Logitech G27 Force Feedback FFB Assetto Corsa - A.S way

[=A.S=]

WheelCheck.. FCM tool Lut Converter.. FFBClip App .. tried them all . . and wasted alot of time testing pretty much every guide out there.. till i was like.. ok . time to think.. and not replicate things like a parot.. and VIO LA.. never been so good before. Let me know how it works out for ya. Enjoy !

G27 Owners .. simplicity is DEVINE!

The confusion and frustration of the G27 not being a good wheel and the pain to set it up correctly.. its your christmas today.. try.. and thank me later.
Without going into once again confusing technical details.. straight forward.

1)
120% in oder to eliminated the "deadzone behaviour" or interfacing bug of the G27 completly - just by doing that

40% - 55% dampening, because otherwise your wheel will turn loose if car isn´t rolling. Standing rubbers still provide resistance on the steering wheel. ( i have to in 70% as the not roll resistance should be higher than the slow roll one)

2)

Under C:\Users\UserName\Documents\Assetto Corsa\cfg in the controls.ini file edit like this:

[STEER]
JOY=0
AXLE=0
SCALE=1
LOCK=855
FF_GAIN=1
FILTER_FF=0.6
STEER_GAMMA=1
STEER_FILTER=0
SPEED_SENSITIVITY=0
MIN=0
MAX=0
DEBOUNCING_MS=5

[FF_TWEAKS]
MIN_FF=0
CENTER_BOOST_GAIN=0
CENTER_BOOST_RANGE=0

[FF_ENHANCEMENT] (or 0 them out.. and do by car individually)
CURBS=0.15
ROAD=0.05
SLIPS=0.1
ABS=0.05

[FF_ENHANCEMENT_2]
UNDERSTEER=1


3)

Create a file Linear.txt file in the C:\Users\UserName\Documents\Assetto Corsa\cfg folder, open it with notepad or editor and add:

0.00|0.00
0.02|0.02
0.04|0.04
0.06|0.06
0.08|0.08
0.10|0.10
0.12|0.12
0.14|0.14
0.16|0.16
0.18|0.18
0.20|0.20
0.22|0.22
0.24|0.24
0.26|0.26
0.28|0.28
0.30|0.30
0.32|0.32
0.34|0.34
0.36|0.36
0.38|0.38
0.40|0.40
0.42|0.42
0.44|0.44
0.46|0.46
0.48|0.48
0.50|0.50
0.52|0.52
0.54|0.54
0.56|0.56
0.58|0.58
0.60|0.60
0.62|0.62
0.64|0.64
0.66|0.66
0.68|0.68
0.70|0.70
0.72|0.72
0.74|0.74
0.76|0.76
0.78|0.78
0.80|0.80
0.82|0.82
0.84|0.84
0.86|0.86
0.88|0.88
0.90|0.90
0.92|0.92
0.94|0.94
0.96|0.96
0.98|0.98
1.00|1.00

There you go, you wanted a linear G27.. there you have it now.

After that, rename the Linear.txt file to " Linear.lut" (lima, utha, tango) and save it.

4)

Open the ff_post_process.ini file and change it to:

[HEADER]
VERSION=1
TYPE=LUT
ENABLED=1

[GAMMA]
VALUE=1

[LUT]
CURVE=Linear.lut

5)

Open the assetto_corsa.ini file under :\Steam\steamapps\common\assettocorsa\system\cfg and make sure the following part look like this

[FORCE_FEEDBACK]
FF_SKIP_STEPS=1

[FF_EXPERIMENTAL]
ENABLE_GYRO=0
DAMPER_MIN_LEVEL=0.0
DAMPER_GAIN=1.0

6)

ingame should now look like this:

Here you can also use 60-70% as GAIN Main and Filter ZERO !!

 

[=A.S=]

i sense frustration for no reason. dont try it.. dont bother.. dont troll

 

[Romano1]

yep, I am sure your random numbers have more feelz than all the other random numbers on the internet that obviously have less feelz.

I gotta say that I love my AC ..... it is by far my favourite racing sim........ but I have always felt that the FF with my G27 was not as good as I felt it should be
I have not yet tried A.S.'s LUT ..... but I have to say that the free Snake oil I found from Rasmus has been extremely satisfying...... makes the FF in AC feel closer to what you get in AMS, which is not as good of a sim as AC but the FF feels more realistic IMHO
I am looking forward to testing out this new snake oil.
Cheers
Romano

 

[=A.S=]

dont worry.. i dont heal cancer nor do i sell free energy devices ..haha - not yet

 

[RasmusP]

yeah right... there isn't enough snake oil in simracing... WE NEED MORE!

