Yggdrasil

[tyleretters]

@coreyr

delete a phenomenon without clearing the rest of the info in a step?

no, that is a good idea. i want to get phenomenon going in the editor too.

---

can you send me the routine you’re using? i’ll test it out.

[coreyr]

Will do after I mess around a bit more! My buddy @kingmetal pointed out that I can probably do a bunch of the routine file creation from inside maiden instead of in text edit on my laptop, then moving the files over to Norns… Looking at one of my tests, there is a whole lot of formatting garbage that text edit decided that I needed in a plain text file, so that explains some of the issues I was having.

I do have things up and running far better than when I sat down for a session earlier today! I had a nice long session last night with Yggdrasil driving my pod64 2 voice skiff via crow with some YPC drums. Gave Lacuna my first try, and it is so cool! Bouncing a voice between several tracks with differing clock divisions makes for some cool bursts in slower melodies.

[tyleretters]

oh awesome! yea RTF files will not work!

that’s a really cool idea to have lacuna kick over to slower tracks. i would’ve never thought of that

[coreyr]

Odd, so I had a few routines working before I went out for a walk this afternoon. The only thing that has changed since is that I installed the new orca-dev branch and slept my Norns. Had to do a system-reset to get sc properly working again. Yggdrasil seems to be ignoring the first line of my .txt files, if I leave a blank line at the top of the file it freezes Yggdrasil but not my system (don’t leave blank lines in your routines), if I double up the first line it seems to work. Editing the routine files in Maiden seems to work but are not available for Ygg to see until after a sleep cycle, if I make/edit them offworld in text edit (being sure that they are saved as plain text) and move them over to my Norns file system they seem to be available right away.

@tyleretters I’m wondering if you can describe your process for building/editing routine files. I feel like I’m so close but just missing something.

Sorry if I’m jamming up this thread, Yggdrasil is just so exciting!

[2197]

minor thing also related to routines: if i start yggdrasil and have default track depth set to 16, and then load a routine which tells a track to have depth 32, the hud isn’t updated (it only goes to 16) and i can’t move the cursor past row 16, but the track continues to play slots all the way down to 32. fixed when changing track depth to 32 in the terminal manually for any track.

continuing on the hud and gui: is there a reason for the space between row numbers and track 1 when the cursor is on track 1? since norns has quite a small screen, would it make sense to use as much of it as possible, and have channels 1-6 in view whenever the cursor is on track 1-3, and start scrolling horizontally when moving the cursor to track 4 and onwards? it somehow more satisfying to see as many channels as possible scrolling in a tracker : )

@coreyr you can set textedit to use plain text format in settings, but i would still recommend to use a proper text editor. i use atom, but there are many alternatives like sublimetext, emacs, bbedit, vim, textmate that will all do a better job than textedit. i have no problem with ygg ignoring the first line of my routines, so i suspect it might be textedit that adds some type of header.

[tyleretters]

@coreyr

Had to do a system-reset to get sc properly working again.

sc sometimes doesn’t boot up for me either. i can’t reproduce why but i don’t think it has to do with yggdrasil. (it was happening to me before yggdrasil existed.)

@coreyr

Yggdrasil seems to be ignoring the first line of my .txt files

could you share one of these files? i cannot reproduce.

@coreyr

describe your process for building/editing routine files

i pretty much always start by copying a run file over and subtracting a few lines here and there. no secret sauce here!

@2197

if i start yggdrasil and have default track depth set to 16, and then load a routine which tells a track to have depth 32, the hud isn’t updated (it only goes to 16) and i can’t move the cursor past row 16

could you try running tracker:refresh() in maiden right after this? let me know if that updates it.

@2197

is there a reason for the space between row numbers and track 1 when the cursor is on track 1? […] it somehow more satisfying to see as many channels as possible scrolling in a tracker : )

yes! there are two reasons.

first and most important: the cursor is always in the exact same position. this subliminal/subtle ux decision to does a lot of work with consistency. this program is crazy with slots lighting up all over the place. having the same “focus” slightly off center on the screen at all times reduces cognitive load and helps you stay oriented.

a distant second reason is to reduce the sensation of claustrophobia. yes, the screen is super tiny! one way to fight that is to give you the sense of space and breathing room. the negative space around the tracks and mixer do just this. this is, i think, partially where that satisfying sensation comes from.

[2197]

@tyleretters

tracker:refresh()

doesn’t update it. maiden just tells me <ok> but nothing happens.

