Yeah, Dan has given me some instructions. I’ll take stock of inventory and clean up all my ports for repository listing maybe later this week.
ok after further experimenting, this 8x canard actually works ok on mm!! (80%c @ 48/512):
seems using in3 instead of 4 is the way to go for now…
u knob is mapped to the push button on the trigger module: have to turn it below .5 then above .5 to trig everything.
edit: also further testing, am able to load ok: 8x 1min long 48k/16b stems (initial ram usage 44mb, after all stems loaded 178mb used) without mm freezing. quite useful for anyone wanting to create asynchronous loop clouds
Nice stress test! But it should work with the gate input… I’ll give this another look later.
1 Like
after further testing, am able to load and run ok: 8x 1min 20 second long 48k/16b stems. there is almost a minute’s pause after hitting patch play button while all 8 load into ram. so with no load-spinner ui visual, it feels like the mm might be crashing… but then it starts running ok(i.e. cpu meter appears in top right). im confident i could load and run even a bit longer samples files, if i unload some plugin sets maybe. have pretty much most of them auto loading. init ram usage at startup = 44mb, after 8x 1.3min long 48/16 samples = 178mb used.
edit: have twice run through a series of 12 patches using different sets of 8 of these large samples, so far no crashes 
1 Like
Very happy to hear that!
I am having personally not so great of time even getting the VCV Rack versions to work? Half the time it plays and half the time it doesn’t for me, and a lot of the code is eyebrow raising.
edit:
The canard really doesn’t play well with LFOs… it really wants a push button it seems like. I mean if that was the original behavior I believe the MM is matching this.
the gate input 4 is just straight up buggy imo. looking back in some of my old pre v1 vcv patches that used this module, noticing im always just using trigger input 3, avoiding 4. im assuming this weird behavior has just been in canard since about forever.
i think the issue maybe have been the “!” in the filename
Weird. Nice find.
Yeah, the code also has a lot of strange conditionals which seem to expect a perfect 0V idle state, which is not reasonable IMO - it also means it can’t work with LFOs dipping below and above 0V which seems like expected behavior.
also i was trying to get an autostart going (+5v atvert plugged into in3) and found another bug where canard will crash the mm if no sample is loaded in, and then a voltage above 0 is hooked into the trigger input 3. this behavior is only on mm, not vcv
i think canard is great at least for the tasks of playing oneshots, or just having async loops going. but with the latter it seems there’s no way to stop a sample while it’s in the middle of playing (although you can restart with trig in 3). would be nice to have that gate working, but understandable if the code is just a bowl of spaghetti
Interesting, I pushed an update to hopefully provide a safer playback guiderail to prevent reading sample data that doesn’t exist. I haven’t tested it though yet.
Gate should allow you to stop a sample while it’s in the middle of playing, I agree, it should work but am sometimes having trouble still, I can dig into the code some more.
just gave it a try and it seems to be working ok, no crashes if no samples loaded while volt above 0 in the trig input 3
1 Like
@etcetc I’m sorry for the basic question, but I would love to try out these newly ported devices. Are they available to use or would it be better to wait til they show up on the 4ms plugins list?
more experiments, am able to get the gate in to work with an lfo (or offset+knob mapping) running the lfo through a comparator that just spits out full on or off, nothing in between. i guess that’s the trick to make canard happy. so i recant about avoiding gate in4. no doubt its wonky code under the hood, but it seems to be working ok more or less w the compare module
I finally got cANARd to v/o track. in 10 seconds today Ha Ha
I must have been really tired last night, and not zeroed some critical knob. 
earlier post deleted.
they are definitely available for any of us to use but you will need the current beta dev 2.0 firmware:
https://metamodule.info/downloads
i grabbed the latest version of bidoo from this github link (you’ll need to be logged into github i think):
1 Like
canard spd gets weird when you are pitching down really low, it starts to only go in steps, not very incremental which is unfortunate, but still usable for experimental things imo
great! Thank you @offthesky
That is helpful to know.
Currently I am mostly interested in octaves, 4ths and 5ths, forwards and backwards as the other harmonic tensions are added in live performance.
I hooked up Befaco’s Voltio to it today with a singe test tone, and it was behaving much more in line with what I was expecting.
Looking forward to diving into this module more, hopefully tonight
@danngreen im noticing that when loading in new samples, while other sounds are playing in the patch, the audio buffer gets stuck until the sample loads in (longer if bigger samples file). im curious if this is something that we might see fixed on mm at some point? would be nice to, during a live performance, be able to load in new samples without the buffer thrashing out.
also @etcetc - found a bug where mm crashes if you have a sample loaded up, and then turn on record (little red button lights up). also crashes if the sample is playing(gate=0) and you start recording. mm seems to work okay if you clear the sample first, then start recording.
on vcv, canard has no problems recording while a sample is loaded or playing, when you stop recording, the module immediately switches to the sample you just recorded, and if gate is above 0, starts automatically playing the recording, which is really nice.
also noticing some cpu spikes while recording (but not the buffer glitching you get when loading in a new sample), one spike(little red cpu icon in top right) + audio glitch, every 2 seconds or so while recording. system set to 128 (512 causes a lot of sporadic glitches with sample playback) and spike when recording is done. maybe @danngreen would know whats going on there(ram issue?).
test patch:
canard recording test.yml (3.5 KB)
a=gate play
b=record
c=clear
u=change root of osc