A lot of people seem to be asking about adding and editing sounds in MK2 and 3, as, on the face of it, it's all samples, right? Well, actually...
The biggest issue with these is that the samples that are recorded utilise the proprietary DCS codec (Digital Compression System, See this talk for more information). To summarise, it can be considered like the Tracker sample players of old, where prerecorded samples are combined and modified in play, but using a compression format to pack more data in. It can be said that the format is MP3 like, in that it uses psychoacoustics to remove the less important parts of the waveform to strike a balance between quality and sample size.
Since there is no public encoder for these,making a new sound is currently impossible*. Copying or trimming sounds is possible, but difficult by messing around with the pointers. So the next time you find yourself thinking "Why doesn't UMK3+ have the 'Rain' callout", there's a lot in the way to get there.
*On a point of pedantry, a decoder for a given compressed format, can be reversed to make a rudimentary encoder, but it would not contain any of the enhancements that serve to improve sound quality, so any effort done towards this would lead to a noticeably inferior sound that could be argued is worse than nothing at all. And before you ask, the source leaks have the various command line parameters for the encoding, and the DSP program for the decoding but not any encoder itself.
No comments:
Post a Comment