Always a pleasure Stefano!
I totally get you, a lot of "random numbers" out there and not only for AC.
I spent a lot of time to create the lut I'm currently using and it's pretty much the default ffb. Just changed the lowest numbers for the crappy G27. It seems I made a lot of people happy with it
So if someone tells me that he tried my settings and likes his own better, I like it!

Or maybe it's horrible, don't know yet

But since you read this thread I have a question that you could maybe bother to answer?
If I use 100% overall ffb in AC and 100% in the car, the ffb bar in the pedals app shows a lot of clipping. I thought with everything on 100% it would be right on the edge of clipping.
Maybe you can clarify on this?

 

[StefanoCasillo]

But since you read this thread I have a question that you could maybe bother to answer?
If I use 100% overall ffb in AC and 100% in the car, the ffb bar in the pedals app shows a lot of clipping. I thought with everything on 100% it would be right on the edge of clipping.
Maybe you can clarify on this?

sure, this has been explained before.
We use an automatic procedure to generate the FF default levels. The target is not to "avoid any clipping" but to keep the level "under normal driving" around 90% in order to allow for enough headroom to convey useful information.There is no reason to avoid clipping altogether, high frequency clipping peaks are totally fine, they don't transmit useful information other than "oh there's a small clip there".
Again, I've said this before, my simulators are the ones that introduced the concept of "clipping" and FF levels with netKar into simracing, no other sim before included a "ff meter with a clip led"... fast forward 15 years and people turned it into a religion without really understanding what it was all about.
On consumer wheels (such as the G27) power is a scarce resource and needs to be used at its maximum to convey the best useful signal possible.. going taleban over peaks is simply wasting signal to optimize it for the exceptional case.

Regarding OP, nothing really much to say.. he proposes a linear LUT without realizing it does absolutely nothing other than waste CPU cycles, a 120% on the driver that who knows what it does (non linear? clipping? well.. it feelz good so whatever right?), filter on the FF signal that removes info and introduces lag and calls it "devine"...
as a G27 user myself all I can say is ... bhaaa

 

[Michael Hornbuckle]

@=A.S=
If you still have a deadzone after using LUT, your LUT is incorrect. Using above 100% gain in the profiler is only going to add to your troubles since you will now need to lower the individual gain of every car. (which in turn lowers the detail of the feedback)
---------------
I just saw you post above in which you called the creator of AC a troll.....

 

[Stephen O'Sullivan]

@kunos I have a second-hand Driving Force GT. After using Wheelcheck and this AC Lut generator, I have been enjoying AC immensely. (I also use Clipapp to check each car's FFB level - "light" for no clipping and "normal" for occasional clipping.) Even on my old wheel, the sim's FFB really shines. It's been fun exploring different cars' driving characteristics around all the Highlands circuits.

 

[RasmusP]

sure, this has been explained before.
We use an automatic procedure to generate the FF default levels. The target is not to "avoid any clipping" but to keep the level "under normal driving" around 90% in order to allow for enough headroom to convey useful information.There is no reason to avoid clipping altogether, high frequency clipping peaks are totally fine, they don't transmit useful information other than "oh there's a small clip there".
Again, I've said this before, my simulators are the ones that introduced the concept of "clipping" and FF levels with netKar into simracing, no other sim before included a "ff meter with a clip led"... fast forward 15 years and people turned it into a religion without really understanding what it was all about.
On consumer wheels (such as the G27) power is a scarce resource and needs to be used at its maximum to convey the best useful signal possible.. going taleban over peaks is simply wasting signal to optimize it for the exceptional case.

Regarding OP, nothing really much to say.. he proposes a linear LUT without realizing it does absolutely nothing other than waste CPU cycles, a 120% on the driver that who knows what it does (non linear? clipping? well.. it feelz good so whatever right?), filter on the FF signal that removes info and introduces lag and calls it "devine"...
as a G27 user myself all I can say is ... bhaaa

Thanks for the detailed answer!
I agree on that religion thing. You said this was explained before but it seems I was unable to find the information. I probably couldn't get through the sheer mass amount of "religious" posts
And I have to admit it feels good to have been right about the "linear-lut".

