Cars Physics reworked for the two Maserati 250F 2019-09-13

[alekabul]

They don’t do anything besides serving as a collider for if the tires pop.

What about the tyres sidewalls? Those don`t matter for handling too? Or they are not connected to rim sizes? Or real life physics don`t apply in that aspect of the game?

 

[mclarenf1papa]

What about the tyres sidewalls? Those don`t matter for handling too? Or they are not connected to rim sizes? Or real life physics don`t apply in that aspect of the game?

Tires are an empirical model, not physical. The tire width parameter doesn’t do anything either (and radius only changes radius - nothing about the behavior of the tire).

 

[Kyuubeey]

I too thought years ago that perhaps the dimensions are some kind of abstract multiplier for the values, so the misunderstanding is okay. However I will be honest and say I won't be surprised if you will still argue against it. Like your whole understanding of everything is based on misunderstandings and little trivia you read online.

 

[alekabul]

Tires are an empirical model, not physical. The tire width parameter doesn’t do anything either (and radius only changes radius - nothing about the behavior of the tire).

Got that, only have problem with tyre width - wouldn`t it affect the contact patch for size and shape? Or is it a 1 point model of a contact patch?

 

[Kyuubeey]

You guessed right.

 

[mclarenf1papa]

Got that, only have problem with tyre width - wouldn`t it affect the contact patch for size and shape? Or is it a 1 point model of a contact patch?

Once again, it's an empirical model, not physical. The tire's represented by equations, not its shape/size/construction.

 

[Rufus Baggsby]

1. Take a look at the Kunos "setup.ini". See the "min=15000" for the ARB? I forced myself to go a bit outside of the "setup.ini" only for the ARB so the car gets rid of the uncharacteristic understeer.
2. Front roll centre of both 250F`s should be exactly what Kunos did as I did not change the length and orientation of any arms.
4. Solid rear axle with no ARB is exactly what Kunos did, but I see your point. And don`t agree with it - the DWB would have change drastically the dynamics of the rear wheels laterally and DeDion is known to have the wheels parallel at all times.

Kunos didn't do everything by the book, which is obvious given that the real 250F didn't have a front ARB. Also, one of the last models they did, the Alfa Stradale, has cut and paste suspension from the Porsche 917 /30 for example.

Where did you get the roll centre information from (besides Kunos)? Access to that information for other cars would be invaluable.

The addition of an ARB to DWBs at the rear was precisely to emulate as closely as possible the lateral rigidity of the De Dion whilst providing the control over wheel alignment and roll centre that a solid axle doesn't permit. These factors in themselves have a great bearing on handling and would, in my opinion, negate the need for an ARB at the front. Personally I don't think it's doing any justice to the 250F to handicap it with the rear suspension of a horse and cart. However, if it was good enough for Kunos... oh no, wait a minute , wasn't your mission to improve the pants handling of the Kunos car?

Don't hold your breath for the Merc W125, it's more of a personal project and a learning curve. I can get it around Brands Hatch GP in 2 minutes now without any detours into the landscape, but it requires a whole different technique of driving, which is why I love it.

 

[Kyuubeey]

Abbo as well had said something about emulating an axle setup with DWB to provide adequate alignment control, seeing as axle in AC doesn't have any toe or camber, while it's very common to at the very least have some toe, and often some camber as well. It does make a sizeable difference.

 

[alekabul]

Where did you get the roll centre information from (besides Kunos)? Access to that information for other cars would be invaluable.

Agreed, it would be great to have that info readily available but still didn`t find any such source. If I see one I`ll share it.

Kunos or no-Kunos I`m okay either way.

 

[alekabul]

the real 250F didn't have a front ARB

After a bit of trouble with my gfx and installing a fine gtx 1660 I looked a bit more for info and turns out 250F had a front ARB for sure:

which is correctly animated in the model:

 

[Rufus Baggsby]

Sorry, don't see it myself, just an upper wishbone. ARBs hardly move so I don't think they would ever get animated.

 

[Rufus Baggsby]

You can see it here though. Apologies, there is a front ARB, you are right.

 

[lepiol]

respect for your background for physical relations in AC. I don't have those because I'm (not yet) committed to digital products. But the reality is different: Your power curve (@alekabul) is wrong, as is the torque curve. the torque curve is much worse! Kunos' curves are met here good.

T2 goes for 240 BHP @ 7,300 rpm and 270 BHP @ 8,000 rpm. Peak torque was at 5,500 rpm.

The bore / stroke ratio was surprisingly heavily oversquare at the time and was therefore designed for high speeds. Kunos' 250F T2 achieves consequently better lap times.

