Tactile Immersion - General Discussion - Hardware & Software

[Mr Latte]

Your massive contributions to tactile tech surely belie that.
Thanks for being the first (so far as I know) to sample this version 0.1 attempt.

Those are about as different from the 4 Dayton pucks in my ShakeSeat as imaginable.

Indeed, tactile is currently being used only for this plus off-idle and near-redline engine RPM.
I am still sorting measurements by piezo contact elements before deploying a Thruster for braking tactile feedback. All other non-visual feedback currently is by harness tensioning and AccuForce.

I have yet to dismantle my ShakeSeat to evaluate Dayton pucks, but measurements to date suggest
their Thruster is capable of neither responding below 10Hz nor sensed below 25Hz at best;
can no more imagine 3Hz tactile than a blind person can colors.
Practically, I was generally unable to control real cars with tire judder below 40Hz or so.

Thanks for the offer.
Based on what I have read, my HP Reverb would be unhappy with ACC's VR on my current PC.
Hardware upgrade prices should become less unreasonable when/if supply chains recover from lunar new year (May?). I dread feeling like a fish out of water, trying to cope with GT3 car demands...
It has been years since my reflexes could keep up with more than mildly modified street cars.

I dont have exciters up and running at the moment, so what I wanted to do was see primarily how your effects operated/responded. Ive spent a good couple of hours faffing about and well impressed with how these can be implemented for whatever transducer/exciter is being used.

Sorting the Hz that feel good on different units isnt that hard but Ive been keen to get a lovely floaty weight sensation and its really nice around the 2-4Hz on the bigger LFE.

I noticed:
Even with all wheels placed to ouput as mono on a single unit the response is still good.
I will have to look in more depth with spectral analysis but one of the effects seemed to be doing the opposite for L/R

I mean increasing the Hz Vs Decreasing the Hz for the two values used.
Will give a deeper overview at a later time when I have looked more into these but I think this while needing some tweeks could be one of the best custom effects I have tried in a while.

Excellent work

 

[blekenbleu]

one of the effects seemed to be doing the opposite for L/R

For the first custom effect (the one that exports proxy_G property),
Left/Right is a misnomer, since left is lateral acceleration and right is longitudinal.
Also, because both positive and negative accelerations are wanted,
but SimHub constrains values to between 0 and 100, then 50 corresponds to zero acceleration.
Consequently modifying any of its response filter settings impacts front/rear and left/right symmetries.

For channels in the second custom effect, I am trying to image how summing to mono must feel,
since frequency would be dropping for one corner while increasing for e.g. diagonally opposite,
based on estimated weight transfers, supposing equal slippages.

As you will have noticed, quite a large white noise frequency modulation effect is applied,
to avoid hearing single tones for tire squeals.

I am of course pleased that you find it worth bothering in such a different than designed configuration.

 

[Mr Latte]

If possible it would be good if you can separate these into individual Lateral & Longitudinal effects
This would give more control over what settings and frequencies we then apply to each.

Your explanation highlights why I was finding a different feel from the steering.

 

[blekenbleu]

individual Lateral & Longitudinal effects

If I understand correctly, that might be two additional Custom Effects:
1) average right and left slip properties for each axle, multiplied by weighted longitudinal acceleration
2) average front and rear slip properties on each side, multiplied by weighted lateral acceleration.

Is there a way to uniquely name custom effects? I have not yet found it...

 

[Mr Latte]

I think you can only add a description to each layer but no I don't think you can change "CUSTOM EFFECT"

Heres what I would like operating similar to what you have.

4 Wheel/Corner Based Effects:
Lateral G
Longitudinal G
Wheel Slip

Then a Mono based version for Both Wheels or Side
Lateral G
Longitudinal G
Wheelslip

 

[Mr Latte]

I turned one of your effects Threshold upto 51 to stop it operating on idle.
Not sure if thats related to ACC

Can do nice idle with engine response curve so no need for other effect operating with idle

 

[blekenbleu]

I think you can only add a description to each layer

What is a layer? Where are they defined, applied or controlled?

no need for other effect operating with idle

This reminded me that Engine RPM effect in AC is no longer active for me when cars do not move.
This is not wanted, but I do not know how to enable RPM effect when not moving.

