Automobilista 2 Custom Force Feedback - Overview & Recommendations

FFB widget shows the effects as you drive.

 

I have tried them all. There is no best file, just a file that will be best for you.

Once you find one you like, stick with it, you can always try others at a later time.
Thats the only shortcut.

You really need to spend a good amount of time with each file tweaking the in game settings to get the best from them.

My personal favorites, are Joaquim Pereira Fullffb v12 follow the recommended settings and you are good to go ( you may want to tweak) , and Kuku Maddog latest file, which takes a bit more in game fidderling to get to my likings.

Stakanov file seems promising and I am excited to see where it goes.

But you may find a completely different file to your liking.

 

Absolutely, the best file is the file you like and that suits you most. He doesn't actually need to spend a lot of time trying all the files. Just try each creator's latest file, drive at least 3 different type of cars, say open wheel, GT and some slower classes, see if you like it and proceed to the next file. After a few days one file will stand out, that'll be the file you want to use.

I honestly thought all of the files are based on the rFuktor 5.0.1, but yeah NuScorpii said on page 1 that it was a new custom file. I could correct the info in the file collection thread, but I don't know if it is that relevant.

 

I see in a lot of files it is recommended to have in game Gain between 50-65 but there is a line in the FFB files that says...
''(df_compression (power (max 0 (* (/ 1 0.6) (- gain2 0.4))) 1.5)) #calculated range 0-1. 0 is result for gain 0-40%. This will keep DD bases at compression of 0, when using high output on the hardware."

Is the compressed signal a big downgrade?
Should we be setting up these files to have gain set to 39 / 40 and no higher?

If i turn my wheel base all the way up and set the in game gain slider to 40 or under, it feels to weak on the custom files

 

I think there is a misunderstanding and I think it has been misinterpreted in the past, at least I hope I'm not wrong with my bad english .....

the original gain in the rfuctor is gain2 and is calculated as a derivative of the original gain slider command....
(gain2 (min 1 gain))
it is therefore identical and linear from 1 to 100, then gain2=gain, 1=1, 50=50 and 100=100 or in unit from 0 to 1, 0=0, 0,5=0,5 and 1=1, where 1 is the max.
If you follow in the code the Gain2, it is applies only and directly to the end of the file in the rack_big and will therefore determine the output in the same way and in a linear way from 0 to 1 (taking into account the master rack that you have set in the file) and here is the line code....
(rack_big (*gain2 rack_big master_rack))


--------------------------------------------------
Instead the downforce_compression thus determined it is a compression that is not applied to define the gain, this....
(df_compression (power (max 0 (* (/ 1 0.6) (- gain2 0.4))) 1.5))
is applied to define how much the gain (gain2 is equal) must influence the determination of the limit of the high_downforce, this formula is always 0 between 0-40 and is exponential between 0 and 1 when the gain2 is between 40 and 100. ..
So when you blend the df_compression between 1 and 100 to determine the limit_df_scale1 the exact opposite (of what you claimed) happens and therefore the DDs are helped when high gains are used to determine the value of the limit1 itself ... and it can go down but not beyond limit2, the subsequent formula in fact chooses the greater of the two limit (1 and 2).....

#limit high and low df
(limit_high_df_scale1 (blend df_compression 1 100))
(limit_high_df_scale2 5)
(limit_high_df (* limit_high_df (max limit_high_df_scale1 limit_high_df_scale2)))

Surely Karsten could explain it better than me, I hope my bad English is understandable...

-----------------------------------

EDIT...

#Blend ...
#(blend <bias> <input for one> <input for zero>)


Then Blend formula works like this:
(blend b v1 v2) = (1-b)*v1 + b*v2
where b is df_compression, v1=1 and v2=100

 

Thank you for explaining that. Yeah i totally misinterpreted what that line meant, so thanks for clearing it up.

 

Thanks again for the advice, however I must underline that I found the differentiation made by @EcoBrum very useful, because in this way I was able to select 4 files of different origins and I am making initial comparisons only between these, I easily understood that many files are based on the same version personally tuned by the person who tuned it and which I will only look at later to evaluate the differences.
I will continue to read you trying to understand as much as possible and if I have an answer, a doubt or feedback I will ask you for it.
Thanks again everyone.

