Hi folks, some time ago, someone posted a patch in the pd's facebook group, about an attempt to implement hardsync.
I do not care about the implementation or how to implement hardsync (I already found a better solution), but what freaks me out is that I copied the patch and it's not working like the post did. He only posted a picture of the patch, but I copied it exactly and it's not working, maybe someone out there could help check if I'm crazy and did something wrong, or if there's a bug or something.
find attached the picture that was posted and my patch. thanks
it works at low frequencies.
i think at higher freqs, you'll start getting errors because the slave phasor's phase is only reset on block boundaries, so you have the potential for 63 samples of deviation.
this is why converting from the signal to message domain and back again is never usually a good idea if you need accuracy.
I see, but in the picture it seemed to be working for high freqs... :P
2015-05-29 2:38 GMT-03:00 i go bananas hard.off@gmail.com:
it works at low frequencies.
i think at higher freqs, you'll start getting errors because the slave phasor's phase is only reset on block boundaries, so you have the potential for 63 samples of deviation.
this is why converting from the signal to message domain and back again is never usually a good idea if you need accuracy.
and the patch seems to be way off, way more than the block boundary, there's something fishy here...
2015-05-29 3:19 GMT-03:00 Alexandre Torres Porres porres@gmail.com:
I see, but in the picture it seemed to be working for high freqs... :P
2015-05-29 2:38 GMT-03:00 i go bananas hard.off@gmail.com:
it works at low frequencies.
i think at higher freqs, you'll start getting errors because the slave phasor's phase is only reset on block boundaries, so you have the potential for 63 samples of deviation.
this is why converting from the signal to message domain and back again is never usually a good idea if you need accuracy.
-
Alexandre Torres Porres -
i go bananas