SimHub ShakeIt Bass Shakers custom 4 corner tire slip

[blekenbleu]

Spoiler: Background and disclaimers

My sim driving is Assetto Corsa with an office chair and wheel stand.
Without a harness keeping one's back contacting them,
back transducers seem of arguable benefit.

Already using SimHub,
having unused motherboard 7.1 sound and obsolete Denon surround receiver,
a ShakeSeat pad (no longer available) was ordered because:

• Sim Racing Garage review
• Dayton TT25-16 pucks were, at the time, in short supply elsewhere

An equivalent can be made using Dayton pucks
with e.g. 40cm square zippered cushion covers and EVA foam floor tiles.

SimHub's ShakeIt tire slip effect does not consider tire loading.
For example, an inside tire may be fully unloaded during cornering,
with very high slippage, but no perceived sensation in actual driving.
Haptic slip effects ideally change frequency based on slip
but change amplitude based on load,
which was impossible using only earlier SimHub support.

This SimHub profile attempts to approximate track sensations from street tires,
when securely harnessed in a fitted racing seat.
Usefulness of such haptic feedback may vary widely among users:

• e.g. Niels Heusinkveld indicated that seat feel is overrated
  and he cued more visually off horizon movement

This haptic wheel slip feedback was IMO less useful
than deceleration and cornering simulation from harness tensioners,
but reduced a sense of "something missing".

Unambiguously directional haptic cues
wants direct stimulation of body parts,
e.g. left and right thighs, lumbar and/or sit bones.

Original 4-puck SRS ShakeKit cushion is no longer available,
but equivalent DIY PuckSeat assembly is not very hard.

Sensations to be simulated here:

• squeal for small slippage, transitioning to moaning with more slippage,
• then shuddering near breakaway,
  with intensities proportional to tire loadings

Implementation augments SimHub Wheels SLIP effect
with Custom Effects

1. latitudinal and longitudinal acceleration properties
2. loaded tire slips, products of tire slip properties times accelerations

Again, because load modulates slip in telemetry,
heavily loaded small slip could not be generated
as strong high frequency tactile effect
until SimHub added Forced frequency option.

Accelerations were originally estimated from speed and yaw changes,
serving as G force proxies.
Earlier SimHub versions sampled speed and yaw
at a lower rate than Assetto Corsa telemetry waas updated,
resulting in some changes 2x others, which code here detects and mitigates.

Loaded tire slip corner effects use products of tire slips and G force proxies:

A complete SimHub ShakeIt profile is available on GitHub.
It is now IMO obsolete.

SimHub slip telemetry in this profile is not portable among all games;
specifically, iRacing provides no slip telemetry.

An upgraded ShakeIt profile based on slip/grip ratio is recommended:

• requires less car/track/sim-specific tweaking
• delivers haptics better correlated with tires reaching grip limit,
  based on slip/grip ratios.
• Improved grip estimation combines lateral and longitudinal
  and is modulated by heave

 

[blekenbleu]

ACC is maybe not the best usage case example neither as we do not see in this example any excessive independent wheel activity.

SimHub's WHEELS SLIP effect has much lower output for ACC than other games,
at least in @dMASS' AMRV8 GT3 Suzuka recording.
As usual, differences between left and right slip are typically less than between front and rear,
but ACC has 4 independent channels, unlike iRacing:

Here are corresponding front and rear loaded slips, from current candidate formulae:

 

[Deleted member 1449502]

SimHub's WHEELS SLIP effect has much lower output for ACC than other games,
at least in @dMASS' AMRV8 GT3 Suzuka recording.
As usual, differences between left and right slip are typically less than between front and rear,
but ACC has 4 independent channels, unlike iRacing:
View attachment 502532
Here are corresponding front and rear loaded slips, from current candidate formulae:
View attachment 502535

The response looks far more dynamic, can't wait to get my hands on it!

