Quoting Mathieu Bouchard matju@artengine.ca:
On Sat, 20 May 2006, thewade wrote:
I didn't know that about [line]. Thanks for the info! I always thought [line] was just [line~] except at the control rate. I learn somethin' new every day.
What's "control rate"? the messagesystem doesn't have a "rate". Its rate is whichever speed individual messages are fired at, which is very much varying, as you can select any rate you want using [metro] or the second argument of [line] or a setting in [gemwin] or whatever else that use "pd clocks" to work. In addition to that, it's possible to remove messages from a stream, duplicate them, etc.
This is what makes the messagesystem completely different from the DSP, in which everything happens at fixed rate (one fixed rate per subpatch).
I thought the non-tilde objects got computed at the samplerate/blocksize rate. So for example if you used [line] instead of [line~] as a signal rate envelope you get stair-steps in your signal with each stair width being one block size.
[osc~]
| [line]
| / [*~]
|
[dac~]
In Csound this is what the control rate was, and you could set this rate using the ksmps value in your .sco file I think. I bet that if you made the block size 1 sample then [line] and [line~] would behave exactly the same, right? Again I am no expert.
My understanding is that the control rate objects have an upper bound of samplerate/blocksize and has no apparent lower bound, but in actuality are computed at samplerate/blocksize.
I could very easily be wrong.
-thewade