+ Drive cleanup

Thank for your replies, I come from electribe 2 and i would like to forget the sample management in this way, I was expecting a better view for sample usage, and no limit for the 127 samples in the soundpool…

The digitakt is the brain of my setup, i make techno live, not just a track, so 1 pattern can be played few minutes. Actually I have 16 pattern wich is my liveset (1 hour of playing).

I had to write down all the samples that i actually use in my patterns in an Excel spreadsheet, and sort by usage or not, Now i know that I use half of the samples in my soundpool, I can delete the other half.

Not a really fun way to manage the samples… When I will have my soundpool full of used sample, I will have to create another project etc… not the easiest way.

I was thinking to assemble all the samples for my bass for example, and choose the bass with the start sample parameter, but it make the computer mandatory for that, I will have to use Audacity a lot, and it’s not the way I want …

Sorry for the long post !

1 Like

pretty sure I’m just going to have to bounce out everything I definitely need and start again from scratch at some point. Some kind of “back up entire device” function would be really useful…

4 Likes

I am quickly reaching that conclusion for myself, fortunately I’m a month in & during the ‘learning process’ I’ve discovered I’m a really untidy person :-). Now I’ve got a better handle on things regarding digitakt +drive & workflow it’s time to backup samples & start again.

just hoping i can fit the live set I have to build on and resample all the samples before deleting everything around it…

oh wow. thats a limitation then…ohkeey.
so, my hopes have been killed :wink: . Thanks for clarification !

Re to my post:

my new Re:
heading here out to whom has a clue about programming etc.:

speaking theoretically:
could a code be written so that “collect samples” would work this way: )???)
all samples in the project are collected
_they get stored within that new project related folder that is created automatically (same name as project)
_ these samples get stored as a dublicate, each with a slightly changed name.
On the ending for example. for example add -“c1” ( c stands for copy)

Q: is this correct, can this be donne by programmers ?..if anybody ever would want to do :wink:

further:
i just opened crunch elk.
i think such “sample collect” functionality would ideally be implemented in the crunch elk editor.
The programm could also be written so that a “kind of a log” is created about the collected samples.
so when a copy of a sample with ending -c1 allready exists, the next copy would be named -c2.
Also could such log folder also allow the user to get insight of what samples he has doubled, and how often a sample has been copied.

again: this is doable by a programmer right ?

next Q:
so where is the problem ?

will elektron ever do ?
would you give Elektron any trust to move something ?
how about the community to set up a fund raiser for mr. @mzero

maaaaybeee this could be a doable path ?
i´d be willing to pay 100$ for this functionality…looking at 50$ to be a more reasonable number

(i´m not willing to setup any dynamics. thats just not my part,
but maybe we here as a group can get something together…?)

  • I REALLY WANT to see a better sample handling within the DT*
    ( i had the DT now for years ! I can overlook its value and functionality vs. other Gear i have or what is availbale on the market. ( having also a OT mkII here, …unused)

it would be a good spent 50$ or 100$ to me. :wink:

1 Like

The “no duplicates allowed” is not a matter of the sample name, but the content of the sample. So “no”, it is not possible until you change also the content of the sample (at least slightly).

The “de-duplication” of samples is implemented on the Digitakt, because the +drive storage is quite limited.

2 Likes

I don’t think there can even be an unload unused samples for 2 reasons.
The first reason, the one mentioned earlier, is that the sample slot being a modulatable destination can create ambiguity about which samples are being used by a project.
Second reason, samples associated with soundlocks don’t seem to appear in the project sample list, creating even further ambiguity about what is being used.

@lostinthemanual I think the likely thing that could be implemented is a backup entire project where all associated sounds, samples, & patterns could be transferred to the computer. Then the +drive can be cleared and your already sorted samples and associated project can be transferred back.

2 Likes

This is the answer.

This is the answer.

ok. all hopes finally gonne then.

So i will delete LOTS of my sample content on the DT and will see how much of my projects still work, and which ones not.

its just a mess on my +Drive.
need to get a better workflow overall.
also since i got a feel what type of samples suits my DT workflow :wink:

Thanks for the very valuable Input “tnussb” !

Backup at least all your sample off the dt first. Because the dt identifies sample though a hash system, You should be able to organize and retransfer your samples and they should repopulate the spots in the projects properly as long as they are the samples you pulled off the dt.

