I have an issue that is bugging me.

I have been using the pure data/firmata with arduino connected to the computer by a set of xbee trancievers. this was nice at one point owing to its simplicity of setup but now its data handing is slowing me down. I purchased a set of Roving networks RN-XV wifi trancievers and they are MUCH faster although no where near as stable.

currently, in windows, i have made the arduinos send their data over wifi to a router that is just for the system. from there i have beenable to use "serial/ip" serial port redirector (demo) to get the data from the router to the comport. This is very un-elegant. it works but is skippy and glitchy because i imagine that low latency live performance was not their primary concern.

are there any objects or techniques, in pure data, to get the ip and port number and dump straight into pd, possibly either directly into the comport object or bypass it and dump straight into the rest of the arduino object guts? today i will be experimenting with the net objects in PD, more, but so far none i have played with can be substituted for serial port (real or virtual) and comport-combo.

i have been researching socat on the linux side but by ubuntu-fu, sucks.

cheers!

Onyx

I am having a hard time trying to figure something out.

I have a looping system i cobbled together that got its timing information from a vline~ whose value was sent to 8 loopers. this has been cool and stable but i want to time stretch the audio so i decided to swap it out for a phasor~.

It kinda sorta works. i clear the table, then i use a timer to set the loop length in ms and use that to calculate the sample size (ms/100*samplerate) and use this number to ride the phasor and read the index in tabread4~. i play something into it and it loops and i have control over the speed of the phasor~ using a slider, but its distorted and grainy and eventually fades away.

I think i've left something out but i cant figure out what. any help would be highly appreciated!

onyx

http://www.pdpatchrepo.info/hurleur/looperexperimentor.pd

this is really starting to take time away from developing.

I know that my problems are all based on my ignorance of how linux works but i really need to install 0.42.5 on this laptop.

i had the 64bit version of the OS but had issues with either poke~ or some sort of 64bit table bug that played havoc with my looping system, so i switched to the 32bit version of ubuntu studio 12.04 but i have found no way to get 0.42.5 onto the system. 0.43.1 is easy but it uses way more resources and is generally more buggy than the former, in my experience. 0.42.5 is just more stable and smoother in all respects, but with all of the deprecated dependencies, it is impossible (for me) to install. i have collected and downloaded and installed, in order every single dependicies but towards the end of the process, they start to break things, which keeps the process from going beyond a certain point.

please! i really need to sort this out asap because it is destroying my productivity. i really dont want to reinstall windows xp, but that is looking like to only way that i am going to get a working system that i dont have to fiddle with constantly. what steps do i need to take to put 0.42.5 pd-extended onto ubuntu 12.04?

thank you in advance.

Onyx

i guess this is a different question from my previous post.

i have just discovered that some objects in the 64bit version of pd-extended, do not work as they should. i need these objects as i have not found anything to replace them (poke~ is the object in question). i have (as seems to be my complete experience with linux) spent days trying to track down why my patch, which worked fine in windows but now doesnt work in ubuntu studio 64bit 12.04. katjav helped me to discover that the 64 bit implementation was the culprit as i ran the patch on a 32 bit ubuntu install on another computer, and it ran as expected.

i have been attempting to simply install the 32bit version but there are way too many broken dependencies, so my next quest is to attempt a 32bit chroot (which i only even know exists because of google). but i am wondering if i should simply install the 32 bit version of ubuntu studio 12.04, which, besides this issue, i am pretty happy with.

is there an easier option? are there any pro's or cons i should take into consideration before attmpting this operation?

thanks in advance. i look forward to making music with linux and pd one day...

onyx

i have been beating my head against a wall for 2 days.

I made the switch to ubuntu studio 12.04 recently, from windows 7. i have been running my patches on win 7 with no significant issues other than typical windows cpu bloat. i have had a running looper abstraction running for months. now, for some reason, trying to run the same abstraction in 0.43.1 pd-extended for ubuntu 64bit, the aspects of the patch responsible for looping, are all out of whack.

once i hit the record then the overdub, the resultant loop is immediately half the length and pitch of what i played, although the loop length is exactly what it should be, so each go around the recording pitch and length blip into nothing but a slight "chip" sound at the loop start.

i have taken some of the main bits of code and made this patch. it works fine in windows but not in linux. are there any other aspects of swtiching to linux that i should be aware of, that will affect my patching?

(actually, the example patch is more screwed than the patch i need help with and for similarly mysterious reasons so i will include it anyway)

thank you in advance

Onyx

http://www.pdpatchrepo.info/hurleur/looplayer.pd

greets all. i have a new ( in a series of new) issues.

