I’ve encountered pops/clicks because of instantaneous fader automation volume jumps,
I usually manually adjust all my automation points to have some variable of small ramp to avoid weirdness..
But it would be interesting if there was a global slew rate setting somewhere (is there?)
And perhaps even on a per-track/per-automation lane basis as well. By default they get assigned the global, but then the user can adjust on a per-rate basis.
Click the fader, then hold shift, and the fader will move in finer increments.
Cubase will actually do that automatically for “instantaneous” automation (because instantaneous volume changes are impossible). You can see that if you zoom in a lot:
It is not configurable, though, and not enough to prevent distortion in certain cases. Well, there will always be some distortion if you modulate a signal with another signal, aka amplitude modulation, but in some cases it might be audible. Tbh, I have never encountered that in real life, though, only in dedicated tests with sine waves.
Cubase’s “Volume” plugin will prevent that, becauses it applies a smoothing function to the automation data, but it is another plugin to instantiate of course.
Edit: Pre-Gain automation funnily is also smoothed (at least when I last tested).
Absolutely not. The faster the better. One of the biggest gripes with using Steinberg’s EQ and automation is that it reacts so slowly even with EQ/Filter transition set to Quick.
Try looking at return time setting under your automation pane prefs (assuming Cubase has that) for when you’re writing.
Perhaps give an example of where this is a problem because I can’t say we’ve ever had issues with pops/click on volume changes and automation is used extensively.
As others have said, instantaneous volume jumps would be a really bad thing, so all DAWs take precautions to avoid that. Here’s an interesting six-part discussion of volume automation.
First, why absolutely not? If it’s a variable preference.
Secondly, absolutely not the faster the better.
Just noticed a bunch of pops and crackles in a bunch of printed delay tracks, after investigating, it was a squared volume difference on the source track, not even a big one.. slanting the change a bit and it went away… This is why faster is not always better.
@GlennO, just to be clear, I’m wanting to slow down what I am referring to as “instantaneous” fader jumps.
I understand. It’s a balance that DAWs try to achieve between avoiding pops while still making automation precise.
As you can see from the article I linked, there are a variety of strategies that DAWs employ to try to find a balance between these two competing objectives. But the fact is there is no single ramp that could be set in a global preference that would work in all cases (lower register notes generally require more of a ramp than high, for example).
So, I believe the thinking is: if the default ramp doesn’t suit your needs, do your own ramp, e.g. with the range tool, which it sounds like is what you’re already doing.
I’m foreseeing this becoming more of a problem now that people are using the range tool to edit automation…….
I think people have been using the range tool for “instant automation” for as long as it existed, and before that, drew that manually. It’s not even a new problem, the people who developed the first analog consoles with fader automation faced the same problem (which they also mitigated by applying a filter to the automation data).
It might just be that it is possibly more noticeable since Cubase introduced sample accurate volume automation some versions ago, instead of a smoothed curve (that might have even been dependent on the buffer size, don’t remember exactly).
As GlennO wrote, it’s always a trade off between precision and possible artifacts.
A quick few reasons off the top of my head:
I can’t say anyone has had an issue with artifacts here but it might be that we used a different tool to solve the same problem. For example, we never use mute automation and rely on either muting the events or volume automation instead.
Volume automation is actually pretty much the only automation in Cubase that is sample accurate. Pre gain is smoothed, the new Cubase volume plugin is smoothed (on purpose), third party plugins do whatever they want, but most do their own kind of averaging, if the developer is aware of the pitfalls, but very likely they are not sample accurate (as afaik JUCE and other frameworks don’t support it yet). Mute automation, as you wrote, is rarely a good idea. I haven’t actually tested clip gain.
I agree though that it could be helpful to have some kind of mechanism to easily slant the “squares” should it be necessary, maybe something similar to the “Tilt Left/Right” handles if you have two or more automation points selected.
That’s great for you and your team, but I encountered it in a clear case that wasn’t even extreme… This isn’t of concern to you? Just because you haven’t encountered it yet?
Personally, I don’t want to have to be thinking about my fader automation creating clicks/pops or having to be slanting/rounding corners all the time.
Apart from preventing clicks/pops, we’re talking about likely inaudible to the human ear differences. Establish to your team what setting will be used.
There can be a settings hierarchy of Global → Project → Track → Individual automation nodes.. Affording some forwards compatibility. Backwards compatibility would be an issue, but I’m not sure how many people are mix-collaborating backwards to people with previous versions.
Having some node adjusters would be good,
But I still think users being able to set their own slew rates Global → Project → Track → Individual automation nodes would be useful.
At the very least, Project → Track → Individual Nodes… Or at the very least (Multi)Track → Individual nodes.
Take into account someone doing very experimental work using a lot of wild fader automation on very bass/mid dense material… If a lot of pops/clicks/zipper noise is being created from the faders - why not let them adjust the slew rate gradually until they all disappear?
imo, faders creating pops/clicks, regardless of how infrequent, shouldn’t be a problem in 2025.
Just curious, what type of automation lane is that? And what samplerate are you using?
Because for me, zooming in, the line as far as I can see, is perfectly vertical.
Any automation lane, no relation to sample rate. Same in C14+C15. Always ~5 samples apart.
It is just a visual representation anyway. even if you see just a perfectly vertical line, you will never get that in reality, not possible.
I’m just curious as to why yours displays a slant of zoomed in.
are you using 64-bit mix engine?
No, but that isn’t really relevant for automation or the graphical display of it.
anyway, it is only 5 samples, not nearly enough to prevent clicks on low frequencies notes… a quick test show that you need at least a slant of one half, better one full cycle of a sine wave (at 80Hz) to have less audible distortion, and then preferably having the automation points at zero crossing. It is of course less noticeable on more complex signals.
