Automobilista 2 Custom Force Feedback - Overview & Recommendations

oh my god it s fun

 

Thanks @Danielkart - Just tried 4 cars for 15 mins so only quick impressions. Note: I'm DD1 50/10/30/0 (base on 100%)

- Definitely onto something re: the rack feel
- The wheel goes super loose (soft) under skids... what feels like 90% to 10% so it feels like it's on ice when the back-end is going round (I think there's a relax value that's adjustable here).
- It seems there's much more 'relax' for the back wheels than the front, so I lose the feeling of front--end understeer vs. oversteer...
- Maybe a one-off thing, but in a quali session it was fine, then very different in the race that followed, like 50% softer.... with the CPU spiking consistently and periodically, almost like there was a bug causing a 100% spike... whether this is FFB calcs or not I'm not sure.

I'll let the experts pick up from here and develop - hopefully that makes some sense!

Thanks again for your hard work.
Rob

 

Here is an experimental file inspired by parts from all the latest tunes posted here.

* Allows very high force at high gain, as it seems from @Stakanov 's last file that might help help g-series wheel bases.
* Experimental bump feel.
* Experimental suspension feel.

"rFuktor 5.0.1.4-experimental".

 

I don’t know why you think this is some kind of competition, it’s just a collection of different techniques and ideas.
Stakanov has always been the best option for gear wheel and good for other low power wheels.

This has never been a debate.
and as i have said , i’ve been in time limited testing to see how this might better suit more powerful DD wheels, as currently i don’t believe it does.

So far i’ve converted the file to have power and steering forces similar to other files .

So far i have got to this stage , but it’s very much a work in progress, but it’s getting there.

there’s still fine tuning needed.
But it may help showcase the extra layers of this file in a way that easier to digest for DD users used to other files.

 

@Karsten Hvidberg , yes the force are more manageable but the wheel oscillating in straight (left right in the center) with high speed and with auto Formula (try gen v10 gen2 kansai or spa).

With this new lines in red oscillating
(gain2 (* 2 (power (/ gain 1.2) 1.4)))
(df_compression (power (max 0 (* (/ 1 0.6) (- (min 1.0 gain2) 0.4))) 1.5))

With old lines the oscillation go away and at high speed rest only a little moviment from bump
(gain2 (min 1 gain))
(df_compression (power (max 0 (* (/ 1 0.6) (- gain2 0.4))) 1.5))

for the rest is all ok with note below

EDIT:
1) Ok for no large bump, no problem and no difference, on the contrary, the road surface is more consistent and coherent with only the microbumps and the changes in level and texture are more evident and clear

2) Ok for micro bump but is usfull rehabilitate these lines and the two variables, because, even if they slightly, make the effect of the jerks less pointed and round:
(left_details (smooth left_details (blend (smooth FL_bump_kerb 0.01) micro_bumps_smooth kerbs_surface_smooth)))
(right_details (smooth right_details (blend (smooth FR_bump_kerb 0.01) micro_bumps_smooth kerbs_surface_smooth)))

Spoiler: Complete Code for Microbump

(left_details (* 0.002 (- FL_load (smooth FL_load 0.006))))
(right_details (* 0.002 (- FR_load (smooth FR_load 0.006))))
(left_details_pos left_details_pos)
(right_details_pos right_details_pos)
(left_details_pos (+ left_details_pos (* (- left_details left_details_pos) (power 0.001 dt))))
(right_details_pos (+ right_details_pos (* (- right_details right_details_pos) (power 0.001 dt))))
(left_details_pos (max left_details_pos left_details))
(right_details_pos (max right_details_pos right_details))
(left_details (* (power (- 1 (min 1 (abs left_details))) supress_micro_bumps) left_details))
(right_details (* (power (- 1 (min 1 (abs right_details))) supress_micro_bumps) right_details))
(left_details (* left_details 0.6 (if FL_bump_offroad 0 1) (blend (smooth FL_bump_kerb 0.01) micro_bumps_scale kerbs_surface_scale)))
(right_details (* right_details 0.6 (if FR_bump_offroad 0 1) (blend (smooth FR_bump_kerb 0.01) micro_bumps_scale kerbs_surface_scale)))
(left_details (smooth left_details (blend (smooth FL_bump_kerb 0.01) micro_bumps_smooth kerbs_surface_smooth)))
(right_details (smooth right_details (blend (smooth FR_bump_kerb 0.01) micro_bumps_smooth kerbs_surface_smooth)))
(kerbs_details (- left_details right_details))
(road_details (blend (smooth (min 1 (+ FL_bump_kerb FR_bump_kerb)) 0.02) (+ left_details right_details) kerbs_details))
(road_details (spring road_details 5000 0.001 1))

3) I like the new suspension out of box

Thanks

 

Just had a quick look at file and have a question. Are these lines correct;
(RL_start_load_rear (/ corner_FL 9.8))
(RR_start_load_rear (/ corner_FR 9.8))
or should they be;
(RL_start_load_rear (/ corner_RL 9.8))
(RR_start_load_rear (/ corner_RR 9.8))

 

Now it works again, the way is right. There are now completely new possibilites. Will now watch and follow Stakanov development. Most of the other files are copies with few ideas. Thanks for you support Stakanov

 

Thanks for the feedback, Stakanov,

I have adjusted accordingly, but in slightly different ways, as you will see in the file.