 

[Mr Latte]

SimHub's WHEELS SLIP effect has much lower output for ACC than other games,
at least in @dMASS' AMRV8 GT3 Suzuka recording.
As usual, differences between left and right slip are typically less than between front and rear,
but ACC has 4 independent channels, unlike iRacing:
View attachment 502532
Here are corresponding front and rear loaded slips, from current candidate formulae:
View attachment 502535

Yes but the differences in the channels is not always that great, I have looked at many recordings.
What I will do is turn off the Auto-Calibration and compare that at a later point.

If comparing your effects to standard, I can use the meter level tools to determine how to balance both. I showed this in one of the other videos for the LUFS. The input gain on the response curve combined with the effects volume.

Simhub should be designed to cope with the differences in various titles.
I believe that is why we are given these controls.

I can see how those alter the output. It's easy to also duplicate an effect to have eg: 8 variants and then using different settings for each, then send those to different outputs and compare the same effect with multiple settings via the monitoring tools.

You can show and look at Live output all day if you want, what matters is what the tactile transducers receive in tone generation and can use from that.

 

[blekenbleu]

Simhub should be designed to cope with the differences in various titles.

SimHub auto calibration works fairly well for e.g. Acceleration properties,
on which @Wotever's experimental-belt-tensionner depends.
Here is something @Wotever wrote about slip and automatic calibration:

In fact most effects values are speed unit (or kind of (speed of the suspension, speed of the slip on the ground ...) So there is no maximum value out of box, so automatic calibration Indeed learns from the telemetry what is the boundaries
Imagine you have incoming values going from 0 to 500, and for instance 99.9% of the values never goes over 80, and very very very rarely you have a peak at 500 (this will vary from game to game, track to track car to car ... ) Then simhub will pick 80 as the actual maximum instead of 500, this will ensure to get a good contrast between the higher effects and the smallest.

... but maximum slip values will obtain for e.g. drifting or uncontrolled spins,
while spin values for peak grip want different calibration.

FWIW, @Wotever wrote this on discord about slip in iRacing on 09/21/2020
"iracing slip is the only one I didn't changed, as there's no slip data"
" It's a basic RPM vs Speed algorythm as Iracing does not gives any wheel telemetry"

 

[Deleted member 1449502]

This is really interesting. I never really gave auto calibration a chance, but after reading this I will be spending some time making comparisons

 

[blekenbleu]

spending some time making comparisons

I have observed telemetry changes seemingly resulting from auto calibration,
but do not know how to reset SimHub telemetry to uncalibrated.

 

[Mr Latte]

I have observed telemetry changes seemingly resulting from auto calibration,
but do not know how to reset SimHub telemetry to uncalibrated.

I believe via the Calibration tab on the right-hand side or when saving a profile you can slide it between "Pre-Defined or Automatic"

It does not appear we can apply this to only specific effects or indeed layers, so personally, I prefer to stick to "Pre-Defined" and manually tune effects.

 

[blekenbleu]

You can show and look at Live output all day if you want, what matters is what the tactile transducers receive in tone generation and can use from that.

I recorded SimHub tactile output for FrontLeft channel of my candidate 4 corners loaded wheel slip effect as a .wav file,
then used Audio-Spectrogram, a Windows Python script, to render that as a spectrogram.
The clip is 3369 to 3812 of dMASS' 10 Sep 2021 amrv8_vantage_gt3@Suzuka Circuit:

..resulting in this spectrogram:

Given that higher frequencies are configured for smaller loaded slip values and vice versa,
this result is as expected.
Since SimHub presumably employs square waves, which have strong odd harmonics.
appreciable energy at 3x and 5x fundamentals is unsurprising.
I have yet to find a quick and painless way to make .wav recordings,
but rendering ShakeItFrontLeft6_1.wav to a spectrogram was easy:

 

[blekenbleu]

I have yet to find a quick and painless way to make .wav recordings

Renee Audio Recorder Pro (Trial) is quite satisfactory.
Trial version is limited to 30 second clips, more than sufficient IMO for spectrograms.
It records directly from a selected output device (no need for an audio capture card)
e.g. whatever sound card SimHub uses.
Output card gets selected in PC Sounds Test

However, despite selecting Mono (under Settings->Files),
it seemingly mixes all active channels on that device.
Otherwise, simple to configure and use.

