Automobilista 2 Custom Force Feedback - Overview & Recommendations

It should not, no.
I did try to increase & decrease tear_scale and I feel that difference very clearly. I did not try further experiments.

Best,

-K

 

Thanks, I will do some more experimenting. Thought it was strange, but when I tested, there was definitely nothing from Tear when I put Scrub at 0. Mmmmm, interesting.

 

Will your file work with low-powered Logitech wheels? If so, is there anything else I need to do other than follow the instructions in the first post in this thread? Thank you.

 

For Logitech I would pick latest tune/variant from @Stakanov posted a few pages back.
Also he might post a new file one of the next days using parts from this latest file too.

 

Yes, I am already in the process of editing together with dear tester friends ... today I have little time but by tomorrow evening I should complete the update and tuning of the previous 4W file in version 5.0.1.4 final (I want to try it and adjust it with as many machines as possible and with as many steering wheels as possible other than Logitech) ... and I think that making the changes to the file (which are shared by several different steering wheels) focusing more on the main code than on the tuning of the variables, is bearing fruit.

@Karsten Hvidberg in addition to the "grip in front and rear wheels" I found another change in the code values that allows you to adjust the effect as you want so that everyone likes it by adjusting its intensity (I will explain it later) ... I'm in turmoil and I really like it ... I hope it's not just a sensation of my head!!!

 

I was scrolling thru the file and saw the following (don't know if this is intentional or a typo or leftover)
1. (sa_delta_rack (* (/ 1 (* dt 333)) (- steering_angle_rack old_sa_rack)))
shouldn't this be 300
2.
(grip_feel_rear (min 1 (/ (+ (* grip_rl rel_weight_FL) (* grip_rr rel_weight_FR)) (max 0.0001 (+ rel_weight_FL rel_weight_FR)))))
see values in red. if it is for rear should it be
(grip_feel_rear (min 1 (/ (+ (* grip_rl rel_weight_RL) (* grip_rr rel_weight_RR)) (max 0.0001 (+ rel_weight_RL rel_weight_RR)))))

 

Whoops

 

Sorry trigger fingers

 

Hey Lars,

Thanks for posting,
I can see why you ask for both #1 and #2, but everything is as was intended and no problems.

However might change it slightly at some point, so thanks for pointing them out.

 

I'm undecided about the center ..... I need to think aloud, and even though I think I know the answer, I think I need comfort and affection from @Karsten Hvidberg ... I specify that I don't want to deceive you and you are not my type!!!

Let's get to the point and assume the example of the file......
center_full_low_velocity 0.8
center_full 0.4
(center_full (blend (min 1 (/ kmh 80)) center_full_low_velocity center_full))
----------------------------------------------------
# 1 - If I go at 50 kmh we have:

(center_full (blend (min 1 (/ 50 80)) 0.8 0.4))

50/80=0.625 which is lower than 1 so "(min 1 (/ 50 80))" is equal to 0.625 and the center becomes.....

(center_full (blend 0.625 0.8 0.4))

So from 0 kmh to 80 kmh the range of the voice "(min 1 (/ x 80))" always goes from 0 to 1
----------------------------------------------------
# 2 - If I go at speeds above 80 for example at 150 kmh (but it is the same for any higher speed after 80) we have:

(center_full (blend (min 1 (/ 150 80)) 0.8 0.4))

150/80=1.875 which is higher than 1 so "(min 1 (/ 150 80))" is always 1.00 and the formula becomes.....

(center_full (blend 1.00 0.8 0.4))

So, beyond 80 km / h, it always remains like this at whatever speed
-----------------------------------------------------
So, which is not obvious at first sight, I would still obtain a differentiation between the center at a speed lower than 80 and the center at a speed higher than 80 even by setting the same value of both center_full and center_full_low_velocity

(center_full (blend (min 1 (/ kmh 80)) 0.8 0.8))

Because....
at 50 for example it would be (center_full (blend 0.625 0.8 0.8))
over 80 would always be (center_full (blend 1.00 0.8 0.8))
------------------------------------------------------
I got it right, no?

 

Almost.

Blend works like this:

(blend b v1 v2) = (1-b)*v1 + b*v2

So "v1" specifies what blend result should be at velocity 0 and "v2" specifies blend result at velocity >= 80. Everything in between velocity 0 and 80 is the pct as provided in "b" between "v1" and "v2".

In your example, since both v1 and v2 are 0.8, the result will be always 0.8.

 

Ok @Karsten Hvidberg ... thank you very much for the explanations ... I was right to think aloud then, I would have made a mistake just following the code ... the mathematical functioning of the "blend" is now very clear ...

 

Such a busy topic...it looks like you are non stop working to improve AMS2 FFB.
What are you or others recommending for a CSWv2 now? Ive got the latest from post 2, is that the file to use and if so what in game settings to start at? Does dampening work as dampening again?

 

Hey @2ndLastJedi,

The last file is 1 page back here in this thread. I would think you would need to use slightly higher gain than my suggestion for Torstein.
Damping is now damping again, yes.

 

Hey @Karsten Hvidberg
Can it be that more gain is needed after experimental 8 vs your latest file? Or are you still using the same settings for your simucube?

 

I increased master_rack slightly in the last file:

(master_rack 0.76)

So the earlier experimentals required slightly more gain in-game compared, yes.

 

I’ve found the last couple of experimental files rather odd feeling, like it’s lost lots of weight and strength of detail, I’m not sure what’s happening or if I’m getting a bug or something, but it’s just strange , feels very vague , soft and mushy.
I’ve not got time to investigate why this is happening, and are working on something else in between, and trying to test different features individually.

 

Your latest is a massive improvement from the older rfukter files i tried. From my little time today I think it gives the dampening to the Formula v12 that I've been missing.
More testing to come.

 

@Karsten Hvidberg
Have you tested your latest file with Karts.
for me they seem undrivable

 

Für alle, die eine ehrliche und hervorragend funktionierende benutzerdefinierte Datei suchen