csnjr@its63b.ed.ac.uk (Nick Rothwell) (09/24/87)
In article <13235@bu-cs.BU.EDU> hen@bu-cs.UUCP (Wm. H. Henneman) writes: >I replaced DMCS with Mark of the Unicorn Performer and Composer >because I count my time as being worth a certain amount of money. I >can't do anything now I couldn't do before (except synch with SMPTE, >but I use an Atari to do that),... Ok, gripe time. I was thinking about maybe coughing up my cash sometime for a Mac and the Unicorn software, but according to Bill, above, it can't follow a timecode. Actually, I expected this, since it *can* generate its own, very sophisticated, tempo changes. [Into the realms of assumption here, since I'm not very familiar with *that* many of the packages in question, but...] Why can't I buy a sequencing package which can (i) let me program all my tempo changes, or let me compose a track in two parts with different (maybe overlapping/crossfaded) tempos, and (ii) follow a timecode? Sounds like heaven to me... Presumably (correct me if I'm wrong) - this isn't possible, because if you're following a timecode, then the tempo comes out, robot-like, from the timecode reader, taking the tempo control out of the sequencing package's hands. I think this is WRONG WRONG WRONG, and a classic case of putting the intelligence into the wrong box. A timecode reader should be responsible *only* for reading the nasty SMPTE code from the tape and transducing it into something like MIDI timecode (MTC), preserving absolute time in hours/minutes/seconds. That way, a sequencing package can impose any number of smart tempo changes in software, rather than relying on the (from what I've seen, primitive) tempo control in the timecode reader's firmware. Put briefly, A SEQUENCER SHOULD BE ABLE TO READ (MIDI) TIMECODE, and A TIMECODE READER SHOULD GENERATE MIDI TIMECODE, for the maximum flexibility. With the situation as it stands, (i) I get stuck with the timecode reader's tempo control, and (ii) lose any such in my sequencer. Or have I got the wrong end of the stick? Admittedly, the reasons are probably historical (the first sequencers being self-clocked or reading MIDI song-pointer), but I think this wants tidying up... In the meantime - if any Mark of the Unicorn people are listening - I'll think about buying your package *IF* you allow tempo control AND synch to tape! How? Well, you assume that the incoming MIDI `clock's from the timecode reader are really a timecode, and build your tempo on top of that - and I program my timecode reader, once and for all, with a single, solid 100Hz (say) tempo, which I pretend is a timecode. Will that work? Gee, sorry, this posting wasn't supposed to be so long or opinionated. Not too much flammage, please, guys? -- Nick Rothwell, Laboratory for Foundations of Computer Science, Edinburgh. nick%lfcs.ed.ac.uk@cs.ucl.ac.uk <Atlantic Ocean>!mcvax!ukc!lfcs!nick ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ "Nothing's forgotten. Nothing is ever forgotten." - Herne