View Issue Details
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0008746||ardour||bugs||public||2021-06-15 21:30||2021-06-16 21:51|
|Platform||Debian GNU||OS||Linux||OS Version||(any)|
|Summary||0008746: Transient detection can be fuzzy sometimes|
|Description||Long story short whenever I press ctrl+arrow keys I cannot be completely sure the program is correctly detecting the transient. I usually check visually and most times it makes sense with the waveform and with the audio, but then I find situations like the pic attached, a transient "detected" clearly way past the right spot, this is an extreme case that got me to report here, but I often find situations in between, a transient that is detected just a little bit before or after what could appear and sound as the right place... and then I get stressed out about the whole track having been cut wrong x (|
|Tags||No tags attached.|
> a transient that is detected just a little bit before
That is intentional. The default is to find the transient and the rewind 64 samples (or whatever the default fade shape is set to).
When splitting regions at transients it is preferable to retain a bit of context (and then x-fade).
||Thanks for the reply, yes I understand that, but I mean a bit that is more than what it is context for the fade, and I say so by the comparison with other detections that are just fine. Anyhow from my experience detection just seems unconsistent, sometimes is here sometimes is there... while visual representation on the contrary seems to be representing the waves right from what I can hear.|