1 Like

There is a way how I clean up a mess on Digitakt with using of elk-herd application and a couple of scripts.
I have a few Live projects on my Digitakt which contain samples I would like to preserve. Say I would like to remove all samples from drive except them.
I copy all samles from Digitakt with Elektron Transfer to my computer.
Then I connect to Digitakt with elk-herd app, go to the tab Project and fetch one of the projects.
I open the console in Chrome and write a line:
$('.bank-name').map(function() { return this.innerText; }).toArray().filter(o => o != "" && o != "UNTITLED").join("\r\n")
All my patterns are called “UNTITLED” by default, so this line gives me a list of sample names from Samle Pool. I copy the result string to a file and then repeat for other projects adding the sample lists to the same file.
Then I copy this file to the directory which I copied all my samples from Digitakt to and name it as “input.txt”
I run a file which contains a bash script (in Windows it can be run with Git Bash):

output_dir="..\_ElektronBackup"
output_file="output.txt"
input_file="input.txt"

sed -i.bak s/$'\r'//g $input_file
> $output_file
rm -r ./$output_dir
mkdir $output_dir
while IFS='\n' read -r line; do
	line=$line | tr -d '\n';
	line="${line}.wav";
	foundsample=$(find . -name "${line}")
	if [[ $foundsample != "" ]]; then
		cp --parents "$foundsample" ./${output_dir}
	else
		foundsample="!!!"
	fi
	output="${line}: ${foundsample}"
	echo "$output"
	echo "$output" >> ${output_file}
done < ${input_file}

This script finds all wav files with names from input.txt and copies them to a separate directory with same directory structure, in output file I can see which files were found and copied. Then I can clean my Digitakt and upload necessary samples back.

4 Likes

Doesn’t Elk Herd already let you purge unused samples? Or is that just from the current project, not from the +Drive?

1 Like

Correct, it just purges from project and leaves it on the +Drive

Thanks for sharing the scripts/code and steps. This is the only solution I’ve ever seen, apart from a pen and paper :wink:. My +drive is full so I’m tempted to give it a try. Maybe it could be incorporated into a chrome widget or plugin to automate it a bit more…

On a Mac, I followed your instructions and had two problems with the script, so I amended the script for Mac OS and got it to work. My +Drive now has plenty of free space :slight_smile:
PlusDriveScript.sh (506 Bytes)

4 Likes

I (and others, I’m sure) are incredibly interested in this. Would anybody be willing to explain how to implement this for the code challenged like myself, or indeed build this into a friendly interface somehow?

Ask away, I will try to explain. You can open Chrome console: use the keyboard shortcut Ctrl+Shift+J on Windows or Ctrl+Option+J on Mac. When you run that one line of js script in the console you will get text which you should copy to a file with the name input.txt. Then you should create a file with extension .sh with that big bash script in it, for example script.sh, the files should be located in the same directory where all your samples were downloaded from Digitakt to, for example in “C:\ElectronSamples”. On Windows you should download and install Git. On Mac nothing special is needed to be installed. Run the script with a command line like bash script.sh. Make sure you run the command in the directory of the script, for this run something like cd C:\ElectronSamples. Also make sure that the variable output_dir refers to a location outside the directory of the samples. I haven’t tested this on Mac, but I believe it works. If it doesn’t let me know I will fix the script…

I once sent to mzero a request for a feature in elk-herd for saving projects with its samples, he replied that he is probably going to include this in the version 3.1. I think it would be natural to have such feature in his app.

1 Like

Would anyone mind helping me with this on mac? I’ve tried two ways of specifying the output directories in PlusDriveScript.sh and each time the output file turns up in the input directory where the script, input.txt, and wav files are located.

First try, the output folder is a sibling to the input folder inside the parent folder Music:

output_dir="../dtused"

Tried taking the long route with the second try (my name replaced here with xxxxx)

output_dir="/Users/xxxxx/Music/dtused"

Also, the output file appears without any wav files, most likely because my .wav files are separated into their own folders as they were on the DT itself. I should probably get rid of the folders and have the samples directly in the file, shouldn’t I? That would explain why the output file appears but no samples?

Thanks, this is my first time running bash scripts from the terminal.