* High gain back down to normal, as it seems g-series did not benefit.
* Experimental rear feel / lean feel.
* Micro bumps adjusted lower and smoother.

"ffb_custom_settings - 5.0.1.4-experimental2"

 

Now 11:30 i am at work ... i will go back home at 18:30

 

I'd like to post all these new files in the 1st post of the thread once they settle down and people feel them somewhat "final". It's great with this variety.

 

Felt experiment 2 was too "pointy" on rear downforce, so here is next iteration.

* Smoother feel of rear end.
* Value "rear_end_feel" added - might be slightly too high at current 0.6, so try 0.5 or even lower if that is the case.

"ffb_custom_settings - 5.0.1.4-experimental3"

 

* Improved suspension feel(not more, but hopefully slightly better, adjust up using "suspension_scale" if you want more).
* Slightly more delta load feel.

"ffb_custom_settings - 5.0.1.4-experimental4"

btw @Stakanov @Danielkart @Kuku Maddog In some of your tunes there is a constant rumble while driving, which effect in the file is changed to give that? (Sorry, out of time to look myself)

 

Just out of curiosity or to be sure that I understand, how work these two new lines....

(FL_load_norm (* FL_load_norm (+ 1 (* grip_rl rear_end_feel (- (min 2 (/ (smooth RL_load 0.2) (max 0 (smooth RL_load 2.0)))) 1)))))
(FR_load_norm (* FR_load_norm (+ 1 (* grip_rr rear_end_feel (- (min 2 (/ (smooth RR_load 0.2) (max 0 (smooth RR_load 2.0)))) 1)))))

Premises this sum
1 +
the negative result on the R_load, ((the min of division of two R_loads smooth or 2) -1) +
the positive value of the half of grip if rear_end is 0.5

Result
If this result of the PREMISES is > 1 then increases the front load norm with multiplication
If this result on the PREMISES is < 1 then decreases the front load norm with multiplication

consequently if increasing the front simulates a lean feel of the grip rear and vice versa

The logic is this???

 

I think what you are saying is correct or almost correct(not sure about the last line reg. front/rear/vice versa).

You could say that:

short_term_rl_load_delta_pct = (+ 1 (* grip_rl rear_end_feel (- (min 2 (/ (smooth RL_load 0.2) (max 0 (smooth RL_load 2.0)))) 1)))

and then:

FL_load_norm = FL_load_norm * short_term_rl_load_delta_pct

That might be more explanatory than what I put in the file.

The idea is that rear tyre increased/decreased short term load should be felt on the steering on that same side, and this is done by multiplying that relative difference on to the front load feeling.


Btw, there is a slight error: "(max 0" should have been "(max 1" in both lines, as it can produce an error right now, not sure it matters.


I posted these experimental files in brief succession as each little change might be something that some of you guys like or hate As small pieces they are easy to transfer over.

 

@Karsten Hvidberg Yes I had guessed well but I mean very badly sorry ... after all if I speak normal English badly by translating everything with google translator imagine how can I be clear to speak English by explaining the mathematical function ... I really suck ...
But I do not give up, love gives me courage and I always study, after having learned the language and logic of your files now I am moving on to the translation of mathematical formulas...


If my employer knew that I study FFB with Prof. Karsten during working hours would send me to Mars with a gravitational acceleration much higher than that of Earth of 9,8 m/s2

 

Ahahahahah ... @easy81 sei peggio del mio amico che ogni volta che sono a casa in malattia mi citofona la mattina al campanello di casa cambiandosi la voce e mi dice "Visita fiscale di controllo apra!!!"

Ahahahahah ... @ easy81 you are worse than my friend who every time I am at home in sickness calls me at the doorbell in the morning changing his voice and says to me "Tax medical control visit open !!!"

 

ATTENTION GUYS!!!!!

Karsten surely by typing mistake left two lines of code active for the same function in the last files 5.0.1.4 Experiment 2 - 3 and 4, delete the second in red and bold (or insert # symbol for deactive) and leave the first in green active!!

(gain2 (min 1 gain))
(gain2 (* 2 (power (/ gain 1.2) 1.4))) <-------- Delete or insert in front # symbol

 

Hi

Sorry i can’t keep up with this stuff at the moment, but quick test of last exp 1/2 and 4 felt pretty good once i put in my preferred scrub and tear etc, and also slightly different yaw signal based on rel weight.

perhaps the different way we like to have scrub work contributes to the road vibes /. rumble.

 

I ll write something about it later

 

Stakanov, what do you think center_full....., reverse the values?
Grip_tyres_smoth, is not good for my file. Even with Stakanov I find it better without. Stakanov you can also post your file without 4x4 grip? To test? There are very few 4x4 cars, and now the ftont axle is connected to the steering, just an idea Karsten your file is very hard and difficult to drive with dd. Too heavy for my taste maybe others feel better? This is not a criticism , just my opinion. All these new files are world class, and world better than before
Regards and good night