A Message For ISI

I recently phoned up Leo Bodnar to arrange a trial of his sim steering system and didn't miss the opertunity to ask what i thought was a pertinent question. Some of you may have read a very interesting article written by Leo Bodnar entitled "Why Force Feedback In Computer Simulators Does Not Work". I asked him whether he was going to write a plugin for Rfactor 2 to change how the force feedback works like his article suggested and his response was, it was something that had to be changed in the simulation software itself and that he was willing to work with sim Software creators to experiment with this idea.

I don't like to stick my nose in to where it doesn't belong or pretend to fully understand something i don't, but i read Leo's article and to me it sounds convincing and as person who is passionate about motorsport and sim racing i want the best out of my software and hardware, So i am writing this to raise awareness of a way to potentially improve the sim experience.

The article ; http://www.racesimonline.com/articu...back_In_Computer_Simulators_Does_Not_Work.pdf

 

We do have a contact page...

 

Am I the only one who finds it strange that someone would make a ffb wheel for software that they think isn't proper in the first place?

 

My interpretation (and it is only that) of his white paper was that he was expressing conclusions and opinions that he had come to after years and years of working with this stuff. In other words I think it's an opinion he came to as a result of doing the work, including creating the wheel. The wheel has been in use by real world racing outfits for a number of years so that kind of coincides with the timing of the white paper. I interpreted it as him explaining what needs to happen in order for the steering feedback to be truly realistic. It's essentially the same as John Carmack, back in the day, writing some manifesto on what graphics card manufacturers need to do in order to take visuals to the next level. Or Michael Abrash talking about what needs to happen for VR (eg, Oculus) to be fully realized.

http://www.roadtovr.com/gdc-2013-michael-abrash-virtual-reality-oculus-rift-presentation-slides/

 

Yeh sorry about that Tim, i just thought posting here would get more exposure for Leo's idea and encourage discussion as well as getting the message to you guys.

 

Very interesting Matt. Thanks for the reply.

 

Forgetting all else...
This could be the white paper to start a radical change in PC simulation effects.
Maybe ISI should look at possibly collaborating with Leo Bodnar going forward on this aspect.
The way I see it, they both have something to potentially gain and nothing to lose.
He has not indicated a willingness to develop simulation software, so it's not about competition.
If his products interact more realistically, he'll benefit through more sales...ISI stands to gain a better understanding of the interaction between hardware and software. It'd be a win-win for everybody else.

 

I don't know if Leo has contacted anyone regarding his ideas or what his intentions are, but his paper goes in to enough detail so ISI could research what Leo wrote about on there own. Saying that i don't know if they already have, im just a consumer trying to use what resources are available to me to try to help improve the product. =)

 

Is realistic ffb something that we want? We aren't able to use our "internal gyroscope" so don't we need a little extra? Just brainstorming here.

 

I really don't care who reaches out first...as long as someone does.
It'd be a real waste of an opportunity if no one does.
I was extremely happy when Tuttle joined ISI.
It meant a super great track guy was there to add his input. My only wish would have been Feels3 going along as well.
Anything which brings the community a better product, I'm for.

 

I agree. realistic ffb has far fewer feedback. try with your car. our ffb must supply to many other effects. we drive only with the steering wheel. real drivers drive with their butt
I hope I have translated well

 

This exactly. The ffb in a sim needs to give ALL of the information through the wheel, where as in a car you are getting information from all different angles and every part of the car. The steering wheel in a car really doesn't give any 'feedback' at all...it barely has any resistance, the bumps and dips in the road get absorbed by the suspension (and your butt lol), and with loss of grip you'll feel the car moving underneath you well before the wheel slips (if at all, depending on which tires lose grip.)

Realistic ffb in the wheel in a sim just isn't going to work, unless you combine it with an amazingly realistic motion simulator so that you get the actual feedback you need to drive. Otherwise, it would just be a little bit of resistance in turning the wheel, and not that much of that.

 

FFB for me is something that is going to be ever-evolving, it gets better gradually over the years (as proven already) and will continue to improve in years to come. Not read Leos paper and not going too either, but if he feels what he has to offer will aid the development of FFB for the Sim racing market then it cant be a bad thing to have a chat, surely he would've done that himself with a Sim company or two by now anyway. If those companies are interested then cool, if not it's no biggy as we know each developer likes to do things "their" way.

