Automobilista 2 Custom Force Feedback - Overview & Recommendations

what setting causes the self oscillation of the wheel?

 

I got the post section wrong, sorry

 

I prefer the previous file. Couldn't get on with the new file. Good to have options and thanks for your hard work.

 

Hi Richard, I'm waiting for my R12 so... what are your settings (Moza software)?
Do you use "Road Sensitivity"?

 

Hi JP i am running moza ams2 profile but with game ffb int up to 100% in game 50/0/24/50

 

Join the MOZA RACING FAN CLUB Discord Server!

 

I feel the same. With two mini bass shakers, one in pedals and the other under seat, I get rid of all FX in FFB, because it don't mix well with other vibrations (the frequencies aren't the same for the same input effect to be simulated) and mask pure steering FFB.

To keep it simple, FX in FFB is useless with good calibrated haptics in pedals/seat/rig.

 

As I said, the more immersion hardware you have, the more other senses are served. Some people adjust the butt kickers as if they were being kicked by a horse. Or the motion as if you were riding a roller coaster. No wonder you can't feel anything on the wheel anymore. For me personally, the feeling in the steering axle and on the wheel is the most important thing to feel the car and the track. To put it briefly, the FFB gives me the connection and the feeling for the car, then there are subtle additions from Motion, buttiker or something similar

 

Could someone with knowledge please explain the difference between the tyre_resistance1, tyre_resistance2 and tyre_resistance3 entries in the custom ffb files plz. I have experimented with them, but it would be nice to have a clear understanding for what they do.
Thank you.

 

FullFFB by JP

This is a full-bodied FFB!!
It is not for everyone, some may prefer a cabin by a bend in the river...

The final goal is to include all forces acting on the car, while been front axle biased (that's where the steering wheel is connected, right? )
It should shorten the incoherence between visual and tactile effects in AMS2, that many identify as a physics related problem (the visuals show information and FFB is completely mute to that, leading to a cognitive dissonance).
If the car moves (track/crash bumps, lateral and longitudinal weight transfers, etc.), FFB should signal that in some way.

It's assumed a linear/neutral profile on wheelbase software/driver.
Some manufactures may add they own telemetry interpretation on top of FFB provided by the game. That could increase (if in phase) or cancel this file's FFB details (if in opposite phase).
Please, control clipping (histogram, last bar, on tyre/suspension HUD), PARTICULARLY if you have a strong wheelbase. That (red) bar can appear on high suspension stress situations (high downforce + high speed turn + hitting a kerb), but just a touch.
This file assumes a (safety) best practice of limiting max torque at wheelbase software/hardware, not by reducing GAIN in game.
Use vehicle-specific gain (changeable while driving) to adjust FFB (low downforce cars may require adjustments, while LMDh and F1 should not).
From v6 onward, vehicle-specific gain has a double impact and is the best method to adjust clipping behaviour.



Typical in-game dials, last version (to get max output dynamic range):

• GAIN: 65
• LFB: 00
• FX: 20
• DAMPING: 08

Still to improve/implement:

• (no roadmap)

v21 (2024.DEZ.13)

• Revised vertical bumps algorithm (including special offroad conditions)
• Revised "weight transfer and details" section
• Revised "rotation" section (less oversteer compensation). Use yaw_gain, yaw_ramp and yaw_relax to control it.
• Adjusted scrub parameters
• (please, read the notes - very important to achieve more pleasure and less pain)

Spoiler: OLDER VERSIONS

v20 (2024.NOV.24)

• Introduced speed dependency and non-linearity to weight transfer function
• Further car rotation tweaks (new algorithm)
• Stronger off-track suspension effects
• Corrected a bug that was cancelling front scrub effects
• (please, read the notes - very important to achieve more pleasure and less pain)

v19.1 (2024.NOV.20)

• (sorry) published v19 has an older version of car rotation formula - corrected now
• (please, read the notes - very important to achieve more pleasure and less pain)

v19 (2024.NOV.18)

• Back to basics, some small changes, some bugs fixes, to achieve a stronger FFB overall
• (please, read the notes - very important to achieve more pleasure and less pain)

v18 (2024.OCT.11)

• New lateral G forces algorithm (implemented non linearity)
• Adjustements in many parameters (rack/effects ratio, fine details, front wheels velocity delta, weight transfer)
• (please, read the notes - very important to achieve more pleasure and less pain)

v17 (2024.OCT.02)

