparison/roll center/instant center and perhaps ignore the small camber inaccuracy...

 

[Kyuubeey]

It seems it was the caster axis. Curves are closer now in fact, and behavior is more believable. Also determined that my old brake bias was indeed the N1 R34 bias: so it should be slightly more forward. New update will roll out later today.

 

[Kyuubeey]

Kyuubeey updated R34 GT-R V-spec Scratchmade Improved Physics by Arch with a new update entry:

Corrected rear geom caster axis, brake adjustments

15.03.2019 Version 1.4

Old setups are not compatible.

- Suspension changes
- Corrected rear caster axis and upper arm wheel joint height
- Result is closer camber/toe curve and hopefully more accurate coasting behavior
- Electronics changes
- Adjusted ABS
- Brake changes
- Adjusted bias
- Adjusted force

Read the rest of this update entry...

 

[Kyuubeey]

Big physics update dropping later today. I wasn't planning to update so soon but the improvement is marked. Thanks again for all the feedback from everyone: it was all useful if I listened to it in the correct way.

 

[Kyuubeey]

Kyuubeey updated R34 GT-R V-spec Scratchmade Improved Physics by Arch with a new update entry:

HUGE PHYSICS UPDATE

21.03.2019 Version 1.45

HUGE PHYSICS UPDATE

- Suspension changes
- Corrected front geometry slightly
- Redid rear geometry based on schematics and measurements: wrongly implemented before
- Implemented S3 front camber arms to light tuned cars
- Electronics changes
- Changed HICAS logic
- Remapped some HICAS curves
- Remapped ATTESA lock curve: more realistic mid-corner, exit and drift behavior
- Drivetrain changes
- Removed front LSD from light tuned cars
-...

Read the rest of this update entry...

 

[Rogerbee]

Just been driving the standard R35 round LA Canyons and it's just sublime!

 

[Kyuubeey]

Just been driving the standard R35 round LA Canyons and it's just sublime!

Damnit, I told my clone to not release the R35 yet! Wait what

 

[Sanch0z]

Tried to beat 8 min mark with weather conditions that more or less represent real thing (Mid Clouds, Air 17°C, Road 15°C). Car became much more stable on braking (seems reworked rear suspension is paying off). Although it is still possible to loose rear on downhill or on high speed corners without rear wing setting (guess that's a R34 feature since ATTESA is remapped). Overall great mod. Can't say what needs to be improved at this point.
P.S. Do you have any plans on making physics for R35? I'd love to drive one.

 

[Kyuubeey]

Tried to beat 8 min mark with weather conditions that more or less represent real thing (Mid Clouds, Air 17°C, Road 15°C). Car became much more stable on braking (seems reworked rear suspension is paying off). Although it is still possible to loose rear on downhill or on high speed corners without rear wing setting (guess that's a R34 feature since ATTESA is remapped). Overall great mod. Can't say what needs to be improved at this point.
P.S. Do you have any plans on making physics for R35? I'd love to drive one.

That's mostly an open diff and stiff ARB thing. GT-R's lift up the inside wheel a bit on braking. The Skylines still need a bit of lock-curve tweaking. I erred on the side of caution and left it just a tad conservative for 1.45, but it should be *slightly* more RWD. It'll take a while to dial it in. Otherwise the base car is completed.

About the R35, I'd thought about it. I'd need to redo everything apart from the base ATTESA code from the ground up, so I don't think I will do that. However if someone is interested in modeling a CBA35 from scratch, get in touch with me and we'll talk. I'd probably be the best guy for the job.

 

[Icky427]

if only game devs could put this much work into the physics of every car in a particular game. imagine forza with its hundreds of cars all being correct to this level. what a monumental task that would be. all this makes me wonder how much is wrong with all the other AC cars...