Neyrinck DD 2 Encode Issue

[Nirvana5253]

Hi everyone -

Just curious if anyone has had any issues encoding ac3s with the DD 2 plugin. I'm using v 2.0.4.2 and am experiencing some strange behavior. When encoding, it appears that the encoder is adding additional "silence" at the tail of the file. The total length of this silence is depended on the TRT of the program material. I've confirmed this behavior in PT 10, 11, and 12 and have also confirmed this with Neyrinck Support but have yet to find a solution / workaround. This silence is added no matter if you decode to line, RF, or data stream mode in the decoder. Also appears to happen regardless of channel count.

Can anyone else confirm this on their system?

Attaching screenshots here as well that show different "tail" lengths (fractional frames). Thanks!

[Nirvana5253]

Anyone able to reproduce this or have any suggestions?

[joachim]

I do not have this encoder, but can you rule out, that the additional silence in the end does not have something to do with AC3 frames are stored in blocks of 1536 samples ?
(So wenever the last block is written, any "missing" samples are filled with silence.)

[Nirvana5253]

HI Joachim - thanks for the reply. You bring up an interesting idea - this would explain the length of the "tail" being dependent on the program material. Unfortunately, I'm not sure how I would go about checking this. Suggestions?

Also - this was using the Neyrinck encoder - seems like something that is designed to work in PT should maintain frame accuracy, no?

[joachim]

I am not an expert, but I would :
select the video from start to end, measure the length in samples ,
divide those by 1536 ,
if there is a fraction behind the comma - the decoder rounds this up to the next 1536 samples.

What are the fps and samplerate of that project in the right picture ? (24 and 48000 ?)
What is the total length of that video in HH:MM:SS:FF ?
I could calculate with my spread sheet, that I made to avoid the 1536-block problem a time ago, because I am unsecure about the consequences. It may be, that that little extra silence end does not matter at all, if the movie is authored.

I hope, some expert will chime in.

[Nirvana5253]

Any others have any suggestions? Neyrinck support communication is spotty at best and still have no fix or workaround on my end. Suggestions welcomed. Thanks!

[quadraphonics]

Two things:
1.) From having done DVD mastering for years and years, your authoring facility should be able to clip the excess off when they mux it to video.

2.) There was an issue with the v1 of this software that would shift the start timecode of the encoded AC3 stream based on the value you have set a the Audiosuite handle. That might be an area to start looking at.

Originally Posted by Nirvana5253

Any others have any suggestions? Neyrinck support communication is spotty at best and still have no fix or workaround on my end. Suggestions welcomed. Thanks!

[Nirvana5253]

Hi Randall -

Thanks for the heads up. We did in this case end up trimming the AC-3s to fit but it still brings the question up of "how do I fix this?". Neyrinck is non-responsive to my support requests now - I haven't heard anything back in almost a week.

I'm using the latest version of the software.

My sneaking suspicion is that this is from the decoder, not the encoder as I took an AC-3 encoded with another program and it exhibited the same behavior - longer in length than the original source.

Anyone who has the Neyrinck setup - are you able to confirm the same behavior?