Hi,
i have a Problem in Cubase 10.5 @Win10 with Zebra and Diva,
when i try to bounce a sample from a simple midi-note-event.
Latest Versions on everything, i tried the VST 2 & 3 Version, same thing.
A clicking noise at the beginning of the sample occurs, on
the recorded *.wav file, because it seems Zebra/Diva starts
outputting BEFORE the midi-note. So the "attack" of the waveform
gets cut off on the recorded file and the amplitudes are not at zero,
producing a clicking noise.
This happens mostly on Percussive sounds and even simple waveforms
It can best be seen on these screenshots:
Having tried other VSTi's, it does not happen there (like Massive starts
its output pretty late it seems).
Any ideas?
Zebra and Diva recording from Midi Problem (clicking noise)?!
-
- KVRer
- 17 posts since 22 Dec, 2019
You do not have the required permissions to view the files attached to this post.
-
- u-he
- 30215 posts since 8 Aug, 2002 from Berlin
Unfortunately, that's expected, but it should be no more than 7 or 8 samples in the worst case.
Like most digital synths (and many digitally controlled analogue ones as well), modulations are updated at a so called "control rate", which means "every so and so many samples" as opposed to "audio rate" meaning "every sample". This is an optimization strategy, using a lot less CPU for things where audio rate does not matter much or at all.
Most of our stuff uses different control rates for different things; for modulations (envs, LFOs etc.) it's commonly "every fourth sample", but for MIDI events like NoteOns it's "every 16th sample". This coincides with our latency reported to the host since we slice everything into blocks of 16 samples. A Note always happens at the beginning of such a slice of 16 samples. This would make some notes appear 15 samples early. Hence we delay them by another 8 samples so that notes which would appear late in the slice we happen early in the next slice, reducing the margin of error to 8 samples in total.
In your screenshot it seems like more than 8 samples, can't say for sure. If that's the case, I'd like to look into it. Maybe there's an issue with MIDI timestamps and the latency reporting.
(but also, maybe it's time to evaluate 4-sample control rate for everything...)
Like most digital synths (and many digitally controlled analogue ones as well), modulations are updated at a so called "control rate", which means "every so and so many samples" as opposed to "audio rate" meaning "every sample". This is an optimization strategy, using a lot less CPU for things where audio rate does not matter much or at all.
Most of our stuff uses different control rates for different things; for modulations (envs, LFOs etc.) it's commonly "every fourth sample", but for MIDI events like NoteOns it's "every 16th sample". This coincides with our latency reported to the host since we slice everything into blocks of 16 samples. A Note always happens at the beginning of such a slice of 16 samples. This would make some notes appear 15 samples early. Hence we delay them by another 8 samples so that notes which would appear late in the slice we happen early in the next slice, reducing the margin of error to 8 samples in total.
In your screenshot it seems like more than 8 samples, can't say for sure. If that's the case, I'd like to look into it. Maybe there's an issue with MIDI timestamps and the latency reporting.
(but also, maybe it's time to evaluate 4-sample control rate for everything...)
-
- KVRer
- Topic Starter
- 17 posts since 22 Dec, 2019
Thanks Urs for the fast reply. Im not sure where to look, too find out if that small part is more than 8 samples long.