Guide for G27 (and other wheels) configuration

[RasmusP]

February 2024 Update:
Please don't read this thread. I know Google will show it, but it's old and filled with partly wrong information from me.
Just use my LUT package instead:
http://www.racedepartment.com/downloads/rasmusps-lut-guide-for-g27-29-and-dfgt.16799/



-------------- Don't read this! --------------

Older edit: After almost everyone prefers the LUT-Configuration, I created a Lut-only Thread with all my LUTs attached in a zip-file in the Download-Area here at Racedepartment.
Have a look:
http://www.racedepartment.com/downloads/rasmusps-lut-guide-for-g27-29-and-dfgt.16799/

Original post:
Hello everyone,
I know there are a lot of settings and guides online for "the perfect settings" but after I spent about 30 hours searching and trying, I really found MY best settings.
They are a combination of different guides (but with the "GAMMA-option" in use) that I will name right now and first:

Really nice and complete Guide on Assetto Corsa Forum
Reddit Thread where my adventure began (and where I got the Link above)
FFB Deadzone Fix Logitech G27
Guide for LUT Generator (can't recommend it but helped me understanding)

EDIT: I thought the LUT was what gave me my perfect feeling BUT IT WAS THE TYPE=GAMMA, not a LUT so I rewrote that whole Guide:

EDIT2: I found out how to write a LUT yourself so I did it and the result is something similar to the gamma-setting BUT less compressed, smoother middle-transition and therefore BETTER! Here is the link to the post if you don't want to scroll down: LINK

TL: DR Lines:
1. Adjust InGame-settings and Profiler

2. Open the controls.ini:
you find this ini in: "C:\Users\%username%\Documents\Assetto Corsa\cfg\controls.ini"
set MIN_FF=0.04, CENTER_BOOST_GAIN=1.3, CENTER_BOOST_RANGE=0.001

3. open the assetto_corsa.ini:
you find this ini in: "%...%\SteamLibrary\steamapps\common\assettocorsa\system\cfg\assetto_corsa.ini"
set ENABLE_GYRO=1, DAMPER_MIN_LEVEL=0.0, DAMPER_GAIN=0.7

4. open the ff_post_process.ini
you find this ini in: "C:\Users\%username%\Documents\Assetto Corsa\cfg\ff_post_process.ini"
set VERSION=1, TYPE=GAMMA, ENABLED=1, VALUE=0.5, CURVE=%doesn't matter%

5. HAVE FUN

6. additional:
- Logitech Profiler:
- overall: 100
- spring: 0
- damper: 0
- untick center spring, NO combined pedals, 900° rotation, TICK "allow game to adjust settings"

- AC InGame:
- Gain: 50% [depends on your wheel but for my G27 it is 50%. 75-125% InCAR]
- Filter: 0
- Min Force: 4%
- kerb, road, slip: 0
- understeer: untick
- Steering Settings: gamma 1, filter 0, sensivity 0!!

For those who want a detailed guide:

After I read very very much about all seen and all hidden settings in Assetto Corsa, I tried everything out, but one of these things was always bothering me:

1. you have a huge dead zone
2. your wheel oszillates
3. deadzone ok but no smooth transition at center
4. just too much force

The solution for me was the GAMMA-setting in the ff_post_process.ini!
Combined with the dampening-setting, the Gyro-setting, the min_force and the Center_Boost!
Short version of what this settings do:
To get rid of the Deadzone you need to combine the very little center-boost with the min-force. But then you will have a little bump in the middle because they have a "hard cut" at zero-force.
If you enable the GAMMA-setting this hard-cut will be transformed into a curve, just like the FFB-Output from Assetto Corsa will get compressed to a curve where little forces are boosted and the high forces are lowered.
Here are a few pictures of that:

1. Wheelcheck Result:
2. How the AC-FFB has to be alternated:

3. All these settings combined:

So you get rid of the deadzone but don't get an oscillating wheel! [Plus you feel more details.]

The Settings in detail:
1. controls.ini:
you find this ini in: "C:\Users\%username%\Documents\Assetto Corsa\cfg\controls.ini"

- Filter=0: This should eliminate FFB-spikes but for me it seems like it just kills little details but does not really smoothes out! So let it on 0!

