Hi there,
I´m doing a sequencer in pd for my modular rig starting with a trigger/gate abstraction that should bang the analog envelopes & friends. A very simple approach starting as trigger and ends up in variable gatelenghts - see patch attached.
If I bang it manually there is no problem. If I bang it with a metro, the gate starts to wiggle a bit in lenght.
Hmmm, guess there is a sync problem with free running metro and dac~???
I´m using the ES-8 from Expert Sleepers as soundcard which is dc coupled, but needs a signal for output at least...
Any suggestions?
Thx, Rainer
Hi Rainer
On Fri, 2018-05-25 at 18:55 +0200, Rabbit ́s . wrote:
Hi there,
I´m doing a sequencer in pd for my modular rig starting with a trigger/gate abstraction that should bang the analog envelopes & friends. A very simple approach starting as trigger and ends up in variable gatelenghts - see patch attached.
If I bang it manually there is no problem. If I bang it with a metro, the gate starts to wiggle a bit in lenght.
Hmmm, guess there is a sync problem with free running metro and dac~???
Not really. The [line~ ] object starts (and ends, I guess) ramps only on block boundaries, while the [metro] sends bangs with exact time periods, and so does [delay]. It happens that the bang arrives sometimes at the start of the block, sometimes at the end, the same with the bang from [delay]. So the number of full blocks that fit between the metro bang and the delay bang varies by one block.
The reason why you don't see that jumping when manually triggering the gate is that GUI interventions are quantized to blocks (or even more coarsely).
If you want to have a (sub)sample-precise envelope, you need to use [vline~] which is able to generate ramps starting and ending at any time, even in between samples. Simply replace [line~] with [vline~] in your patch. You will notice that not the length of the envelope is wiggling now, but the position. This is due to [tabwrite~] being able to start writing to a table only at block boundaries, too.
Roman
Thanks Roman!!!
2018-05-25 19:42 GMT+02:00 Roman Haefeli reduzent@gmail.com:
Hi Rainer
On Fri, 2018-05-25 at 18:55 +0200, Rabbit ́s . wrote:
Hi there,
I´m doing a sequencer in pd for my modular rig starting with a trigger/gate abstraction that should bang the analog envelopes & friends. A very simple approach starting as trigger and ends up in variable gatelenghts - see patch attached.
If I bang it manually there is no problem. If I bang it with a metro, the gate starts to wiggle a bit in lenght.
Hmmm, guess there is a sync problem with free running metro and dac~???
Not really. The [line~ ] object starts (and ends, I guess) ramps only on block boundaries, while the [metro] sends bangs with exact time periods, and so does [delay]. It happens that the bang arrives sometimes at the start of the block, sometimes at the end, the same with the bang from [delay]. So the number of full blocks that fit between the metro bang and the delay bang varies by one block.
The reason why you don't see that jumping when manually triggering the gate is that GUI interventions are quantized to blocks (or even more coarsely).
If you want to have a (sub)sample-precise envelope, you need to use [vline~] which is able to generate ramps starting and ending at any time, even in between samples. Simply replace [line~] with [vline~] in your patch. You will notice that not the length of the envelope is wiggling now, but the position. This is due to [tabwrite~] being able to start writing to a table only at block boundaries, too.
Roman _______________________________________________ Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/ listinfo/pd-list
-
Rabbit ́s . -
Roman Haefeli