Automobilista 2 Custom Force Feedback - Overview & Recommendations

@Karsten Hvidberg ... I have tried in every way to keep the double center but I have not succeeded, the comparison with a single value together with the "blend" remains superior in consistency and coherence ... moreover I have not been able to remove the items relating to weight and to the grip both in the yaw and in the center, taking them off is pleasant on the one hand but makes the sensations more vague on the other ... for the rest I managed to keep all the news related to 5.0.1.4 ... if I can I always try to follow your development.
In this way I was able to overcome my personal feeling compared to the excellent 5.0.1 ...

The conservation of the power_steering is only facade, because if you convert the values you get the same results as your last one file ... why do I do it??? Just to quickly change from damping to power_steering and see how the file reacts ...

Another deepening deserves the topics "grip, strech and slip" and because "scrub and tear" are inverted, but tonight I'm exhausted and I'll make it clear another day!!!

Thanks for all master ... and good night to all!!!

 

Thanks for posting this file . Is this compatible with the current public beta build ?

 

Thanks as usual @Stakanov , you have outdone yourself as always, the file is beautiful and the indications together with the alternative values are useful ... compared to your excellent first modified version (very direct and decisive) I now feel the softer and more sinuous tires, always informative at the same time to a greater balance between the front and rear sensations.
For this reason you added stretch in the name?

 

Thank you for your thoughts and added information, @Stakanov!

And thank you for all your efforts, despite circumstances, a true inspiration, and taking this whole ping pong experience between us all to another level!

I am really looking forward to giving your files a go!

 

I'm currently using it on the public release candidate with no issues.

 

Yes, of couse
Regards

 

It is not a fancy name but it is precisely the use of the strech to influence the grip and other voices ... the game or its physics processes its Strech value for each wheel FL, FR, RL and RR ... in the top of Karsten's file we have how this code is after utilized or managed...

Spoiler: 1° Block

(FL_stretch (power FL_stretch (power (abs FL_stretch) -0.025)))
(FR_stretch (power FR_stretch (power (abs FR_stretch) -0.025)))
(RL_stretch (power RL_stretch (power (abs RL_stretch) -0.05)))
(RR_stretch (power RR_stretch (power (abs RR_stretch) -0.05)))

The strech above indicated in 1° Block (with different front/rear power coefficient) is oblivious linked to the fuction tyre_strech (2° Block where are all the component for the karsten strech tyres) ... where are even the grip function, but even here i remenber that in original code, in the front the grip_fl/fr is absent...

Spoiler: 2° Block

(tyre_fl_stretch (* front_tyre_stretch_feel grip_fl FL_stretch angle_d_l wheel_diam_FL 0.5))
(tyre_fr_stretch (* front_tyre_stretch_feel grip_fr FR_stretch angle_d_r wheel_diam_FR 0.5))
(tyre_rl_stretch (* rear_tyre_stretch_feel grip_rl RL_stretch angle_d_l wheel_diam_RL 0.5))
(tyre_rr_stretch (* rear_tyre_stretch_feel grip_rr RR_stretch angle_d_r wheel_diam_RR 0.5))

I solved the intrusiveness and the management of the grip on both fronts (front and rear) by first leveling the strech equally on the first block of code (increasing or reducing strech), and then I can define how much this modification impacts on the grip and others inserted in all the lines of the second block of code (the items multiply each other and therefore affect each other) ... so now by increasing or reducing the strech I also affect the grip and all the other items within the second block ... this has brought not only the possibility of management but also the homogeneity of the values obtaining less oscillations because there is no difference between the wheels and fronts.

The values to choose in 1° Block are as many as 0.025, 0.050, 0.075 or 0.25, 0.50 and 0.75 ... and so on ... and each one has a different and homogeneous result allowing the values coming from physics to do the rest ... or at least I hope so!

I created and inserted a new variable to manage it "strech_tyres_power_feel"

 

@Stakanov This latest file is extremely heavy for me in cornering. Loads up a lot. I have change a lot of the normal lines to counteract it but is still heavy. If I lower GAIN, it is just too light around the centre. I will try increasing power steering value to see if that helps.

 

Yes @Michael Enright is only to be adjusted using the alternative values according to your tastes and wheel or try using or insert the values of your favorite file

btw the main items for modify this heavy in corner are: power_steering, power_steering_angular, velocity_res_scale velocity_rot_scale, velocity_rot_limit_coef ...

and for general load we have even this items: tyre_resistence, limit_high_df, exaggerate_load_feel, static_force_reduction and the two variables exp_static_force_reduction ...

i insert this note in the post of the file ;-)

 

@Stakanov amazing job, I have a TMX and It's the first time i feel the same feel of rfactor2 in ams2. Very punch and dry feedback, without been too heavy, squishy or rubbery.

I'm using your previous file (rfactor 5.0.x.x), with values 72-85/20/60/15. And all cars at 100% gain (I prefer modify the global gain, it's better or worst than modify the % gain per car?).

I'm going to try your new file and read about which values in the file I have to change to fine tune. Amazing job all of you, dudes.
One question Stakanov, I read that you modify the file using reference from users using DD, but this file is made for low end tier wheels like g29/tmx, no? Conceptually which properties are you changing from the base file and why? Just short and simple explanation, i don´t have knowledge about simracing physics.

