Discussion in 'Automobilista - General Discussion' started by Renato Simioni, Mar 9, 2016.
I totally forgot about caster, you're totally right!!! My mistake
But Marc Collins said that on some cars they were changing them so that their MaxForceAtSteeringRack would be in a certain area and with other cars they just let them be with whatever they came too...That's the weird part I'm not getting.
Thanks for the reply, Domagoj. Are forces from different cars the same in the virtual world? Ignore force feedback for one second. If telemetry detects MaxForceAtSteeringRack @ 2750 in the middle of a corner for a truck, and then telemetry also detects MaxForceAtSteeringRack @ 2750 in the middle of a corner for an open-wheel car, are both cars in the virtual physics world (regardless of FFB or user) outputting the exact same virtual-physics force for that moment?
Answer to your question: Yes!
All of this would be much more clear if the sim simulated power steering. Since it does not, we end up with a kart and a giant truck affecting the steering rack in the same way at certain points/conditions. I am non-plussed as long as they both feel and respond authentically!
What do you mean by simulating power steering? I feel people are making this subject over complicated, or maybe I don't understand it the proper way. But it seems so simple to me...
The game calculates all the forces (geometry, forces, torques... physics) properly, and, if we had the most powerful and perfect steering wheel in the universe, we would feel the proper feedback from it. Sadly, that's not the case, and our wheels have limitations. So we rely on MaxForceAtSteeringRack to scale down the output and make it fit the range of our wheels. Some cars need more scaling down, some less. But if you have an extremely powerful wheel, you can set MFASR equally on all the cars (probably set it as the highest of the values), and you will experience the difference of "steering wheel weight" on all the cars. Am I right?
Also, isn't the MaxForceAtSteeringRack function already a form of power steering?
Setting the MFASR the same for all cars SHOULD be ideal but there are 2 major problems for that:
1. Power Steering isn't modeled so, for example, a Truck and a Formula 1 car may have way more powerful FFB than something that is supposed to have more powerful FFB (etc. GP2 or Formula Renault 3.5 car). The truck with it's huge MFASR output numbers is a great example of how wrong basing FFB strength purely on the in-game's steering rack forces can be.
2. Some cars had their physics changed in order to meet a certain MFASR number
... or something like that (according to Marc Collins) so these cars will be outputting incorrect FFB forces relative to the cars which are not edited/"scaled".
P.S. I guess you can say MFASR is a "form" of power steering - a super simplified version - but, if going by that logic, then you can also say reducing your control panel or in-game FFB strength is a "form" of power steering. In real-life though, power steering affects not just overall forces strength but feel and character/behaviour as well; the steering is affected in a much more complex way then just an overall-force slider.
So the way I see, from two 2 major problems, there are actually 0.5 problems. The first one mitigated by increasing MFASR pretending it's a power steering (not really, thus the 0.5), and the second one does not actually exist, as the differences pointed out by Marc come from real differences in F3's suspension geometry (as Domagoj explained). But then, whatever...
Cheers for helping me understand those things.
The two versions of the F3 DO NOT explain any significant difference in the MaxForce values. Compared to the whole huge variety of vehicles (we can't even say "cars") in AMS, the F301 vs. F309 would be so close to identical that, obviously, no other vehicle would be more like each one than the other one. Yet, the MaxForce values are -2200 and -3000. At +2000 we have the heavy as a boat (by car standards) Camaro. Is this part of a pattern? No, because the little Mini is also +2000.
The go-karts are in the +475 to +550 range. The Formula V12 is -2000 (closer to the F301), but the Formula V10 is -3000 (the same as the F309). Is the FV10 really closer to a F309 than it is to the FV12? The Marcas is +3600, but the Super V8 is -4000.
Conclusion: ignore these numbers because the wizards at Reiza know what they are doing and somehow get all these cars to feel balanced and authentic, including relative to each other. If you need to tweak (I never do, they are all perfect out of the box), adjust them up or down as you please using the Realfeel controls. You will go mad before you find any pattern or consistency in that variable.
Yes but how are you supposed to know what to raise the MFASR to? If I want to use the same MFASR for all cars so I can drive all cars with perfect FFB force-difference relative to one-another then that's impossible because cars with real-life power steering will be way too powerful since the game doesn't model power-steering and therefore the raw steering-forces the car is outputting will be way too high for user's end. If raising the MFASR to compensate for no power steering modelling, how are you supposed to know what to raise it to in order to have that vehicle correct relative to all the other cars?
