Yes the steering is too light in ac with gyro off. i totally agree on that.
I can still feel the under-steer effect in AC with Gyro enabled. I consider that effect to be confirmation of going beyond optimal slip-angle; ideally, one should rarely encounter it. The only other title that I am aware of that uses such an effect is RaceRoom, maybe PC2 to a lesser extent.
It seems AMS and rF2 rely on increased resistance leading up to front grip-loss where a plateau can then be felt in the steering.
I have been using gyro since the day It was avalaible.
I remember that understeer effect was more noticeable un some cars like some Lotus , the mx5 ,old F1 , and a few more. Probably the only one GT3 cars doing It was the Audi and the Lamborghini.
Car dependant, maybe is because if my settings.
Perhaps running a bit more Dampening / friction / filtering reduces the under-steer effect in AC. In some titles, having more road-bump detail seems to define the transition (reduction in resistance) better; I do use some of that in AC but, it’s very noticeable in PC2 when using the FX slider to increase bumps.
“I noticed exactly that ever since watching the first ever OSW videos while researching it for a year before I purchased mine. I thought this was just due to the OSW being in early development and not refined yet (ie. powerful, fast but unrealistic behavior) - unfortunately I was wrong. And, yes, it feels like cheating for me too. I’ve been harping about this for probably two years now (way too strong centering, the wheel does the work while you just hold on, feels like cheating, etc.). People say they can save slides better, yes, because the OSW is basically cheating for you.”
You do know that this is actually a realistic effect and has to do with the caster on the car. With the right amount of caster on a set-up at the correct speed you can straighten out a out of control real car with just a little touch on the steering and correct throttle application or non-application… Caster cause steering centering which means it will cause the wheel to track straight when you let go of the wheel. This is also part of why drifting is possible… With a normal car setup it is much harder to drift than if you have a car with more positive caster enhancing the self steering effect.
FFB in the way that the games deliver can enhance this self centering effect depending on the force levels you are using and the filtering because if you get the overshoots and oscillation just right it will use the caster and the slight counter steer overshoots to keep the car stable. So in the end it is a combination of real car behavior AND FFB that causes the effect… The DD wheels are just fast enough do this really well.
Another note… FFB delivery is not position based but the the amount of force the game tells the wheels is needed is generally based on the difference between the optimal position of the wheel and where you actually have the wheel (which is position based).
It’s not realistic. You have to correct slides yourself in real life. If you simply let go of the wheel, cars don’t save themselves. The opposite lock applied naturally by the vehicle’s physics without the addition of driver input will not be enough. You have to “work the wheel” yourself in real life.
From everything I’ve researched and all th FFB “gurus” I’ve spoken with over the years, they all say our FFB software and hardware are setup for position based control/output not torque based and that torque based would need re-working from both the software (games, firmware, etc.) and hardware.
Yes you are correct that you normally and most of the time correct slides with throttle application and steering control. But physics wise a car set up just right for it and circumstances in just the right manner, a can correct itself with little to no input from the driver. In real life it is rare but in most cases where the cars on iRacing are correcting themselves there are probably some other input forces at work as well, i/e throttle application, braking, just a touch on the wheel, ect…
In reality that is the holy grail of car set-up in real life is to have the car just right so that it feels as if the car is thinking what you are thinking and even when it gets out of shape it comes back to you in the way you intend… Like it has a mind of it’s own correcting itself and mitigating your errors on it’s own.
With the way I have my wheel set up it is very Natural in that respect, I do have to guide the wheel in MOST circumstances to correct an oversteer moment and if I do it wrong I do pay the price. I could never get the Skippy to self correct itself without a little (or in some cases a lot) of input from me.
It is actually kind of both Position and Torque based The software generally will use some sort of positional differential to determine the level of force signal to send to the wheel. Whether that is determined through a telemetry model that determines torque (as in iRacing) or is arbitrarily set by the programmers (or a combination of the two) is up to the developer. But the wheel is actually delivering that signal in the way of direct torque based on the signal strength provided by the software. This is why you can actually Turn OFF the FFB with the eStop and still drive on a Simucube as the positional feedback loop is completely separate from the Torque application.
