Your Lordship: Please give us an untouched pure physics ffb, not that consumer thing.

It wasn't ideal for some cars to begin with and placebo or not feels even more intrusive and heavy after last release as if amount of gyro (dampening) effect increased.
https://www.assettocorsa.net/forum/index.php?threads/osw-owners-thread.28560/page-51#post-811257
At the moment DAMPER_MIN_LEVEL=0.20 in assetto_corsa.ini seems to work better. But SimSteering is quite different indeed it does something to FFB signal from what I've heard https://forum.virtualracing.org/sho...-impressions?p=2183132&viewfull=1#post2183132

And this one even weirder as oscillatory behavior is affected by supposedly only stationary DAMPER_GAIN
https://forum.virtualracing.org/sho...chiedene-Sim?p=2211393&viewfull=1#post2211393

 

g27's dont die, they just take a nap.

 

I don't know, I have not noticed any difference at all.
The DAMPER_MIN_LEVEL does not work with simsteering, although one guy at the last link says he notices it, because game damping is not implemented. So I don't trust too much in what people say they feel. I asked John Beeson from simsteering about it when I bought my wheel because the damping was not working when stopped and he told me game damping was not implemented at the firmware, it will be implemented at the future.

 

It wasn't DAMPER_MIN_LEVEL but DAMPER_GAIN (stationary setting) that accordingly to the poster completely took care of his oscillation. Can you test this? Really curious how it works.

 

It does nothing at all too, I even have an error at the log file saying DAMPER CREATION FAILED. It was working with the accuforce I had before and it stoped working when I installed the simsteering. I asked another guy with the same wheel and it didn't work for him too, so I asked John Beeson and he told me it was not implemented at the firmware yet.
But gyro is not working by adding the damping effect, because if it worked that way it would not work with my wheel too, it is really generated by the gyroscopic effect of the wheels at the physics, that is why I think the cars with such a high damping because of gyro must be like that in real life, that or there is an error at the inertia data for that cars, that I fell quite unlikely.

 

Back tt.

Thx for not only reading my thread.

Things about "filter" never was objekt of topic, only used by not involved user. As I mentioned earlier, AC´s FFB feels for me very well and I can imagine, that the finetuning has been performed with the same dedication and feeling, as your tribute to Prince - Purple Rain.

But please let´s go back from feelings to naked facts with my hope, that you like to bring some light in the darkness of unanswered questions of my deep research. What we have until today:

1.

2.

3.

4.

In summary:

Devs have no steer torque data from manufactures
100% FFB is calibrated @ Lotus 72 levels
No power steering based FFB, only tricked out with the steer assist thing
The whole FFB programming is a product of imagination

If all this is true for today, I have to apologize. Your devs are pure wizzards and I only have one request. Please do a little DD / OSW in your magic pot.

 

Devs have no steer torque data from manufactures
Correct, but not because they refused, we did not ask. Would they give it to us if we ask? No idea.

100% FFB is calibrated @ Lotus 72 levels
Nope. You quoted me, there is an "IF" there. 100% is calibrated car by car automatically on Vallelunga Club with a system that analyze the signal during a lap trying to obtain a similar level for every car. This is not always possible because max values are not the complete story, the way and the speed a car goes to the max and the variations when the car is cornering generate different "feels"

No power steering based FFB, only tricked out with the steer assist thing
It's not tricked out, Steer Assist is a way to influence steer torque that comes from reality. It's not there to simulate power steering, it's there to simulate steer assist.

The whole FFB programming is a product of imagination
I don't know what that means. But in AC, as said countless time, the bulk of the FF signal (the one you get with all the "effects" to 0) comes straight from the physics and is sent to the device. Basically, with the exception of the multiplier for the signal level there is no FF specific value in AC, nothing to tweak, inside or outside the code, the only way to get a different FF is to change the physics of the car. So there is nothing like "optimized for this and that" other than the level adjustment to make sure the FF is using the entire range for all the cars in the default settings, nothing more, nothing less.

