View Single Post
  #13  
Old 06-27-2020, 09:01 PM
RTGraham RTGraham is offline
Member
 
Join Date: Dec 2017
Location: New York
Posts: 51
Default Re: Can Anyone Verify / Advise? - Plugins / Delay Compensation / Timing / Downbeat

Quote:
Originally Posted by JFreak View Post
It's the other way around... plugins report latency to PT mixer and PT mixer handles delay compensation. If the plugin gives wrong report, then it will be off. Make sure all of the plugins are updated.
Thanks Janne.
I understand that the plugins are responsible for reporting their latency to the host so that the host can compensate accordingly by delaying the signal from other plugins to correctly synchronize with the latency-induced audio.

I don't believe that's the issue here.
For one thing, many of the plugins that exhibit this issue are zero-latency plugins, including the A.I.R. Filter Gate and Pumper that are bundled with Pro Tools, and xfer's LFOTool.
For another, in my test scenario the audio is processed correctly as long as there are no other latency-inducing plugins *before* these rhythmic processing plugins; but if there are any latency-inducing plugins *upstream* of the Filter Gate or Pumper or LFOTool, then the timing of the rhythmic processing is skewed.

The appearance is that part of how these plugins operate is that they must be getting timing information from Pro Tools. The Filter Gate, for example, does its rhythmic thing even while the transport isn't running, but once you press play the timing jumps to match up with the grid. (Or at least, it matches as long as there's no other latent plugin in front of it.)
It doesn't matter whether playback starts on a measure boundary. Even if you start playback on the 768th tick of the third quarter note of the measure, the Filter Gate will jump to attempt to synchronize with the grid. That indicates that the plugin isn't just being told "OK - GO! And reset to the beginning of your pattern!"
It's also being told *where* in the pattern the "GO" is occurring.

So it's not just about whether they're reporting latency correctly, or whether latency is being compensated. It seems to be about whether the plugins know where they are in rhythmic time - whether they know the current timestamp accurately. And having other delay-compensated plugins in the session seems to affect how that's being communicated to them.

Hopefully that makes sense to you. And if there's something obvious that I'm misunderstanding about this, or some other way that these plugins are able to do what they do without needing to know the timestamp, let me know.

Also - have you followed the example steps I listed in my second post in this thread? It should be pretty quick, and will let you hear exactly what I'm describing.
Reply With Quote