prgm beta

  • here is short info if you want to test prgm 0.0.3. here is the link to the files etc: http://monome.org/docs/aleph:bees:sharing:tracker

    this is still an early beta so be gentle! the main issues that are highest on the bug fix list:
    - split edit and play modes so that you can stay on one step and edit while the sequencer is playing
    - timing, the trig is sent as the last parameter (after modes, flags and other parameters) so there can be some weird (but hopefully funky) timing shifts when there is a lot going on
    - secure boundaries! you can still end up in uninitialized audio buffer territory, its very loud and sounds very bad so keep levels reasonable
    - cv output glitch when too many calculations per frame

    I´m happy to get input and ideas in this thread!

  • looks amazing. but, I'm going to have to wait to test this out in the new year since I'm playing live with my aleph a lot and don't want to mess with the current scene it's running. :)

    eager to hear how others get on

    also congrats @Test2 on the first non bees app!

  • congrats indeed, and thanks for sharing your progress in that matter!

  • sweeeeet!!!

  • i've managed to almost blow up my speakers with a hectic blast of noise but apart from that can't get a sound out of it..what am i missing?

  • @binksman connect sound to input 1 or 2, set a step to modes 'recin0' or 'TGrecin1' and the trig to 1, press sw3 (play step) to record to that tracks buffer. on the same track, set any step to modes 'one', 'loop', or 'scrub' to playback the recorded buffer. I will have a look and se if its possible to prevent uninitialized buffers from playing, that's a good input (maybe merzbow will reject), hope your ears are ok! oh, and I need to rename the record modes to recin1 and TGrecin2!

  • Thanks @test3 I will give it a try tonight

  • @all thanks, I hope it is the first of more to come!

    just added a small update on the sharing page, 0.0.4
    - audio buffers are now initialised to zero, still there could be a scenario of ending up outside boundaries so continue to keep levels moderate(!)
    - fixed a dc pop bug in the TGrecin2 mode
    - tweaked mode selection and the scrub mode

    enjoy!

  • anybody had success using this?

    trying today and i cant get any sound output so far

  • do you have to hold "play step" to record? or press start and stop?

  • these are the parameters that needs to be setup (can be done in any order...)
    - set at step to a record mode (row 2: sw1+enc)
    - set the step trig to 1 (row 1: enc)
    - set a record time (row 3: sw2 to enc, up to 16 seconds)
    - adjust the input gain knob so that the led is active to the incoming sound

    then press play step (you don´t need to hold it down, though that maybe is a good idea to implement for recording!) or send in a foot1 ot cv1 trig.

    that will start the recording to that tracks buffer, for playback set an step on the same track to one, loop or scrub, and don´t forget the trigs and time settings! all step parameters are parameter locked at the moment.

    it will also work when you have a clock input to cv1 or foot1, this way you can set one (or more) steps to rec mode and others to play to continuously playback sequenced fragments of the incoming sound. (!) in this version you have to pull or stop the clock when you edit...

    in the play modes you have also a parameter on the other page, usually its start time for and its controlled by enc's.

  • thanks for the suggestions, gonna try this right now

    i've been using foot1 as a trig

  • i'm hoping spend some time with it over the holidays, will share the tearing cacophony of sinew and bone that i make (just kidding...my stuff's pretty..)

  • hmm @test2 i had been getting flashing leds all along and now i realize sound was passing thru but i had my mixer levels low because of the volume warning

    i cautiously boosted the volume on my mixer (aleph has been at max this whole time)

    i finally heard audio input

    i've lowered/tweaked the volume a variety of ways and it is distorted thru the entire range of the volume encoders, yet the highest setting is still extremely low

    just in case my aleph/mixer/cables were the issue i switched to a BEES scene with lines and found no problem

    this is beta so i know there might be hurdles while using it...just wanted to describe what's happening to help you troubleshoot

  • also

    still no output from the buffer
    if the source stops, sound stops

    it might be helpful to have a description of where "step" "track" & "row" are located and edited because i might be confusing them

    i think row refers to each of the four mono tracks arranged in a box on the screen
    step is represented by the number in the lower left corner

    is that correct?

  • I agree it is confusing, I´m still not 100% about what to call these things myself :)

    the idea with this app is having separate 4 tracks, on screen they are placed to match the placement of the encoders, so enc 1 is always track 1 and so on.

    what I like to call "row" is the row in each "box" of track parameters, oh my god it's getting more confusing still! :)

    on page PRGM:
    trig
    mode
    time

    on page PARAMETERS:
    it´s still a mess :) but turning the encoders on this page will change the parameter, if the mode has got one and then it will also indicate on the screen, so only mode one, loop and scrub and HOLD.

    step is the sequencer position, play step (sw2) is a bit strange but it´s the only way to play a step without moving the sequencer forward, which was needed in this version because the way play and edit are using the same counter. sw3 is step next without playing the next step!

    in tape recorder terms, recin0 and TGrecin1 are the record heads for input 0 and 1, and one, loop and scrub are the play heads. the strange but also nice thing is that you can sequence all of these on the same track, along with the cv modes, so one step may output a cv trig thru the cv out and another a looped buffer thru the audio out...