The determination of the yaw rate remains to be checked at Kunos' T2

 

[alekabul]

respect for your background for physical relations in AC. I don't have those because I'm (not yet) committed to digital products. But the reality is different: Your power curve (@alekabul) is wrong, as is the torque curve. the torque curve is much worse! Kunos' curves are met here good.

T2 goes for 240 BHP @ 7,300 rpm and 270 BHP @ 8,000 rpm. Peak torque was at 5,500 rpm.

The bore / stroke ratio was surprisingly heavily oversquare at the time and was therefore designed for high speeds. Kunos' 250F T2 achieves consequently better lap times.

The determination of the yaw rate remains to be checked at Kunos' T2

There`s one thing you got right - I`m not into this for ratings.

If you insist on the "real life" racing data being far from my research then try this - find and summarize lap times from the 50`s and compare those with times of AI on accurately recreated AC track in similar weather and track friction. Prove me wrong by revealing mismatch of data available to everyone.

I gather you like that constant car wobble that Kunos did. Well, maybe a matter of taste.. I hated it and got rid of it. Not a biggie. And most certainly not a "masterclass" of any sort.

 

[mclarenf1papa]

There`s one thing you got right - I`m not into this for ratings.

If you insist on the "real life" racing data being far from my research then try this - find and summarize lap times from the 50`s and compare those with times of AI on accurately recreated AC track in similar weather and track friction. Prove me wrong by revealing mismatch of data available to everyone.

I gather you like that constant car wobble that Kunos did. Well, maybe a matter of taste.. I hated it and got rid of it. Not a biggie. And most certainly not a "masterclass" of any sort.

I could make a van do the same lap times as historical data - that's the easy part.

 

[Kyuubeey]

This dude tried to prove that the NSX springs are too soft by posting random AI laptimes on a track the NSX has probably never even seen. I can't even begin to fathom how the logic in the system works.

 

[binijh]

@alekabul I race this car in real life (the blue & red, with the monegasque livery, #56), happy to provide evidence. Would be keen to work with you in figuring out how to perfect your mod?

 

[alekabul]

Would be keen to work with you in figuring out how to perfect your mod?

Most of all I'm interested in tires editing for the cars now. It is crucial to get them right or everything gets messed up.
Which of the tracks available in Assetto you were driving on? Can you get me a cornering speeds breakdown for one of those in the dry so I can get to adjusting the tires? And what tires do you use IRL?
Of course everything else which you see incorrect I'm ready to edit if necessary.

Cheers!

 

[binijh]

@alekabul Need to figure out if I have any Vbox footage and what data I can get you, but as a start, some things below:

Reference lap times:
- 2.05 in the dry at Nurburging GP (without the Schumacher curves)
- ~2.30 at Silverstone GP w/ the motorcycle chicane
- ~2.55 at Spa, no configuration change

Tires:
Need to check with the team but think running Dunlop R5s in the dry. What I can say from driving your V6 on AC is that the tires are slippery but predictable in a high speed slide, which is nice (probably more similar to period rubber... than modern rubber). Slower speeds they slide too much and light up too easily.

Handling:
Car is much too unstable in general, a common issue I find in sims of classic cars (have tried most of the classics on AC that I have driven and they're all too wallowy, twitchy and unpredictable). These things are brilliant driving machines, totally predictable, totally approachable to start sliding them around. Around twisty circuits you'll find 50s/60s cars that keep up with modern sports cars that are running wide, modern rubber.

Car is particularly unstable under power on an exit (the actual car sits back on its heels and keeps pretty still if you step on it on an exit) and much too unstable at speed - the car is totally planted if you're on the rev-limit on top gear. Feels totally safe. Haven't played with the set-up much, could be I need to fiddle with the suspension on AC. Will trying softening the rear a bit.

Brakes:
- Lock up too easily - Spa you can step on it pretty hard after the Kemmel straight and only at 75/50m from 175/200m break-point do lock-ups become a serious issue.
- Cold brakes pull left or right unpredictably (classic maserati drum issue) - would be cool if that can be simulated.

Engine:
- Feel like the torque curve starts a little too late. Car starts to develop decent torque at 3500/4000rpm. Need to look at it more closely.

This guy is a good reference view of how it drives - legit 250F and a good driver:

- though ignore how much the steering wheel is moved on the straights. The 250F wheel has a lot of give in the center you can wobble the wheel without moving the tires.
- Here also at Monza:

 

[binijh]

@alekabul Correction the tires are now Dunlop L15 550 (204 compound) on the front and L15 650 on the back. See here: https://www.longstonetyres.co.uk/vintage-tyres/dunlop-racing.html