Options for tactile feedback setup.

Hi, i'm about to embark on my journey into tactile feedback, but I have to be honest and whilst I think I have grasped the basics things are pretty unclear as to what my options are, after watching You tube, reading the posts here and on reddit etc, there's appears to be a huge array of option and a fair amount of conflicting information.

My rig is a self built 8020 profile and pretty damn solid, and my chosen seat (Sparco R333) is quite padded with no real options to mount any exciters directly to it.

A couple of pictures of it for reference:

IMG_3090.jpg

IMG_3091a.jpg


As far as my options go, I'm more interested in quality than budget, that doesn't mean my budget is unlimited, it more of a case that I'd pay the extra to get the setup right at the outset rather than try and do it at a budget and then end up paying more in the long run due to upgrading.

With the rig being solid I think I have pretty much written off the budget end of the market, I don't think I'd get enough feedback and the money would be better spent towards better kit.

Ultimately my goal is for as realistic feedback as I can within my budget, do i go the whole hog or do I start with a more simple solution and improve it as I become more familiar with tactile in general, I'm not sure at this point and the more I read the more I'm confused. If I had to put a figure on budget I would say £1500 ish is my limit.

I appreciate that I will need to isolate the rig as well as the seat/pedals as well as buy cables and connectors, so I'm not including any costs of these in my budget as they are necessities and ultimately the cost of these will depend on what I end up buying.

To that end I've come up with a few rough ideas but I'm not sure which is the better option.

Option 1. (probably overkill and the most expensive)

4 x Buttkicker advance, one per corner

2 x NX1000D or NX3000D for the 4 x advance

1 x Buttkicker (not sure which) mounted between the 2 seat rails which will be isolated from the rig, plus amp to suit (probably another 1000D)

Option 2. (Similar idea but cheaper)

Pretty much as Option 1 but replacing the 2 x front advance and seat unit with Aurasound AST-2B-4's or similar, and suitable amp, keeping the 2x advance and nx amp for the rear.

Option 3.

2 x Buttkicker Advance, setup for left and right or front and rear 1 x Nx1000D, again not sure which under seat unit & amp

Option 4.

2 x Buttkicker Advance, 1 x Nx1000D setup for left and right or front and rear.

Any comments or suggestions are welcomed, I may be way off with my thinking, I really don't know at this point, I'm guessing that I might be best just starting with a single unit + amp and build from there.
 
Now, explain to me how a vibration-reducing substance is fully applied to a material that then magically tones down the frequencies causing ringing, reverberation yet supposedly does not reduce the output of other frequencies?
Words are funny things... "If tin whistles are made of tin, then are fog horns made of fog?"

For a plate being used to transmit vibrational energy from a transducer to a substantial mass (e.g. our butts),
then adding damping to that plate reduces transmitted energy to the extent that
  • the plate was acting as a spring instead of a rigid object.
  • damping material added mass
    • lowering the resonant frequency of that transducer+plate,
      which may increase output for some frequency of interest
 
Upvote 0
Basically, you are saying you have no real idea the impact it has on his setup but it's not offically approved by you :)

You gave little to no context what that -6db was about so one could only assume that in context you thought it might be what could be lost for his solution.

Sometimes you have to stop with all this 'I dont want other people to get the idea they can just do what he did.....'. It worked, you want proof, do it yourself.
The -6db is in reference to applying DSP based PEQ, as stated if you determine the frequencies causing the problem then you can attempt with DSP to reduce only those frequencies. -6db was only used as a figure nothing more. Had he of had DSP then it’s possible the issue he had may of been controlled with no application or physical changes to the rig. This way you are also certain you are not dampening other frequencies felt sensations and the opposite without he DSP you can boost others if necessary based on personal preferences.

The application used was overkill and clearly he did not try other alternative solutions.
It’s not for me to prove anything, just highlights the lack of trying different methods sone will do.
 
Upvote 0
Words are funny things... "If tin whistles are made of tin, then are fog horns made of fog?"

For a plate being used to transmit vibrational energy from a transducer to a substantial mass (e.g. our butts),
then adding damping to that plate reduces transmitted energy to the extent that
  • the plate was acting as a spring instead of a rigid object.
  • damping material added mass
    • lowering the resonant frequency of that transducer+plate,
      which may increase output for some frequency of interest
No do a practical test with measurements
Words are funny things... "If tin whistles are made of tin, then are fog horns made of fog?"

For a plate being used to transmit vibrational energy from a transducer to a substantial mass (e.g. our butts),
then adding damping to that plate reduces transmitted energy to the extent that
  • the plate was acting as a spring instead of a rigid object.
  • damping material added mass
    • lowering the resonant frequency of that transducer+plate,
      which may increase output for some frequency of interest
Theory , show measurements from actual tests or 3 plates with different amounts of The material used, let’s have physical tests.
 
Upvote 0
You know the only reason I say prove it is because you keep asking people to do just that, as shown again in the post above, you want people to do tests for you, with three plates, physical test, etc, etc, etc and show measurements. To date I have not seen any measurements at all for your stuff but they are the best out there. It's not hard to test the drop off on different parts of the rig. I only see anecdotes about it, nothing substantiated. Hey, I am not doubting you get good results but understand you ask a lot from others, more than you have given yourself in terms of sharing data and measurements.

