Jump automation -not quite- instant?

I used TestGenerator to audio file as well. Sine, 440Hz.

If one of 3rd party plugins has a developer that likes to communicate you could ask them about it.

It’s certainly different, using different plugins.

I tried the same test (the simple saw wave riff I did with a low pass) using an arturia filter I have, and the “blip” is much less pronounced, almost not there, than when using the other plug-in.

We have shown that automation can indeed be sample accurate (except for mlib’s system), so the root cause is most likely to be found in the algorithms of the plugins themselves.

Steinberg is not very communicative. So, again, find a developer of plugins who is and inquire there about the reason.
Feel free to come back here and relay your findings. I am sure others are interested, too.

1 Like

Tried again with a sine wave. I am not getting the same results as you do.
This time I ran Cubase Pro 12.0.70 in safe mode with both preferences and 3rd party plugins disabled.

I did a few more tests and here’s what I found.
If I set the two automation nodes to the same Start value I get a roughly 10ms fade (see above post).
If I set the Start value of the second node 1 sample after the first, I get a result that corresponds to the Volume Automation Precision setting in Studio Setup.

Still strange. If I set the automation nodes to the same position I still get an 8 sample ramp, which corresponds to my settings.

One thing I noticed is that your long ramp ENDS at the automation node, where my ramps START at it.

Let’s start to look at “weird” things: I use 48kHz sample rate. Can you try with 48kHz as well? EDIT: Never mind, just read your prior post that you use 48 as well.

1 Like

I also use 48Khz

If I automate a paramter of a VSTi directly, in this case Vital, instead of the Cubase DAW parameters I get a different result.

This is a volume drop of approx. 85 dB, automating Vital’s built-in volume paramter.
Takes some 128 samples (suspicious number).

IMO that just strenghtens my theory that the length of response to automation depends either on the VST3 plugin standard or on the algorithms used in the plugins themselves. The DAW has nothing to do with it.