 

Thanks Stakanov - latest file works gr8 with old G27 wheel!

 

Thanks for your feedback. I'll sort the custom files by the file they're based on then.

 

Thank you very much for your work, I get along very well with the Moza R16V2, it has taken the entire AMS experience to a new level!

 

I am using your file on a Moza R16 V2 and I am very, very happy. Thank you!!

 

Has anyone tried both Moza V1 and V2 bases? Any difference in force feedback?

 

I don't think there are big differences between V1 and V2. The software has been expanded to all Moza bases. Of course, a lot of it is marketing to sell the new bases. The engines are still the same. Other factors have more influence on the differences. The FFB software from the manufacturer must certainly be taken into account. How well is the signal processing of a base processed and output with the built-in chips. Encoder resolution is also a factor. This means Moza R5/R9/R12 works with 32768 (15 bit) pulses per revolution. Moza R16/R21 operates at 262144 (18 bits) per revolution. For comparison, a Simucube 2 Pro with Mige engine works with 4000000 (22bit). Do you need that? I think yes and no. You won't feel any difference from around (18Bit). But it can help if you use filters on the base because then there is more to calculate and your base can process it better without losing or overlapping other feedback. Everything will also become smoother, more even and more precise. In the end, as with the FFB, personal taste decides

 

I think it also depends on the base, R21 was from my understanding, slightly different from the rest on v1 already (according to SimRacingGarage review, it was smoother and didn't have some negative traits Barry experienced with v1 R9 and R16). So I wouldn't expect my R21 v1 to be that different to my v2 what comes to FFB, the differences are more in stuff like front mounting possiblity and such.

ps. R12 was using similar tech than R21 from the start I think

 

Yes, as I said, I don't think there is a difference between V1 and V2. But there are differences between the bases. I had the opportunity to test an R9 V1 and noticed clear negative differences compared to my R21 (both tested with 9Nm). But of course there is also a different price that you have to pay for it

 

Thanks for the info
I've been trying to find this sort of information about the logitech direct drive wheel but for some reason its not well documented or looked into. I have had this wheel for a while and really like it but the lack of information about it is kind of annoying. Plus the silence about any addons still so far after release is making me reconsider my purchase.

Meanwhile, Moza is spitting out new products left and right

 

Yes, there are manufacturers who unfortunately don't provide any further information, which I think is a shame, or do they want to hide something? It wasn't clear with my previous Fanatec dd1 either. The encoder resolution was never confirmed by the manufacturer and was only assumed to be native 12 (bits) which was scaled up to (15 bits).
I can recommend a Moza if you like a lot of feedback, it's more aggressive and louder out of the box than a Simucube 2. But I wouldn't go below R12 for a Moza. I consciously chose an R21 alongside my Simucube2 Pro. I like both, the softer and the louder. Of course, a Simucube2 can be adjusted even more precisely and the Mige motor is the gold standard for me. But you won't go wrong with a Moza base

 

I have a Moza R9 V1 and I‘m very happy with it. It works great with ACC, Rfactor2 & LMU (for me the game got strange FFB updates, but thats only a personal feeling) & Raceroom (my favorite FFB).
It also works well in AMS2, but I get some oscillation with it if I try to use a (very) high FFB level. It‘s much better with Daniels custom FFB file!
I don‘t know if it‘s a problem with the Base itself or maybe a result or my relative lousy mounting on a wheel stand (don‘t have room for a cockpit).

 

Hi, I tried to make the change but I feel all the bumps stronger.
My problem is: at low speeds the HIGH curbs can be felt on the steering wheel, but if you go fast you can't feel them (for exaple if you cut corners too much 1 Variante or Variante Ascari in Monza).
I have Fanatec CSL DD 8nm.
Help me?
Thank you

 

I recommended these changes for weak bases. If you drive fast you don't have enough feedback. You mention an important factor here. You don't have too little feedback, but it is overshadowed by the forces that are too high and the fine details are lost. Ams2 generally generates high forces and sometimes they are too high. There are bases that can compensate for this wonderfully and bring out even more details. Unfortunately, the old Fanatec dd bases with the FFB software are not included. But they can also do a lot of good things with the right in-game and base settings. Test different settings to find your sweet spot