Omg , quick somebody give this man some code!!!
Not all heroes wear capes. Some wear hats and glasses.
elk-herd is now open source!
Mind you… it’s 75+ files and 17k lines of code, written in the pure functional language elm*… so may not be the easiest thing to wrap your mind around.
:dizzy:
But, if you’re intrepid… have a look. The README.md
and CONTRIBUTING.md
files have lots of info:
Now that that’s done… my next task is Digitakt II support…
(*) Don’t be scared: elm is a lot of fun to code in once you accept that it’s nothing like C++ or JavaScript or Python!
Super cool, thanks for sharing and putting together the readmes and documentation! I’m finding it really interesting, and instructive, to look around in it. 17k lines! What an undertaking!
Super impressive ! And open sourcing it is just wonderful. Thank you for your work
Wow, this is awesome. Looking forward to this!
Great , i will buy again a big beer . Was hoping for that . On OG Digitakt it was a gamechanger. Thanks
Holy shit this is awesome. Def gonna buy you a beer or two. Elektron pay this man!!!
This is really good news, knowing that my DT II projects are more future proof for being compiled into a set !
Incredible!
I’ve been using the current version with the DT2 with reasonable success. It’s a lot more reliable than the official Transfer app
One challenge however is the size of the sample list on a 20GB +Drive even half full. I’d love to see some improvements on that if possible
What reliability problems are you experiencing with the official transfer app ?
Documented verbosely here
Where I finally got to is described here
Summary: Folders with over 100~150 samples (don’t know the exact number) makes the Transfer app hang and spin forever
I don’t know how anyone has filled the 20GB on the DT2 with this limitation
Amazing! Can’t wait to try it out
That’s awesome ! Looking forward to it !
Okay brave Elektronauts…
elk-herd 3.3b2
with support for Digitakt II
This is a beta release after a ton of work, and while it works in my studio… you should exercise some caution:
- Back up your project by saving it on the unit, and then after you load it into elk-herd, saving it as a file on your computer.
- When you send it back to your device, check that it is all there and is correct. Save a copy as another file on your computer at the same time, so if you later find and issue you can send them to me I can see what went wrong.
Changes in this release:
- Support for Digitakt II w/it’s larger sample pool, track flexibility, and stereo samples.
- Major re-working of low level structures to support the DT II and make future OS updates easier to support.
- Many small UI tweaks, including a collapsing left side and ability to alphabetize items.
Possible issues:
- On Mac OS 13 & 14 have a bad bug with USB MIDI and SysEx messages. See post below for details if this is you.
- Sometimes not all the names of the factory samples are fetched, and the samples show as
???
If you see this, just fetch or open the project again. If this happens often, please let me know. - FireFox on Mac OS just outright crashes on the first pattern sent back to the DT II. This is known bug with FireFox on Mac OS with any SysEx over about 60k. Developers know, but no fix yet.
- FireFox on Linux works! BUT, you need to bump the kernel buffer size with something like (need more research to find out how to make this a permanent change to one’s configuration):
cd /sys/module/snd_seq_midi/parameters/
echo 800000 | sudo tee output_buffer_size | sudo tee input_buffer_size
Okay, ready?
Live site: elk-herd 3.3b2
Code: elk-herd on github
Bug reports: here, or on github issues
Of course, if you’re just needing to get stuff done with your Digitakt original, and don’t want to fiddle beta testing, the released elk-herd is still here:
— Mark
New forum, new Elk Herd
What a time to be an Elektronaut!
Amazing work, can’t wait to give it a spin
Nice … I’ll bet that was painful though.
Thanks for all the time you put into this!
- ~9300 lines changed
- made in 160 commits
- over 5 weeks
Given that the whole program is ~18k lines in 80+ files… It was a major undertaking.
The low-level structure re-write was long over due. It is much cleaner now, and no longer requires me to hand edit SysEx dumps to support each OS release.