Automobilista 2 Custom Force Feedback - Overview & Recommendations

@Karsten Hvidberg I think maybe what @Mayron38 is trying to say is some medium to fast frequency vibrations happening in the center when the cars are static, parked in the pits. I noticed it. But then as soon as you start driving it disappears. Only happens on cars that aren't moving.

I noticed with this file and low FX values the signal is so clean it makes me overconfident and I've noticed running low or zero LFB makes it feel more natural. I'm really surprised with the new rFuktor version. It shares many of the qualities that made me enjoy the @Joaquim Pereira files and the "rack force" file.

TBH I've been testing a lot with LFM race GT5 at Campo Grande because there are many corners where you can have weight transfer or momentum shifts under trail braking and slowly loose the rear of the car.

I noticed that my fastest laptimes there include some slip angle so I'm noticing to dirve fast I need to feel the car sliding and wanting to correct, but instead of trying to correct it all the time, I need to manage the slip, feel how much slip I have and figure out when it's safe to push it out of the corner, or if it's safe to hold more the trail braking or release it, having some slip and feeling the car sliding just enough to go the quickest, but also not too hard to the point where I bleed time off, or worse, I get into a spin because spent too much time over the limit.

I've also noticed the tire behaviour where if I make a corner with the car sliding for longer time, maybe too long, at the end of the slide the tire limit becomes smaller and smaller due to the heat generated in that moment. After a corner with some long duration sliding and slip angle the car will have much less tolerance by the end.

Even though I need slip angle to go quick, I must pay attention to how long the tires slip otherwise even on low power, I can make it oversteer from throttle input alone due to the heat generated previously.

So far I've been caught off guard a few times with trail braking and some kerb instability. 3rd sector the car can slide for longer and there is some acceleration oversteer.

All this is felt through FFB. On my tests I noticed while I can correct faster and be more on edge with higher LFB values because I just feel everything that tiny bit faster, it sometimes disguises some small things with others. Meaning I'm less consistent because sometimes the car is starting to slide but I mistake it as some other force or bump or small weight transfer.

I've also noticed that having 45 gain in game vs 60 gain in game (with different wheelbase gain % to match physical forces)... the FFB does feel different. And the way I drive. With a higher signal I feel more detail and the signal feels a bit more noisy but less "peaky" while with lower signal ingame things are less compressed and signal peaks can use huge nm from the wheelbase, to the point where if I'm about to spin into a hairpin like T1 I've had the Formula steering wheel jump out of my hands and I couldn't catch it back.

I'm really surprised at how much stuff I can feel and get out of the Default+ profile just with gain and LFB settings alone to get all sorts of different feeling FFB; before going into FX or Damping which tbh I would prefer at zero.

The new rFuktor with some FX does remind me of many things I enjoy on Default+ but sometimes idk if I can feel understeer as well as before. I like that it feels very clinical and clean in a way that you fight less the car and the physics and instead it focuses a lot on useful clean information for driving. However it requires a lot of time to get used to a custom ffb file. I would say in this file I'm more comfortable not going past FX20 or FX25. Over time I've started to enjoy more and more a cleaner file, even if pushing hard I don't do my absolute fastest lap time, I am able to replicate consistently very close laps.
(Ginetta GT5 at campo grande my TT is #3 1.38.3, in online races I've dont 38.5s, and many laps between 38 and 39.0 flat without messing up too hard, which is some very welcome consistency)

Sorry for the long post, just sharing my experiences and findings if anyone finds this info useful to enjoy AMS2.

 

You felt that very well. The Topspeed 2 has that mechanical feel of the steering axis, as you can read in my changelog. This means it works with a little friction at the base and a little damping in the game. This change naturally also has disadvantages. These disadvantages include a more muted feel, which can mask more subtle FFB details. All these adjustments to a base—friction, damping, inertia, rotation speed—ultimately dampen the FFB. But to achieve a natural feel, it's sometimes necessary to use a little of it. Additionally, the different systems (base technology, gear/belt/dd/hybrid) come into play. For this reason (and several others), there are two files, because in my opinion, just one file cannot possibly compensate for these differences. For example, a gear base is terrible with my dds file or with the defaults, and nothing can fix it. On the other hand, the gear/belt/hybrid base file is a joy to use with a gear base and a completely new experience. People underestimate these differences, and that's why there are so many different opinions.
But these disadvantages are less significant ( to damping , friction) than the advantages it brings. For this reason, my FX recommendations are now a bit higher. You now have several options to adapt it to your needs.
1. Increase FX even more
2. Increase this value in the file (kerbs_surface_scale 4.0)
3. Reduce friction at the base and dampening in play
4. Stick with the V7000 Topspeed

 

Hey Carlos,

I'm curious what things you would suggest to make better reg. the understeer, if you have any ideas?
Other ideas welcome as well!

