By
Grump
Tue Jan 15, 2019 7:44 pm
I too, would love to see something like this, or maybe...a JJOS4, a live performance oriented variation that would allow importation of JJOS2XL sets, without the live audio looping features, but instead would allow live recording of multiple Drum/Instrument/MID tracks via the main window, either quantizing on input, or re-quantizing a track without stopping the clock, including editing the swing, and maybe adding a global swing parameter. Also, Simulsequencing, with a limit of 32 tracks per sequence, but the second sequence is spread across pad banks C and D in Pad Mutes section instead of just playing the whole thing with no access. No editing for the simulsequence, and if there are multiples of the same command, the active sequence takes priority and filters out the duplicate event to avoid event stacking and glitching.
Quantized pad play of patterns on/off functions would also be awesome.
If the clock has to stop to permanently record the events into a sequence, this is ok, just being able to change sequences, and record multiple tracks while the clock is going would be a godsend for live performance, as well as some folks' modes of composition.
Also, adding a probability parameter for each track, as in the probability of any note-on event taking place. If probability is 100, the events all happen normally. If probability is 0, the events are all muted on that track. If the probability is 50, it is a 50% chance that any event will happen or be muted.
As for taking sampling to a new level... I'd rather see the addition of access to the noise source, because when you have access to noise source, you can create sample and hold functions, random note generation, probability functions, humanization of quantizing, random smooth and stepped envelope/modulation functions, all kinds of fun stuff that don't take a lot of processing, but are powerful creative tools. Once the noise source is there, comparative logic functions are kind of a given too, because they're already in use. They're actually already in use now, anyway, so accessing those should not be difficult. That's what sets the recording thresholds on the sampling section, and in the compressors, or the choke groups, or the cycle functions for the pads. Logic functions are AND, OR, XAND, XOR, NAND, NOR, Min, Max, etc.. and they are super fun to use. There's a random LFO shape, so I'm guessing that it uses a noise source. Lots of the aforementioned functions are just comparing a measurement of the noise source and a given value. Connecting those same already existing processes to other processes in the OS should not be too difficult or require a bunch more code.
I would bet that people would even be willing to pay as much as double the price of OS2XL, just for the OS to be able to record multiple tracks into the sequencer on the fly. Also, creating tracks of arbitrary loop length, as long as they are shorter than the total sequence length. They could just loop at whatever length they are and when the sequence reaches the end of its length, the tracks all restart at the beginning of themselves and loop where they loop, just like they would when when you play from the beginning of a sequence normally. Ideally, the sequence can never be shorter than the shortest track. If someone wants to create or edit a sequence length that is shorter than the longest track, the sequencer should ask if the user wants to split the tracks and create a new sequence with the remainders of the tracks in the previous sequences in a new track that is the remaining length of the sequence before the split was made, or just discard the remaining length and all of the tracks/info. Splits would add note-off commands to any notes that were playing before the split.
For some, this may seem crazy, but I don't care, and I don't want to hear "Well it doesn't do that, so figure out another way.". I KNOW it doesn't do this stuff, that is why I am saying something about it. It SHOULD do this stuff. People will pay for it to do this stuff. There is NO REASON for it not to be able to do this stuff, except for not enough people expressing interest in it being able to. The hardware can handle it, no problem. With as much buffer space is used to record audio for the live looping stuff, or sampling audio, reallocating 1 or 2MB (which is a TON of MIDI info) of sample memory as an event buffer to enable the on-the-fly edits and multiple tracks recordings should not be even a second thought if it enables these features. Lots of other sequencers do this stuff already, but I'd much rather use the MPC1K with these VAST IMPROVEMENTS TO USABILITY than having to buy another hardware sequencer.
If this isn't something JJ can do, maybe someone else could???