a Mono based version

I apologize for the delay in responding, but my brain was locked into wrong thinking
and needed to sleep on it. Here are a seemingly clearer (to me) concepts:

• At one extreme, e.g. spinning on wet ice, there is high slippage but nearly no net force,
  although there may be high rate of yaw change, which is my proxy for lateral acceleration.
  IRL there may be no tire squeal or judder. This remains an unsolved problem.
• A better approach, even for current 4 wheel/corner based effects,
  translates lateral and longitudinal accelerations into corner loads
  • Corner loads are gravity + (aero downforce-lift) + square_root(lateral**2 + longitudinal**2)*center_of_mass
  • If a car center of mass were at pavement level, there would be no weight transfer,
    and (ignoring aero) tire vertical loads never change.
  • At rest (ignoring aero), each corner has roughly equal gravity load.
  • One or two corners may have increased weight from accelerations,
    which is implicitly transferred from diagonally opposite corners.
  • At the limit, one corner may have all load (e.g. braking while cornering).
• A mono signal should be controlled by the weighted sum of each corner load times slip.

By considering only positive corner loads (if negative, then car is overturning),
Custom weight transfer effect converts from lateral and longitudinal accelerations.
Positive-only values allow users to alter SimHub response filter.

Ideally, weight transfer should consider distance from center of car to center of mass.
For example, in a drag racer with mass concentrated near the rear tires,
weight transfers more readily than if mass were more centered.

 

[Mr Latte]

What is a layer? Where are they defined, applied or controlled?

This reminded me that Engine RPM effect in AC is no longer active for me when cars do not move.
This is not wanted, but I do not know how to enable RPM effect when not moving.

If a car is at a stop but the engine on idle then its RPM is still active at whatever the idle RPM value is for the car being driven. We just need to have an output for the RPM within that range., We can apply two "Layers" with different Hz to also alter the "engine beat".

"Layer"
Mmmm well it's a term some have used for applying multiple effects to operate together as a single effect. The "Layers" could have specific controls in thresholds or settings to operate for low or high telemetry values. An example would be generating a different response for a minimal slip to a secondary one for high slip values, or something like small bumps to larger bumps. Mainly giving each scenario its own felt character output rather than one single effect.

If we are using an installation comprising of a larger lower bass capable unit and also a smaller or more capable mid/higher frequency based unit. Then we can apply a "Layer" to operate on the more suited unit (depending on its role and Hz used). For some, layers depending on Hz used we may want it going to both units, as even if the mid/upper frequency smaller unit may struggle to deliver a lower frequency being sent, it will still generate its harmonics and with more detail than say a piston based shaker would.

So what we get is a more advanced effect that is generated on more than one unit, with layer components that deliver improved or more immersion. We can further control what each unit generates by an implemented crossover but we configure that so each unit blends well (if desired).

I apologize for the delay in responding, but my brain was locked into wrong thinking
and needed to sleep on it. Here are a seemingly clearer (to me) concepts:

• At one extreme, e.g. spinning on wet ice, there is high slippage but nearly no net force,
  although there may be high rate of yaw change, which is my proxy for lateral acceleration.
  IRL there may be no tire squeal or judder. This remains an unsolved problem.
• A better approach, even for current 4 wheel/corner based effects,
  translates lateral and longitudinal accelerations into corner loads
  • Corner loads are gravity + (aero downforce-lift) + square_root(lateral**2 + longitudinal**2)*center_of_mass
  • If a car center of mass were at pavement level, there would be no weight transfer,
    and (ignoring aero) tire vertical loads never change.
  • At rest (ignoring aero), each corner has roughly equal gravity load.
  • One or two corners may have increased weight from accelerations,
    which is implicitly transferred from diagonally opposite corners.
  • At the limit, one corner may have all load (e.g. braking while cornering).
• A mono signal should be controlled by the weighted sum of each corner load times slip.

By considering only positive corner loads (if negative, then car is overturning),
Custom weight transfer effect converts from lateral and longitudinal accelerations.
Positive-only values allow users to alter SimHub response filter.

Ideally, weight transfer should consider distance from center of car to center of mass.
For example, in a drag racer with mass concentrated near the rear tires,
weight transfers more readily than if mass were more centered.

