Hi Guys, not sure if this is old news, I'm kind of falling out of touch with the scene
But I just came across this video which shows a project of Imogen Heap's in which PureData is core feature.
She is attempting to make algorithmic music, to accompany your jogging.
Long time!! I'm exciting to have a reason to have to come back to Puredata, and maybe start contributing to this community again.
I have an hackathon occurring next weekend where I am determined to pull samples from Freesound.org into Pd.
I have found this: https://gem.puredata.info/pd/pd/Members/patrick/pdfreesound/view
I have managed to load the py/pyext lib into Pd,
But I can't connect because I can't figure out where to input my username and password (or really how to work around this patch at all really.
Does any one have any experience connecting freesound to Puredata? I don't mind if it happens internally or from another avenue (i.e. via Processing or Supercollider, e.t.c.)
I'm determined to make this work somehow, so I'll post any results.
I had a conversation with the creator of py/pyext last night. really cool guy.
I'll email Patrick, (the guy behind the freesound abstraction).
Hopefully he'll have some advice.
But if anyone can help me on my journey....
Right, so in linux, everything is fine and dandy, I can open pd and put in arguments by simply tapping out
$ pd -nogui
When you install Pd in linux, it behaves like a proper application (it's installed right into the OS, the app is in the bin folder e.t.c. e.t.c.)
so running it from the command line is nice and fine.
However on the mac I am stumped.
I can use the command Open to open the application package, if i reference to it in the Applications folder, I can use ./Pd-extended if I root right through the packgace contents. But the former method doesn't allow me to supply arguments, Open gets confused, and the latter method allows to write arugments, but I have no idea if they're passed through -nogui certainly has no effect and it doens't open any locations to any pd files I may supply it.
So what's up? How do I do this?
As easter's just been and I'm always late with everything, put this into a pd file:
#N canvas 344 398 492 385 10;
#X obj 111 110;
#X obj 151 110;
#X obj 129 137;
#X obj 131 167;
#X obj 130 193;
#X obj 163 194;
#X obj 194 194;
#X obj 98 195;
#X obj 66 194;
#X obj 41 167;
#X obj 227 169;
#X obj 262 147;
#X obj 13 139;
#X obj 134 225;
#X obj 134 252;
#X obj 135 275;
#X obj 135 300;
#X obj 106 324;
#X obj 80 346;
#X obj 166 324;
#X obj 197 348;
#X coords 0 -1 1 1 85 60 1 100 100;
*** EDIT: Post #7, below has the solution attached ***
I'm trying to make an abstraction that makes improvements on [grid], namely the ability to record and playback motion.
Perhaps there is a standard way of doing this in programming science. But I've got round it by using arrays that record vales and an array that records [timer] values, for playback.
Would somebody, (for the sake of the pd community) go through the patch, it is neatly written, and try and stump out why, playback doesn't seem to be accurate. it gets worse the faster a motion you record into it. Leaving me to think that maybe it has something to do with the limitation of the control rate, and the frequency of information that is trying to be recorded.
How can I get around this. Let's talk about it... It would certainly be a handy feature to have!
GOP = Graph on Parent
I'm doing the move canvas along hack to have multiple screens, but unfortnately, when I switch position, the last graph on parent display comes through and thus there is horrible overlap. I'm thinking if i [t b b] a message to clear current canvas, and then reopen a canvas at new location, that might get rid of it.
i don't know what 'donecanvasdialog' is all about but it seems to refer to the canvas. setting every value to 0 doesn't work.
Anyone in linux find that [dumpOSC[ and [sendOSC[ objects don't seem to exist, even though the library is installed (pd-extended) and oscx library does exist in /usr/lib/pd/extra?
even pure data, at startup tells me the library is loaded all other libraries load.
what can I do about it?
Long time no (little) contribution from me in the forums, I apologise, I have been making the most of my summer holidays... But you will all no doubt be pleased to know that I'm getting myself back on track again.
I was never good at doing file and list and string related stuff in pd, and I know some of you have the knack for it. I probably would be able to work it out if I spent hours an hours randomly going through the help-browsers abstractions, but I'm willing to give your generosity a shot.
Now here's my problem.
I want to have a text file that contains my scale in midi numbers, i.e.
then in pd i want it to read through the file, one line at a time appending the line number to the beginning so it sends out the messages
so that i can [route] them into into their respective abstractions that will [mtof] and feed it into [osc~]
can you help?
thanks so much
sorry for not keeping up appearances in the forum lately. I have been extremely busy.