• Improved weight transfer algorithm
• Improved brake and general front suspension feel
• Improved surface transitions algorithm
• Parameter change in car rotation counter-action
• Parameter change in rear suspension
• (please, read the notes - very important to achieve more pleasure and less pain...)

v16 (2024.SEP.22)

• Introduced a counteracting yaw force to compensate the (excessive?) forces from tie rods
• Introduced rear suspension contribution
• Adjusted brake feeling (tighter feeling)
• Adjusted weight transfer function
• (please, read the notes - very important to achieve more pleasure and less pain...)

v15 (2024.AGO.25)

• Introduced dynamic adhesion gamma
• Adjusted front scrub effect
• Minor adjustment in bumps amplitude
• (please, read the notes - very important to achieve more pleasure and less pain...)

v14 (2024.AGO.18)

• Further improvements on front adhesion algorithm
• (please, read the notes - very important to achieve more pleasure and less pain...)

v13 (2024.JUL.20)

• If you are supersticious, consider skipping this version
• Dynamic front load got corrected (formula error)
• Significant change in front adhesion algorithm
• Significant change in weight transfer algorithm
• Car rotation implementation now only considers front axel forces from internal physics engine (they are enough to get it )
• (please, read the notes - very important to achieve more pleasure and less pain...)

v12 (2024.MAR.14)

• Increased front axel complexity/layers
• Elimination of vibrations around the center
• Code cleanup/simplification
• (please, read the notes - very important to achieve more pleasure and less pain...)

v11 (2024.FEB.18)

• Substantially revised front axel algorithm (better tracking, better weight transfer feeling)
• Substantially revised rear axel algorithm (to mimic rF2's manners)
• No more pseudo-gyro forces (rack forces already capture that very well)
• Several adjustments (including removed excessive oscillations)
• (please, read the notes - very important to achieve more pleasure and less pain...)

v10 (2024.JAN.31)

• Introduced ABS effect
  to adjust, change, cancel: (ABS_gain 0.4) and/or (ABS_freq 0.4)
  (different wheelbases/manufaturers may require changing those values to get desired effect)
• Improved near-steady-point feeling

v9 (2024.JAN.13)

• New formulation of several FFB parameters, with particular emphasis on surface details
• More resilience to FFB signal saturation
• Reduced harshness on several dimensions

v8.1 (2024.JAN.01) (corrective version)

• Corrects a bug in car rotation (yaw)

v8 (2023.DEC.29)

• Better near-center load feeling
• New adhesion/grip formulas
• Introduced tyre stretch component
• Introduced offroad texture (canned effect)
• Reintroduced scrub effects
• Overall adjustments in parameters weights

v7 (2023.DEC.05)

• Some parameters adjustments, mainly loss of grip (was way too strong, by mistake)

v6 (2023.DEC.04)

• Renewed front suspension load algorithm
• Improved (loss of) grip feel (both axles)
• Revised vertical G-forces effect
• Revised car rotation feeling
• Harmonization between car classes through Vehicule-Specific Gain (including mods)
• Vehicle-specific gain now has a double effect (CAUTION: requires less change to make a bigger difference)
• Cancelled scrub (canned) effects (those coming from physics engine still there). If needed, change "scrub_scale 0.0" (was 0.8 in v5, 2.0 in v4)

v5 (2023.NOV.26)

• Improved high G-Forces cars feeling (previously, clipping occoured)
• Improved loss of front grip feeling
• Improved car rotation and rear loosing grip feeling
• Adjustments in braking and vertical bumps definition
• Implemented surface transitions effects (track-kerbs-offroad transitions)
• Reduced standstill force
• Reduced scrub effects

v4 (2023.NOV.07)

• Improved lateral forces (including those caused by crashs or bumps)
• Improved vertical forces feeling
• Improved kerb feeling
• Improved offtrack feeling
• Improved scrub effects
• Improved engine effects
• Retouched gyro forces
• Reduced (not eliminated) wheel hands-off oscilations
• Removed both kerb and offtrack (canned) enhancements

v3 (2023.OCT.17)

• Adjusted overall grip feel (before, gain slider had a strong negative impact when set too low)
• Improved longitudinal weight transfer
• Increased minimum grip feel
• Added (back) independent/direct scrubs effects
• New in-game dials suggestions: 70-00-35-10
• Adjusted and tested on Moza R12 (other wheelbases may require further fine tuning from users)

v2 (2023.SET.27)