 

[blekenbleu]

I recorded SimHub tactile output for FrontLeftchannel of my candidate 4 corners loaded wheel slip effect as a .wav file,
then used Audio-Spectrogram, a Windows Python script, to render that as a spectrogram.

That Python script generated spectrograms to 20kHz;
an Internet search yielded a hit for low frequency Python spectrograms,
using librosa (which required installing many Python packages) to simplify,
so I whacked this:

Spoiler: custom Python spectrogram script

# https://www.earthinversion.com/utilities/efficiently-compute-spectrogram-in-python-using-librosa/
import sys
import librosa
import librosa.display
import matplotlib.pyplot as plt
import numpy as np

def plot_spectrogram(audio_path, argv = ''):
    # play with hop_length and nfft values
    hop_length = 128
    n_fft = 2048
    y_axis = "log"  # linear or log
    fmin = None
    fmax = 2400.0
    cmap = 'jet'
    bins_per_octave = 22
    auto_aspect = False

    (data, sr) = librosa.load(audio_path, sr=None)  # no resampling

    # Plot spectrogram
    D = librosa.amplitude_to_db(np.abs(librosa.stft(data, hop_length=hop_length, n_fft=n_fft)), ref=np.max)

    fig, ax = plt.subplots(figsize=(8, 3))

    img = librosa.display.specshow(D, y_axis=y_axis, sr=sr,
                               hop_length=hop_length, x_axis='time', ax=ax, cmap=cmap, bins_per_octave=bins_per_octave,
                               auto_aspect=auto_aspect)

    ax.set_ylim([fmin, fmax])
    ax.set_aspect(0.005)
    ax.set_title('loaded slip spectrogram')
    fig.colorbar(img, ax=ax, format="%+2.f dB")
    plt.xlabel('Seconds')

    print('Graphic interface...')
    plt.show()

    plt.clf()

    return

plot_spectrogram(sys.argv[1], '')

.. and a simple .bat file to enable drag and drop tactile .wav files from Renee Audio Recorder:
python e:\my\Audio-Spectrogram\spectrosa.py "%1"
.. yielding e.g.:

Hmm, if that spectrogram is accurate, the High effect (lower Hz) limit should be raised...

 

[RangeyRover]

Not wanting to drag discussion away from the OPs own solution so please say if wanting these posts removed or if intruding?

I re-ran things, still game audio coming through Simhub. (Part Of Your Approach?). So what I did was to disable ALL GAME AUDIO via game menus to determine only what your custom effect is doing regards just tone generation with the slip.


Perhaps I need to do something else, but so far this is my findings using your effect with it set to Corners. I have it outputting on 2x large BK for front wheels and 2x TST 429 for rear wheels. Not ideal for the test but its just to get feedback for all 4 wheels.




Audi RS5 DTM 2020 / Suzuka

Example 1:
Auto Calibration Bug?
Here the car is stationary, yet look at the activity, all wheels are outputting feedback, so it appears the sensitivity is far too high or not working correctly with this car at least? Any slight movements in steering, seem to present more of a surface/contact sensation, rather than more useful feedback of when slip is happening or increasing.

Example 2:

Here we see in a spin all 4 wheels output to the higher Hz set for the effect. So what we have representing the tyres and mass of a whole car spinning is very little sense of true depth in felt low frequencies. I understand you have primarily entry-level transducers for the corners but herein lies the problem you face.

How do we generate, more authentic slip effects if using 44Hz - 64Hz? Offering 20Hz of range (including, using the best of the budget transducers abilities) but really producing limited felt sensations with lack of depth and true variation of the felt sensations from the effects lowest-highest telemetry values?

I do see independent wheel operations which are good and maybe needing some work to help improve the sensitivity issue? Although part of that may be to do with the (Auto Calibration). Also, assuming I am using this correctly? However, I would say if referring to, the "quality of the feeling" being generated its a bit meh for me and I desire more authentic slip sensations giving greater range and much more satisfaction. It depends to what you are used to I suppose.

