Delay compensation broken

[jturner]

I have seen a couple of other threads pertaining to issues with delay comp, but most of them seemed to involve more elaborate scenarios, so I thought I would post this. I just bounced a track, and used a TDM Trim plug in to flip the phase and null test the resultant track. It was 4 samples off-the amount the delay comp was claiming it was compensating. Turn off and on delay comp and hit play...silence. So delay comp seems to stop working occasionally. Which is kind of bad. Back to 7.3.1 it is.

Dual G5 2.7g
OS 10.5.7
PT HD 8.0cs3

[M2E]

Yeah, I've had this same thing. Sometimes it works and sometimes it doesn't. I thought I was the only one.

WOW....

[Eric Lambert]

I know what you're referring to and I've seen it a lot - it's frustrating - but I haven't seen it as much in PT8. Quite a bit in 7.x.

[25ghosts]

Originally Posted by jturner

I have seen a couple of other threads pertaining to issues with delay comp, but most of them seemed to involve more elaborate scenarios, so I thought I would post this. I just bounced a track, and used a TDM Trim plug in to flip the phase and null test the resultant track. It was 4 samples off-the amount the delay comp was claiming it was compensating. Turn off and on delay comp and hit play...silence. So delay comp seems to stop working occasionally. Which is kind of bad. Back to 7.3.1 it is.

Dual G5 2.7g
OS 10.5.7
PT HD 8.0cs3

Yes, ADC needs constant surveillance, especially when one is instantiating a plug-in during playback. That really makes ADC go bananas. The digi delays like short delay are breaking ADC 50% of the time. Toggling ADC normally relieves this problem but there are occasions where I need to remove the instantiation completely.

[Jeffrey Talbot]

Not that I doubt you, but is it really possible to discern the 4 samples difference just by ear? That does seem like a very tiny amount of difference.

And not that ADC isn't broken and shouldn't be fixed... that's not my point... just that without your test, would you have known the timing was off by 4 samples....

[Giles Reaves]

Originally Posted by Jeffrey Talbot

Not that I doubt you, but is it really possible to discern the 4 samples difference just by ear? That does seem like a very tiny amount of difference.

And not that ADC isn't broken and shouldn't be fixed... that's not my point... just that without your test, would you have known the timing was off by 4 samples....

(jumping in on someone else's thread)
In one situation it can be obvious - parallel processing. Any other situation, I would never notice it.

[jturner]

Certainly 4 samples is basically a negligible amount. I only discovered it when I was phase-null testing a bounce to verify that it was true to the original. However, the problem isn't the 4 samples. The issue is that ADC wasn't working, which could lead to a lot of bigger problems.

Interestingly, when I went to the Tiger/PT 7.3.1 partition on my machine, I had the same problem. And I have been doing this bouncing and phase checking for years on this machine and never seen this behavior. I only recently created this Leopard/PT 8 partiotion, and now I have the problem on both!

[WernerF]

I've always had this problem. I'm on P T's 8 now with the the same, having to toggle ADC on and off to make sure that it is working after making a new track or instantiating a plugin, problem that I've had ever since ADC was introduced. Nothing has changed. I've just gotten used to doing this all of the time just to be sure that ADC is working.

[Jeffrey Talbot]

The issue is that ADC wasn't working


I totally agree..

[WernerF]

It's always been partially broken. The good part is, even though this should not be the case, that there is a workaround.