Automobilista 2 Force Feedback Overview & Recommendations

I'd rather compare it to reducing the steering angle with the base. There, too, there's this increased directness of their steering axis. And I know some people who reduce the steering angle with the base. The disadvantage of this method is that your powers will increase and change significantly.

 

Only Reiza can answer that, but usually steering lock is linear, while sensitivity settings are logarithmic/exponential.

Edited: and steering lock does change the max amount of angular travel while sensitivity keeps it

 

You know, since I've had a VNM base, it's become even more irrelevant to me whether it's linear or not, or whether it needs filters or not. And who tells me what linear is or should be? Or how linear should feel, like in Ams2 or like in another simulation? Nobody can tell me that. If it feels better for me, then I'll do it or change it. Whether other people feel the same way or not is beyond my control. I share it because it's true for me personally, nothing more, nothing less.

 

My answer has incomplete at first and still incomplete (must be something age related ).
Let's try again:

• Steering Lock: defines the maximum angle the wheels will rotate at full lock (Monaco and Spa may require different values)
• Sensitivity: changes the steering responde curve from linear (50) to something logarithmic (>50) or exponencial (<50) (it's a matter of personal taste)

 

Yes, I agree with you on almost everything. But only almost There are things that aren't personal taste; there are things that make it better

 

Based on this and previous comments, I'm beginning to suspect there is something more going on with regards to your wheel or FFB setup external to AMS2. There should be zero latency in regards to steering and FFB response, and you shouldn't need to adjust steering sensitivity outside of 50% to compensate for this.

It would be interesting to hear what your rig setup is - Operating System, which DD bases you tested with, what your hardware setup is, etc., basically any external factors that could influence why you're having latency issues with FFB because that clearly shouldn't be the case.

 

Everything's fine with my hardware. I'm not talking about increasing steering sensitivity, but rather reducing it. No problem, you can stay at 50%, I won't

 

I was planning to try this properly, but it's not usable on VR where you need to see your wheel in the game as the rotation get's so out of sync with how much you are actually turning and what you see.

There was a strange thing though, when I changed the steering linearity value to 25 and went onto track, I got impression that let say 5 degrees of physical rotation is a lot more degrees of wheel rotation in game visually as if lowering the value would made the steering more sensitive and not less. I was on VR, so I had to change the linearity back immediately.

 

Yes, of course you can't use that for VR. I'm talking about the non-VR mode.Maybe you have the option to test it without VR. And yes, you noticed correctly: reducing the steering sensitivity makes the steering more precise, faster, and more progressive, and it adapts better to the visuals.I've reached a value of 25-30%. If you take an open-top Formula car, you'll notice in the garage that 50% steering sensitivity makes it very sluggish and slow to steer.Too slow for my taste, and also too slow compared to other simulations. I don't like to compare, but in situations where it doesn't work for me, I use this option. As I said, for me, reducing the steering sensitivity works better (without VR), maybe not for others. But I take the liberty of making it better for myself

 

In my view, that's a graphics rendering bug.
Decreasing sensitivity makes the virtual wheel turn less than the real one. The opposite is also true.
Yet, on the screen, the effect is reversed.
I've comment on that some times in the past. While driving, it doesn't affect me. What affects me is that such a simple thing doesn't get corrected.

 

Yep, I thought it could be graphics bug and not apply to physics, I just didn't have time to actually try driving the car to see if it was the case.

 

Have you tried adjusting steering lock in the car's setup? I don't know about you but my steering wheel does exactly what I make it do, there's no lag or 'sluggishness' unless it comes from me.

Also, when you calibrate your wheel, did you correctly hold the wheel at 90 degrees in the second part of wheel calibration?

This is sounding less like a 'physics' issue and much more like personal preference with car setup and feel. And it feels like you keep moving the goalposts.

 

Finally tried this properly. Drove the BMW gt3 gen2 on Laguna Seca bunch of laps and flipped the linearity between 25 and 50 to be sure I'm not having some placebo effect because the change of feel thru the wheel was subtle, but it was definitely there and a improvement. It reduces the slightly "mushy" and soft feeling and FFB felt more connected to road.

The thing I was more surprised was the amount of steering required to make the corners didn't feel to have changed much at all. It's like on the default linearity, the game tricks you to apply more steering input that is actually needed (from the telemetry view, I could clearly see that the steering input was much more on the 50 than with the linearity of 25 even though physically I didn't notice clear need to turn the wheel more).

not sure what is going on, is the 50 actually even linear at all? They would definitely need to fix the rotation on the screen for this to be universally usable. I really hope Reiza devs would actually take a look and investigate this as whole (including fixing that bug, but also how the steering linearity setting has been implemented).

 

I've just tried the same test and honestly, couldn't pick any tangible difference between steering sensitivity and changing the steering lock angle. I even played around with camera settings etc to try and reproduce the "boaty" feel and just couldn't figure out how to break it. This is in the Porsche Gen2 GT3 at Brands Hatch. Did several laps with a bunch of different configurations.

The only thing that really made any difference was switching to the bonnet camera, but surely nobody is using that as their reference point.

Relevant settings (for direct comparison with everyone else)

Camera settings are:

High Speed Shake = On
Head Movement = 80
Legacy = No (also tried on, was a little weird but adapted within a lap)
G-Force Lateral = 10, Long = 25, Vert = 10
Show Helmet = Yes
Helmet Depth of Field = No
Last 3 settings = 0

FFB
Type = Default (not Default+)
Gain = 60
LFB = 0
FX = 20
Damping = 0

If anyone experiencing the boatiness/floatiness can post their versions of these settings I would love to try and replicate it to see what's going on

 

I changed my recommendations yesterday (excluding VR) in my files. If I sense something is wrong, I'll pass it on. Whether people use it or not is up to them. The reduction brings me some advantages personally. More pressure, more center, and more stability on the steering axis. The connection between the car and the track increases, and thus also less "boat feeling." I'm not saying the "boat feeling" disappears, but it's definitely reduced. As I said, it's definitely an advantage for me. Or to be more precise, there are several advantages to this.

 

@SpaceYam You are right, but you may not been looking in the right direction
Take the graph below (x= 0 no steering input, x=1 full steering lock):

Red Line: you have a 18º steering lock (x=1 => y=18º), sensitivity = 50 (linear)
Blue Line: you adjusted sensitivity to a value >50. You still get x=1 => y=18º but now all the smaller values are amplidied (more sensitivity to steering input)
Green Line: You have only adjusted the steering lock to 24º (sensitivity = 50). As you can see, there’s only a small difference compared to the blue line across most of the usable range, but as the input increases, the difference from both the red and blue lines becomes noticeable. Now the full steering lock reaches 24º

 

Ok so from a technical perspective, yes that all makes sense.

However, to quote DanielKart:

From a practical point of view there was no difference in the way the car "felt" aside from less movement in the steering wheel. No difference in lap times. No "boatiness." I couldn't even perceive any difference in the FFB itself - I can only assume a more complex FFB profile with more information will feel different but I dislike too much information in the wheel, hence why I use Default and not Default+.

And on this point, the game needs to be compared to real life, not other simulations. Too many people are caught up on "It's not the same as AC/RF2" when in reality those simulation models have flaws too - they might have been gold standards in their times, but times change.