- MIN_FF=0: That means, that the "zero force" is still zero force, theoretically it's THE setting to get rid of the deadzone BUT I experienced, that you need to get up to 17% for the G27 and then don't have a deadzone anymore but case 2.: your wheel oscillates.
If you go down to about 12% you have case 3.: the deadzone is still there, but little. Problem: it doesn't feel nice... 0 Force in the Deadzone and then BAM, FORCE.
My thought about that:
The "MIN FORCE" doesn't have a "Curve" for the lowest force. It just does a hard cut which ends up in my experienced results. (See Picture above: 3. combined)

- CENTER_BOOST_GAIN: In the Guide "FFB Deadzone Fix for Logitech G27 1.2" this setting is used.
Explanation: If you imagine the Wheel, the FFB begins at a little bit right and left from the center-> "deadzone". That is because the force that IS THERE, IN THE CENTER, is too low for your wheel.
This setting BOOSTS the FFB in the RANGE of "CENTER_BOOST_RANGE"

- CENTER_BOOST_RANGE: as mentioned, this is the range of the boost that applies on the FFB FROM THE CENTER. Theoretically you just need to "Boost the deadzone away". But this does not work, I tried EVERYTHING!! (combined with GAMMA this transforms to a "curve"!)

I combined this 3 settings in every possibility but you have always ONE problem: the "center transition" as in "you steer from left to right" gets a BUMP in the middle, because of the boost.
I tried to lower the boost, widen the range, give a little more MIN_FORCE. But nope, you always gets this unrealistic "center-bump". (so you need the GAMMA!)

I have to give the author much credit, it is a really nice setting but it's just not enough for me.

2. assetto_corsa.ini
you find this ini in: "%...%\SteamLibrary\steamapps\common\assettocorsa\system\cfg\assetto_corsa.ini"

- FF_SKIP_STEPS=0: If set to 1, you get half of the FFB-Refresh-Rate. So if you have a decent hardware, go with 0! Some people get an error on servers because their CPU is too weak. Then go with 1. It doesn't make a HUGE difference, but the more FF-Inputs, the better, right?

- ENABLE_GYRO=1: Honestly, I don't understand this settings but it smoothens the wheel without taking away important informations!
If you read about that setting: It is for Direct Drive Wheels as the Logitech G27 (not belt drive like Thrustmaster for example) and interprets "the wheel as a Gyro on certain speeds".
If your wheel is a little bit "shaky" overall or oscillates: go with 1!

- DAMPER_MIN_LEVEL=0.0: This means, that the InGame dampening (a few versions ago there was a slider?!) is OFF, when the Engine says the damper to be zero. If you increase this settings, your dampening will ALWAYS BE ON, even if the engine does NOT give the input for it.

- DAMPER_GAIN=1.0: That is simple. If the engine gives the input to dampen the FFB-Signal, it is indeed dampened. For my G27 I like it on 0.7 but if your wheel feels a little "numb" you can lower or even deactivate it. I find my wheel too shaky when on 0.0!
Sidenote: ALWAYS USE THIS SETTINGS FOR DAMPENING, NOT THE LOGITECH PROFILER DAMPENING!!!


3. ff_post_process.ini
you find this ini in: "C:\Users\%username%\Documents\Assetto Corsa\cfg\ff_post_process.ini"

- Version=1: No idea, just leave it!

- TYPE=GAMMA: I don't know how it works exactly but it is THE setting that got me stunning! I drawed a Curve about how it feels for me and it makes sense but honestly: no idea...
Here is the picture:

- VALUE=0.5: Well in one guide the author tells you "it is on 1 at mine, no idea, just leave it" BUT IT IS IMPORTANT!
It determines how strong the FFB gets "compressed". So on 1.0 it does NOTHING, on 0.1 you feel absolutely no details, it's like ultra-power-steering.
For me about 0.5 is the optimum for having details but no deadzone.
THIS IS THE SETTING THAT ELIMINATES MY DEADZONE AND LET IT FEEL REAL!
What is important here: you have to lower the global FFB in order to have "normal forces". The lower the value, the lower your global FFB! If you put VALUE=0.1 and global FFB on 100% you get broken arms for sure...

- CURVE=%doesn't matter, not active%": Well this is just the settings to take the correct LUT. BUT: I don't use a LUT and I can't recommend it with the G27. Tried it and it was just awful!
IF you want to activate a LUT you have to change TYPE=LUT (thanks to Ross Garland for giving me that info!)


Conclusion:
I know that was a lot and I know it's not perfect but maybe it will help you with your settings and increase the fun you have with Assetto Corsa.
I switched over from Project Cars but was so mad with my wheel that I thought about buying another one. The Deadzone was so annoying but the graphics, sound and gameplay were better and I really wanted that game to get working!