 

The DD users feel and express sensations things that the low end does not feel ... if the low end does not feel it I do not see why it should not implement it, so much I do not feel it ... then I add the tuning of the values that we both feel and share ... in the end there are only those few irreconcilable values between a DD and a Low/Med range, but it is only the intensity that is different ... obviously using a cheap Logitech G920 I chose the values for the Low/Med range but I indicated in the file what to change for DD and High (only 5 voices would seem particularly different).

About the file ... Karsten's original takes data from physics, when it serves all 4 wheels, but when it does not only serve 2, it implements some differences between the front and rear axles or between the front wheels and the posterior ones and simulates some posterior voices by taking the data from the previous ones ... finished everything develops its code by merging the results in the momentum rack and in the output ...

I have only extended the data collection always from all 4 wheels and not just from two, I have standardized the physics of the front and rear wheels only where this was different (grip and strech), and therefore not I simulate the rear with the data of the front but I take it directly from the rear because now calculated ... the rest does not change and everything goes into the rack and finally into the outpout.

 

@Michael Enright , @Roland Baltes and all with the problem of heavy wheel during curves ... i carried out some tests to reduce the lateral forces of the steering wheel when cornering at high speeds ... I found that, among others and after "velocity_res_scale", the item that most reduces the effect and the lateral force is the following with this increase ....

(vel_rot_limit_coef 1.40) ----> Original is 1.00

and with this above you should already feel the lightening clearly .... indeed you may need to fill your forces a little and only in this case let this item rise a little to rebalance (optional no mandatory)
(vel_rot_scale 1.20) ----> Original is 1.00

Only for correctness should be specified that this is a setting that is used in a more extreme way by @Danielkart with values of 1.50 for the first item and 1.25 for the second ... but I find it applicable to the low/med range without problems in size 1.4 and 1.2!!!

Let me know if it helps you, if you apply them both or only one and with what value ???
You can test this on any file not just mine, the rule is the same ... because, if I'm right, maybe we can also standardize this weight and speed characteristic between both DD/High and Med/Low wheels ...

Thanks in advance and good night to all!!

 

Hello, everyone!
Unfortunately I haven't been able to test Stakanov's new custom yet because I'm on vacation until tomorrow. Yes you can increase the (vel_rot_limit_coef ) and (vel_rot_scale) values. With (velocity_res_scale I would not go below 0.3, you lose a lot of information as a result. Tomorrow I will try to eliminate this hard and heavy without losing this fantastic feeling. Good evening

 

Thanks Stakanov. I will give it a try a bit later.

 

After initial tests with Stakanov's file, I can confirm the hardness and heaviness of the steering. But I already have a solution for dd wheels. It will be a combination with some changed values. If I can, I'll post these changes to you today. The commuting is still too extreme at dd, but that shouldn't be a problem either. And the 3 point is the very big bouncing of the cars, also on the curbs. But everything is easy to improve, I hope I can post you a suggestion for dd wheels today. Then Stakanov can customize it in his custom
Many greetings

 

Sure you can Daniel .... for now the curbs I have already reduced a little and also the two parameters for cornering hardness ... i still WIP for yours and Michael suggestions ...

(kerbs_surface_scale 7.8) -----> Original 7.5
(kerb_bump_scale 3.4) -----> Original 3.0

(vel_rot_scale 1.20) -----> Original 1.0
(vel_rot_limit_coef 1.40) -----> Original 1.0

 

Try these values and give stakanov feedback please. This is for dd wheels
(power_steering 0.35)
(power_steering_angular 1.5)
(tyre_resistance 2_lo_df 1.3 )
(tyre_resistance2_hi_df 1.2 )
(vel_rot_scale 1.2) this increased value is also good for the middle
(vel_rot_limit_coef 1.4)

These settings give about the power of my custom file without becoming extremely hard and heavy. I will also post the jumping of the cars and the commuting for dd wheels
regards

 

Ok i go to test ... but you mean in my file so you increase the power steering to 0.30 - 0.35 .... and the angular from 1.66*power_steering to 1.50*power_steering ... right???

(power_steering 0.35)
(power_steering_angular 1.5)
(tyre_resistance 2_lo_df 1.3 )
(tyre_resistance2_hi_df 1.2 )
(vel_rot_scale 1.2)‎
(vel_rot_limit_coef 1.4)

 

I still have small suggestions or insights for the game settings Gain, LFB, FX, Dampening.
Each simulation has an ideal gain value, try to keep to this, with ams2 this value is about 60-65%. You can do a simple test. Go on the track with a gain of 95% in the game, lfb 0, FX 0, damping 0 and on the base with a gain of 5%. Now see what happens to the ffb graph, it will take effect and you have the full ffb variety. Now do the same with gain in the game 5%, lfb 0, FX 0, damping 0 and at the base 95% gain. Almost nothing will happen with the ffb graph, you have almost zero ffb. That's why it's very important to choose the right gain in the game, if it's too little, you lose ffb feedback. If there is too much, it goes into clipping. That's why I always adjust the gain on the base depending on the wheel, not via the vehicle-specific gain, otherwise I'm out of the ideal gain in the game again.
LFB anything over 0 will cause sway on dd , sway is bad for vehicle control , even worse with shifter. For non-dd bases, lfb might be useful. FX I think the ideal value at dd is between 30 to 40%. Dampening, I always try with 0% additional dampening, because I don't want to suppress the ffb too much. Maybe that helps a little for dd users
regards

 

Yes Stakanov