@chapelier-fou It will be a problem with the order of operations...... which you might thoroughly understand, but if not then this is good bedtime reading...... http://puredata.info/docs/manuals/pd/x2.htm

You probably need to use another trigger [t b b] on the right outlets of [tabplay~] to make sure that the [stop( message arrives at the inlet after the player has been re-started.
The player is probably getting ...... stop, play...... when it needs..... play, stop.
You can check by adding a [print] and look at the terminal to see the order of messages and bangs.
David.

@jasper13 There will be ways to smooth and scale your acceleration data.
You will want to reject "small" changes in data (within a certain timeframe) as well.........
Then you will want to do something like this to "return to normal" settings.....
reset_norm.pd
David.

@amazingrolo The audio settings dialog changes how Pd talks to the soundcard driver. If the souncard doesn't like the settings that will hang Pd (Pd will be waiting for the soundcard to say "ok thanks" and the soundcard will be saying "eh..!"..).
If you want to change the block size you will have to do it for the soundcard as well. "64" is standard.

Pd schedules audio processing ticks in 64 sample chunks anyway, so I am not sure there is an advantage to using [block~] in a heavy patch. In fact I would think that changing block sizes would have a cost.
If your patch is not "live"..... you have no connection from adc to dac..... increasing the Delay (mSec) wil help more. It will increase the latency by using a bigger buffer.

A few mistakes are possible using [block~] or [switch~]..... with [throw~] for example......
http://puredata.info/docs/manuals/pd/x2.htm chapter 2.4.4 onwards.

I have had trouble while changing [block~] parameters ..... but once it is set in the patch (patch saved and re-opened or audio toggled on/off) it is fine. That is painful when you are experimenting though.
David.

@svanya Acoustic waves are Longitudinal and liquid waves are far more complex. Water waves are bounded by the interface between gas and liquid as well. There is not really a common relationship.
http://www.acs.psu.edu/drussell/demos/waves/wavemotion.html
Have a look at the PMPD library again for some inspiration maybe......
There are a few examples of wave-like particle behaviour in there, and some with ideas for accompanying audio generation.
David.

@svanya I tried it in the bath when I was young....... but you need a backwash..... an undertow.... for them to break.
If that makes sense (which it must) it might help...... but in Pd logic terms I am not so sure.....
David.
PS.... I know that is a stupid reply.
You cannot have a waveform that "curls back" underneath itself...... as @weightless says.
But maybe you need to create a sort of "building tension" in your sound....... followed by a prolonged "crash" and then calm.
I don't mean to be rude, but the "tension" part needs to sort of "suck" and continue under the breaking "crash".

@youaresound Ok.
2012 MacBook Pro. (apple) UK
1234567890
qwertyuiop
asdfghjkl;
`zxcvbnm,.

!@£$%^&*()
QWERTYUIOP
ASDFGHJKL:
~ZXCVBNM<>

@youaresound Complete (probably) list with graphics..... https://en.wikipedia.org/wiki/Keyboard_layout
David.

@dopamine333 No, you have done nothing wrong, it was built for "extended" so libraries are missing in "vanilla". I thought that I had zipped up all the necessary components, but......
Here is [compare-any]...... compare-any.pd
Will it then run?
[list-something] I cannot find, and the import object can be ignored / deleted.
David.

@tmarcondes [openpanel] is asking the operating system to open a window....... and the instruction will be incorrect for android or iOS. PdDroidParty claims to have an abstraction that works for Android.... called [loadsave]. See here, under "abstractions"...... http://droidparty.net/
David.

@quiethorn In case you ask..... here is the same, but with adding an index and its value as well.
The "removing an index" is the same as my previous post......
this.zip
David.