When I alter the Delay field for an expression map, should I expect it to scale depending on the tempo of my playback? Because it doesn’t seem like it’s doing that. I have to change the value depending on the tempo, otherwise my Spitfire tambourine rushes or drags.
Like, it shouldn’t really be in milliseconds, right? It should be a percentage, I think. Or at least, it should be milliseconds per q=100, or something like that.
It should be in milliseconds, yes. All such features are.
Negative delay with samples always works with all tempi. The recording of the original sample always has a certain attack time in the original recording. The number of ms of attack in the original recording doesn’t change as a result of tempo.
100ms seems quite high for percussion though.
The number of ms of attack in the original recording doesn’t change as a result of tempo.
No, but how I compensate for a dragging sample would have to scale based on tempo, yes?
Looked at another way, there is some duration of silence in the sample before the sound begins. That duration would have to change based on my tempo.
Obviously if you edited the start offset for notes to actually have your MIDI note starting early by, say, 1/8 of a beat, that will be by the wrong amount depending on the tempo. That’s why manually editing note start offsets to compensate for late starting samples is not a good way of going about things. The whole point of the negative delay is that you can then have all of your notes quantized so they are starting exactly on the beat.
Yes there is some duration of silence in the sample before the sound begins, but no that amount of silence in the sample doesn’t change with the tempo. Samples don’t play any faster or slower at different tempi unless you are enabling timestretching in the sample player, which isn’t generally done because it is quite intensive. So with time stretching off, as is the norm, there is always the same number of milliseconds of “silence” at the beginning of the sample regardless of tempo. It only makes sense to do timestretching with things like measured tremolo or things that you want to last a specific number of beats like precorded patterns or gestures (ex. runs). For an unmeasured tremolo or individual notes, timestretching generally makes very little sense.
OK thanks, I think that my confusion has sprung from the first note dragging during playback, presumably due to the negative delay. Successive notes do play on time.
Thanks!
Do you have a note start offset on any of those notes? Any manual note start offset will completely override the negative delay for that note configured in the expression map.
Nah, I just entered the notes and applied the unmeasured tremolo notation.
As I wrote, I figure the first one drags because it would technically have to start before playback begins, due to the negative delay specified in my expression map.
Ahh so you have it in the first bar. Then, yes, this would be a problem. In this case I would leave an empty bar at the beginning.
It’s not even placement in the score that’s the issue. If the note occurs at the beginning of whatever passage I’m playing (e.g., select notes, hit P), then it’s going to drag. Seems like Dorico’s playback engine isn’t engineered to handle delay that takes a note into negative playback time.
Not a big deal, minor annoyance.
Ahh yes, I see what you mean. Yeah, when hitting P it might not be able to take the negative delay into account for that first note. The exception would be if the plugin itself was able to do read-ahead and reported a PDC value back to the host. This is how NotePerformer works, for instance.
