On 2010-02-03 08:20, Andrew Faraday wrote:
I think it's an historical thing, most audio inlets which might benefit from a signal level message have the [sig~] built in. I understand this might not always have been the case so some more old school PDers use it, and they remain in the help files. I don't know if there's some minor technical benefit to using the [sig~] object, but I've still to find a need for it
i don't think there is any technical benefit. the benefit is mainly in the non-ambiguity it offers. a message and a signal are 2 fundamentally different things, so an object could well accept both signals and float-messages on a single inlet (e.g [z~] comes to my mind). in order to allow this, you need a converter, which [sig~] is (and [line~] as well; but [sig~] probably _does_ have performance benefits with regard to [line~].
that was back then in the golden age of Pd (right after you could hit "delete" in the [f] box two(!) times without getting a crash)
since the conversion between numbers and floats seems to be an often demanded feature, it was eventually introduced as a built-in (breaking objects like [z~]). sigh.
in order to create a gatom controlled sinewave, you had to do
| [sig~] I [phasor~] I [cos~] I
fgmasdr IOhannes
PS: as we all now, the above patch (and it's modern equivalent [osc~] will produce ugly clicks when used like this; isn't it time to request a built-in [line~]?