I have finally goten my looper to do what i want. on my interface i have a joystick that, when flicked in direction "one" will resize the looper tables to 30seconds, and clear any audio within them (i am using some of Katja's looper code for the loops). then when clicking in the other direction once, a spigot opens for the choosen looper (of , to allow recording while a timer is initiated. click that again does a number of things.
1. it send another bang to the timer, whose output is then converted into ranges and sent to a vline for use in controlling the playback of the loopers.
2. it closes the spigot that was opened with the first click so that all subsequent loop record clicks simply record within the given length range without interferring with the loop length.
3. closes the spigot that controls the audio going into each looper.

the problem is that it wont record any audio for the first loop. it records the loop length prefectly but i must click again to instantiate recording of audio at the new loop length.

i "think" that there must be some order in which these things are instantiated. i assume that the "resize" must be done just slightly ahead of the end of the audio recording, but i could be wrong.

is there an issue with recording to a table while it is being resized? if so, is there a way to resize the table that is being recorded into, quickly enough that there isnt a glitch? any insight whatsoever is most highly appreciated.

cheers!!

Onyx

http://www.pdpatchrepo.info/hurleur/looper.zip

greets all. i have a (possibly) simple question...

how do i determine the largest number that has occured in a stream of numbers? i want to measure numbers in a fluctuating stream of numbers, but i want the highest number that has occured within a given time frame, to be shown or saved. i would be very surprised if there isnt a super simple way to do this that illudes me.

thanks in advance!

Onyx

greetings. I have a question.

i want to know if it is plausible to use a mic on an analog (or digital) pin on the arduino (using this http://www.sparkfun.com/products/9964 and this http://www.sparkfun.com/products/9816) and convert the resultant analog signal back into audio within pure data?

My idea is to bring the stream into PD then "sample" it by creating a signal rate midi-like on/off for each sample. so instead of

605
620
650
765
700
675
630 (etc)

i would make the stream be seen as

605
0
620
0
650
0
765
0
700
0
675
0
630
0

i feel that, theoretically, this would produce samples that i could discern as audio information. or...am i overthinking this and it is really much simpler?

thank you in advance.

Onyx

i figured i'd post something where i'm not begging for answers for a change.

i read a post on this forum from someone who had discovered that minimizing the patch (in windows) saved somethng like 40% cpu but i was skeptical having just switched over to linux, but i needed to crank up the windows version and my vst rig for a number of shows this weekend. this system usually hovered around 90% cpu but for this show i also have a GEM viz that i figured i'd either not use or brace for the core meltdown. but i decided to test this theory and it works! i am able to run NI Kore and the GEM viz and the cpu hovers around 60%!!

just thought I'd share this tidbit. its has saved my ass and made my day.

Onyx

greets all.

I am having a fundamental brain fart moment. I am creating a looper that i want to record the first loop into a delayline ( and if directly into a table is better, please let me know, and why, and i will gladly go that route) where i will use the [realtime]object to get the length of the relevant recording, then use that to resize a table and loop it with tabplay~, but the whole thing has turned to gibberish.

at the root, i just dont understand these objects or their functions properly yet. any insight into what i am doing wrong, would be greatly appreciated. my hope is to create a counting function to go along with this patch (which i plan to have 4 syncronized copies of) that will only access the [realtime] function on the 1st go around going into the delwrite~. there will eventually be 10 tables and 10 tabplay~'s all set automatically to that length. as the counter abstraction counts up, the count will access the table/tabwrite~/tabplay~ combo for that number, up to 10. an "undo" function will mute down this list of tabplays and redo will unmute, but once it returns to 1, it goes back to the delwrite~ and creates a new loop that determines the length of all the others.

before any of that can happen, i really need to get a grasp on creating dynamically resizable tables and accessing them instantly.

sorry for such a noob question but this is my first foray into the audio objects. I've been swimming down at the midi end of the pool.

thanks in advance,

Onyx

http://www.pdpatchrepo.info/hurleur/looper_alpha.pd

Greets all. headache problem.

for some reason, i can not get pd-extended 0.43.1 to load Gem. i am running on puredyne linux. i have tried putting gem into the startup, to the path, both individually and together. I have downloaded the latest version and copied it to the directory, which didnt work. i tried linking to the folder in my home folder...didnt work. i copied all of my patches to the pd-extended directory which didnt work. i have no idea what to try next. i dont know if its a 0.43.1 problem, a puredyne problem, niether, both or something else entirely.

also, any one have opinions about 0.43.1 on linux yet? yes, no, maybe? or should i go back to 0.42.5?

any advice is, as always, much appreciated.

Onyx

greetings.

I am investigating using netsend to send patch data from one pd instance to another, to make use of both cores of my cpu. I have replaced most of my internal patch wiring with [send] [receive] objects. is it possible to [send] to a [netsend] and [recieve] from a [netrecieve] or am i fundamentally mis understanding what these objects do?

I am trying to shuttle about 20 streams of data to another instance. how should i go about formatting these individual streams (which are currently [send] objects) so that they reach their intended destinations.

thank you for any insight

Onyx
www.indiegogo.com/beatjazsystem

greetings.

