#325

While trying to document the timing drift, I think I may have figured out what’s happening, or at least have found where the drift is actually coming from.

I played a 16 note phrase from polyearthsea clocked by PNW. I also used Pam’s to clock a kick and hat pattern in tandem with polyearthsea. Let it run for 20 minutes, and nothing slipped. What I discovered was my timing discrepancies were with the clocked stretch sample player on my 301. While running a 4 bar drum loop that was recorded at 84 bpm, and clocking it to something like 90 bpm, this is when I start to drift in and out of time. It takes about 3 minutes to start to notice the drift, but at least I ruled out polyearthsea as a contributor to the issue.

More investigating to come…

Found it! There is a shift parameter in the clocked player unit that was the source of slippage. I had in set to a non-zero value thus every trigger it would be slightly off.

#326

I’m trying to get polyearthsea on my ansible to work with Just Friends.

Will it work with a crow with pullups enabled? At the moment, my i2c bus is set up like so: crow > ansible > JF

I’ve set up voices to go to Just Friends as per the docs, but nothing seems to be happening!

#327

yes, should work with this setup.

to confirm i2c is working properly: are you able to control JF from crow?

also, in addition to mapping voices you need to enable i2c, 6th button from the top in the rightmost column on voices page: https://github.com/scanner-darkly/polyearthsea/wiki/i2c-control-and-voice-assignment

make sure to set sound/shape switch on JF to sound.

#328

i2c is working properly—I’m able to control JF from crow.
i2c is also enabled within polyearthsea, and JF is in sound mode.

Interestingly, when I enable i2c in polyearthsea, JF briefly makes a sound (‘the I changed mode’ sound), which leads me to believe that Ansible is also speaking to JF.

However, mapping voices within polyearthsea still gives me no sound output from JF. It’s very odd!

#329

couple more things to try:

  • map same voice to both physical outputs and JF and confirm you’re seeing it output on physical outputs

  • on the voice assignment page in the JF column try toggling the bottom button - it switches JF between synth and regular mode (although even if it was in regular mode you should see envelopes being output…)

#330

wow! i’m blown away! I need more time, more txo’s and coffee!

#331

hi @scanner_darkly!
I’m trying out polyes with my earthsea monome module and JF via i2c, everything working great. one question, is it possible to control earthsea clock via teletype this way?
I have ES, JF and TT i2c connected to a TELEXB.

#332

unfortunately, no. since you’re controlling JF from ES, polyES is acting as an i2c leader, so it’s not able to receive commands from teletype. a workaround would be to use a trigger output from teletype into a clock input but this would require running polyES on one of the modules that has a clock input - ww, mp or ansible.

1 Like
#333

thanks for the reply, that’s what I thought :disappointed:

#334

one potential workaround is to go the other way - clock teletype from es:

  • select a pattern that will serve as the leader/clock pattern
  • enable only voice 8 for this pattern
  • record a note (one note is enough, doesn’t matter which note)
  • set the pattern to be the leader pattern - enable clock only, not reset

now other patterns should be synced to the leader pattern when you linearize them.

now to get the clock out:

  • open voice assignment again
  • select voice 1 and unmap it from the module outputs
  • select voice 8 and map it to the module output 1
  • connect the gate output on ES to a trigger input on teletype

[you can essentially use polyES as a multiple tap tempo clock source using this technique…]

1 Like
#335

that is and interesting idea :thinking: polyES as a multiple tap tempo clock source, I’ll have to try this!
thank you for your wisdom and amazing work! :raised_hands:

#336

with this approach you could even create an irregular clock - just record it in the leader pattern and that’s it!

1 Like
#337

Hmm, I gave these two things a go, but still no output from JF.

I wonder if anyone else has tried a similar setup (crow + ansible + JF) and has it working?

#338

sorry it’s been a hassle…

a couple more things to try:

on voice assignment page: both the jf mode and i2c mode buttons should be brightly lit.
the 5 upper buttons in the rightmost column (above i2c mode button) should be brightly lit - these are device mutes.

on output parameters page: check if jf outputs volume is somehow set to 0.

are physical outputs mapped to same voices working?

#339

Both JF mode and i2c buttons are brightly lit, as are the 5 above i2c. JF outputs volume is up all the way, and the physical outputs mapped to the same voices are also working.

#340

very strange. the only other thing i can think of - is there anything running on crow that might also be sending to just friends?

will try to test jf+crow+es set up this weekend.

#341

I don’t think so! I had that same thought, and tried running a few different scripts on crow to see if it was conflicting at all—different scripts didn’t seem to affect the lack of functionality.

#342

Basic question - If I run this firmware on Ansible am I still able to switch into the stock Ansible apps or would I need to reinstall the original Ansible firmware to be able to access Kria, Meadowphysics, stock Earthsea, etc.?

#343

it replaces the original firmware completely, so, yes, to switch back to the original apps you will need to reinstall the original firmware.

1 Like
#344

partial success. i’m able to reproduce the issue with ansible running polyES + jf + crow setup.

i was able to get it working for a short while while crow was still running the initial first script by running druid and executing:

ii.pullup(true)
ii.jf.mode(1)

i tested with orca’s heart first and the above fixed the issue. what’s strange though is that switching OH to a different i2c device and back (which also tells jf to set mode to 1) made it stop working again, and repeating the above didn’t work until i powercycled.

with polyES same thing - executing the above fixed the issue, but then turning i2c leader mode off and back on again would result in no sound again.

i then tried updating to the latest crow firmware and now i can’t get it to work using the above method. the good news is - it’s definitely solvable, just need to figure out what’s going on. need to see what crow is doing under the hood.

2 Likes
A user's guide to i2c
(ansible) 3.0.0 release