Regarding the F3s...I wasn't speaking specifically about the F3s; I thought Marc said that there was a time when Reiza would edit a car in order for it to achieve a certain MFASR area but now Reiza just leave the vehicle alone and whatever the steering rack numbers it hits while driving are what it is - no editing vehicle physics soley to raise or lower it's steering rack forces. If that is what really did happen with certain vehicles, then those vehicles' steering rack forces are going to be inaccurate relative to the other cars that didn't go through this operation. Maybe I'm just misunderstanding what Marc said about some of the cars' physics being edited soley for the purpose of hitting certain steering rack force numbers (rather than just keeping whatever the physics engine spits out regardless how wrong or right, desirable or undesirable, it may be) but that's what it sounded like.
Many, if not most/all vehicles in real-life with power steering have different suspension design & geometry to what they would have if they were designed without power steering. That's a reason why power-steered vehicles with failed power steering are generally much more difficult to drive than cars designed from the beginning without power steering. That's probably another reason why the sim-vehicles that use power steering in real-life output such humungous steering rack force numbers in the sim - because their suspension design & geometry is made to use power steering. So, when we play, instead of driving a car with no power steering, we're actually driving a car with broken power steering, and that is a big difference.
So the system in general is flawed but I'm not pointing specifically to Reiza or even the physics engine used - almost all sims, if not all, don't model power steering (apparently iRacing does but I don't know if it's modelled in a true attempt to model power steering or just some guy simply lowering an internal FFB strength slider for each car, lol).
Will try to answer more later, but something like this never happened.
Just want to ask Reiza how much I need to increase MFASR for a Bodnar DD wheel from the default. Seems I need to at least double or even triple it.
Best answer to that could be given by
Yes thanks would be interesting to know. I use the Bodnar V2 with AKM54 motor 26nm.
How could the karts have such tiny forces? Even rental karts give me sore muscles and need strength to turn, let alone "real" karts.
What is really getting calculated with the MFASR? Why is it in a unit that cannot be converted to Nm...Is the MFASR a different force alltogether than steering Nm? If it's the same, then how come it cannot be converted to Nm?
Who said it cant be converted to Nm?
One thing, it is force on steering rack.
To see how it can be converted to Nm would require lookout in code but why is that so useful? Like saying oh yeah i'm driving my bicyle with 70Nm of momentum? I mean for 99% of people having force expressed in this or that measurment best one is probably percentage.
As for karts i'd say that is due to chassis flex not being modelled. Karts dont have traditional suspension but rely on chassis tubing flexing. When you turn your wheel in kart, inside rear wheel lifts up. Imagine when you are sitting in it, you are the force acting down and resisting inside lifting high up, which theb results in chassis flexing which you are doing with your hands. Hence crazy stiff wheel in a kart...
@Spin what is your point/suggestion?
Or you are just ranting on in your opinion flawed ffb or what?
Tracking FFB forces in Nm seems like something of a hobby for some or perhaps it's about bragging-rights for others; it's certainly a popular topic in the iRacing community with a significant DD-wheel user-base. Some of that crosses over to other titles as more users move to DD-wheels and I see the same discussions in the AC/rF2/R3E forums.
Well then guess we'll add that on todo list but do not expect to see it developed for AMS.
In AMS we are using realfeel and it is what it is, we can't change it...
I think MFASR is a linear, translational (or whatever the English term is) force applied to the rack. If you want to convert it to torque (Nm) you will need the radius of the pinion. Force times length equals torque.
I have no idea how realfeel deals with that in its code, but I would guess it uses MFASR and pinion radius together in order to scale steering rack forces into suitable torques for our wheels.
Even then, FFB output has to be measured at the steering wheel to really make it something we actually can feel and understand any comparison to RL forces; using a lever mounted on the output-shaft and scale to measure force and applying the appropriate formula based on wheel diameter . . . it takes some work to get accurate numbers where it really matters.
To me, it's about how the FFB works in translating the connection with the virtual car and less about the actual numbers or, even the accuracy. I suspect the output data just appeals to the "numbers" crowd now that we have more robust hardware.
Separate names with a comma.