Hi! I’ve just found that Digitakt II (probably also Digitakt 1) don’t read correctly mp3s.
It infact don’t take account for the gap present at the start of every mp3 file: this means that if i chop a 4 bar loop at a certain bpm in a DAW it will not be in time on Digitakt II, expecially if i try to use the Grid machine.
This is really frustrating because if Digitakt II was able to read the metadata with the exact value of this initial silence gap, the file would result in a perfect 4 bars sample.
Absolutely but when quality is not a problem and storage space is, it could be interesting to use the capabilites of mp3s.
What’s the relevant forum thread where i could write?
This is actually a known issue or something, there’s a thread where they talk about this stuff and that trickles back to elektron. It’s not MP3’s specifically, it has to do with something else, probably the way it attempts to negotiate a zero crossing slice on a stereo file which will obviously not often have aligned zero crossings between L and R waveforms. This is just my guess at what’s going on though, some people believe it’s a “bug” but the behavior is pretty consistent.
I think it does this with any sampling of fixed length. Like you could import a sample with a bunch of silence before and after and manually trim it down to size but that sounds like a headache. Digitakt 1 which is mono doesn’t do this, so one would assume it has to do with the stereo capabilities.
There are several existing reports about this, and someone did a pretty good job of illustrating the actual issue using the waveforms themselves. I’ll find you a link so that you can add your complaints to the proper sections of the forum, maybe that way they’ll find a way to do something about this.
I see, but this is about resampling inside the machine.
My problem occours when i import an mp3 via Elektron Transfer and the silent gap that all mp3 have at the beginning is not eliminated automatically by the machine like a DAW would do.
Sure, but it’s just a matter of analyzing the metadata in the mp3 and removing the gap to obtain the correct length.
I think this could be an easy to solve issue.
Gaps at the start of MP3s aren’t noted in the metadata.
Honestly, like other’s have said you’re just better off using an uncompressed format for creating music with devices like these. There’s too many variables with how different apps create MP3s (CBR, VBR, Bitrate, masking, etc) that it’s only going to lead to more issues than you really want to mess with when you’re writing.
In all seriousness though, using MP3 or any other lossy format doesn’t change the space taken up on DTII. There’s nothing to be gained, and by the sound of it, there’s a lot of hassle involved in trying.
What about other formats like .ogg ?
What about dropping a .mp4 movie so that you get the audio directly extracted?
There is no limit to such request.
It’s extremely easy to get samples in the right format. There are apps and scripts, and many ressources to learn to use them.
And hard drives are extremely cheap nowadays, so storage space is not a realistic argument.