 

Hopefully the "steer assist thing" is no longer used because that is what makes FFB non-linear and would therefore give a lot of troubles for people with more powerful wheels. The "thing" was used in the past to give high downforce cars weighty feel on lower speed too. Unfortunately that makes powerful wheels oscillate like crazy. I haven't opened any ACD-files to see if it's still being used but as said, hopefully it's not used anymore.

If it still is, a way to bypass would be much appreciated for open wheeler lovers, I guess..

The "steer assist thing" should not be mixed with actual power steering. The function ingame is just a way to boost or weaken low forces.

Missing power steering simulation only forces you to run lower gain for tracks like Nordschleife because the bumps are just way too rough to handle without any filtering. By using the word "filtering" I don't mean the filtering you can find from FFB Options because that does not do the trick. Actually I'm not even sure how it's supposed to work.

 

It could be completely placebo on my part (I'm prone to imagine things like that..), but this flexibility or rubber band effect in the steering that kind of makes it harder to pinpoint the car, for me gives a perfect impression of the flexing of the tyres sidewall during cornering. With the v10 tyres this is more pronounced, and I'm really really fond of it, specially the heating up of the tyres I can now sense through the steering.
Limited to the experience of my rl family car, the steering actually fels like that. During cornering of a constant radius curve and although the steering wheel is practically at a fixed angle, the car still seems to move about due to the sidewalls of the tyres flexing over the undulations of the road, and this flexing of the sidewalls you can directly feel in the wheel, its a constantly bloating and ebbing force on the steering wheel, more pronounced with the high shoulder winter tyres, and makeing the steering little bit imprecise. With the low profile summer tyres its not as noticeable but still there.

In my opinion, the FFB is perfect really.
That said, I'm also looking forward to upgrade to a DD wheel (sometime) in the future, and my judgement is solely based on driving with the T300RS

 

Wouldn't it be good to take these into account (if possible)?
Just surprised that you didn't even ask for that or measure it (in case the cars were driven).

Sure, most consumer wheels don't have the range to really reflect the torque, but with the common DD wheels, you have enough of that to simulate just about everything.
Of course, getting the torque right wouldn't mean that it feels exactly the same as in the real car, you would need to select the appropriate wheel size as well, but that's up to the user.

Also regarding the FFB clipping. Wouldn't it be possible (and desirable) to get rid of that completely? Doing something new where you can control the FFB on the wheel without signal clipping but also without the limitation in resolution and force (up to hardwares limitations, of course).

On a side note, I forgot that in the early OSW days the original creater (Bernhard Berger) of the OSW worked on an "API FFB mode" for AC, which was working with the Argon drive.
Once the MMO firmware came out, this effort stalled since he wasn't a regular AC user.
iirc, the code is available on github, though.

 

The problem I see with trying to get the steer torque data from manufacturers is the fact that some might give it while others do not.

I don't expect any manufacturers to have this sort of data for older cars, meaning whilst we might have accurate data for a few modern cars (and the chances are it won't be of any relevance to the majority of users with normal wheels anyway), we then have a large amount of vehicle in which the devs just have to guess.

 

You are not the first to give that explanation. It should actually be possible to find out if that's the case by modifying setup parameters like camber and the pressure.
It should also feel different for different types of tyres and tyre geometries.
Unfortunately, my GT3RS wheel can't deliver such differences and thus it wouldn't really make sense to investigate.
To me a heavily underpressured and heavily overpressured car feel mostly the same in the FFB, they just have different grip properties.
I'm very curious to find out if I can "feel" more differences here with the Lenze.

The other explanation was that most gear- or belt-driven wheels have a lot of damping, be it by design (thus the hardware) or by the FFB signal processing and that that is the case that the FFB with the DD wheels can feel differently as you do now see the unfiltered signal, which in comparison might feel "wrong" at the start.

tbh, I can't judge if any of these two explanations is correct.
The only thing that I do know for sure is that with a DD wheel, you can get the representation of the FFB signal undamped and unfiltered, with a linear torque curve and almost no external or internal artifacts.

