Still experiencing delay compensation bugs in 11.3

[Eric Lambert]

Originally Posted by Ienjoyaudio

Anyone else still getting this?
Need a fix to continue using my workflow.

Oh man, I've written novels about it. In fact we spent an hour looking into this just yesterday. It's ridiculous.

[KSeaton]

Originally Posted by PT Lover

I can confirm this behavior on HDX 11.3.

Workarounds (for now):

Create a send on the VI feeding the aux in question (orange ADC value I would think).

Or, bypass (Ctrl+Cmd+click) the ADC dly value on the VI track, and the cmp value should jump up to the correct value.

PS: haven't done this in a large scale, so don't know how well it works in big sessions.

EDIT: Bypass the dly value, NOT the cmp value.

Will try it out in a larger session and see if it holds up. What a pain in the *****...

[The Fez]

I'm not using any aux tracks but the rhythm trax slide off the click generated in protools by the end of a 3 minute song. Something is severely wrong with delay comp.

[KSeaton]

Originally Posted by PT Lover

I can confirm this behavior on HDX 11.3.

Workarounds (for now):

Create a send on the VI feeding the aux in question (orange ADC value I would think).

Or, bypass (Ctrl+Cmd+click) the ADC dly value on the VI track, and the cmp value should jump up to the correct value.

PS: haven't done this in a large scale, so don't know how well it works in big sessions.

EDIT: Bypass the dly value, NOT the cmp value.

This appears to be a pretty good workaround for now in larger sessions too. but you need to identify all the problem tracks and bypass the dly value. can't vouch for how accurate everything is overall at the sample level. maybe this narrows it down to the interaction between Aux inputs and VIs followed by a bpm synced plugin?? Really hope this gets sorted soon...

[Niilo Nuori]

Originally Posted by KSeaton

Really hope this gets sorted soon...

We have been waiting for basic delay compensation issues to be fixed for close to 2 years now. It seems to me that delay compensation is not build into the PT sw architecture (PT did not use to have it contrary to all other DAWs) but it is poorly patched on it, tweak here and another there, hard to keep it all together. Maybe a rewritten PT 13 will fix these rudimental issues finally.

[Stig Eliassen]

Originally Posted by KSeaton

This appears to be a pretty good workaround for now in larger sessions too. but you need to identify all the problem tracks and bypass the dly value. can't vouch for how accurate everything is overall at the sample level. maybe this narrows it down to the interaction between Aux inputs and VIs followed by a bpm synced plugin?? Really hope this gets sorted soon...

True, it is a pain to keep chasing problem tracks throughout the session. I didn't check sample accuracy, and it was a small session, but it sounded tight to me. I haven't seen this issue, as I still keep all my VIs in VE Pro.

There is a nasty storm going on here, so I've shut down my studio. I will run some more tests when able, probably in a couple days.

[Eric Lambert]

Originally Posted by Niilo Nuori

We have been waiting for basic delay compensation issues to be fixed for close to 2 years now.

2 years? It's not worked properly since its introduction many, MANY years ago.

[DC-Choppah]

What you guys are describing is consistent with how ADC works to my knowledge and this is not a bug.

By default, Virtual Instrument tracks have ADC suspended. That allows them to make a sound as soon as the person hits the keyboard. MIDI events always come out immediately. The PT defaults are set up for performance where it is assumed that you want a VI to respond immediately.

You can see that Instrument tracks have had ADC suspended because the track delay has been forced to be 0.

So as you add plugins anywhere, to any tack, that add long delays, ADC will simply add the delay to all tracks to compensate - except for tracks were ADC has been suspended, which is what happened to your Instrument track.

To force delay compensation to be applied to a track where PT has suspended ADC, you do what PT Lover says. By command-ctrl-clicking on the track compensation window you are forcibly applying ADC to that track.

This is similar to the OPs problem where ADC had been suspended by PT to keep the tracks at low latency. But then by overriding low latency, ADC was now applied.

Just be aware that now that you are using ADC on a Instrument track, you will have latency when tracking if you use long delay plugins anywhere in the session.

You can change the default preferences to force ADC on these Instrument tracks so you don't have to chase them all down.

Under Preferences/MIDI/Delay Compensation...

Check the box next to: MIDI Notes and Controllers

[Stig Eliassen]

DC-Choppah, there are quite a few misconceptions in your post.

What you guys are describing is consistent with how ADC works to my knowledge and this is not a bug.

By default, Virtual Instrument tracks have ADC suspended. That allows them to make a sound as soon as the person hits the keyboard. MIDI events always come out immediately. The PT defaults are set up for performance where it is assumed that you want a VI to respond immediately.

Instrument tracks doesn't have ADC suspended by default. How could they? It would sound awful with any significant system delay present.
Forget recording scenarios for now - this is about sync'ed playback, regardless of the complexity of the session, and types of plugins involved.

You can see that Instrument tracks have had ADC suspended because the track delay has been forced to be 0.

So as you add plugins anywhere, to any tack, that add long delays, ADC will simply add the delay to all tracks to compensate - except for tracks were ADC has been suspended, which is what happened to your Instrument track.

Let's take Xpand2 followed by Soundtoys Tremolator. That is a delay of 43 samples. Compensation value is 0, although I have an aux with 6400 samples of delay in the session. The timing of the notes are fine, but the MIDI sync'ed tremolator is way off. This is a major problem.
If I bypass the delay report on that instrument track, the cmp value jumps up, and tremolator syncs properly to the session tempo. This is a major problem, and a horrible way to work.

Another thing is that even though the cmp value can have a value of 0 on the instrument track, and sound fine, then how come the notes play way too early when I bypass it? Bypassing zero compensation shouldn't matter. Something isn't right here.

To force delay compensation to be applied to a track where PT has suspended ADC, you do what PT Lover says. By command-ctrl-clicking on the track compensation window you are forcibly applying ADC to that track.

This is for audio tracks only, and to keep ADC enabled on record-enabled or TrackInput-enabled tracks.

If you Ctrl+Cmd+click the cmp value, you are effectively bypassing any delay compensation for that track. It will turn gray, and play back too early.

You can change the default preferences to force ADC on these Instrument tracks so you don't have to chase them all down.

Under Preferences/MIDI/Delay Compensation...

Check the box next to: MIDI Notes and Controllers

Yes, delay compensation for external devices, and has absolutely nothing to do with my or KSeaton's issue here.

The bottom line is that MIDI sync'ed plugins behaves badly, and we're forced to find workarounds to make it work. This is a major problem.

We really shouldn't have to think about this at all, in my opinion. It should just work.

[DC-Choppah]

Originally Posted by PT Lover

Instrument tracks doesn't have ADC suspended by default. How could they? It would sound awful with any significant system delay present.
Forget recording scenarios for now - this is about sync'ed playback, regardless of the complexity of the session, and types of plugins involved.

Actually, this is the core of the problem you guys describe. The instrument track HAS been isolated from ADC. It is NOT being compensated for.

When I learned how to work with PT11 and ADC, one source was the book:
ProTools 11 Music Production, Recording,Editing and Mixing by Mike Collins.

On page 249 he quotes Avid Solution Specialist Simon Sherbourne saying:

'Instrument Tracks and Auxes that contain a virtual instrument with MIDI routed to it always show 0 as their compensation, and MIDI is always kept in time dynamically at the point of playback.'

Zero compensation value means that ADC has been suspended during playback. This explains all the symptoms you guys are describing.

I am with you that what you are describing is a playback issue, not a recording issue. Very different than what the OP's problem was, which was about low latency monitoring during recording.