One query I have, is that users will also apply similar Hz to other effects and how then does a person distinguish "actual slip" activity from those other effects (when or if) we apply standard approach with simplistic single layers to represent effects?

My Own Perspective?
I'm keen to work with people that can bring advantages with how the telemetry is used but I do not think there is much desire for specific car profiles being needed. Possibly for engines yes, not however regards slip as it's making things too complicated. We just need effects to work well with each sim and then apply the "effect gain volume" appropriately to suit users' preferences and variations in different sims or for specific cars.

I commend the work being done but I think 4 wheel installations are not the way to go and people are better seeking to buy into the more technically accomplished hardware with front/rear installations combining TST/BK . This will achieve a much better level of immersion in quality and range of frequencies. Having that is much better than having 4 channels with independent feedback but limited by the points highlighted.

However, I am also interested in what would be possible with effects that offer excellent individual wheel feedback on an installation with higher quality hardware and applying multi-layers to better represent the effects generated sensations.

So Im still keen to take on what you guys are learning/sharing regards the telemetry activity but apply to that my own experience or approach in creating the effects felt sensations.

Please try with the F3. It's not setup for the DTM2020
Which are tail happy nutters
I just tried it with them. The effect is horrible.
I'm afraid that responses and profiles have to be tuned by car.

 

[RangeyRover]

To me, I am a bit perplexed as it appears you're just maybe trying to be over technical on the least technically accomplished hardware.

I think you are missing the point of trying to get detection of events. Each persons rig/budget and hardware determines Audio out. Detection is a constant. Audio output and waves are specific.

 

[blekenbleu]

if that spectrogram is accurate, the High effect (lower Hz) limit should be raised...

tweaked:

Here are both front left and right loaded slip tactile effects:

Raising High effect Hz from 30 to 60 sounds better thru home theater subwoofer..

 

[Mr Latte]

Please try with the F3. It's not setup for the DTM2020
Which are tail happy nutters
I just tried it with them. The effect is horrible.
I'm afraid that responses and profiles have to be tuned by car.

Yes I will get a look at it sometime.
Can you show us the differences each car makes in the generated output?

If applying this approach. I wonder how many people want or can be bothered with needing to tweak/tune responses and settings for each car? Sorry it just seems a lot of hassle, not for me.

Standard effects settings are working fine for most people, we can, of course, get more creative with how we apply different layers, so yes I gotta question if it's worth it all and I cant say I was like WOW with the approach your taking.

Each person to their own thing I guess.

 

[blekenbleu]

I wonder how many people want or can be bothered with needing to tweak/tune responses and settings for each car?

Yes, that is an issue for slip, unlike e.g. RPM and acceleration effects,
which SimHub better equalizes among games.
Literally, people who want or can be bothered may be
about the same as those who work on car setups.

The extent to which your campaign for standardized transducer installations minimizes a need for tweaking audio settings cannot compensate for evident variations among cars and games' slips, even ignoring iRacing's lack.
What we can attempt, without changes in SimHub:

• minimize the number of telemetry settings that need to change among games/cars/tracks
• clear and repeatable procedure for adjusting those settings
• for those without RaceBass, a repeatable procedure for adjusting SimHub Frequencies settings, independent of telemetry

 

[blekenbleu]

I whacked

This Python script and .bat file are now on GitHub.

 

[Mr Latte]

Yes, that is an issue for slip, unlike e.g. RPM and acceleration effects,
which SimHub better equalizes among games.
Literally, people who want or can be bothered may be
about the same as those who work on car setups.

The extent to which your campaign for standardized transducer installations minimizes a need for tweaking audio settings cannot compensate for evident variations among cars and games' slips, even ignoring iRacing's lack.
What we can attempt, without changes in SimHub:

• minimize the number of telemetry settings that need to change among games/cars/tracks
• clear and repeatable procedure for adjusting those settings
• for those without RaceBass, a repeatable procedure for adjusting SimHub Frequencies settings, independent of telemetry

Yes already we have had work done on determining which effects in each sim need perhaps different settings. It needs more of a deep dive but this is something I will use the beta testers for, giving feedback from their own preferred sims/titles.

