Automobilista 2 Custom Force Feedback - Overview & Recommendations

JUST STOP GUYS !!!!!

Move to PM please. Nobody is interested in this conversation.
Both your egos are way too big

 

Deleted

 

.

 

.

 

Deleted

 

Update:

* Simplified & improved load calculation.
* Added missing credit where due.


Adjust "turning_momentum" (set at 1.75 currently):
- Down, if you feel the tightening under speed is too high. 1.5 seems really good too.
- Up, if you just can't get enough of that.

Also friction might be worth changing depending on your wheel base.
If you are on a DD or other high-end wheel, you might already have friction set, so then adjust "friction" lower.

Attached custom ffb file "Silver Bullet v5":
(I've added same correct and super simple approach to the car balance - feels awesome, might post later with another fix too)

 

Just a heads up guys, not sure what's changed, but these latest files aren't working well with the latest update. I swapped and changed so many times last night i can't remember which file did what, but i found one of them caused numbness on most corners which needed lfb to be dropped way down introducing looseness.
Got it somewhere near after a lot of trial and error in the file but still not quite right. I'll try again later and update now I've had some sleep.

 

I put a late update out last night that reset an experimental variable I had been playing with. Made it too heavy in the corners. Maybe that is what the issue was for you with mine? You can find the updated file in the usual place.

 

.

 

Yes this . You are obviously both very intelligent guys . . . but right now you are both doing yourselves a disservice and frankly making yourselves look ridiculous .

 

Group hug?

 

I think that was one i tried @NuScorpii , about to jump on again and will update as i go.

 

First test
Heavy messing v2.1
Changed a few things, but still heavy numb cornering.
Not much road feel.
Onto next one.

 

It works really well with the new build from Reiza that I'm sure you guys will be testing very soon too, @NuScorpii silly question, but do you see any pattern on Reizas FFB development? Cos its always some minor adjustments here and there and a file works well, or it seems!? Is there anything like damping, tyre scrub ongoing improvement? Thanks

 

Really? I know I'm probably stating the obvious here but have you tried turning FEI back to 100?
It feels really responsive to me at 60.35.55 (the file he linked earlier today with the 9.7version)

 

2.1 or 2.1a? There is a bug in 2.1 fixed in 2.1a.

However I went back to 2.0 and there is definitely a slight difference for the worse that I will take a look at later where some cars feel too light and seem to skip a bit over the track.

 

It does seem like there are continued small developments that require minor tweaking of the custom files. But there doesn't seem much of a goal or move to solve some of the issues Karsten and myself have highlighted.

 

Hey guys @NuScorpii and @Karsten Hvidberg could you please help or give a hint?
I try to reduce the massive mid corner clipping which was introduced in the last versions of the Reiza FFB. Is there a way?
The problem is that more or less all cars with a more heavy SAT (high downforce) seems to be clipping in high SAT situation and so the details even harsh curbs are all smoothed out. This means a very dull feeling in the corners although the FFB is only at 50% but the graph goes very calm and the expected spikes form a curb or from some scrubbing are completely lost.
I included already the following part in the new 0.9.7 custom file, but it seems only helping to a small extend, so I'm not sure if that really complete or to much.

Maybe I should try completely without the clipping part. Maybe it could also be already smoothed out in the internal Reiza algorithms before we get the FFB signal.

Code:

###########################################################
#
# Clipping
#

#allow 30% headroom

(headroom ( - 0.7 gain))
(clipTo ( + 1.3 headroom))
(clipTo ( if (max 0.0 headroom) 1.3 clipTo))

# Sig clip output

(sg_p 0.6)
(sg_s (+ (* 2.0 (- 1.0 sg_p)) (* 2.0 (- clipTo 1.0))))

(sg_f (+ (/ sg_s (+ 1.0 (power 2.718 (* -1.0 (* (/ 4 sg_s) (- output sg_p)))))) (- sg_p (/ sg_s 2.0)) ))
(output (if (max 0.0 (- output sg_p)) sg_f output))
(sg_f (+ (/ sg_s (+ 1.0 (power 2.718 (* -1.0 (* (/ 4 sg_s) (- (* -1.0 output) sg_p)))))) (- sg_p (/ sg_s 2.0)) ))
(output (if (min 0.0 (+ output sg_p)) (* -1.0 sg_f) (output)))

(hist_output (* output (/ 1 clipTo)))
(output (* output gain))

 

Does this happen in latest "Silver Bullet"? I would think it works fine in that since it blends the forces depending on loads as best it can, so the result should not add up like that, I would think.

I don't have access to the latest if it's a new beta release.


Cheers

 

Yes i was using the old v2.1
I tried the v2.1a and it's loads better, however i still prefer the old v2.0, get my fastest times with that one.