• Low Force Boost Slider (LFB) is now functional (before, was fixed at 15)
• Damping slider is now functional (was fixed at 20)
• Improved lateral weight transfers
• Improved track bumps feeling (now includes all vertical G-Forces acting on the vehicule)
• Longitudinal forces/weight transfer now depend exclusivelly on car G-Forces (from physics engine)
• Better front wheels grip differences/discernement/separation
• New recommended in-game gain settings (58-15-35-20)

v1 (2023.SEP.13)

• Original version

FILES:

 

There are 3 areas for tire_resistence. Additionally, tire_initial_resist and tire_resistence_timing are also included.
Tyre_initial_resist is the initial resistance of the tires when turning.
Tyre_resistence 1.2.3 is the resistance of the tires when turning, the higher the more resistance. And finally, tire_resistence_timing for a longer build-up of tire resistance. However, in my experience, tire_resiatence_timing should be left at 0 because otherwise there will be anomalies in the steering (too much delay in the steering).
Test everything and decide what feels best on the wheel.

 

Thx Daniel, I understand the fact they deal with the resistance and build up of resistance, but why are there 3 of them? TyreResistance1, 2, 3, what is the difference between them? Do they represent the force on the tyre at different track grip levels or different tyre heat levels?

 

I don't know for sure why there are exactly 3 tire resistances. You would have to ask the producer of the Original_dateien (Karsten Hvidberg). I always imagined it with 3 zones, i.e. on the outside in the middle and on the inside of the tire. Maybe it's your theory? But in the end, I don't really care, the main thing is that it works

 

I can't help thinking the second thing you change here is a direct result of trying to fix what the first setting does...

 

Thx again Daniel, I had not considered the different parts of the tyre, that seems the more likely scenario.

 

Sorry, I meant, "steering lock" (there's no steering ratio). Will change that.
And they are two different properties. One affects the transfer curve (linearity) and the other the absolute max turning angle of the wheels.

 

Is this for belt driven wheels or DD? I think you mentioned you're getting a Moza iirc.

 

Yeah I knew what you meant, it just seems that you're having to increase the steering lock in cars because you've moved the sensitivity to 100%

I tried your file and yeah for me to use it it would need scaling somehow with the steering lock - default lock it was like cycling over cobblestones on a bike with solid tyres (I can't remember which track but it was one of the Brazilian ones with long sweeping bends on it) Upping the steering lock made it feel better.

TBH with a geared wheel like I have that has a sloppy centre the only way really to drive it fast, especially RWD cars that are going to oversteer when you plant at the apex means less force because trying to correct the steering when it's going from effectively no force -> heavy is much harder. It'd be like trying to modulate an accelerator that took hard push for a bit and then suddenly went light - you'd press it too far every time you reached the light bit. Then if it went heavy again you'd be thinking whatever information someone is telling you you're getting it's more of a hindrance than a help.

I noticed too light and you end up spinning simply because it becomes difficult to simply point the wheels forwards and accelerate in a straight line, but too much and now the difference means you overcorrect or undercorrect. There's a sort of sweet spot (which may vary from person to person I suppose) but which, sadly is not really "nice ffb" and nothing like reality.

Maybe these 'light heavy light heavy' custom files work better with a DD wheel that doesn't have a sloppy lack of force at top dead centre.

 

Hello everyone I'm new in the world of AMC 2. The first thing I want to note is that an amazing simulator, you can finally delete ACC
I, like everyone else, wondered about feedback, but the suggested options on the forum, unfortunately, are not suitable for use. Going out on the grass, the effort should be minimal, in all the settings that I tried, wheel slip or loss of connection with the roadbed does not give any information. I can't go to TC 0 because I don't feel the moment of the wheels breaking. I decided for myself that for the time being I leave the default Custom settings for my ALpha 15 nm, where I can safely drive on the vehicle 0, and I get on the grass or tearing off the wheels, I feel the lack of friction force.

 

Sorry, sent you a private message about this, but I'll expose the idea here, nonetheless.

I believe there is a "max torque factor" and a "manufacturer factor" into how FFB feels.
I believe "belt" or "DD" don't qualify as factor/category on it's own, but I know what people mean (is more like a "torque range" categorization, low-mid-high).