    I will try and make this more intuitive and it would great to get some input on this based your workflows!

  • prgm 0.0.5 is ready, it is still a lot of work to be done but this version have these improvements:

    - edit and play are now separated so you can edit away and listen to the sequence at the same time
    - sequencer can now be trigged into audio rate using an external trigger (please try this!), and it is much more solid becuase the whole thing is "onchip" now
    - an embroy to a wav mode is there as a suggestion on how this would be implemented, it does not work yet but it is getting closer...
    - input recording distortion is fixed
    - it now has 1-32 steps and you can select some common measures
    - ui is fixed on the parameter page, it now shows when a mode is available for editing and dims the text if there are no parameters that are editable
    - the ui is also reworked so that you can use more as an instrument, and this will be further improved later on...

    the zip includes all files and folders that are needed, the sample files are there just to examplify how the ui will work with wav's later on but folder structure is needed to start the app without any errrors

    link to files and info, see the first post!

    edit: oh, and it takes about 10-15 seconds to start after the boot!

  • wooooooooo yeah

    i'm studying for a class but afterward i'll load 0.0.5 up and see how it goes

  • whats the new record button?

    i see that foot switch will be included in future update

  • @gli foot 1 will be record again but it needs to be reworked now that edit and play are two different counters, when it's back there will be a fast way to setup recording with switch 4 and encoder 4 that will place the app in record mode directly to a selected track without editing any steps in the sequencer.

    the way things are now you still need to set three step parameters on the env page. encoder 3 is step play now so when you have setup a step to a record mode then make that on screen and move encoder 3 in any direction. encoder 3 moves the playhead to the step you have on screen and plays that single step, you can also trig the steps by cv1 in as before, then recording will start when the play counter reaches a trigged step that is set to a record mode.

  • thanks for the tips

    so i have to have one track in "recIn" mode...and at least one track in something else (like "one" or "loop" for there to be trigged sound?

  • where the vid of this again?

  • @gli replace 'track' with 'step' in your last reply, yes.

    @binksman, something was posted in the thread 'Aleph tracks and videos' but that was just some simple sound example, not very instructional.

  • any chance of a newer vid? os audio examples?

  • here it is..

  • been in love with that and rewatched a million times to try figuring out the previous incarnation

    but i agree, a newer video would be great since some controls have changed

    doesnt have to be an instruction video at all

    the sound demos are awesome

  • @gli thanks for the kind words! new videos have to be later on, sorry!

    Added a new version 0.0.6 beta, see link in the first post.

    this is new:
    - selectable inputs
    - one aux per track with level control using sw4 + encoders
    - new mode: [aux master]

    this version is a first step into aux/fx territory, and it also opens up the possibility to route outputs to inputs internally, including the aux master input, which is a mix of all outputs with level attenuation. With that, a new type of modes is added : [aux modes]. These can be different types of audio (or cv!) processing, the first one is a basic but useful one called [aux master].

    note. I have only tested this version using the noise mode, so please let me know if you get any problems with audio quality/distortion in some scenario, and 'In 1+2' input is not tested, I only had one cable(!)

    and about aux... the mode [aux master] requires a trig in this version, the trig is required to let the bfin know it's time to check which dsp to run. Later this could be "included in the mode", eg a trig sent already when selecting an aux mode, this maybe makes more sense because it is thru only...

  • hi,
    so i'm finally getting around to trying this out. Is it possible to use without CV? I can't work out how to change the speed of the sequence or the step it is playing? (apart from the scrubbing the steps with an encoder? Am I missing something?
    thanks

  • hej! the sequencer only works with a trig to cv input 1 (there is no internal clock yet!) the next version (0.0.7) will have more "aux modes" (a delay and a filter are ready as of writing, plus more internal routing/feedback options) so it will be more useful also without the sequencer, more like "lines" type of effect setups will be possible too.

  • Ah I see, are you planning to add an internal clock? Would make it super useful for those of us without modular set ups :)

  • would it make sense to be able to have a sequence of effect chain setups, using a foot-switch to advance to next?

    each step can work kinda in a similar way as a preset in bees, though the i/o settings are not sequenced now, but they could be, I´m just not sure if this kind of setup makes sense or not... glitch/pops could maybe be a problem when moving from very different setups in this way.