People are not in the habit of seeking to have dozens of profiles for a single sim, then dozens more for another sim just to cover their favourite driven cars. Ask anyone now that has been into tactile for many months or years.

They generally make do with a few changes. So my position on this, is that if certain "custom effects" need such specific tuning then it overcomplicates things and my point that this approach is then being taken with people mostly with budget/entry-level hardware (hence 4 corners installations). So these more complex effects are then being greatly restricted by the limitations of that hardware and the potential in what they can accomplish in felt sensations is still very limited. Like I said being over technical on the least technical hardware.

What I can tell you is that the feedback I have gathered so far, is that very few people feel the need for per wheel slip. They are much more focused on having a "higher quality of tactile effects" than per wheel sensations. This is even before we take into account the issues with installation and crosstalk that are common problems with 4 way installations.

I have used your own effects so far on the best hardware, (TST 429 & LFE per corner) with some adapted effects layers and I do see potential with it, but I'm not willing to have to rely on specific custom effects for dozens of cars just to gain the slight benefits per wheel sensations might offer. Not when we can still achieve very good slip sensations based on front/rear in mono.

Sure I am still keen to see how you improve things and appreciate your dedication....
We have different perspectives which is okay but I can't ignore the reactions people are having with the multi-layer effects and using better quality hardware to bring more enjoyable and immersive tactile.

It's like the opposite, we want the best or more advanced hardware to be used but with the simplest way for people to fully enjoy its potential on various sims or cars with as little fuss as possible.

 

[Tom_Hampton]

What I can tell you is that the feedback I have gathered so far, is that very few people feel the need for per wheel slip. They are much more focused on having a "higher quality of tactile effects" than per wheel sensations. This is even before we take into account the issues with installation and crosstalk that are common problems with 4 way installations.

I agree with this as well. F/R is much more interesting / useful to me. I can pretty well tell inside / outside from the situation, 90% of the time....and visual/audible cues. Which is the same as in a real car.

 

[RangeyRover]

What I can tell you is that the feedback I have gathered so far, is that very few people feel the need for per wheel slip. They are much more focused on having a "higher quality of tactile effects" than per wheel sensations. This is even before we take into account the issues with installation and crosstalk that are common problems with 4 way installations.

I have used your own effects so far on the best hardware, (TST 429 & LFE per corner) with some adapted effects layers and I do see potential with it, but I'm not willing to have to rely on specific custom effects for dozens of cars just to gain the slight benefits per wheel sensations might offer. Not when we can still achieve very good slip sensations based on front/rear in mono.

Sure I am still keen to see how you improve things and appreciate your dedication....
We have different perspectives which is okay but I can't ignore the reactions people are having with the multi-layer effects and using better quality hardware to bring more enjoyable and immersive tactile.

For me, properly setup it's worth half a second a lap, especially in open wheelers.
I'm not sure why my experimental work had been compared and found lacking?
especially to a car I didn't advise it for.
I'll continue to work on this for my benefit.
@blekenbleu thank you for sharing. for me your approach was far superior to simhub.

 

[blekenbleu]

properly setup it's worth half a second a lap

Yes, it's hard to convince others of the value of approaches that they misunderstand and disdain,
but I also do not doubt that a F/R "RaceBass" approach is rewarding for its advocates,
and I feel somewhat obliged to @Mr Latte for his pioneering and ongoing efforts.
Consequently, I will continue to fiddle with front/rear loaded slip telemetry solutions,
despite lacking an appropriate tactile configuration.

The 4-puck ShakeSeat solution is IMO both satisfactory and elegant;
consequently, it will continue to be my primary tactile solution for wheel slip.
Minimizing the amount of tweaking required for car/track/game loaded wheel slip setups
seems likely to be a journey of discovery, AKA process of illumination.
I suspect that telemetry tweaking for 4-wheel and front/rear solutions will be similar,
so long as both are deployed with the goal of exploiting traction limits.