Sam

[coreyr]

@zebra thank you for the reminder.

I am on a fates (pi3b+) running 200106

Sam is up to date according to the file manager

I’ve been chopping loaded tape files for a bit and Sam is starting to act up again. It was working just fine for a handful of samples but is now spitting out files with long silent tails. I was not monitoring in maiden, and now that I am I can’t get the error to repeat. I was thinking that it was an issue with very short sample lengths being saved, but it does just fine as long as the short sample is the first thing saved from a loaded tape.

Where I am running in to issues is trying to load a tape and then cut several short samples out of it. The first short sample will save properly. However, when I move the start and end point of a loop to later in the loaded file and attempt to make a short sample to export (0.50 seconds for example), Sam will save a longer file (2+ seconds in this case)

Maiden is not returning much, I don’t know if it is because I am connected via Cyberduck as well?

Here is what is coming back

Summary

1

matron

sc

script load: /home/we/dust/code/sam/sam.lua

cleanup

script clear

pset >> write: /home/we/dust/data/system.pset

script run

Engine.register_commands; count: 0

___ engine commands ___

___ polls ___

amp_in_l

amp_in_r

amp_out_l

amp_out_r

cpu_avg

cpu_peak

pitch_in_l

pitch_in_r

script init

reading PMAP /home/we/dust/data/sam/sam.pmap

metro_stop(): pthread_cancel() failed; error: specified thread does not exist

metro_stop(): pthread_cancel() failed; error: specified thread does not exist

Thank you again for any help. Appears that this is only happening if the second (third, fourth …) sample I am trying to save is shorter than the first. I can seem to get around it by re-loading the longer tape file to save the second(+) sample.

[Justmat]

Yeah, that read out doesn’t say much/anything of note This is probably a script problem, and I will have some proper time for it tomorrow.

edit: Ah! pretty sure I know what’s going on. I’ll try to get a fix out tomorrow, sorry for the troubles!

[coreyr]

Thank you very much Mat! No real troubles at all, grateful for all your work on your scripts. Quite an inspiring suite of programs.

[Justmat]

Looking at this now. I see a few areas for improvement and will have an update ready this weekend. I may PM you for additional info, if that’s cool

[zebra]

200106

oh ok, that will do it (re: changes in disk write implementation)

@Justmat i think you were following the PRs on the disk worker threads, but it was a pretty complex change so i’d certainly any feedback or reports of weirdness

reason i ask here is, Sam is just the kind of thing that would be affected / can push the envelope of the write-to-disk feature.

before changes, disk-write was executed immediately and would stall event handling in crone. it would also take super long to write (minutes, for the whole buffer) so the hang could be long and look like a crash.

after changes, you should be able to queue up a practically unbounded number of disk-write “jobs” that execute in the background, much more quickly. (~few seconds for whole buffer)

if we need to, we can pretty easily add a completion callback mechanism.

[Justmat]

I pushed a small update. It removes a couple unneeded lines, and adds auto loop resizing when loading samples.

[coreyr]

@Justmat, I’m getting an update fail attempting to update from the maiden file manager:

‘Updating project “sam” failed
update failed: worktree contains unstaged changes’

Should I attempt a de/re-install?

[nitefish]

Same here.
But when I tried again, I had “already up to date”…
Edit: @Justmat, great tool!

[Justmat]

Should be ok to delete/re-install. I’ve gotten this before with maiden updates, but like @nitefish, the next attempt says it’s up to date not sure what’s going on there.

[Tim052]

Got the newest version and seem to be having an issue where the sample isn’t looping worked fine in the old version and latest firmware with fates. Also possible UI suggestion move the record to ALT + K2 it’s too easy to acidentaly clear your recording while messing around with the loop parameters.

[Justmat]

I’ll take a look at it this morning.

re: UI, I agree. I made sam really quickly, and there is very little to the UI. Maybe I should add some indicators on screen for what key2/3 are doing.

edit: @Tim052, i’ve fixed the looping bug. starting working on a more informative UI now. should have something up tomorrow.

[Justmat]

update v1.1

• various screen improvements
• various bug fixes

[papernoise]

Sorry if this is a silly question, but in which way is this a slicer? Can I create slices from a recorded audio?

[Justmat]

Yep! You can load up a long audio file, select start/end points, then save that smaller selection as a new sample/file.

[papernoise]

Oh that’s what you mean by slicing! Still, I can totally see how this can be useful! Thanks a lot!

[Hlp]

Was wondering the same and i think the best interaction on norns for now is sampling with sam, and then loading up a folder of those at timber player where you can actually slice non destructive.

[Hlp]

An update to choose save location would be cool to make/fill folders for timber!

[Justmat]

Just pushed a small update that changes the way files are named. We’re now using the same text entry system that the params menu uses. I think this is way better

[hypnosapien]

I got an error in maiden when I tried to update, but unfortunately, I didn’t take note of it. Tried updating again and maiden says it’s already up to date. But when I save a file, it just gives it the 000x.wav name. What am I doing wrong? Thanks for the ongoing work on this!

[Justmat]

Hmmmm, I removed the file that keeps track of the sample count, so it definitely shouldn’t be using that naming scheme.

Maybe try deleting the script and reinstalling? I have definitely had the error, then already up to date situation happen in maiden. Not sure what the cause is.