Well even before assembling it, I can tell you you will be shocked. I knew the size (~12×12×20 cm³) and weight (~10kg) of my servo upfront, but holding the thing in my hands I was like "are you fscking kidding me?!?".

IIRC, the "small" Mige, which most people select, is even bigger and heavier.

 

Why wouldn't they? I don't see a reason for this to be top secret. Both the values before and after power-steering.
The active steering aid systems like you see on quite a few modern cars (which is similar to the speed-sensitive steering that most sims provide) could be a different story.

Or calculate or measure. (Well, the latter would depend on the availability of the car, of course.)

 

I'm not suggesting its top secret data, just that I doubt manufacturers will actually have this data for anything other than brand new current cars.

And for everything else, how do suggest they measure it? Maserati 250F, GT40, Miura, any of these older cars - I don't think Kunos are in a position to be able to call up the manufacturers and get access to these (again, if they had for modelling before). And you can't exactly walk into the local showroom either.

I just think as nice as it would be for the few with powerful enough steering wheels to simulate each cars steering torque, its not worth the effort. If they start AC2.0 and start all vehicles from scratch again, then sure why not give it a go.

But go back to over 100 vehicles? Nah, I'd rather they spend their time on more significant things.

Just use the multiplier and a bit of common sense - modern power assisted vehicles are quite light, older vehicles with no assist, wide tyres are much heavier, so crank it up.

 

That (and the price) is what is still holding me back. My Raceroom RR3033 (although modified and strengthened) really does not look like it will stand any DD wheel...
However, to have linear range of torque up to something like 10 or 12 nm, that is really tempting...

sry for off topic...

 

I think every manufacturer since the introduction of power steering has numbers for this. For the older cars the biggest problem (apart from the main will to share data) might be that it could be on some old sheet of paper hidden in some folder in some project data storage and therefore could take some researching effect to get to (which costs work time and thus money).

Yeah, hence my last sentence. Regarding the measurements, I don't think it would be necessary to measure the torque while driving, at least if you consider that the tyre model is "correct".
I would rather determine the efficiency of the steering system (or the force losses) and the play in the system. That could be done without driving, even without wheels.
Since these are not the most important values, I guess that's the reason why it hasn't been done and might not be as important for cars without power steering.
Although for some older cars (e.g. older American muscle cars) simulating the play in the steering would be an interesting effect.

But overall I think it would be more interesting for cars with power steering support and for validation.

Depends on what you are after.
e.g. for Porsche, the steering in their car is very important. Every time a new 911 comes out there are reviews praising it for the feel of the steering and how connected you feel to the road.
Thus it must be different than in other cars and personally I would like to feel these differences.
Of course there are differences in the feel of he cars in AC, no question about that, but if there are differences that originate from a different steering setup, I would like to be able to feel that as well.

Quite frankly, this is not my job to do. Here I absolutely agree with what Stefano wrote about.
Neither FFB clipping nor the differences in the force output should be something I as a user should have to spend any time on.

There are two smaller Lenze servos available (I hope I didn't mess up the product numbers):
MCS09F38: 3.1 Nm holding torque and 15 Nm peak torque, 5.2 kg
MCS09H41: 5.5 Nm holding torque and 20 Nm peak torque, 6.1 kg
These are considerably smaller (and lighter), but I would need to investigate which PSU you would need for that, it could be that there is a limitation there.
Also I can't tell you any prices, you would need to ask at Reimer Servosystems for that.
Mige has smaller servos as well, but I don't think they were ever used for an OSW thus far.

 

Using 1:1 steering rack forces would be a bit problematic because you'd need to match your wheel rim too.

Just to give you an idea, I had to raise amperage by 1A and gain by 15% when I replaced my tiny 275mm diameter rim with 325mm one in order to get close to similar feel. And 15% on DD-scale means ~50% increase in gain...

The wheel rim size is THAT big of a deal.

