Create proper samples files in Live for the DT

Hello everyone,

I am using the DT for a year now.
I also have a lot of hardware gear i use with Ableton Live for production.

Now i’d like to export all the individual sounds i have in Ableton, into my DT (i have dozens of kicks made with Kick2, lots of basslines…), but i’d like to avoid using the DT sampling function, and instead learn how to properly export individual samples .wav files from Live.

I was wondering if you had a few workflow tips for this ? Especially about the volume of the samples. I am unsure if the DT normalizes the volume of the sample files.

Thanks

Big up :wink:
Interested as well. Not so much from Live but more like the workflow of File Transfer (C6, Transfer)…
I have a DT ordered that should arrive soon… I have started to read the manual as well as watching YT tutorials… The thing that gets me overwhelmed above all IS exactly this process as well as the workflow for library/samples/project management :slight_smile:
Thanks for the inputs if any,
V

1 Like

Hi guys, I can’t find an official answer right now but I’ve read somewhere that samples were normalized when imported into the Digitakt. It’s probably better to export at max level tho, to avoid unwanted noise after normalization. Also the Digitakt works in 24bits/48kHz, so I try to use this format when the source material allows me.

As for the workflow, I don’t use Live so I cannot help you @extaz93. But @vasidudu the sample management is very easy from a computer. With Transfer you can have an overview of your Digitakt +Drive and simply create new folders, drag&drop files into these folders, or even import whole folders with samples in it. Easy as managing files on the computer itself I would say. On the other hand it’s kinda annoying to re-arrange thing on the Digitakt itself, but still very possible.

Hope this helps! :slight_smile:

1 Like

Something to consider is that Transfer app takes just the left channel of stereo samples and doesn’t sum to mono, so you may want to sum to mono first depending on material…

2 Likes

@AldoVino
THX a lot for the feedback. Much appreciated :wink:

@Open_Mike
Good to know. So better to bounce WAV to mono 24bts/48kHz prior to uploading to DT? @0dB or -3dB?

Sorry about all the questions :disappointed:

1 Like

Thanks for the answers !

Friends, I am a mastering engineer by trade, musician as an enthusiast. There is nothing to be worried about re: normalization on the DT. If you have gain-staged correctly during the capture process, export your sounds at whatever level you like - the noise floor they have is the noise floor they have. Normalization does not change the noise floor.

But don’t take my word for it here is the definition of it from the trusty crowdsourced encyclopedia:

Audio normalization

Audio normalization is the application of a constant amount of gain to an audio recording to bring the average or peak amplitude to a target level. Because the same amount of gain is applied across the entire recording, the signal-to-noise ratio and relative dynamics are unchanged.

6 Likes

It appears that the DT does auto-normalise samples to 0dbFS (see this thread Optimising Samples For Digitakt / Normalization).

Regarding workflow, the Transfer app is your friend:

Various sample formats are supported (.wav, .aiff, .mp3, and more…). The exact list of supported file
formats may depend on your computer OS and its built-in codecs.

The files are converted and re-sampled automatically using high-quality algorithms if they are not already in the native format supported by your device (typically 16bit/48kHz, mono).

From the Transfer User Manual (page 6).

2 Likes

Live’s audio exporter offers 48Khz 24bit Normalised Mono tracks.

I would recommend using short file names so that the names are legible in the DT’s small display system - I’m not sure right now what the limit is but think around ten characters.

The other thing to bear in mind is whether you want to work with sample chains. My own experience suggests that it is an excellent way to pack more than 128 samples into the DT’s ram storage. The ram is limited to 64Mb and if your sample size is small (as most drum samples are) then the limit is not the 64Mb but the 128 sample limitation. By say packing eight samples into a single chain the equivalent number of potential samples loaded into RAM increases to 8 x 128 = 1024.

I recommend either packing similar types of samples together - i.e. a series of kick drums or exporting a kit-chain of kick, snare, ohh, chh, tom, ride, kick2, snare 2. The advantage of the latter - particularly if you keep the order of equivalent sample types the same in each chain is that an entire ‘kit’ can be switched using Ctl-all (of course you will need multiple kit-chains set up the same way to do this).

The main drawback of this approach is more time is needed preparing the samples for export. If you decide to do this you would need to take care to ensure that the placing of each sample in the sample chain is the same - which generally means that some RAM space is wasted but personally I think the trade off is worth it.

The main caveat is that at present the DT does not provide a zoomed in display of the samples in memory making it difficult to ‘dial-in’ the correct sample start location.

My own solution is to create the sample chain with eight drum samples all equally spaced, and then to use the DT sample start knob to dial in 0,15,30,45,60,75,90,105 offsets which in this context takes you to the exact start point of each sample in the chain - these values can be p-locked within each track which is great. If Elektron were to implement a FUNC sample start option to jump in discreet increment values then this approach could be a dream to work with. Given that other DT knob values such as tuning use the FUNC key to jump in discreet units, this would be extremely easy for the DT’s engineers to implement. However I fear that these same individuals are now most likely working on the successor to the Model:Samples so this may never happen.

4 Likes

Quick FYI: Sample format is 16 bit 48kHz mono and it does not normalize upon uploading samples, only when recording.

However, the Transfer application converts whatever you throw at it to the correct format.

I usually normalize one shot samples to 0, easier to dial it back than to gain once inside the unit.

7 Likes

Any quick way to normalize your samples when exporting with Live?

Thanks for the infos!

Has anyone tried to use or record say a 24 bit 96 hz sample and re-record that sample to another device at 16 bit 48 hz or is conversion the better option lets find out.

thanks, this answered every question i had about this subject