PlanetZ forums - scopeusers.com

<a name="planetz-file"></a><a href="http://www.j9ksys.com"><img src="/forums/images/file_icon.gif" border="0" alt=" File"> File</a><BR> <a name="planetz-tag"></a>Type: beta<BR> <a name="planetz-tag"></a>Pulsar Version: Pulsar 3.x<BR> <a name="planetz-tag"></a>Requires: Modular 2<BR> _____________________________________<BR><BR> these modules are almost done. i just have to fix a couple of things.

when hooked up normaly it acts like the decimator set to 8 bits and the sample rate in bypass. i found that by rearranging the cables it can make a whole new range of sounds and waveforms.

the modules are polyphonic but have a few problems if the polyphony is over 4 or 5.

j9k

<font size=-1>[ This Message was edited by: j 9 k on 2003-05-08 17:14 ]</font>

those are simply cute
they are much more interesting than simple bitcrusher as they allow bitshifting and other mess.
thanks j9k!

btw logical operators like AND, OR, XOR etc. could work very interestingly with those... pretty please

Hi,
would either of you care to explain how these work?

MSB & LSB are Most and Least significant bit, I know, but I don't really understand what the modules are doing (I don't know much about digital sound, it has to be said.)

I experimented with these modules in the Patch that j9k posted with his 'to knobs x5' module, but seemed to get the same output no matter what settings I was trying.

Any explanation of what these modules actually do, or tips on how to use them would be much appreciated...!

Cheers,
Ben

Ben

try loading both A2D and D2A, connect source to input of A2D, connect output of D2A to wherever you want your signal.
now connect Msb out on A2D with msb in on D2A and so on to Lsb.
now it works as hardwired 8bit bitcrusher.
if you want (for example) wire it as 4bit crusher you should route 1 out of A2D to Msb and 1 in of D2A, 3 out of A2D to 2 and 3 in of D2A and so on.
you can try much wilder wirings like reversing bit order (connecting msb to lsb etc), this gives UGLY digital distortion.

btw, i felt realy limited when i tried to do something with those devices using 4way switches.

Great, thanks Michu, I'm beginning to get the idea now, having followed your instructions and played around a bit more last night.

Forgive me if I'm labouring this, but I'm keen to understand what's actually going on here, and maybe to help explain it to other people who are as baffled by the marvels of digital audio as I am.

I'll try and explain what I think's going on - please correct me if I'm wrong, or if I'm using the wrong terminology.

The audio signal that passes through a digital system is made up of (in a 16 bit system running at 44khz) lots of 'slices' of audio, one for every 44thousandth of a second, each one made up of a 'word' which is 16 bits long.

What these two modules allow for is both bit reduction, and 'bit swapping' (I don't know if that phrase exists, but its more accurate in this context than bit shifting, I think.)

The A2D module takes a 16 bit signal, takes the 8 most important of these bits and splits them up, outputting the 'most significant bit' from the MSB output, the next most significant to output 1, then to 2 and so on.

The D2A signal takes the incoming signals from its 8 inputs and 'recombines' these into an 8 bit word, and then outputs this as a 16 bit signal.

In its most straightforward usage, where the MSB in A2D passes to the MSB of D2A, out 1 to in1, 2 to 2, etc, the signal is simply recombined 'in the correct order', so what we get is a grungier 'lo-fi' version of the original signal, this being because we are only hearing 8 of the 16 original bits - a lot of the detail has been lost and we hear a dirtier, grainier sound.

Moving on from here, if we simply start to remove some of the cables which join the two modules, we get a progressively more 'lo-fi' sound as more and more of the detail is removed - removing the link between the 2 MSB's has the most pronounced effect as we're taking out the 'most significant' part of the signal, wheras removing the LSB links only results in a slight reduction in quality.

Where it gets interesting is when you start swapping the cables around, for example connecting MSB from A2D to the LSB in D2A. When doing this, the digital signal is effectively recombined 'in the wrong order' resulting in all manner of unpredictable distortion. Some of the original signal still gets through, but often in an almost unrecognisable fasion!

OK, I know that this is more or less OK, but I'd be really interested in any corrections to my assumptions or to my terminology - I'm pretty hazy on how A/D D/A stuff works in practise, but I'd like to know more. Maybe this is a subject for a topic in the Pulsar Study forum.

Cheers,
Ben
_________________
<a href="http://www.modularsynth.co.uk" target="_blank">www.ModularSynth.co.uk</a>


<font size=-1>[ This Message was edited by: Ben Walker on 2002-08-13 07:15 ]</font>

Thanks for the post. These will come in handy...

Try putting a modulation signal through this and you can get some interesting results.