 

It's hard to put into words because of what FFB and personal feel is. However I'll just share more details.
Today I've recorded some laps. now LFM moved to brands indy and I've just hot lapped there. With my usual Default+ (0FX) it was a bit noisy and ofc rFuktor much cleaner even with 20 FX. After some practice I did a quick Default+ lap but my fastest lap came from rFuktor with little to no adaptation.
Just as a test I came back to Campo Grande, and I almost instantly matched my P.best there from Default+, using the rFuktor file.

However in Campo Grande, during some of the long corners it exposes some things:
Sometimes I accelerate too soon into understeer out of a corner without noticing.

Sometimes when I get a small feeling of lack of grip or lack of weight in the corner I get it, it tried to show me understeer. Othertimes I'm just too confident. With this file I go instantly quickly just by focusing on trying to keep the car on the line, but if I try to push I end up pushing too hard without really feeling it as abruptly as with the Default+.

However the Default+ shows me so much stuff, sometimes I'm correcting "ghost" stuff literally (compared to rFuktor). Much smaller and tiny slips from the car are transmitted to me. In Default+ I get more of a "visceral fear" driving on edge and on the limit all the time, I end up ignoring some of the "warnings" from the FFB, and I have to ignore them to go fast, otherwise I'll do very "defensive" laps where as soon as I felt some small % of instability and slip I reacted defensively to it.
With Default+ I have to push through that, and then comes a problem, some tracks are noisy, some tracks I might push too hard and spin.
Even now in Brands while I've got one of the fastest lap times my race went into the gutter because I didn't feel the rear end get light in trail braking into T1 and ended up spinning.

The same thing can happen on rFuktor where many of the initial forces feel smaller. Specially compared to Default+... it's like D+ with 0 LFB still has some LFB in it while rFuktor is true zero LFB. Much less vibrations and shakings from all random things, less FFB from weight transfers? or slip angles?
IDK... but the feeling is much more precise, with rFuktor, if I feel I have to correct the car, I really must correct it: in rFuktor if I feel oversteer happening, it is really really happening for real ahah.
Sometimes I notice the car is in a 4 wheel slide, 4 wheel drift, common when taking speed into or out of a corner, and the ffb in rFuktor isn't going all bonkers trying to correct the wheels that hard or anything, while on Default+ it would be.

Now it's very tight and hard because I've not delved into rFuktor settings, idk what I would think of LFB on it to make it that little bit crispier and snappier like Default+, and by doing so I actually destroy what I like about it: it's accuracy.
At the same time, with it I've raced a bit less, because I start understeering more, I just take more speed everywhere, I don't feel "danger" the ffb feels more stable and planted all the time.

Soon I will edit and share the Brands laps and also Campo Grande laps, I should go edit those now... then you can see my inputs and also how my wheel moves.
In both cases I'm doing similar lap times which is already something very nice. But now to test consistency under pressure and to be able to race, push the car without overstepping over the limits is a tricky thing.
In Default+ the tendency is under driving, and I push over the FFB sometimes I get used to ignoring some of the small effects or vibrations and I mess it up from overdriving or pushing too hard and missing not feeling something.
In rFuktor I'll naturally push it hard without knowing, and it's so obvious when you're in control vs not in control of the car... that my tendency to push makes me understeer out of corners, overdrive and do inputs earlier or harder then what they should be because the car is still quite stable therefore no crazy FFB warnings go off, cleaner signal etc.

This has been quite an interesting experiment, but very time consuming ahahah I've spent days and days just driving... and to test consistency over long time it's even harder, so many variables.

PS - this is from the Ginetta GT5 experience, with LMDh I think I did some runs at watkins for a race but the low speed corners ffb on rFuktor were again, quite light, I was pushing way too hard into corners and I needed some D+ to keep me in check.

Edit: ended up sharing the video on it's proper topic, here
Need. More. Videos! C'mon guys!

 

Thank you so much for this very informative feedback on those profiles! Very interesting.
I will have some fun trying that combination and see if something can be easily improved, have a few ideas.

Best,

-K

 

I can't wait for a day when some sim will implement direct integration with wheelbase, setting all the configs (both driver / in-game) for each specific vehicle. That are predefined by developer, to represent similar feelings that car's steering wheel would provide (with some small force adjustments to compensate steering wheel size). Since FFB is NOT about personal preference.
That would be a game changer.

### I have a lot of experience with real cars. From some supercars to garage built ****boxes.
And it is normal that you as a driver can be happy with the steering feel in some cars. Yet feel nothing in other ones. And since simulators pretend to to simulate everything - this part should be covered as well. While it affects how driver works with a car through the steering wheel.

