-
Dolby E – bit off topic
So this is my issue. I need to take a previously mastered HDCAM tape that has closed captioning and dolby E ( track 3 and 4 ) I need to capture it, edit some material and then send it back out to tape. I need to preserve the closed captioning and the dolby E. The HD closed captioning is solved by the MXO2. I can keep the HD CC and send it back to tape, I can even convert it to SD line 21 captioning if I down convert. But what about the DOLBY E. How do i preserve the dolby E in and out. Do any AJA products do this. Basically it’s data in and data out but how?
Tony Manolikakis
Rev13 Films -
9 Replies
-
Do you need to make edits that change run time or cut points? That is, does the audio need edits?
If not, simply insert edit picture back onto tape. leave the audio alone.
Dolby E is designed to be edited to a degree (you can make insert edits if the deck is set up right) – but it’s also easy to ruin.
-
I am pretty sure you will have to re-encode the edited tracks at a Dolby facility to get the Dolby E encoding.
Warren
Warren Eig
O 310-470-0905email: warren@babyboompictures.com
website: https://www.babyboompictures.comhttps://www.babyboompictures.com/BabyBoomPictures/AFX.html
https://www.babyboompictures.com/BabyBoomPictures/KnitWits_Movie.html
https://www.imdb.com/name/nm0251670/EDITING REEL: https://www.babyboompictures.com/BabyBoomPictures/Editing_Reel.html
TITLE DESIGN: https://www.babyboompictures.com/BabyBoomPictures/Titles_Reel.html -
Anything that resamples the audio will kill Doby-E. There is no way to keep the dolby-e intact when you are capturing. You need to decode it out to discreet tracks and then re-encode it to dolby-e when you are done.
-
Tony,
Juan is absolutely correct, re-sampling will ruin the Dolby E metadata.
However we do editing on the DOLBY E stream daily with Final Cut Pro and the AJA Kona 3 card. At times we also decode the signal, mix the audio as necessary and then encode back to tape ensuring levels and metadata remain at the same values.
It is immediately obvious with the FCP waveforms if the Dolby E packets are out of alignment with the video frames. Correctly set and aligned decks should negate any corruption when editing the signal back to tape.
The gotcha with the Kona card is to ensure you capture using embedded audio from the SDi as the AES turns the signal into PCM noise due to re-sampling.
-
[Juan Salvo] “There is no way to keep the dolby-e intact when you are capturing.”
Why not? It is just a data stream embedded so you should be able to capture it into FCP and back out. Capture doesn’t re-encode PCM data.
In theory Dolby E was designed to be editable, but in practice I wouldn’t try it with FCP. Part of the problem is that FCP doesn’t pre roll the source before an insert, just does an instant start at the edit. This creates clicks in normal PCM audio as it needs to do a cross fade edit. With a Dolby E stream, the cross fade should be turned off on the tape machine, but even so, I wouldn’t recommend inserting into a Dolby E stream with FCP.
-
Matt
So going through SDI you are able to capture the Dolby E data and then send it back out? In your other scenario – where you remix the music – you require a Dolby E encoder. I assume there is a setting required on the deck ( F500 for example) for it to receive the SDI audio streams as data or is this automatic. Thanks for the replies.
Tony Manolikakis
Rev13 Films -
Tony
So going through SDI you are able to capture the Dolby E data and then send it back out?
Yes, we are. Providing we capture the audio from the SDI input and not the AES. Our capture is set to 24 bit. We use a Kona 3 card. Some other cards re-sample the audio.
In your other scenario – where you remix the music – you require a Dolby E encoder.
Yes, we do require both a decoder and an encoder. The decode introduces a 1 frame delay and so does the encode which we compensate for as our delivery requirements specify that the Dolby E signal be 1 frame early on tape.
I assume there is a setting required on the deck ( F500 for example) for it to receive the SDI audio streams as data or is this automatic.
We set nothing on our DigiBeta or HDCAM-SR decks other then the audio edit transition to cut mode to minimise the risk of corruption at the edit point by destroying the data packets. Some people still deliver Dolby E to us where the data packets are not correctly aligned with the video frame. We have equipment to analyise the data stream but we also use the waveforms in FCP for a quick look to see if we are in the ball park.
Below is an excerpt from a Dolby document available on their web site.
Dolby E is a data stream that contains up to 8 channels of audio. The Dolby E signal is divided into frames. The frame rate is user selectable and is designed to match the frame rate of any accompanying video signal. The Dolby E data does not take up all of the available space within the time period of one video frame; at the start and the end of the video frame an error of zero data is produced called the Guard Band. The presence of the Guard Band allows the Dolby E to be edited and switched by standard broadcast equipment without affected the audio data within the Dolby E stream.
In order to maintain the ability to edit and switch Dolby E signals it is essential that the Guard Band position overlaps the switch point to prevent data corruption. -
..and finally the BBC provide a setup guide for the F500. We use the SRW-5500
https://www.bbc.co.uk/commissioning/production/docs/HDW-F500.pdf
-
Do you know if this works with a Kona 2. So the audio waveforms fall within video frames when viewed on the timeline? The one frame off set I understand, dobly E requires one frame to decompress so it must be one frame ahead of the video frame.
Thanks again for you comments.
Tony Manolikakis
Rev13 Films
Reply to this Discussion! Login or Sign Up