@tyleretters

two reasons.

thanks! it’s very interesting to hear the motivation/thoughts behind certain design choices. even though i knew passively the cursor is always at the same position on the screen, actively/explicitly knowing it suddenly makes yggdrasil more easy to navigate : )

[tyleretters]

@2197

doesn’t update it.

blast. i’ll log the bug and get it fixed. the bugs in this things are so gnarly.

[2197]

don’t know if this information helps, but if i set default track depth to 8, restart ygg, load a routine.txt that sets track 1 and 2 to depth 32, and track 3 and 4 to depth 16, then the hud track numbers on the left gets updated to 16 rows. track 1 and 2 gets dots indicating more rows, but i can’t navigate any further down (either by key arrows or entering 1 32 in the terminal: “1 32 is out of bounds”).

[tyleretters]

v1.0.16

• FIX Copy and paste slots works again.
• FIX Copy and paste tracks works again.
• FIX Copy and paste samples works again.
• FIX Lucky phenomenon are no longer unlucky.

---

these bugs had all four

[2197]

i know the demo video is a haddaway reference, but it also reminds me of the intro of this amiga classic:

Jogeir Liljedahl - Physical Presence | ProTracker

[tyleretters]

oh wow i’ve never seen this before. this SLAPS.

[2197]

hey!
recently did a re-flash of my norns shield while trying to figure out a bug together with @unicity 's fourtunes script. i’m pinging them here because the bug there was also related the clock somehow? (see maiden error message below).

now when i’m running ygg, i always get a graphical freeze after a few (2-5) minutes of using. it has happened four times in a row. the sound is still playing, but i can’t enter any commands or navigate the tracker. i can press k1 to enter system menu and it is not frozen, but when going back to ygg (k1 again) it looks like all channels are highlighted and the whole gui is frozen and can’t be interacted with.
maiden gives me this error:

lua: /home/we/norns/lua/core/clock.lua:82: /home/we/dust/code/yggdrasil/lib/graphics.lua:383: attempt to index a nil value
stack traceback:
[C]: in function 'error'
/home/we/norns/lua/core/clock.lua:82: in function 'core/clock.resume'

i can enter params edit menu and change the clock tempo for instance, and it changes the playback tempo. i can also start/stop with k2, but not with the keyboard. any clues? : )

[tyleretters]

can you share your latest run files from when it happened? i’ll try to reproduce.

[2197]

sure!
as you can see, very basic stuff. i mostly entered notes via the editor.

1st:

Summary

1 synth;v;2
1 synth;v;1
1 synth;v;3
oblique
1 synth;m1;12
1 synth;m2;12
1 synth;m1;32
1 synth;m1;62
1 synth;m1;42
1 clock;2
1 clock;2
1 clock;1
1 clock;.25
1 d;32
2 clock;.5
sync
2 synth;v;3
2 synth;m2;12
2 synth;m1;32
3 sha;1
3 shadow;1
sync
3 clock;.25
3 synth;m1;32
3 synth;m2;12
3 d;16
1 d;12
1 d;18
sync
1 d;12
sync
4 d;32

2nd:

Summary

1 clock;.25
2 clock;.5
1 d;32
2 d;24
3 d;16
4 clock;1.5
1 synth;v;3
2 synth;v;3
3 synth;v;3
4 synth;v;3
1 1 c4

[unicity]

Not sure if it will help, but I’ll mention that I was only able to reproduce @2197’s Fourtunes error when I reflashed and then ran Fourtunes. Even then, the issue goes away on its own on my shield the next time I run the script. Not so for @2197. I’d never seen this behavior until updating my norns system to the latest version, but I still suspect it’s something I’ve done.

Also, like here it doesn’t seem to be caused directly by clock but rather an attempt to index a value that is nil (in my case it’s passing a nil array when calling a musicutil function).

[2197]

thanks for clarifying! i just saw “clock” and had a reaction thinking “not again!”, but i see now how it’s not related.

[unicity]

Oh I think “not again!” is still valid here - it’s the exact same line from clock that shows up in the trace, and it appears to be caused by an unexpected nil crashing the party.

[tyleretters]

interesting.

@unicity did you restart a few times or just relaunch the script?

[unicity]

@tyleretters as far as I recall I can “fix” Fourtunes simply by reloading the script. And once that takes place, the problem doesn’t seem to reappear.