JUDDER
We could apply a "Layer" that artificially generates a unique felt response for when the "high value" is at 80-90%
This gives the user a feedback sensation to highlight when they are close to the upper limit.

The more mathematical and science side of it I leave up to you.
Any options you offer, then they can be tried and tested.

The lateral and Longitidunal G from the effect you had felt active but needing to be separated.
Barcelona is an excellent track for testing G

Look forward to what you come up with but, having someone good with the code side of things for the activity of the effects, will let me work on trying different things with effects felt sensations.

 

[blekenbleu]

The lateral and Longitidunal G from the effect you had felt active but needing to be separated

So, what is really wanted, at least for capable mono transducers, are two separate effects,
to which e.g. separate tactile frequencies can be applied, then mixed to output(s).
That bothers me on the math and science side, but nevermind.

Thanks for provoking me to rethink my 4-corner solution, though..

It has finally occurred to me that 2 channels lateral and longitudinal bipolar accelerations can
be instead split into 4 channels of non-negative properties, at least 2 of which will always be zero,
so that custom effect's filter settings may be applied e.g. to compensate higher and lower G limits
without provoking asymmetry.

 

[blekenbleu]

4 channels of non-negative properties

That and new math help make 4-corner loaded tire slip more credible, particularly tuning
by adding gamma and reducing gain for acceleration properties as well as loaded slip effects.
This required corresponding changes to 2DOF harness custom serial device profile.

Now, on to hack out separate slip effects separately loaded by lateral and longitudinal accelerations...

 

[Mr Latte]

So, what is really wanted, at least for capable mono transducers, are two separate effects,
to which e.g. separate tactile frequencies can be applied, then mixed to output(s).
That bothers me on the math and science side, but nevermind.

Thanks for provoking me to rethink my 4-corner solution, though..

It has finally occurred to me that 2 channels lateral and longitudinal bipolar accelerations can
be instead split into 4 channels of non-negative properties, at least 2 of which will always be zero,
so that custom effect's filter settings may be applied e.g. to compensate higher and lower G limits
without provoking asymmetry.

Sorry for delay in response. I was struggling today with major migraine and sickness.

Will soon try that new 4 corner loaded/slip effect

Just my thoughts here, other effects that are directional can be utilised to operate down to a mono-based response for the general cars "chassis". Of course, then with what we are discussing. A response would be generated for L/R and ACC or DEC to the same unit.

A single unit can still represent a good sensation for these individual scenarios but having them as individual effects, lets us then apply their own operating controls and felt sensations. Also, to then apply if desired additional layers for representing conditions regards "low "and "high" values output by the telemetry.

With the Effect Spatialization
A user can define to apply the effects for the different possible options.

Corners
Left/Right
Front/Rear
Mono

So I think we want individual control for the effects but be able to apply them for different potential installations.

Limitations:
The drawbacks with limited frequency range transducers that most people own, with only @50Hz of good feeling range to use. Is that effects like these will be very active (if representing each). So how does a user apply felt sensations for these that produce their own character but also can then work with other desired effects like bumps, rpm, gearshift etc.

A single unit is limited to what it can also do, (less is more approach) but because of how Simhub is not limited to channels like Simvibe or other options. There is nothing stopping people even with budget transducers having additional channels for specific effect roles. Multiple mono installations are possible to experience more effects but having less effects all being output on the same channel.

I guess we can see how we manage that but I point this out to highlight why I strongly advise having operational use of the 1Hz-200Hz to give us more ability to better apply individual felt sensations for different effects.

Keen to see what you bring but quite happy to mess about if we need to try various ideas or potential solutions.

 

[BKLFE]

Some logical fallacies here. I did not come here to explain my lifestory and ask advice about it. In fact, I repeateadly mentioned I do not want to talk about it.

Don't you think someone who owns an apt does not know how to turn to soundproofing when faced with a situation like this? I have 200 lbs MLV on my door to shield from corridor noises. So the advice you think you are giving is nothing but pointing out the obvious.

I also know how it will go.
Next will come: Sell your house and go somewhere else.

Yeah.

I brush my hair.
I cook an omelette.
I go to mall.
I sell my house.
I buy another house.

Right.