Leo drawing up a proposal and going to one of these companies is a bit like ISI/Kunos approaching a Car manufacturer for a license to me, similar sort of territory.

 

I must disagree that ffb needs to differ from reality. You dont need to account for the sensations in the butt and put them into the wheel. When you drive a good car in the highway at 150 kph in a turn at constant speed you can feel in the wheel the stability that it provides when you compare it with an old small car. The loss of grip can occur at very diferent speeds depending on the grip. If we just related it to a lateral acceleration we couls not drive under different conditions or with different cars.

Enviado desde mi GT-I9505 usando Tapatalk 2

 

I haven't read Leo's FFB document in a while but from what I recall he wasn't looking to make the forces more realistic but actually less realistic where it would do some canned stuff to simulate the rear wheels breaking loose. IMO this is already present in rF2 (feeling rear wheels breaking loose) as long as you don't crank the forces so high that you get clipping. In rF2 (and rF1 with RealFeel) the wheel will turn into a slide when the rear lets go. Just like a real car does. Everyone praises GSC for great FFB when most of this comes down to the inclusion of RealFeel out of the box.

I personally like how RealFeel and rF2 handle FFB. It simply transmits the steering arm forces to the wheel without any canned effects which is like a real car. I would rather keep it simple like this rather than having 300 parameters to tune for each car you drive with 300 opinions all over the forums about which settings feel more realistic even though 95% of the people haven't driven a race car much less a road car on track or even a road car beyond it's limits on the street. Also, I hear a lot of people say the FFB needs to be tied to the tires and not the suspension. What people forget is the tires are connected to the wheels which are connected to the suspension which is what impacts the forces on the steering arms. If the rear wheels brake loose then the load changes on the steering arms. No need to fake stuff. If a car feels odd it is probably due to the tires or suspension. Just like on a real car.

 

exactly! real racers or in the real life when you drive your car, you don't use the estercittive sensibility( the ffb from the steering wheel), or better, you don't use it a lot! the most fondamental sensibility that the driver, specially racing drivers use, is the propioceptive sensibilty that passes via the the inner ear and Spinal-Talamo-corticale via !
Real drivers don't feel bumps, oversteer, understeer..... through the steering wheel ffb, but through their bodys!
Real drivers feel the car, or the car moviment, or micro moviments using their propioceptive sensibility, Chinestisia, pallestesia( Via spinal-Talamo, spino-Corticale and the via from the inner ear to the eyes then to the Talamo then to the Corteccia)!
Then, what is the mean use of the steering wheel?! the mean use of the steering wheel is the correction that driver makes! so the driver feels the car's moviments through his body using the esterocettive/propioceptive sensibility, then, he use the steering wheel to correct the car's moviments!
With the simulation, that doesn't happen like this, because we use the steering wheel to feel the car and in the same time we use the same strument( the steering wheel) to correct/react to the car's moviment!

 

You cant compare the feeling from a road car with soft suspension, low grip radial tires, "slow" ratio steering rack, and probably power steering to say the steering feedback from any racing car, simulated or real.

 

I have been reading the paper an I strongly disagree in one of the basics.
FFB depends as ANY REACTION on the position/time input you apply. The required force to turn the wheel 10 degrees with a determined acceleration profile does not depend on who does it. it would be the same for any oerson doing the same thing.

The guy has a lack of physics concept. CONTACT FORCES are consequence of deformation and not oppositely. They depend on stiffness and damping. It is not the same as gravity forces. When he says that he is applying a determined force he is wrong. You try to move the wheel and you get a resistance from it.

However accounting for all aspects to be included in the simulation a lot of work needs to be done. The wheel properties should be taken into account especially for quick wheel turns. I am not aware of how many things are calculated in rf2 with respect to rf1 but it is a huge improvement. ISI development engineers will know what they are calculating and what is being simplified. It would be great to know if there is still room for imorovement.

Enviado desde mi GT-I9505 usando Tapatalk 2