  • no, i think it's the sequencing bit that's interesting (i.e. that's what the system is based around!) jumping from one preset to another would be weird if you wanted to have regular record steps for example.
    Just a replacement for the CV input would be enough, some kind of pulse that you can set time interval of. In the current version is there a way to limit the amount of steps by the way? I couldn't work out how to do it, is it fixed to 32

  • thanks for the input! it´s been on the list for a long time but I think (this is my current best guess) that is requires quite a deal of knowledge around hardware clocking/interrupts, for a stable, jitter-free clock, and to be honest I don´t know how this would be implemented in a good way. the external clocking is very fast though, it can be clocked up to +500hz at the moment, though that is of course of no conciliation when that's not an option(!)

    limiting number of steps is actually in the code but I had to remove it from the ui to make room for the aux level controls, just need to find a new knob/encoder combo for this one and it will be back again.

  • one other thought, if you can control the amount of steps, it would also be cool to be able to choose the starting step. That way you could set steps 1 - 8 to be one set of actions, 9 -12 as another for example, then would be like sections you could move between, or jumble up in the overlap :)

  • maybe a step to midi note map could be used for this kind of functionality?! another way could be an A-B mode that splits the sequence in two parts (edit: maybe two equal parts + a way to set how many repeats before moving changing part?!? there is a similar kind of functionality in the sam-16 sequencer from snd), if a clever way can be found to setup the ui for it...



    Added 0.0.7 now so it´s it is possible to test the new modes, some things that needs to be fixed are not ready yet though!
    - knob scaling/response; the feedback, level and cutoff frequency controls needs to have exp/log response and better smoothing, this is still not ready, so a big glitch and pop removal (due to encoder value jumps) is no 1 on the to do list
    - aux mode trig, still it is necessary to set a trig manually and play the step once when changing modes in order for the change to take effect, this is for the aux modes mostly...
    - improving the algorithms, better phase invert, try to add inverted feedback to the delay, try to add a combined level+q control on the bpf so that is works more like a resonator, there is also an idea about "virtual output" an example would be sending the feedback thru the physical output for processing and back again, while at the same time sending the total output internally to another track
    - screen glitching, the text output on the screen is a hack, needs to be rewritten to avoid lost pixels and also there can be text dropouts when moving encoders

  • EDIT: removed the 0.0.7 shortly because I had a problem with my unit, but it was not dsp related so the files are UP again and they are OK.

    ORIGINAL:
    I tested it yesterday, just fine, but today something got burnt in my aleph when I turned it on, and better safe than sorry I removed 0.0.7 until it´s safe to say that it did not cause this problem!!!

  • terrible news!

    sorry to hear...hopefully there's not permanent damage :(

  • @gli overloaded one of the audio inputs, but everything checked out OK after I got help from monome :)

  • Really nice to hear that you guys still working on OS. Really need little more clocking possibilities (Clk in) and maybe couple more original scenes not audio, but cv based.

  • just uploading an updated 0.0.7 now, finally it is possible to load and sequence samples, however it´s very crude at this moment, but still, yay! :)

    I have included a couple of samples in the zip.

    BIG NOTE, it takes 2 minutes approx for the included samples to load over spi, and while this is happening the aleph looks rather dead but it will light up when it´s done.

    if you want to try your own sounds, just dump them in the data/prgm/samples folder, all files in this folder will be loaded as samples but it only supports headerless mono signed pcm16be files --> 16 bit / 96khz / big endian for now, you can use ffmpeg or similar tool to convert sounds, longer sounds will be difficult at this moment, compare with the ones I have included.

    the mode is still called 'wav' and the samples will appear with their file names on the parameter page, so just scroll and select, the order is still confusing but the limit is 40 samples (there is no technical limit really, I could set it to 240 instead but…) or SDRAM limit (there is no memory checking in the app yet!!!)

    I´m still not able to remove the the timer used when loading samples, so the overall timing of the app on external trig is less tight than last version, but it will be fixed later.

    also, a new mode called GLOBAL is added, use it to set sequence length, and then it can be removed, the setting is saved globally and not connected to the sequencer.

  • is there internal clock on this version? (to use without cv?)

  • no, but it should accept anything resembling a pulse trig!

  • not sure what that means.. can i just send it audio? what does a pulse trig sound like?

  • there is a sample called trig included in the zip! all the cv input does is looking for an upwards change in level of a certain amount, then it stays in this state until a next upward change is coming in, downward changes are detected too but they do not cause a trig event to happen.

  • can we have tutorial vid please?

  • there are still some things left to do on the sampling, after that I think a tutorial can be a good idea too :)

  • yew! looking forward to it!

  • minor update (0.0.7)
    - added sample memory checking (limit is now 128 samples or 32MB)
    - improved sample loading (4 bytes/ms)
    - added level control per step for samples and noise
    - step time now syncs with sample length

    still a few things left to do before 0.0.8 and tutorial though...