Have you ever bought a house? Where is the guarantee that you will not come across another animal on the new one? Rinse and repeat?

I don't need advice from random people on the net who have no access to the facts and I don't need someone try to take a moral highground on something he does not know nothing about and he was not asked for in the first place. I just asked help with amps that he knows about and I did not expect such an a$$ remark.

This does not diminish all the contributions he has made on topic, but it also does not stop me from labeling him an a$$, that I did not outright do, but thought about.

Now if you care to hear about the story so you can judge properly, I am more than willing to do that. But I think that would be OT and disrespectful to OP and topic followers.

Besides, company responded, issue resolved. I also contributed, albeit small, by confirming the amp choice was right.

And update!

WOW BUTTKICKER! Thanks! Many many thanks for restoring our sanity!

The sound reverbating to the other side of the wall is crazy. And a killer when fed a 15 Hz sine wave.

I have 2 buttkicker LFEs attached to concrete ceiling. I have only used one until now.

I ran this only at the night time, feeding it low frequency sounds. No music. First week, the person did not get where it was coming from. But I could hear them waking up.

2nd week I have this on timer with a 15 Hz sine wave. Multiple 30 sec runs at different times of the night. When I checked, they always woke up as this makes jackhammer sounds. This is also when they realized I was making this and they went ape **** crazy with their noises. I don't really care as I am tormented many years. I have a million dollar view from my bedroom, but I sleep in some laundry room because of these animals. I kept on doing it.

Then they called building security and management on me. I kept on doing it. Then they called the police, I denied it and kept doing it.

Now 1 week I have peace again in a long time. They have stopped all noises and I have stopped buttkickers.

MANY THANKS BUTTKICKER!!!!!

Some people are cowards. They will run at the first sight of confrontation even when they are right. Some people are assholes, they will judge others in their condescending tones. You can't fight a psychopath by being the nice guy. For those of us, when all civilized options end, there is buttkickers. A million thanks!

 

[blekenbleu]

I think you can only add a description to each layer but no I don't think you can change "CUSTOM EFFECT"

Not only that, but

• SimHub disallows separately handling two mono Custom Effects in Sound Output

Consequently,

• laterally-weighted slip channels are output as Custom Effect Left and Right
  • that might be helpful for those who get front slip via FFB and want only tactile rear slip..
• longitudinally-weighted slip channels are output as Custom Effect Front and Rear

Something which became obvious during this exercise is that wheel slip parameter values have a larger range than do others.

Since math and physics for these is at best dubious, the prototype profile will just be attached here.
Updated 4 March: loaded slips use proxy load and slip properties, which can be tuned to user preference. Specifically, included slip properties are tuned for IMO realistic slips,
which are only rarely near 100 (uncontrolled).

 

[blekenbleu]

I strongly advise having operational use of the 1Hz-200Hz

I have yet to sort whether/how SimHub supports changing frequency without also changing amplitude.

The Dayton pucks on which I sit easily deliver 300Hz to my sit bones and thighs.
Above that, they deliver far more sound than tactile sensation.

 

[Mr Latte]

Okay, so just to prevent any confusion.
You have no description name for each effect added

So 1-4 Below
Can you give a description or name each for its role, thanks...

 

[Mr Latte]

I have yet to sort whether/how SimHub supports changing frequency without also changing amplitude.

The Dayton pucks on which I sit easily deliver 300Hz to my sit bones and thighs.
Above that, they deliver far more sound than tactile sensation.

I believe an effect with single Hz given or using "INPUT" mode will increase the amplitude of the Hz applied. "OUTPUT" will alter the gain and frequency between the Base & High Hz vales set.

 

[blekenbleu]

Okay, so just to prevent any confusion.
You have no description name for each effect added

So 1-4 Below
Can you give a description or name each for its role, thanks...

Sorry, I do not know HOW to add descriptions or names..

In your first image:

1. The first (not really) CUSTOM EFFECT, with 4 green squares, emits proxy_G properties used by all other CUSTOM EFFECTS and my 2DOF harness tensioner.
2. The second CUSTOM EFFECT, also with 4 green squares, is my 4-corner effect for ShakeSeat
3. The third CUSTOM EFFECT, with 2 horizontal green rectangles, is the longitudinally-loaded slip effect with separate front and rear outputs which you may mix to mono.
4. The final CUSTOM EFFECT, with 2 vertical green rectangles, is laterally-loaded slip effect, also with separate left and right outputs which can be mixed to mono.

