2017/10/12 00:16 - "You, you , YOU, yoU"
Think some of you might get a kick out of this (caveat I haven't done any spoken-word improv in a long time, so I can't really vouch for the quality of my words):
The pd part:
The recording was done using an old ps2-jack logitech footpedal (which I found out really only has 4 electrical lines out: red, black, white, and green (directly connected to the potentiometers) so I routed them to my arduino and sent those signals (once normalized) to 2 effects (one to the left and one to the right channel of a stereo recording plus as second line (vocal) to the right input (using a really cool) double 1/4" female to 1/8" stereo male adapter cable I picked up at Guitar Center.
So I sent in 2 signals, left and right. Right I left clean as vocal and recorded to its own stereo track and left I divided and sent to a second stereo track with then left being a delay-feedback line (DIY2 abs) and right being a pitchshifter (DIY2).
So as I played my left foot ramped (from 0 to 1) the delay and my right foot ramped (also from 0-1) the pitchshifter.
Very happy with the results.
Found my self thinking it is kind of representative (given the vocals) of what might be called "folk acid-rock".
Tally ho. And Good Luck in all your endeavors.
Would love to hear what folks think.
Hope you enjoy it or at least find it entertaining.
Ciao for Now.
How to get the dollarzero value of parent patch?
@cuinjune interesting, but i don't get the point.
If i open your root patch twice, it will break the separation, because abstractions inside the first and the second root patch will all communicate with the same instance of root patch. In your first example with the second instance, in the second example with the first instance.
If you don't open root patch twice, you don't need $0 at all and can just use [table root-array] and access this directly.
What am i missing?
Also note that instead of passing $0 through creation argument, you can just use inlets and outlets for perfect separation.
Writing audio to a table like delwrite~
I wonder if anyone has come up with a way to write audio to a table in the same way that delwrite~ writes audio to its buffer. That is, you write n samples to an empty table and when it gets to the end, it keeps writing to the last position of the table while shifting everything else so that at any moment the audio in the table represent the last n samples that have been sent to it.
I'm trying to do a freeze patch that records a live input to a table, but if I bang a tabwrite~ object upon freezing, it's going to take a while to fill the freeze table with sound (according to the length of the table) and that produces a gap when switching between the live input and the frozen sound. Besides, the sound written to the table will be the sound that has reached the tabwrite~ object after the bang, and I'd like to make use of that length of audio that has been fed to the table before I hit freeze.
I reckon that if such a method of table writing is possible it would be useful for other things too, like tape-style multi-head delays etc.
Thanks in advance for your help.
oscformat and oscparse... /integer not routing as expected
@whale-av Hello David! Thanks for the reply. A bit confused about this, still.
So, here’s a message sent by TouchOSC:
list page4 multifader1 1 0.193394
My attempt at routing this…
[oscparse] | [list trim] | [route page4] | [route multifader1] | [makefilename %d] | | [route foo]
Getting the error:
makefilename: no method for '1'
Might have misunderstood the patching itself, though.
Here’s what it looks like in Pd.
Newbie question about variables
Just remembered-- here's a pattern Frank Barknecht posted to the user list some time ago:
Here we route based on the value of "x". The
[route] object chops off the 'x' value, which leaves us with "z" for the output. Then it's just a matter of adding the desired value to z.
This works well for two (or maybe three) variables-- that is, as long as it's easy to remember the position of "x" and "y" in the list output from "pack". For more complex conditions you can throw an
[expr] in between the "x" and the "route".
PD-2-ThoughtMap (a patch and its children as a plantuml-usecase) (Linux Shell Script/Package) (v0.1.1 update below)
When I run the command prompt I get:
100% Complete... grep: .pd: No such file or directory grep: .pd: No such file or directory grep: .pd: No such file or directory cat: /root/pd-externals/pd2thoughtmap/res/HTMLHEADER.txt: Permission denied cat: /root/pd-externals/pd2thoughtmap/res/IMAGEHEADER.txt: Permission denied cp: cannot stat '/root/pd-externals/pd2thoughtmap/res/jquery.js': Permission denied cp: cannot stat '/root/pd-externals/pd2thoughtmap/res/jquery_plantuml.js': Permission denied cp: cannot stat '/root/pd-externals/pd2thoughtmap/res/rawdeflate.js': Permission denied pd2thoughtmap/pd2thoughtmap: line 253: defaultbrowser: command not found
Change route with a Toggle
Is there a way I can change the route of a signal by using a toggle?
I was trying to use the object route but it only changes the route depending of the value.
What I want to do is that when I activate a Toggle, all the signal goes from route a to route b and viceversa.
Hope somone can help me. Thank you!
I have a very strange problem right now.
i have 3 dataflow strings passing through a route object using
three |x set $1( messages.
it looks like this.
so far no problem as you see the numbers come out of the route object as they supposed to.
but if i now want to go on working with the numbers the flow stops.
i have no idea why this could be.
connecting things directly to the route object works fine. working with the data before the route object is also no problem.
but in the second instance after the route object the flow stops abrupt.
someone has an idea?
How do I pack multiples arguments into one sendOSC object, and unpack it on the other side?
OSC messages can contain (usually do contain) routing as well. So you can send for example.........
The slashes denote the parts of the route in the same way as for a folder structure in your operating system, and you can prepend First_Message/First_Part/ to a message as you have done already with the "send".
For incoming messages you can then route the data you are expecting using OSCroute maybe (I am not sure which object... for vanilla, or extended?).....
I am sure that you can also route osc messages in a similar way in the other program, as it is part of the OSC format specification.
"First_Message" etc. can be any word you wish, containing symbols and numbers, but it is best to start the message routes with a symbol....... so woof23 for example.
Playing back a sequence of text in Pure Data.
@LiamG, the first is not hard to pull off, but I would want to gather the keystrokes in the format that qlist uses to read, time - send - value, even when PD is not solely active. Is there such an application for OSX that allows you to have two programs 'write-active' at the same time? So that the given text-writer can write in his preferred word processor while perhaps at the same time writing these into a background-running pd patch.
@whale-av, I didn't know of the
realtimeobject, that would certainly make creating the relevant qlist-'score' inside a PD patch do-able.
Given my own programming limitations, I think I'll try to make a patch that would write the timemarked qlist-score as a patch for the write to run in the backround if I can manage to have both a word processor and PD active at the same time.
Thank you for your help so far.