#207

Hi @dan_derks,

checked out the arpeggiator in beta4 (video around 34:12). Here again some of what I notice can be due to my ignorance but it seems there are some weird things going on as well:

  • it seems there is a hold and a start/stop arpeggiator key (the 2 buttons besides zilchmo 2) but the behaviour is not clear to me, especially in conjunction with the alt key(s); e. g. you can set the pad to hold mode by pressing the start/stop key (the one directly besides zilchmo 2), screen then shows pause and as soon as you press pads you’ll get an arpeggio

Does it make sense to document what I found over at Github or is this a building site with known issues anyway?

#208

hey! good eye — the button has changed since the video (again, apologies for zero beta docs, will be doing a spike on them in the next week), but it should work like this:

  • press arp button (there’s only one, in between the traditional “pad loop” and the leftmost zilchmo 2) to turn on arp mode
  • hold pads and they’ll arp, release and they’ll stop
  • if you are holding pads and press button again, they’ll enter a hold mode and continue arpin’
  • press more pads while in hold mode to add those pads to the arp (alt+pads removes those pads from the arp)
  • if you press button while pads are arpin’ the arp will pause
  • press button while arp is paused and they’ll start playing again
  • alt plus button to clear the arp

if that’s not happening, then there’s a bug — i’ll be able to investigate any repro steps next week. thanks so much!!

3 Likes
#209

Right. I got confused because I also messed around with the loop/1-shot key at the same time as using the arpeggiator…

One more thing: If you are in arp mode via menu the following is possible (feature?):

  • press ALT/alt (global or local) and press arp key
  • screen says pause
  • press pads, create arpeggio and set bank to hold
  • arp key is not on
  • press arp key to light up key
  • press arp key to stop arpeggio

And final edit: This is so much FUN!

1 Like
#210

hey! the UI changes do feel better for sure. i too was thrown by the change in k1 speed re: entering the menu, however i was able to figure it out eventually and it makes the pages that do use k1 as shift feel so much better. if it’s still 100ms in beta 4, you could maybe go up to 125-175, but dropping that time down helps a lot for me personally.

some other stray bugs that i wrote down in my notes, not making a github issue for each as i don’t have steps to reproduce any of the following, only the last one was really bad:

-the local shift per bank doesn’t seem to work work in all cases you’d use shift for

-gird gesture to trigger filter mode for arc doesn’t work in bank with an active euclid mode

-small chance looping pad gets stuck retriggering forever

-random delay send seems to occasionally produce a feedback loop independent from the delay controls (requiring immediately ripping your headphones off level feedback lol). i was trying to tap the grid controls to zero out input and feedback but with random delay send (and maybe random feedback if that’s an option? can’t remember) it just kept coming back

sorry haven’t been able to spend as much time with it as i want to but it’s so much fun.

#211

Hi Dan, was a downsampling/lofi effect on the roadmap for CC? I remember some discussion in one if the earlier threads.

#212 
engine.name = "Decimator"

-- sample rate
  params:add_control("sample_rate", "sample rate", controlspec.new(0, 48000, "lin", 10, 48000, ''))
  params:set_action("sample_rate", function(x) engine.srate(x) end)
  -- bit depth
  params:add_control("bit_depth", "bit depth", controlspec.new(4, 31, "lin", 0, 31, ''))
  params:set_action("bit_depth", function(x) engine.sdepth(x) end)
  params:add_separator()

if you have otis & add this to the beginning of the script i think that’ll do it for the the input. there are some options to abuse the delay to downsample the output

3 Likes
#213

Oh yeah I love the otis implementation. I’ll give it a try - thank you

<3

#214

lemme know if it works ! i meant to slide that into my fork at some point. if it spits out error messages you can try plopping the param stuff into init()

#215

Didn’t work for me unfortunately, and I’ve got the latest controlspec.lua ( commit 5559390) :sob:

#216

will be dropping back in tomorrow for some concentrated work (been social-distanced temporarily moving to the east coast this weekend), but real quick:

@ypxkap, dope, thank you for all those notes! i know what half of them are immediately, excited to explore the other half :slight_smile:

@yoyosandshoes, here are some quick delay settings to get really nice downsampling on output (if the live-sampling folks want decimator in there, i can totally make that canonical, as well!):

  • set delay to free
  • set delay time to 0.001s (use k1 hold to fine-tune)
  • set delay fade to 0.001(use k1 hold to fine-tune)
  • rate: 12 (or to taste)
  • feedback: 96%

what i like about this “abusing the delay” approach is that you get to dial this effect in twice (once for each delay line) and that you can create presets for these settings and that you can choose to send only one pad or a few pads or all the pads to it :slight_smile: @andrew’s m4l devices make great use softcut in this way, as well!! it feels really neat that softcut can cover all of these bases.

@madeinspace, is your matron error the exact same as was described upthread? and this repro script returns the same errors? or is there something else happening as well?

this “clear it before there’s anything in it” sequence is a known bug, but forgotten bug! thanks for re-surfacing, super glad you’re having fun!! :sparkles:

4 Likes
#217

Just installed beta 4 and WOW. so cool to see the opz integration. They play together really nicely. Thank you very much for making this.

quick question (kind of a norns noob but maybe others have similar q)

how do i change between recording buffers? Right now it’s at L1 and i can’t seem to find the button combination to change to 2 or 3. If it helps, I’m running this on fates.