 

Absolutely!
The problem here is that these days wheels are interchangeable and many users make use of that feature (I'm planning to do that myself as well).
Except for a few cases (mostly the Fanatec Formula, BMW GT2 etc.) you can't know the wheel size.
The user-friendly way would be to do it like the setting for the virtual steering while, thus define in the options which size your rim has.
Or to create an app with which you can define it yourself. Actually if you only modify the in-game FFB gain, you can write that part yourself already.

One of the reasons why older cars had huuuuuuge steering wheels (sometimes in the range of 360mm to 420mm diameter).

(The other one was precision, turning speed wasn't that big of a deal, since the cars weren't as snatchy.)

 

At first, my heartfelt thanks for spending your time to quench our thirst for knowlede.

As you said earlier, you´re not happy with this situation too. May be if your arm is long enough and positive thinking ist strong enough, in future is a way. Nothing ventured nothing gained.

Very sorry for wrong quoting.

An automated calibration system based on Vallelunga Club data explains much as well. Especially why on NOS all over 75% main FFB is in clipping.

"Tricked" wasn´t meant bad.
Obviously that´s the reason of wrong feeling using linear output hardware ( DD ). More about this in my bottom line.

I think it´s the same theme as above.

Many thanks for your tireless work even here in board. I can resume, that AC´s FFb is in principle pure physics, but modified via steer assist and calibrated by automated systems. I would like to underline again, that the FFB feeling of all non ps cars ist very great and even securtiy protection from overmaxed FFB hitting walls etc. is great. Hitting walls in Iracing as example, with a 30 Nm DD, your wrists became broken.

The prob using DD are only ps cars. It´s very unnaturally. In real ps cars the max torque is very, very low and, and that´s about ist, it don´t increase very much. Standing, rolling, racing - the torque is near equal, but the level of information is high. Driving a ps car in AC the torque grows much to much and reducing this, in the middle is nothing more. I think this is based on stear assist manipulations to increase the feelings of lowtorque wheels. May be not only. As wrote by other users, an option for disabeling power steering was not bad.

I don´t know enough about programming, but is ist possible to give deeply interested users an option for calibrating per car like curves option in Photoshop.



Not in post process ini. Lut´s are not the way. What I mean is something like above as ingame app, together with a graph like FFBclip. This instead of automated Vallelunga Club, for pedantic users like me. Optimal FFB for every car / track. Dreaming allowed.

 

Most makers of DD-Steering systems have said that there would need to be a completely new FFB-API created in order to take full advantage of the hardware's capabilities; Leo Bodnar wrote about it, Berney Villars and Pax both echo those views. Short of that (excluding legal barriers), the best FFB we are likely to have will come through game-FFB or wheel firmware/drivers. That's why I'd like to see SimCommander-FFB support for other DD-wheels; it would open up another way to create and tune FFB when the game-FFB is lacking in effects or just feels wrong.

While most Racing Simulations do a good job of providing FFB for use with DD-wheels, there may be some benefits through enhancements, filtering methods, or fine tuning that are not currently available. As someone that uses both an AF and OSW, I can't over-state the benefits of SC4-FFB enough in it's ability to translate game telemetry into FFB output with titles such as Dirt Rally and Pcars as it really can be "Game Changing". That's not to say that it makes corrections to the underlying game-physics but, it clearly does better at translating those physics into FFB output.

Anyone that has used any FFB steering wheel with a really bad profile or improper settings, or no FFB at all knows what influence FFB can have on our perception of a virtual-car's handling.
I believe some reasons why more people aren't voicing their support for broader DD-steering support in SimCommander are (obvious cost factor aside):

• Many iRacing/rF2 players that have used an AccuForce with SC4-FFB don't play games that benefit the most by using it
• Most DD-users haven't experienced playing various games with an AccuForce and SC4-FFB

Personally, I'm currently very happy with AC-FFB using the OSW with cars on v10-tire model. I do think it may be a bit easier to find the point where the tires convey the rubbery feeling a bit better in iRacing, AMS or rF2 but, that gap is reduced IMO.