I am investigating using netsend to send patch data from one pd instance to another, to make use of both cores of my cpu. I have replaced most of my internal patch wiring with [send] [receive] objects. is it possible to [send] to a [netsend] and [recieve] from a [netrecieve] or am i fundamentally mis understanding what these objects do?

I am trying to shuttle about 20 streams of data to another instance. how should i go about formatting these individual streams (which are currently [send] objects) so that they reach their intended destinations.

thank you for any insight

Onyx
www.indiegogo.com/beatjazsystem

Greetings.

Is there a viable substitute for [change]? Something that can hold back the torrent of sensor data as well as change does, but without the "stopping repeating numbers" aspect?

What's happening is that I am using it to select notes but if I quickly jump from one note to the next (this is before the data has been turned to midi) my patch plays the existing note once, before it plays the next note so on fast passages or with drums especially, it "flams" between two notes. When I take change out, the problem disappears, but the torrent of sensor data that change was holding back, slows everything down to a crawl.

Any insight? And by the way, my project would not be so far ahead were it not for this forum. Thank you. I hope to contribute one day, once I figure out where everything is. Cheers!

Onyx

Greetings all.

I have made really great progress with my controller project www.beatjazz.blogspot.com. Now its getting into the fiddley stuff.

I have been using [value] to sample the stream of breath data and use it to generate a velocity value, but I noticed that because it gets "stored" first rather than simply triggered in realtime, it is always one note behind, ie, if I play a note with a particular velocity, that velocity won't trigger until the following note because the value object is triggering the stored value

Is there a way of triggering or banging a value in Realtime, from the stream of sensor data? Something like [value] in that it will give one value per bang, but be in Realtime?

Thanks in advance,

Onyx

does any one know how to address multiple arduino fios simultaneously? specifically 3 fios coming into one xbee acting as a base station. or where i might look for the answer?

thank you,

Onyx

I am trying to find out if and how to create an arduino fio/xbee based wireless network of multiple fio/xbee's? 4 to be exact.

I am currently building a wireless system consisting of 2 "hand units", a wireless mouth interface and a base station xbee interfacing with pure data. www.beatjazz.blogspot.com. i already have the first fio/xbee communicating with the base station and pure data using pduino and all is perfect. this weekend i pick up the other 2 arduinos and fio's. all 3 arduinos need to connect with the pure data at the same time.

the instrument i am building is a wind midi controller except that instead of it being one horn like instrument, allthe left hand keys are in a wireless unit attached to the left hand and same goes for the right hand. the mouthpiece is worn like a headset microphone and is a pressure transducer for breath and a FSR for lip pressure. they must all work together as if one instrument. most of the coding has been done, andnow is time to split that code to the multiple wireless nodes.

it would seem that since all the xbee's have their own id's that that would translate over to the arduinos they are connected with and the whole network should come into the computer thru one serial port, which should solve that problem. I am wondering; how i might interface the 3 arduinos to pure data/pduino? theoretically, shouldnt i be able to name instances of the [arduino] object, say [arduino left], [arduino right], and [arduino mouth] so that each fio has its own instance? if so, how do i recognize the individual fio's and make use of their individual id's if that is the case at all?

Thank you in advance for any insight anyone might share.

Onyx

Greetings all. I am in the midst of a heavy project www.indiegogo.com/tronbeatjazz

I am building a wireless controller with a pd based "brain" (ie all the processing is done on the computer. Arduinos take care of getting the data to the computer). But I've hit a roadblock. I can convert button presses I to notes and control their velocity somewhat, but I can't control duration dynamically. How does pute data interpret duration data? I want to be able to hold a key, and blow into the pressure sensor and have those two data streams come together to form a midi note with dynamic velocity data like cc#2.

I thi k I need to construct a midi message from scratch. Make note isn't cutting it unless there is a way to dynamically create duration there.

Thanks in advance,

Onyx

does anyone know if such a thing exists anywhere i can download it? I had a set of flashcards i made for the vanilla objects for iphone,but it just got accidently erased so i will either make a new set and/or try to find an existing one.

anyone?

hello,

I have a problem. I am attempting to place PD in between my wx5 wind controller and everything else(windows xp sp2 running PD 0.42.5 extended), which includes fl studio as synth host. my goal is to re interpret the wind controller data before it gets to the other programs, and modify that data. for instance, gestural transposition using touch osc, and using certain note combinations to trigger events.

the problem is, from what i can ascertain, that the breath control data(cc#2) and the lip conrol data(cc#18) are very densely packed and it "seems" to be confusing PD. I have been trying to split them but it doesnt seem to be working. the value of the cc#18 data sometimes intersects with the cc#2 data and controls it, and vice versa.

even when i connect [midiin]->[midiout] directly, pd isnt sending all the midi data and it is not going thru to the synth host. it shows midi activity, but doesnt send any data to the synths or that data is jumbled and the notes come out in spurts. does anyone have any idea why this is happening and what i could do to possibly fix it?

onyx