 

Unfortunately, that's just wishful thinking. It might be possible if you had just one base for everyone. The differences between the systems are currently too great. That will take years, or perhaps it will never happen. Everyone wants to make money, including the base manufacturers.
Think about how it used to be when there were only belt and gear bases, but no DDS. The simulations were much easier to set up with the settings. Today, there are countless different bases of very different quality (from DDS to gear), and on top of that, the manufacturers' internal software varies greatly. In my opinion, it won't get easier, but rather more complicated

 

You can find, in the public domain, real racing drivers with completely different opinions about the same car—whether it's real or virtual. I'll go further: take anything you want from the real world, and you'll find differing opinions about that too. Human (or animal) experiences are always subjective in nature.

Yes, by objective measures, the FFB of a vehicle is determined and constrained by physics; it isn't a matter of opinion. But what we like and what we need varies across the population.

When I drive my car, the seat-of-the-pants feeling comes from information that's not present on the steering wheel. I'd be lost driving my Audi A6 in a simulator if all I had were the rack forces.

 

Live for Speed only had gain slider I think, even RRRE doesn't have many options but in RR i need to do use very odd wheelbase settings to get it anywhere near usable ffb, otherwise it's just huge damping and weight. LFS is more self aligning maybe closer to rack force with some extras.
In both cases I feel these ffbs feel great on weaker bases with less tweaking but as you go for high end DDs you expect more and you don't get that much more. ACC for me was the opposite, while weight felt nice on G923 it was rly hard to feel slip angle and oversteer unless I had a strong DD base. (and even in AC to some extent).

In AMS2 there is modularity so that you can get best of both worlds, FFB is more open ended, more moddable and OFC that is like opening the can of worms, but in the end (albeit with old physics) I was able to do #2 world Time Trial lap with GT1 using a G923 and later with a DD1 I can still have the same competitiveness regardless of wheelbase quality and power. The FFB can be adjusted to work with all bases.

The thing is content variation, if I wanted to go test a new custom ffb file it could take me just so much time it's insane to go through all the same cars tracks w same setup as before. (before even touching the settings sliders).
In my view in most cases the best ffbs are the ones that feel kinda right from the start. Without too much messing around.

And btw high power bases will always have their own software filters and they will always require those because as soon as you want to play some older game or you're playing anything with an FFB designed for weaker bases, on a strong base the details will always be amplified way too much.
It's not like the entire world can afford or even starts sim racing with 20nm direct drives... if that were the case and even then some bases will always feel a bit off compared to others.

 

Good point. Yes, the "old" simulations were partly tuned only for gear and belt bases. Those Sims are still fantastic today with those bases. But unfortunately, not with the DDS anymore, and it can feel awful. LFS is a good example of this.

 

"rFuktor ? experimental 4"


* More resistance on slow turning.
* More dynamic range when turning.
* More dynamic range at slow speed by adjusted LFB to be true 0 LFB when set at LFB 0.
* Disabled "relief" by high downforce squashing the tires.
* Slight adjustment to the constant rack damping.
* A new tire rolling effect.

You will have to adjust your settings just very slightly compared to 3.1. Slight change to gain and possibly LFB.

 

Bonjour, j'ai essayé votre ffb V7000 Topspeed 2 avec mon Logitech wheel pro, les sensations sont bonnes mais vos réglages sont beaucoup trop forts, bien pour se muscler cependant , pour ma part, j'ai réglé ma base sur 8nm, l'amortissement sur 5 et le filtre sur 6 comme conseillé. Dans le jeu gain à 55.

 

The force at the base depends very much on personal preference. For many DDSs, I recommend between 10-12 Nm because I think it's a good compromise for people. Personally, I use between 9-11 Nm, depending on the car. In a simulation, larger forces are used than in reality. But since for many people the wheel is the only or most important connection for the FFB, I think it's fine. Of course, if you're working with immersive hardware, things can be completely different

 

Thanks mate, I like your custom FFB a lot. Took some time to adapt but cannot drive without now.

Its much easier to do good lap times now.

I am using a Moza R9 with GS wheel. Your recommended Pit House settings are fine, except I prefer slightly lower forces (might be related to the GS wheel I am using)

Just for clarification. In your file you state:

# Game Force Feedback Intensity: 100% (Depending on the car)
# Professionel Game Force Feedback Intensity: 100% (Depending on the car)

There is no setting called "Professional Game Force ..." in PitHouse. Or do you propose these settings for "Professional Racers"?

Thanks for your hard work!

 

Maybe this setting isn't available on Moza R3 to R9, or it was changed with an update? But don't worry, it's the same and linked. It has nothing to do with a specific setting

 

my brain must be fried. where do i find these custom files that are recent??

 

Your Custom FFB File Collection | Reiza Studios Forum

 

Thanks Karsten for this awesome FFB it feels just right to me on RWD cars but with the JCW mini it's on low speed to much force especially when driving out of the pitbox.

 

Hey man, I have to give you immediate feedback on this file. It is simply perfect. Love the resistance on slow turning, something I always felt lacking. Dynamic range changes also feels complete.

 

I will try this.
Did you lower the gain in-game to check if it became better?