Simhub - bass shakers Wheelslip

No, in the profile I upload I use a lot of effects, and can feel all. I don't use Simulated Road Texture, Speed/Wind effects and Lateral G-Force only, and specific IRacing things.

And I disabled autocalibration, because I don't like it: soften a lot of effects here, at least in this SimHub build.

 

Cool, good to know

 

I still think there is no good wheelslip simulation. None of the proposed effects in this topic work flawlessly. The Rnageyrover custom file works okay, but it depends way too much on cornering forces, which makes everything vibrate even under perfect grip conditions. Try driving a formula ultimate, endless grip, but my bass shakeres start vibrating excessively in corners.

Actally the standard Wheelslip works perfect for this formula car, but then you try it in a porsche cup and it starts vibrating excessively in corners with grip as well.

I'm afraid something really is broken in AMS 2 telemetry.

Also the other proposed solution with the plugin has the exact same issues.

I will try and seen coming weeks if I can tweak something together, but I'm afraid it really is an issue with AMS2 itself

Edit: I was mistaken. The Viper plugin version does work. After a lot of testing, I will stick with it. It works very well. My custom effect was not set up correctly.

 

I agree with @Shodan0000 assessment of my effect. I don't drive open wheelers with high aero very much, so it's pretty much untested in those areas. I think I need to do a version that has controllable sway and surge input, so that can be tuned for the users requirements. I will say that in my opinion you have to tune shakers a bit per car, so if you are changing cars often that can be a PITA.

 

Hope you didn't see my post as criticism, I really like your settings and appreciate the effort. I also don't drive open wheelers, but last weekend a friend wanted the formula 1 experience So we drove Ultima on Interlagos.

Indeed with tweaking I'm sure it can become a perfect custom effect, although for now the viper mod already works satisfactory.

 

Not at all, it matches observations others have made, and it encourages me to make it adjustable. I'd love to make it automatic, but that seems maybe a step too far.

 

Could you elaborate on your settings for Viper? I followed the previous guide posted and found the bass shakers only come on when spinning the tires in grass or when the car completely loses grip and spins.

RnageyRover's solution seems to perform much better. The shakers come on early, but it at least gives some indication there is going to be a problem soon. Viper only seems to turn on using the posted settings after the problem has already happened or if you are just smoking tires.

 

Fact is viper plugin provides accurate and immediate values for tyre lock/slip.
Since I don't use bass shakers I do not know how to integrate it with viper plugin.

Another fact is (and I've been very repetitive about this) that raw data TyreSlipSpeed is listed as obsolete by Reyza and from what I could test it gave no correlation with longitudinal tyre slip or lock.

I can see a very interisting formula as used by RnageyRover, and I would suggest him to introduce viper values in the place of tyreslipspeed.

I've heard that another option (not so accurate as viper but functional) is to add the internal tyreslip effect introduced recently by Wotever in Shakers

 

Viper will not activate when tyres are not slipping, rnageyrovers file does, the issue with the latter is that it is unpredictable and some cars vibrate too soon and too much.

You can tune ofcourse the sensitivity of viper effects, but it is made to create feedback when tyres slip, not before. So great for immersion not so great as warning.

 

After messing with it over the last few days, I came to that conclusion as well. I am working on implementing something that uses slip angle. Right now I have a crude formula that calculates the car's slip angle and it looks promising. I'd really like the bass shakers to activate when the tires start to scrub.

I think for it to work really well, the slip angle needs to be calculated for each tire. This seems to be non-trivial since it is highly dependent on tire geometry and the values needed aren't in Shared Memory. I have found simplified formulas, but even then I need data like tire width and aspect ratio.

Edit:
If anyone wants to give the slip angle setup a try here is the Java Script for a custom effect:

Code:

//Slip Angle

var lower = 3
var upper = 8

var slipAngle = Math.atan(Math.abs($prop('GameRawData.mLocalVelocity01'))/Math.abs($prop('GameRawData.mLocalVelocity03'))) * 57.3

if(slipAngle > upper){
    slipAngle = upper
}
if(slipAngle < lower){
    slipAngle = lower
}

normalizedAngle = (slipAngle - lower) / (upper - lower)

if($prop('GameRawData.mSpeed') > .1 & Math.abs($prop('GameRawData.mLocalAcceleration01')) > .5) {
    return normalizedAngle * 100
}

The lower and upper set the sensitivity. The values here seem to work across a handful of cars. I have attached a picture of the custom affect settings as well. A known issue is the shakers coming on at very tight low speed turns. Better integrating velocity, acceleration and angular velocity would probably solve the problem.

