I’ve just got hold of a Roland SC-155 that I want to use paired with my Digitakt for some Dungeon Synth and DOS game style composition. I want to work within the constraints of a DT and the SC-155 initially, and then I will track everything out into Ableton Live.
As a result, I want to create samples of all of the drum hits and sound effect patches from the SC-155, and then use the DT’s sample tracks to work with those elements, freeing up the 8 midi tracks for the pitched instrument patches.
The SC-155 outputs sound at 16bit @ 32 kHz, and currently I have it running through the DT’s inputs, which is then transferring over Overbridge into Ableton. I want to create a sequence with each drum hit one after the other equally spaced and panned on the SC-155 all the way to one side, record the audio into Ableton as a mono track, export the file to Audacity for chopping and labelling, and then import the samples back to my DT for future use.
Presumably, because the audio is going in to the DT and being sent via Overbridge, the audio is being resampled to 16bit @ 48 kHz during this process. If that’s right, I presumably should then be working with the audio file at 16bit @ 48kHz at every step in the process after that point, right? If my aim is to capture the character of the sound as exactly as possible, should I be doing things in a different way?
Any comments would be appreciated, I’ve tried searching around but I feel like I lack the correct terms and language to ask the right questions.
As far as I have understood you correctly, you fear a coloration of the original sound because the converter of the Digitakt works with a higher sample- and bit rate? I wouldn’t worry too much about that. As long as you use the DA converters of the original and wire a clean signal into the input(s) of the Digitakt at a normal level (maybe send through a neutral sounding DI box, if necessary), you basically have the ‘dirt’ of the original in the signalflow, because the converters at the output have then already caused this coloration. Strictly speaking, you would now have to resample the signal much more cleanly to retain every nuance (usually with an even multiple of the original signal’s sample rate to avoid interpolation/aliasing errors)… but that’s almost a question of ‚faith‘… and shouldn’t really matter much. In any case, I have been able to retain the sound character of old samplers (E-mu, Akai, Casio etc.) in this way and have not been able to hear any significant difference myself or received any ‘complaints’ about it.
That sounds ideal then, thanks. I was just trying to make sure that I wasn’t making any serious and obvious errors by doing what I outlined above out of ignorance of any ‘best practice’ when it comes to grabbing a big batch of one shots like this.
It should work out, don’t worry. In your case, you could record the signal in Ableton at 48 kHz or 96 kHz. The bit rate should ideally be 32 bits (if post processing will be involved, or normalizing etc.). Downsampling to 48 kHz and 16 bits would then be ideal in terms of quality (and calculation of the algorithm). BUT: When you get right down to it, there are other factors involved as well (converters/audio interface, cables, PSU‘s)! First (and sometimes foremost), for example, the algorithm you want to use to convert the bit rate if you want to convert from a 24 bit recording to 16 bit, for example. As I said, this quickly becomes a ‘question of faith’ and thousands of people have already discussed and argued about it.
Thanks, those are things I can track and test now for sure. I’ll do a couple of sample tests, something like 16 hits, and I’ll vary how I have things set up, import them into the DT and use my ears to compare them between one another and the sound of the SC-155 sequenced.
Spent this morning sampling the ‘Standard’ drum kit on the SC-155, just working in 48kHz and 16bit in Ableton, exporting and then opening in Audacity, labelled all the samples, and then amplified everything to -0.3db before exporting them out as individual files.
Quickly knocked this cheesy loop up, slapped a glue compressor on the master and here’s what it sounds like.