 

[Mr Latte]

Okay thanks will give this a go.


The 2nd image I posted shows how to edit the description.
Easy to miss but simple, place your cursor at the end of the description of the active effect, the pencil icon will appear.

You can also add your own group (folder) to name it and then drag/dop inside related effect layers.
Drag the effects with the icon to the far right.

Example Below
Here I added folders for each as I will try a couple of different layers with different settings for the effect. Then route these to the channels for the BK and TST.

Keeps things tidier when you start to run into multiple effects types or layers for each type of effect.

Query
Does "Effect 2" operate the same as 3&4 combined (Lat & Long) per wheel or differently?

Can you explain more, why we would use it instead of Lat/Long G?
Just want to grasp their operational differences.

 

[blekenbleu]

Easy to miss

I have a knack for missing the obvious..
Even with that hint, it took me far too long to sort that stylus icon must be double-clicked to edit

Does "Effect 2" operate the same as 3&4 combined (Lat & Long) per wheel?

No, for example when mixed down to mono it can yield 4 separate tones at any time,
because an individual load is calculated for each wheel and applied to its wheel slip,
and slips while cornering can be unique per wheel.

On the other hand, e.g the lateral effect totally ignores longitudinal loads and applies the same load to both slips for one side of the car. Since only one side can have more load at any time, resulting signal will be stronger for that side and dominate (unless side loads are equal, but then tones more similar). At the same time, wheel slips for either side are effectively averaged, so no more than one tone per side and less opportunity for more than a single dominant tone at any instant from the mono mix.

The same holds true for longitudinal, except that it is more common, in my experience,
to have more difference between front and rear slips than between left and right, so longitudinal
slip effect should more often produce two simultaneously distinct tones. Mixing to mono with lateral slip effect may yield 3 tones, but I have not perceived such. Frankly, my sit bones and thighs have acclimated to directional haptics from a 4-puck cushion and do not sort tones mixed to mono driving all 4 pucks identically.

At the moment, lateral and longitudinal effects have a refinement for wheel slip parameter processing that has yet to be back-ported to the 4-corner effect.

You can also add your own group (folder) to name it and then drag/dop inside related effect layers.
Drag the effects with the icon to the far right.

I understand neither groups nor layers;
perhaps they do not apply to my haptics usage...?

By analogy, based on long-ago experiences as an audiophile, I can theoretically appreciate your pursuit of extended dynamics and frequency range, but I subsequently focused on directional effects with multi-channel surround in home theater, although current circumstances preclude adequately exploiting either.

 

[Mr Latte]

Appreciate the updated info on the effects code/operation.

The other point to ask:

Take the perspective of a large LFE unit being used with low Hz to generate a deep response in feedback to represent higher slip and the additional loads.

A user could apply 4 corners based effect, but instead of it being routed to 4 independent channels, be sent only to one channel.

My understanding of this is that the user then gets active responses from 4 wheels and with this it will compound the Hz used increasing the amplitude of matching frequencies being generated.

In a sense like a .1 LFE chanel with movies.

The Mono based effect option.
Is only outputting an average of some sorts from the code, creating one output from the 4?

Groups/Layers
You can add a Group with the tab on the right.
Its basically creating a folder to place similar effects into
Layers are additional effects, just that we create them to combine and output as one effect but not necessarily to the same channel.

My own usage of layers is often related to applying or tailoring individual frequencies the effect may consist. This is done to generate effects suited to specific units but can be looked at a way of making better performing effects for the intended transducer/exciter/shaker.

We tailor the creation of the effect to be split into multiple effects (layers) but can then independently control the lower bass portion to the mid/higher bass portion. The approach is similar to having a woofer/midrange/tweeter but using effect layers to represent these individually, lets us better shape the felt character all these layers then generate on the hardware used.

We may want the threshold of the deeper bass to be different, we may want the output of the mid-bass frequencies to be higher or lower. So this is the point of additional layers to customise more the output an effect generates and to then apply the outputs tailored to the hardware it is designated for.