This is just a wrap up from other people's posts and guides but I didn't find a guide here, where you don't need to login to the Assetto Corsa Forum or even there, where all these settings are packed up so you can just read it down and get your settings.
And most importantly: I nowhere found any mention of this GAMMA-setting in the ff_post_process.ini so alone for that setting I wrote that all down.

I made this post to give all searching souls another thing to try and hope this makes even a few people happy

Let me know if it helped you!

Kind Regards,
Rasmus

 

[Larry Thomas]

@mymisslaracroft I have to agree with RasmusP, do not use LUTGenerateur. I used it before with my G27 and I felt it was alright but....
Then I got a Fanatec CSL Elite and tried it with LUT Generateur but for me it was simply wrong.
I looked around and found RasmusP's solution which I tuned to my wheel. The result = Fantastic and RasmusP's solution can be tuned to every wheel. Give it a go. If you have a problem RasmusP is extremely helpful. You won't regret it.

 

[pwrsky]

Thank You!!!
use on g27.

 

[mentalham]

first of all ,thank you for your G27 suggestion.
it is very wonderful
can you give me some suggestion about G27 for rFactor2?

have a great day

 

[Peter Stefani]

G920 works well with the No_Clip_But_Stronger, and I had to slightly modify the DZR curve to the way I like it. Geat work RasmusP !

 

[Peter Stefani]

Hey RasmusP, are you actually using the FFBClip app ? I started using it just to get a more detailed view of what was actually happening, other than the one small bar in the "Wheels" app widget. I'm in the 911 RSR at Spa, and using the Eau Rouge uphill curve to gage the max forces. The chart was showing the small amount of clipping happening during that phase of the corner, with the No_Clip_But_Stronger setup, 50% Menu, 100% in-game. So, then I thought to give the FFBCLIP app a try to see what effect it would produce.
I set it for Dynamic On at 110%, and Dynamic Intensity (ramp up) at 300. It did a nice job of limiting that clipping at Eau Rouge, and seemed to keep the overall FFB at around 108 ~ 110. I kind of like it, and it somehow seemed to add more front load Mz effect while braking.
Do you have much experience with the app? If so, what settings do you use typically.

 

[RasmusP]

Hey RasmusP, are you actually using the FFBClip app ? I started using it just to get a more detailed view of what was actually happening, other than the one small bar in the "Wheels" app widget. I'm in the 911 RSR at Spa, and using the Eau Rouge uphill curve to gage the max forces. The chart was showing the small amount of clipping happening during that phase of the corner, with the No_Clip_But_Stronger setup, 50% Menu, 100% in-game. So, then I thought to give the FFBCLIP app a try to see what effect it would produce.
I set it for Dynamic On at 110%, and Dynamic Intensity (ramp up) at 300. It did a nice job of limiting that clipping at Eau Rouge, and seemed to keep the overall FFB at around 108 ~ 110. I kind of like it, and it somehow seemed to add more front load Mz effect while braking.
Do you have much experience with the app? If so, what settings do you use typically.

I don't use it honestly. I did use it to get the graph but I didn't really see a reason to keep it. Clipping with 50% is just not making any sense. Sure the graph might spike into the limit but maybe it's wanted, so you don't break your hands with a dd wheel. I guessed around quite a lot but decided that even if there might be clipping with my config, I wouldn't care since it feels great.
The dynamic mode however does two things:
1. You get a really nice ffb all the time
2. You lose the muscle memory effect

The problem with a constantly changing ffb level is that your muscles won't learn to "read the car". My settings where created while I was trying to be competitive and therefore I would never use the dynamic mode. If you just want to have a good time, then use it, it's great

 

[Peter Stefani]

Yeah, I did notice that I wasn't feeling so much a loss of traction while using it. But it does feel nice on some of the weaker cars. Thanks for you opinions.

 

[kevbaz]

Has anyone tested this setup for drifting? And if so would ye recommened?

 

[RasmusP]

Has anyone tested this setup for drifting? And if so would ye recommened?

Go to my downloads and just use the "recommended" LUT. I'm drifting every now and then and wouldn't see why it should be less great for it.
My aim was to make the wheel as responsive and quick as possible while maintaining details as much as possible.
One thing I could imagine for drifting though is that you might not need the details for "driving on the grip limit" but instead just maximum rotation speed for chances of direction.
That way you lose some feedback from the car but the main problem for drifting is that the logitech wheels don't spin quickly enough.

The "gamma like LUT" goes in that direction. Killing details by reaching high ffb levels a lot earlier.

What do you use currently?