I have taken extremely well to xubuntu on the eeepc.
Synaptic Package Manager always showed Puredata, pd-zexy and Gem packages, but installing them didn't seem to load in any of the externals, and PD didn't have the GUI niceities of extended. - I had pd vanilla.
So I removed it and added
deb http://apt.puredata.info/releases jaunty main
to my list of sources, but synaptic package manager still didn't care to offer pd-extended for download.
So I used the command
sudo apt-get install pd-extended
and it all installed, lovely.
But synaptic is unaware of this package, and I really wanted to keep a clean system where everything goes through synaptic.
Have I missed something, or was pd-extended never built to deal with synaptic. I thought apt-get and synaptic was one and the same?
I talk about wanting a buffer that is continuously updated and being able to read from it how I want (much like a delay line, but where the index position doesn't change.
Well I DID IT! and here is the 60 tables version.
Including a nice a method to play your 1 second chunks or feed in exactly what sample accross all tables) you want it to play!
(When using the metro-random you can create some great on-the-fly musical compositions just by layering a direct stream of your voice on top of the rhythmically randomized chunks of pre-recording) and some drums and away you go!
can we as a community come up with an efficient way of continuously updating an audio buffer. Rather than the method commonly used that is to simply overwrite the entire last buffer.
So what I mean is that we have a table of say a minute worth of samples, and, from fresh, it is constantly recording into that minute (until a minute has passed we have blank space in front) then once it reaches the end, it starts again at the beginning replacing the sample that was there exactly a minute ago.
The only way I can think of doing this is by splitting the 'minute buffer' into many smaller buffers, so they are dealt with in chunks (or blocks), like 60 one second each tables. Then have a system for recording into each table in time ([metro 1000]-_x[+ 1]?) and then either have a way of importing the data from every table (ordered correctly) into the larger minute table and have your player read from that, or have a pseudo buffer that is just the system of playing through the tables in order (or not, if you so wish).
but that doesn't give a very high resolution.
is there maybe a neater solution with [block~] that I'm simply over looking?_
Itunes has lost me now i discovered it can't use FLAC. Some albums just shouldn't be data compressed.
I have two questions for our more wise linux-audio community.
What's the best iTunes alternative?
And... What's the best open source iTunes alternative that will run on mac and GNU/Linux?
Because I'm kind of keen on sharing the same player on my mac as my GNU-top, but hosting my library on a flash drive. However if there is one distinct forerunner that happens to be Linux only (Amarok?) Then I might be happy to settle for two different players.
I remember seeing a patch contained somewhere within the browser of pd, that (amongst other GUI tools) allowed you to pick a colour and it would tell you it's -1.16022e+07 value... I can't find it anywhere. I need this because I want to make a
low CPU VU.
(it's not hidden somewhere within x_all_guis, that only because I'm a dumb f^ck can I not find it, is it?)
Can anyone think of a cheap USB foot pedal with a reasonable resolution that I can use to feed (at least 127 steps) of data into Pd Linux. MIDI is fine but second resort.
I know I can go into the arduino route and hack anything. Something which I intend to do sometime when I figure out my rigging needs, but until then I keep putting off using my feet and for the meantime would love to have something to play with that's durable and convenient enough to lug around!
I have a few 8 minute 16bit stereo wav files I want to load into pd. So I set my read flags to [read -maxsize 3e+07 -resize ...blah blah blah( and everything loads dandy, works fine.
Only, in total there are 467MB of sample data to load into PD allocated memory. My only concern is about being able to play this abstraction safely (crash-free in a live environment) from my netbook (of 2 gig RAM) and limited CPU.
On the big and beastly mac Activity Monitor shows pd to to have eaten up about 900MB RAM with around 1.2GB in virtual memory. This goes up and down according to how much I actually load in.
So my worries are, why the 900MB of RAM use? is that because its upsampling the 16bit into internal 24 or 32bit processing? And the Virtual memory, that's only requested by PD as a precaution isn't it? And provided there is sufficient RAM, no disk swapping will take place, will it?
Performance is lovely. I'm just worried about pushing the -maxsize so far.
On a side note, think about how much having this sampling power would have cost only 10/20 years ago!! (or the size the computer would have had to have been... netbook! pah!)