1 Like
#218

no worries @okyeron took me through some steps that fixed the issue, the repos were out of sync with the main norns repo, working now :slight_smile:

#219

Go on top of L1, press K3 to see length and offset, hold K1 and move E1.

1 Like
#220

was digging through the menus looking to see if filter resonance was buried in there anywhere and discovered random record %. not sure if this is new or not but really nice addition!

patch info: ipad running aum with @burns_audio’s iOS mutable instruments ports. fugue machine triggering @junklight’s ioniarics polyrythmic arpeggiator triggering mutable. seemingly huge potential here in using the 4 fugue machine playheads at super slow speeds to create longer, evolving chord progressions. synced to norns via link, where live record random was set up around 40%, all three banks acting as different types of looper/delays.

because of link you can hit the zilchmos to set pad length to 16th and then double the length to get tempo locked loops. creates the possibility of CC being the tempo locked “musical” figure while the source music is just absolute chance based, basically the exact opposite of every other approach ive taken with live sampling. you can kind of hear what i mean at the end of this clip, the synth arp is actually norns while the source continues playing whatever out there timing it feels like.

some small bugs i hit––i’m not comfortable enough filing separate bug reports for these yet as they weren’t bad enough to break the creative flow, and have no idea how to reproduce!:

––filter crashed on bank a. pressing the 3 keys to open it on the arc didn’t work and then going into the filter params page also didn’t allow changes to that specific filter only. other two were fine

––“restore default start + end based on pattern ID” grid gesture becomes unstable when using multiple buffers. seems like the start gets set to after the end and nothing plays, becomes impossible to reset.

––saving + loading presets seems to be busted—the presets never save and norns hangs on scanning when trying to change scripts. i have to reset to fix, after which preset does not appear to be saved. (complete guess, but maybe this is because i have the beta and the regular CC installed at the same time?)

––somehow, i stopped playback on euclid bank a. if this is a feature, no key combination seemed to restart it.

––on the loops page, the “end” number in the top right doesn’t scale correctly with the record buffer slow downs.

and a few other tiny UI thoughts/requests:

––if it’s possible, it would be nice for the normal 500ms k1 window to work when you’re ALREADY IN the params/norns menus, but leaving the 100ish-ms one when in the script. i occasionally lock myself out via slow reflexes and as far as i know there’s no real reason to have the shortened input time in the menu section.

––in random record toggle, seems like grid should give visual feedback when recording (like it’s doing on the loops page)

––expose resonance controls somehow. on the filters page, the … on the bottom left makes me think they might be already there and i just don’t know how to scroll to them?

––add “playback direction” as a param target in rnd (maybe could just fit in as a third boolean param for rate?)

sorry for typing forever, i’m all pumped up on fake bell sounds

5 Likes
#221

yooooo, me too – that capture sounds fantastic!!

thanks so so much for the deep testing! re: some of the bugs you hit…

@ypxkap (tagging because this was an edit), i’m totally unable to repro this, if you have any steps you can share? i tried with both live and clip, setting different pads to different buffers. they always restore as expected (to their original positions to the nearest 1/16th of the buffer’s total length).

this sounds like a softcut control crash, which feels weird, lol. super curious if you’re able to repro and what kind of load it takes to repro. i kinda wonder if that’s what happened with the looping pad not being able to turn off loop and the delay feedback loop (since there’s no feedback adjustments being made by rnd)? keep me posted!

is this collections? or pattern presets? either way, i haven’t heard any persistent trouble with these from folks and they’re working on my local devices – maiden error output would be clutch. it might also just require clean installing the script (delete the script and delete the cheat_codes2 folder from data).

blurgh, this just means the clock coroutine driving euclid for bank a is failing, which isn’t a script-level thing. i can build in some restart hotkey, tho.

re: the other UI thoughts:

I misspoke earlier – the default K1 hold time is 250ms and it’s wild how much of a difference shortening that to 100ms makes for performative gestures. i might be able to put a callback in to check _menu.mode state to see if the SYSTEM menu is open and if so switch over to 250ms, but otherwise I could just override K1 except when it’s on the cheat codes main menu – so you can only use K1 to get back to SYSTEM if you’re on the top-level script menu. with some documentation, this should be easy to acclimate folks to, but curious what others think!

current behavior is to generate random direction with random pitch (so, min: 0.5 / max: 2 is generating forward/reverse between half-speed and double-speed). would you like the option to only generate random pitches in a specific direction?

1 Like
#222

Sorry I can’d find the info, which line do I need to replace on fates to run the beta? Thanks

#223

there have been mixed results, so i’m not entirely sure at this point – it might just be best to hold for the public release + fates OS update (unless @madeinspace can share the steps?). but I really don’t want to create any support load for @okyeron just to get some testing done :slight_smile:

1 Like
#224

Sure, of course. Thanks!

#225

hm… i think in the present released version, it is possible to blow up the SVF by raising the cutoff frequency too high, and higher resonance will make that more likely. but limits on parameter ranges should prevent this.

if the filter didn’t blow up (e.g., still audible with post_filter_dry at minimum,) but you can’t control the filter, then that’s quite weird and i can’t think of an explanation from the backend perspective. (The Razor suggests a lua-side problem.)

3 Likes
#226

Ok so, ssh into fates and folow these steps:

cd norns
git pull
git checkout main
git pull
git submodule update --recursive --init 
cd ~/fates
./norns-stopall.sh
cd ~/norns
./waf clean
./waf configure --enable-ableton-link
./waf build
sudo reboot

it worked flawlessly for me, but use at your own risk :wink:

1 Like