 

[John Peake]

Very new to AC with a seconhand G27. I was trying to figure out why everything felt so wrong compared to real life. Your guide has litteraly saved mw hours of work.
Can't thank you enough, cheers

 

[RasmusP]

@Kek700 I know my post wasn't really "correct" since I couldn't find any real explanation etc and figured everything out on my own back then.
But giving me a "disagree" after 2 years is a bit rough don't you think?
Or did you miss the right button?

If not, would you like me to edit my posts? I tried to make it clear in the opening post and in the posts afterwards that I was wrong and nobody should do whatever is in the first few posts

 

[Kek700]

I have been carefully been going from the start of this article that you have most painstaking prepared, looking with interest, with the view of applying it to my Accuforce,
Decided to revisit my FFB for this wheel, seeing if there any improvements to be had.
Reading every entry from the start. It was not my lucky day, if i had accidentally pressed any of the 5 comment button i wold have not offended you, i must have hit the disagree button number 6 button by accident when scrolling though.
I must be more careful, i am now wondering how many times i have done this. ( old wobbly fingers )
sorry

 

[RasmusP]

I have been carefully been going from the start of this article that you have most painstaking prepared, looking with interest, with the view of applying it to my Accuforce,
Decided to revisit my FFB for this wheel, seeing if there any improvements to be had.
Reading every entry from the start. It was not my lucky day, if i had accidentally pressed any of the 5 comment button i wold have not offended you, i must have hit the disagree button number 6 button by accident when scrolling though.
I must be more careful, i am now wondering how many times i have done this. ( old wobbly fingers )
sorry

They are bloody small indeed!
Regarding the ffb with accuforce:
I don't think there's anything you can really improve. With the ts-pc I had for a week and now my csw 2.5 there's really nothing better than the default.
You can only make it worse.

You could however activate the gyro and set the min damper level to 0.2 to get a bit of gyro-dampening at higher speeds.

 

[Kek700]

Well, maybe I was lucky and not unlucky

You have simplified, probably many hours in to one
sentence.

I was going to look into my LUT setting, I created those from iracing wheel checker. With the thought of changing or modifying them.

As my dad always used to say bad luck can quite often be good luck.

Many thanks RasmusP

 

[Kek700]

That's the problem with a forum, email etc etc.
Sooo difficult sometimes not to offend, as I am some what dyslexic even when I attempt to type proof, just in case, I still miss lots of obvious mistakes.

 

[Kek700]

A follow up....

I had interpreted LUT incorrectly ( settings were incorrect ) redid that. did what you advised above + extracts from your article wrt AC settings.
Had to reset my basic wheel setting to match.
So far very pleased with the improvements, will need a bit of time to re adjust to these settings with
maybe some minor tweaks. probably to wheel force and damping. thank you.

 

[Kek700]

@RasmusP

[HEADER]
VERSION=1
TYPE=GAMMA
ENABLED=0
TYPE=LUT
ENABLED=1
[GAMMA]
VALUE=0.5
[LUT]
CURVE=newlut.lut

I seem to have only Lut set, can you see anything obviously wrong in my ini file.
I was expecting gamma and lut to be set???

 

[RasmusP]

@RasmusP

[HEADER]
VERSION=1
TYPE=GAMMA
ENABLED=0
TYPE=LUT
ENABLED=1
[GAMMA]
VALUE=0.5
[LUT]
CURVE=newlut.lut

I seem to have only Lut set, can you see anything obviously wrong in my ini file.
I was expecting gamma and lut to be set???

There should only be one "type" line. You can't run gamma and lut together as far as I know.
So your config should look like this:

[HEADER]
VERSION=1
TYPE=LUT
ENABLED=1
[GAMMA]
VALUE=0.5
[LUT]
CURVE=newlut.lut

 

[Kek700]

cheers for for your help.

 

[Kek700]

Finally got it sorted, decided to stick with LUT, took quite a while to get it all correct, looks straight forward, not for a DD wheel though. Still have my doubts about wheel checker, as for gamma apart from continued back and forward testing with loads of variables, especially not being able to quantify each stepped meaning of Gamma.
Finally got wheel checker to give what appears to be reasonable results, decided to go with that.
I would test with wheel checker, looked at results, bugger, left some variables on the wheels programming, eventually took all programming from the wheel except for FFB strength, finally a result that made some sense.
There is a need, for a way of testing the response from a wheel with quantifiable meaning.
To be honest, it is all a bit suck it and see.
I would say, after the placebo effect wore off, i have an improve result, but that is a feeling not a
quantifiable end result.
cheers for your input @RasmusP