Mika Assetto Corsa uses damping for the Gyro effect, in fact, if you increase the slide in directInput it hardens at high speed
Hello everybody. I registered in the forum just to comment this because I remembered this topic reading the webpage for the new Fanatec BMW m4 gt3 wheel. They say that the new electronics of Fanatec are compatible with can bus protocol. Maybe this is the beginning of the change you mentioned Mika?.
Keep wioth the great work Granite!!!
CAN bus is actually just a different comunication protocol that has is mainly used in the Automotive system management area… the ONLY reason why the Fanatec M4 wheel is Can is so they can use it in the real car as well and have it communicate correctly… The use of this Bus system won’t have any affect on FFB itself BUT it will bridge the gap between Real life wheels and the Sim Arena as if the devices as in the Fanatics Case use the same addressing someone can (in this Case) jump out of their M4 GT3 and take the wheel inside and do some laps on the Sim and then Jump Back into the Real car all with the exact same wheel.
Why don’t we use that communication method"CAN" in the Sim world?
I proposed this in 2014. Bodnar/Beeson said they were working on something and asked me to remove my iracing post, as my proposal was a blueprint for what they had in mind, lol. I did remove the post, but in hindsight, should have left it. But nothing ever came of it, as Bodnar changed direction, I believe, at least what I deduced after feedback from John Beeson.
As far as I am aware, the Bodnar SS2 series and SC2 Ultimate has CAN interfaces. I will study the hardware a bit to see what chip is used, but naturally I will not share any trade-secrets here, other than just basic comments. (Not that I have much yo share in this regard anyway)
It won’t take much to make it work in sims though, simply a CAN (from wheelbase/button box) to USB converter (for pc side). Of course the USB converter will need to include CAN bus transceiver chip and then the necessary software, but will be easy to do.
Maybe I should do such a button-box project:)
2 Likes
I think you need to make one Beano!
Then you can share trade secrets
I might get stuck onto EagleCAD a bit over the holidays and finish a little design I am busy with, Joe. But the main idea for CAN would be the pretty simple and robust high-speed 2-wire comms link, if it works in the automotive environment, it will be super-good for use in especially the motion-sim environment.
Sounds good.
It would be nice if there was a standard between wheel mfg.
Maybe Can could be it? 
I wonder what Leo is up to these days. I hope he is doing well!!
My idea was to have pedals, button-boxes, wheels etc all on CANbus, but going one step further, having a proprietary CANbus ->PCIe card and custom API to do away with MS Direct Input…
Next step would have been iRacing native Linux support, opening the door for mods to the Linux kernel (pre-emptive and gaming patches) to have an almost RTOS running iRacing.
Le sigh. Maybe one day.
Nope, this is just communication protocol between the wheel leds, buttons etc and the wheelbase for them. To get real time system where a CAN protocol could be an important part and an improvement for the force feedback, you would need to ditch Windows also entirely which is not a real time operating system.
One would need a real time OS and hardware + game/sim that is developed to function on that OS. In this kind of environment every given command and position would need to be achieved in certain time frame and the simulation would be with the given boundaries, lag free (or perhaps better say, the lag cannot be longer than what it is allowed to be).
The challenge for this is that the system would need to be more or less predefined, meaning that the solution is fixed to certain hardware and the software is developed for that particular hardware. This is a bit difficult to distribute, to sell for large audiences as it is only meant for one purpose, racing simulation. Such system would be much better for the purpose it is made for than a general purpose computer, but price tag also would differ.
1 Like
Well it seems Tommi thought alot on this subject.
Thanks for the input!
In a perfect world maybe, . . . not quite there yet. 
Interesting conversation although, I wonder what would be possible within the confines of the current PC environment (FFB API limitations not with standing). The hardware has certainly evolved whereas, the API, not so much.
Yeah, Tommi is right. The real advantage of the CAN would be in professional simulator environments where the simulator itself is also a realtime system running on a realtime OS and where the torque commands and position feedback can be closed with a realtime connection. Definitively not a direction in where any of the current consumer simulators are going, but it is used in some professional simulators.