Hi,
Summary:
I find that active Melodyne passages don't cancel in a mix that otherwise nulls perfectly with its original. Is this a bug in the Melodyne plug-in timing somehow?

Background for those not familiar with the null procedure:
I'm working in Logic 9 and regularly bounce and then re-import finished mixes and sum the phase-inverse of this finished mix against the active mix itself. I do this immediately. Thus if the software is working correctly, and the bounce is, I get pure silence as the bounce reverse-phase file cancels exactly.

I've noticed now that Melodyne, if active in the mix, does not null correctly. I don't understand this, but it's obvious because it stands out in an otherwise perfect null. For example, if I've got a short vocal section in an instrumental, and the vocal track is through Melodyne plug-in, then when I null the finished mix all I hear is the vocal; the rest of the instrumentation has cancelled and disappeared. Mixes without Melodyne active don't have this problem.

So apparently Melodyne is having some problem with maintaining the same timing in each pass; or it's having a problem with Logic's plug-in delay compensation; or...something. But I consider this a bug at this point and it's annoying because the purpose of the null is to be certain that the mix is what you intended. When things are still audible it's hard to be sure of that.

Thanks...

Steven Rowat

I can duplicate what you did with nulling mixes on another DAW (not Logic) and get perfect cancellation everytime, so I suspect your problem is with Logic or Melodyne's interfacing with Logic's delay compensation.
I used Melodyne Editor V1.30 VST Plugin.

Thanks; not good news for me; if that's where it's happening then Melodyne/Apple might have a hard time sorting out whose responsibiity it is.

But at this stage it's still possibly a quirk of my system/song default setup, so a confirmation would be nice before I consider filing a bug report w/Melodyne.

Anybody out there who can test it on Logic?
(Not excluding Celemony, of course. . )

I'm using Melodyne Editor 1.2.0, 32-bit
Logic 9.1.5
Mac OS 10.6.8, iMac 3.3Ghz, Intel.

s.r.

Hello Steven,

I haven't heard of this issue before, so I doublechecked with the team. In their QA tests they regularly ensure the sample accuracy of Melodyne and we can't reproduce your issue. If your experience differs, could you try this with the latest 1.3 version of Melodyne? Also, have you set your buffer size to 1024 samples? And is this a realtime bounce (as recommended) or an offline bounce? Does it also happen without any previous editing in Melodyne?

Thanks Claudio,
Buffer is already 1024. No, it's offline, so first I'll do a real-time to see if that helps. If not I will install 1.3 and try again.

Also, please what do you mean by "without any previous editing in Melodyne" -- do you mean does it happen if I transfer audio into Melodyne but don't edit it in any way? Or do you mean a bounce before there is even a Melodyne transfer? Or something else?

Thanks
Steven R.

Yes, this is what I meant. I was pondering if the issue could be related to some timing edits. But the more I think about it, this shouldn't affect the sample accuracy in any way. So forget about this question.

OK, but it was easy so I did it anyway as part of a test in a new project.

I cannot reproduce my results in the new project, so I'm going to give up; the problem is in a project that is very complex and has been through many convolutions over the years; I'll probably never figure the source of the problem if it's specific to that project. So I'll just recheck those sections with my ears. .

For the record, on the new project I did the following:

1. Opened a different template.
2. Imported a stereo background mix and a mono vocal.
3. Tested and got perfect nulls for:
a) the stereo bg
b) the stereo bg with plain vocal
c) the bg with vocal played by Melodyne (after transfering the vocal to M)
d) the bg with vocal pitch-changed and moved in time in Melodyne as it is played back.

s.r.