Please feel free to put my butt down whenever I say something incorrectly. Studying engineering and really interested in these kind of things and always happy to learn how things really are!

 

[=A.S=]

@=A.S=
If you still have a deadzone after using LUT, your LUT is incorrect. Using above 100% gain in the profiler is only going to add to your troubles since you will now need to lower the individual gain of every car. (which in turn lowers the detail of the feedback)
---------------
I just saw you post above in which you called the creator of AC a troll.....

This is correct and what i thought aswell. Remember, AC is not the only sim the G27 is tested at here.
So, i isolated the G27 as hardware piece.. and looked FIRST at this device as such. Sure AC and other sims provide secondary option for tweaking, but what is the problem with the G27 itself to begin with, right?

Thus .. i ran FCM and Wheelcheck playing with differnt Logitech settings. INTERESTINGLY.. if i moved the overall strength slider from 100% up to 125% incrementally.. the deadzone in the result curves disappeared at around 125% completey.
Now this is interesting, good..but also bad (for reasons you mention), because i knew now (and confirmed with driving tests), that i CAN INDEED get rid of the deadzone just by logitech settings. (and it makes sense tbh. if you think about the gears, mechanics and the electro-motor magnetical phasing..anyways).

But i sticked with it and tried to reverse compensate with game options the additional "strength", which can lead to "rattling" and minute "oscilations" on the simulator car wheel (if you start driving or break at slow speeds noticeable - and other regimes). And it worked perfectly.

This was my approach and it seems to be working. Trust me, if i say.. i have spent hours and hours and hours testing everything one can just find on the internet.. testing all "snake-oils" advertised myself first.

As Rasmus pointed out - there are 3 main problem with that G27:

- deadzone shaky feel too much (can be eliminated in Logitech profiler pre sim)
- oscilations if setup wrong (after adjustments in the sim)
- rattling plastic-ish noises and behaviour (after adjustments in the sim)

 

[=A.S=]

Summed up: This is what i use now for those who want to try it.. i can talk all day long, but you guys testing it on yourself is only way to transport the message:

ff_post_process.ini

[HEADER]
VERSION=1
TYPE=LUT
ENABLED=1

[GAMMA]
VALUE=1.5

[LUT]
CURVE=Linear.lut

controls.ini (Documents\Assetto Corsa\cfg)

[STEER]
JOY=0
AXLE=0
SCALE=1
LOCK=855
FF_GAIN=0.7
FILTER_FF=0
STEER_GAMMA=1
STEER_FILTER=0
SPEED_SENSITIVITY=0
MIN=0
MAX=0


Linear.lut (Documents\Assetto Corsa\cfg)

0.00|0.00
0.01|0.01
0.02|0.02
0.03|0.03
0.04|0.04
0.05|0.05
0.06|0.06
0.07|0.07
0.08|0.08
0.09|0.09
0.10|0.10
0.12|0.12
0.14|0.14
0.16|0.16
0.18|0.18
0.20|0.20
0.22|0.22
0.24|0.24
0.26|0.26
0.28|0.28
0.30|0.30
0.32|0.32
0.34|0.34
0.36|0.36
0.38|0.38
0.40|0.40
0.42|0.42
0.44|0.44
0.46|0.46
0.48|0.48
0.50|0.50
0.52|0.52
0.54|0.54
0.56|0.56
0.58|0.58
0.60|0.60
0.62|0.62
0.64|0.64
0.66|0.66
0.68|0.68
0.70|0.70
0.72|0.72
0.74|0.74
0.76|0.76
0.78|0.78
0.80|0.80
0.82|0.82
0.84|0.84
0.86|0.86
0.88|0.88
0.90|0.90
0.92|0.92
0.94|0.94
0.96|0.96
0.98|0.98
1.00|1.00

assetto_corsa.ini (steamapps\common\assettocorsa\system\cfg)

[FORCE_FEEDBACK]
FF_SKIP_STEPS=1

[FF_EXPERIMENTAL]
ENABLE_GYRO=1
DAMPER_MIN_LEVEL=0.1
DAMPER_GAIN=0.7

Let´s keep in mind, that the driving sensation is a very subjective immersion and diverting from person to person though. I only focused on the 3 main issues FIRST.