 

The main goal of my plugin is to detect locking or spinning wheels caused by too much braking or throttle inputs.
And with the included Overlay you have a direct feedback where you can see it.
The second benefit is for setup work: brake balance, ABS, TC, ARB, differential settings.

The question is what feedback a bass shaker should give.
My plugin eliminates most of the lateral slip, only in extreme steering angles or if the car completely slides sideways you have it in as a part.
That means with my plugin you get only feedback for locking or spinning wheels and not for understeering or oversteering caused by too much steering inputs.
Extreme understeering for example can work with the plugin, because at some point the tyre rotation speed is slowed down by the high slip angle. But I think that is not really useful.

Further for having bass shaker feedback on lower slip values of my plugin, then that is in my opinion something for the response curve.
With my plugin you have a range for locking wheels between 0 and 1, in percent 0-100%.
If you can feel the bass shaker at higher output values only, then you have to change the response curve that for example 5% from my plugin producing 80% output to the bass shakers.
But that is only my theory, because I have no bass shakers

 

I think my expectations were wrong for the plugin and that is on me. I was hoping it was going to help determine any type of wheel slip including lateral slip. However, as you said, it is primarily for feedback when the tire is locking or slipping due to no TCS or ABS. For that, it works perfectly. It is useful to give the sensation of burning out tires or locking the brakes. I am working on implementing a script that will indicate lateral slip in oversteer and understeer situations. I think the combination of the two will give good feedback with bass shakers. I am also putting together overlays to indicate when you might be near the limit of the tire.

 

Using the Viper plugin, I get rumble in AC, ACC and GT7 but not in AMS2. Using the demo dashboard, in AMS2 the Ratio VelX/Speed stays at 0 but the other 3 boxes are working. Anyone else run into a similar issue?

 

I just tested my plugin yesterday with AMS2 and it was working.
Which versions do you have (AMS2, SimHub, my plugin)
I tested with the current AMS2 Beta, SimHub 9.1.12 and my plugin 1.4.3.
And then can you please search for "viper" in the available properties in SimHub and check if there are tyre diameter values and if yes, check if they can be correct, the values are in meters.
Further the plugin uses the properties mLocalVelocity01 from raw data and SpeedKmh.
Then you can check, if these properties in SimHub have correct values while driving.

The ratio for VelX/Speed is normally small (below 0.1 most time), the default limit for detect tyre diameters is max 0.001.
The ratio is only needed for the tyre diameter detection phase. If the tyre diameters are detected, they are saved to a file and the ratio then normally never needed again.

 

Thanks for the quick reply. I can confirm I am using the same sw versions for AMS2, SimHub and the plugin.

As for the SimHub values, SpeedKmh has data but I don't have mLocalVelocity01, only sLocalVelocity01. The tire diameters are listed but don't have values, just a "-" currently. In the demo dashboard, none of the boxes turn green, they are all red (though throttle, brake and speed are active and have bars/numbers populated in the dashboard).

I used the .json from the .zip which is pre-populated with a lot of AMS2 tire information. However, I also renamed it to see if maybe that caused any issues but it was re-created afterward and nothing changed.

 

This is the complete data file for AMS2 v1.5.3.1build2438

 

This should be the same file as in the plugin release, because I added your file there

@TronLi
Is it possible that you selected the wrong game in SimHub, because sLocalVelocity01 is no property of Automobilista 2?
Or do you have other plugins installed?

 

Could you verify?
The installation data file is 3KB while this one is 17 KB

 

I had SimHub set to use UDP for AMS2 instead of the default. I changed that setting and noticed that all of my replays went away, so I fired up the beta and it immediately worked.

sLocalVelocity01 turned to mLocalVelocity01 when I changed from UDP to the default SimHub settings.

Thanks for the help! Works great now.

 

What installation data file do you mean?
If you use the plugin without copying the data file from release, then there is no data file. The data file is then generated if tyre diameters will be detected and SimHub closed.
And in the latest release I added your file, I only changed the formatting like it is saved by the plugin itself and added an example from another game, but all your data is in.
The plugin itself writes further a settings file if you change the limits in the plugin settings.
Maybe you mean this file which has only some lines, but I think it has only 1 kByte.