About the plate, it's funny because had he used 10mm plate instead of 5mm he would not have had the issue and would not have had to correct it - and you wouldnt have worried at all that he is using 10mm plate even though that 10mm plate is possibly worse than his solution. So I think you are just arguing because it's different, not on the merits.

I dont know if he has DSP but the average guy doesnt and has to solve it in another way. As I said before.. Pick a frequency that you use a lot and lets say it was a culprit for resonance. Are you really going to give up good energy at that frequency so you dont have to dampen your plate? It doesnt make sense.
 
Upvote 0
Theory , show measurements from actual tests or 3 plates with different amounts of The material used, let’s have physical tests.
With materials on hand, comparative vibration swept spectral responses could be made for:
  • 3.3mm aluminum cantilever
  • 3.3mm aluminum cantilever with damping material
  • 14.8mm plywood cantilever
  • no cantilever
but if that will not be satisfactorily actual, then I will not bother,
since new piezo vibration sensors would need to be assembled.

FWIW, 3.3mm aluminum weighs 1.08x 14.8mm plywood of the same length and width...
 
Upvote 0
You know the only reason I say prove it is because you keep asking people to do just that, as shown again in the post above, you want people to do tests for you, with three plates, physical test, etc, etc, etc and show measurements. To date I have not seen any measurements at all for your stuff but they are the best out there. It's not hard to test the drop off on different parts of the rig. I only see anecdotes about it, nothing substantiated. Hey, I am not doubting you get good results but understand you ask a lot from others, more than you have given yourself in terms of sharing data and measurements.

About the plate, it's funny because had he used 10mm plate instead of 5mm he would not have had the issue and would not have had to correct it - and you wouldnt have worried at all that he is using 10mm plate even though that 10mm plate is possibly worse than his solution. So I think you are just arguing because it's different, not on the merits.

I dont know if he has DSP but the average guy doesnt and has to solve it in another way. As I said before.. Pick a frequency that you use a lot and lets say it was a culprit for resonance. Are you really going to give up good energy at that frequency so you dont have to dampen your plate? It doesnt make

I am not the one using science jargon or principles to then make claims of what materials or how a solution may perform based on calculated expectations. Like the BK Concert plate being problematic (nonsense) or stating that wood materials are more optimal, (yet none of the manufacturers use wood accessories/applications) to install tactile.

So those that are talking of such recommendations or in scientific terms, well I say then let us see the data or highlight from personal experiences to back up the points being made by yourselves.

There are far too many factors and a change in one element could have a big bearing on other factors. For instance, the thickness of the plate, its size, where the transducers are positioned, what the output of effects are, what gain is being applied, what isolation if any has the rig.

What is being debated is if or how Dynamatt or similar has an impact on the frequencies being generated by the transducers. As I pointed out, Andrews's own installed hardware is already frequency limited so we need to look at the 1-200Hz range or a bit higher to maybe test this. Sure he may not have felt much loss or difference and that's not being debated, but his application is using what output frequencies with his effects from the 1-200Hz range? No indication was given, nor was any indication of the frequencies that were causing the reverberation. things that would have been interesting to know.

Is he or anyone here, in a position to say with confidence or certainty, his example of application will not affect the detailing of any bass frequencies or hinder someone else's builds performance with their installed hardware?

As for controlling audio, DSP is used with many professionals in the industry to enable exact control of individual frequencies and then determine in dB how much +/- dB to apply to them. Can you please tell me a better way to have control of frequencies, or what needs to be proved by this method when already lots of people apply DSP to configure the felt output on their rigs to suit their own preferences/rig seat materials and to also reduce issues with noise/resonance or reverberation?

I am not saying installation is not important or not also a factor or that DSP is going to always be a solution on its own. What I am saying is the DSP gives the user much more control.

It appears you have taken a stance, that applying vibrational control materials to the primary contact object, based on Andrews report has no drop in generated frequencies detailing or output. Yet you want to question the professionally used methods to controlled drops in amplitude using DSP as being detrimental to performance?

When you guys can prove to people here that Dynamatt makes no reduction in the felt frequencies and that it can be fully applied to the surface of an object between the transducers and the rig. Highlighting no concerning drops in performance within the 1-200Hz range. Then I will congratulate you and admit I am wrong on this, is that not fair?

My method or approach is to base things on past tests or experiences, using various specific hardware and trying different approaches regards how to mount or install tactile. My experience with such materials leads me to believe it is not an ideal way to sort this issue.

Do I claim to be an expert or always right, no not at all, you will never see me talk of myself in that way and people that do talk more privately with me will tell you I always put myself down as just an average guy from a small town, nothing special about me. Now you are welcome to search the forums over the last several years where the advice or methods I recommend were not working well or people criticized as inaccurate or performing badly. I never have an issue admitting if or when I am wrong on things.


My Installation Recommendation
Here is how I would attempt to install tactile to a profile-based rig...
Andrew may or may not be using isolation for his plate, this was not even raised or noted and in my experience, decent isolation (not cheap bobbins) to a plate (preferably steel) with DSP would be enough to control or remove resonance issues and be the basis to ALSO get improved performance and personalized tuning from whatever transducer hardware was being used.

Consider using vibration control materials for reducing vibes into floors etc. Not as a method applied directly between the transducers and rig frame. Look at other possible solutions and try more than one solution/attempt/method to determine for yourself what seems better.

Your Experience May Vary
 
Last edited:
Upvote 0
Back
Top