- deadzone shaky feel too much (can be eliminated in Logitech profiler pre sim)
- oscilations if setup wrong (after adjustments in the sim)
- rattling plastic-ish noises and behaviour (after adjustments in the sim)

 

[=A.S=]

@=A.S=
---------------
I just saw you post above in which you called the creator of AC a troll.....

I don´t know, who that person was.. i only responded to what i see and read by him. He created AC.. nice.. but he did not create the G27.. and its problems He does have some "expert attitude" though..one can tell.. nothing new to me.. i have had "experts" (coders *rolleyes*) before. (technics or physics is my field of "expirience"- lets say) . .. comes with the branch i guess, hehe. PS: i did learn car mechanican at Mercedes and produced Beamers at BMW later (old days of life). Did drive amateur rally too. Just as reference.

 

[=A.S=]

anyways.. no harm in sharing..i am first time happy with the crap 27.. thought, some of you might like it too.. if not.. just ignore it..and stick to what you have or had .. peace out.

 

[RasmusP]

Tried your settings, and they work.
I like mine a bit better, but each his own, no harm done to anyone

To your settings in the profiler:
I use exactly the same approach for Dirt Rally, F1 games etc etc.
All games who don't have a good minimum setting.

So yeah, your settings work better globally, for any racing game.
For AC however I prefer to stick with the 100% to not alter anything I don't know about exactly what it's doing.
But I know what the LUT is doing, so I play around only with this, for AC.

Negative Point: I had to play around with AMS and R3E again

Maybe you should create a post in the upload section. Putting your LUT and ff_post_process.ini in a zip-file and include the screenshots and a little text-file with all the explanations.

Easier to understand etc. for someone who just stumbles upon this thread

 

[=A.S=]

Thanks.. i tried your thread.. 2 times (to make sure i did not read nothing wrong) .i tested your suggestions..hoping.. "this guys seems confident and knowing what he talks about" - UNFORTUNATLY for me (dont know why) ..it created a "mess".. totally un-drive-able.

PS: I do consider, that not all G27 are and feel the same by default ..wear and tear? who knows.

 

[=A.S=]

Bottom line.. G27 is a upper class kids toy.. but i am afraid i have to built my "man wheel" (DLD)

 

[RasmusP]

Thanks.. i tried your thread.. 2 times (to make sure i did not read nothing wrong) .i tested your suggestions..hoping.. "this guys seems confident and knowing what he talks about" - UNFORTUNATLY for me (dont know why) ..it created a "mess".. totally un-drive-able.

PS: I do consider, that not all G27 are and feel the same by default ..wear and tear? who knows.

Strange though... I could understand if you don't like the feel in general as it's a bit too "lively" in some terms. A bit too light and too detailed for this little wheel and therefore not exactly "like a real car" but you feel everything and can be fast while still maintaining kind of a "real car feel".

In the end it doesn't do anything else than just raising the low forces a bit, but in a "exponential" way to stop the wheel from oscillating.
Are you sure you didn't use more than 100% in the profiler while testing it? Even 101% create a total mess!

But hey, as long as you are happy don't change anything! We are just trying to share something to try and help. Not to start an argument about what way is better!

 

[=A.S=]

indeed.

 

[Jim Liak]

I don't understand why you are using a 1:1 lut file.
As it is now you aren't making your wheel linear , you just multiply 1*1 . I am not sure that you understand the concept of the post process.
As it is now you aren't using post process.
You indeed fixing the dead center with 120% profiler but you are importing huge none linear behaviour on the top of force spectrum twhich you let them there.
Also the 65% filtering is immense, killing all the minute forces that g27 has problem supplying them in the first place.
Really strange approach but if you are satisfied or making other people too then it's better to share than not.

Thing is and that's why @kunos interfered with his known attitude (that we love ) is that when people give some guiding they need to be completely familiar with the mechanisms otherwise they will make some people confused.

Cheers

 

[=A.S=]

@ Rasmus

This is what i meant earlier with "i worked back-wards" (in rough estimated only demonstrative principle pictures)

yours (hardware suffers by nature in force.. THUS the delayed pickup entry and deadzone)

mine (motors have enough "juice" to cover the 0-point, but need to be "tamed down" afterwards INGAME)

Get it now? Simplex ! but not "simple" to explain it appears.

 

[=A.S=]

@ Jim Leak Filtering you can disable.. it is just personal preference (taste) to eliminated "spikes" and add "tire-flex / grip" lag

You can keep Filtering at 0 ..at it is not relevant for sovling the 3 main issues meantioned above.