Hi there, > From: Hendrik Jan Veenstra <h@...> > > My guess (from experience with other synths) would be that every time > you have to send an entire byte with all bits set correctly for all 3 > functions. So when setting function 1 to have the value "3" (both > bits on), you cannot simply send a byte 0000 0011, as that would set > functions 2 and 3 to have value 0. Instead you have to merge the > values for all 3 functions and send that as a byte. So when all > functions are set to value 1, you should send 0100 0101 (bits 6, 2 > and 0 set to 1). Unless it's possible to subtract (or add a negative number) certain values in order to change the whole 8bits. so, 69 + (-1) = 68, 69 + (-5) = 64, 69 + (-64) = 5... i.e., 0100 0101 + 11111111 -------------- 0100 0100 0100 0101 + 11111011 -------------- 0100 0001 Can the environment do this kind of stuff? with regards, -- Lachlan Deck ldeck@...

Thoughts from the mind of Lachlan Deck, 02-04-2002: > > From: Hendrik Jan Veenstra <h@...> > > My guess (from experience with other synths) would be that every time >> you have to send an entire byte with all bits set correctly for all 3 >> functions. So when setting function 1 to have the value "3" (both >> bits on), you cannot simply send a byte 0000 0011, as that would set >> functions 2 and 3 to have value 0. Instead you have to merge the >> values for all 3 functions and send that as a byte. So when all >> functions are set to value 1, you should send 0100 0101 (bits 6, 2 >> and 0 set to 1). > >Unless it's possible to subtract (or add a negative number) certain >values in order to change the whole 8bits. "Unless" what? I don't get what you mean. Something is not possible unless you can subtract? Is that what you intended? But what is 'something' then? >so, >69 + (-1) = 68, 69 + (-5) = 64, 69 + (-64) = 5... > >i.e., >0100 0101 + >11111111 >-------------- >0100 0100 > >0100 0101 + >1111 1011 >-------------- >0100 0001 > >Can the environment do this kind of stuff? Not directly, but you can work around it. First of all, there is no such thing as a negative number in the environment, so adding negatives is not going to happen anyway. That means that somehow you have to store your negative values as positives -- i.e. -5 as +5. And that in turn makes the above simple: use a transformer set to subtract, and use the positive value (from e.g. a fader) as a meta event to set the "subtract value". I.e. dragging the "subtraction fader" to e.g. 5 sends a meta event to a transformer that's then being set to "Sub 5". Easy as pie. The only complication is that often you immediately want to use this subtraction value -- i.e. you have some parameter/fader value set to e.g. 100, you change the subtraction fader to 5, and now you'd like to get an output signal of 95. The way to achieve this is to have the subtraction slider first set the transformer's "Sub" value, and then generate a "bang" that bangs the "100" into the transformer, resulting in a 100-5• output. Not very complicated, really, once you get the idea -- I've done this dozens of times. It would be very convenient however if transformers somehow remembered the last value they received at their input, and if there then was some sort of meta event that not only sets the operations/conditions but also recalculates the result with the last input. A kind of meta-set combined with a bang, all in 1 message and operating within one transformer. Would really save a lot of trouble in such a (quite often needed) scenario. But... uhm... how does all this apply to the original problem of combining multiple bit-level messages into one byte? As far as I can tell, my 1st remark, quoted above, is usually correct: when sending a multi-msgs, consisting of a bit-OR of messages, you have to combine them each time you send the message, since otherwise you'll reset some of them to zero. I haven't gone into the details of how to do this (yet) since I wasn't sure if that was indeed the question... The basic approach would be the same however: generate some "Add xxx" value (meta event "set transformer operations lower value") and bang the other messages' old values to recalculate the result. -- Hendrik Jan Veenstra <h@...> Omega Art: http://www.ision.nl/users/h/index.html