Full mapping is in channel 1-12 order: 2.0 Lt, 2.0Rl, Silence,Silence, 2.0 M&E Left, 2.0 M&E Right, 5.1 L, 5.1 R, 5.1 C, 5.1 Lfe, 5.1 Ls, 5.1 Rs.
I think if my track orders were wrong they would have flagged that. I understand if I mixed up the 5.1 L & R with the Ls & Rs it would effect things but L & R always have consistently higher levels than the surround channels and that's exactly what I'm seeing.
I understand the broad strokes of loudness measurements it's just not something I do in my Avid system and to reiterate if I had a gain structure problem then the stereo mix wouldn't have passed either IMHO. I know anything is possible but if tone is right it doesn't leave much reason to suspect gain structure and most certainly points to the source material.
Bottom line is if I send you bars and tone and they are at the correct level then where would a gain structure problem be happening. If channels were multing under the hood that would effect tone level. In the abstract I suppose Avid could somehow be not playing fair under the hood but given everything looks the same on the source side as it does my timeline I see no indication of that.
It's a situation where the network QC would rather stay in the problem and clearly isn't looking for a solution. As I've said before square one ask me to deliver bars and tone not a sample clip of dubious origin. As has been pointed out loudness issues of such a subtle nature point to the mix not the workflow signal chain in most cases.
---In Avid-L2@yahoogroups.com, <bogdan_grigorescu@...> wrote :
From: "bigfish@... [Avid-L2]" <Avid-L2@yahoogroups.com>
To: Avid-L2@yahoogroups.com
Sent: Wednesday, November 1, 2017 3:54 PM
Subject: Re: [Avid-L2] Today's QC corner
The mapping of their example .mxf Jpeg2K file is impossible for me to determine in Avid. There is a strange structure to the mxf clip where there is something like a "master mxf clip" that references individual mxf files of stereo audio and video and anc located in a folder labeled "media". The master clip when ama linked to avid has tracks all over the place. If I ama link to the individual audio mxf files the follow the spec pattern.
I'm not a protools operator is it possible that protools could open these individual audio files directly to run a loudness measurement? The stereo tracks pass it is the 5.1 stems that are being flagged and I've been given nothing specific other than out of range. If my gain structure was off then I doubt the stereo tracks would have passed loudness testing.
---In Avid-L2@yahoogroups.com, <bogdan_grigorescu@...> wrote :
From: "bigfish@... [Avid-L2]" <Avid-L2@yahoogroups.com>
To: Avid-L2@yahoogroups.com
Sent: Tuesday, October 31, 2017 6:14 PM
Subject: Re: [Avid-L2] Today's QC corner
Also if I make my own bars and tone they pass the jpeg2K export and link back perfectly. So when they claim that I have a gain structure problem they should just take my bars and tone file to prove I don't.
I agree if this was our mix we'd have the mixer address it but it's there sample media and they gave it to us as avid media files for the test. Once I figured out how to open the mxf sample file I did make mixdowns of their tracks which maintained -20 tone level in Avid. For the last test I used that audio not the stuff from the avid media files and still they reject it.
Me thinks they just don't want it to work and they'd rather do the mezzanine conversion themselves. Their solution was we deliver a proresHQ file. Doesn't really address gain structure but I seriously feel like they don't want to fix this. If you say someone has a gain problem why not ask for reference tone to start with? Just boggles my mind we live in a time when a network would rather complain about loudness without starting with square on. I do realize that loudness, LKFS/ dialogue norm is not the same as reference tone measurement but given they supplied the material and telling me I'm altering the gain somehow when tone stays at the right level is just absurd to me.
---In Avid-L2@yahoogroups.com, <cutandcover@...> wrote :
On Tue, Oct 31, 2017 at 10:24 AM, Bogdan Grigorescu bogdan_grigorescu@... [Avid-L2] <Avid-L2@yahoogroups.com> wrote:not sure if audition will be able to read your mxf file, you may need to extract audio and then analyze it.BG
From: "bigfish@... [Avid-L2]" <Avid-L2@yahoogroups.com>
To: Avid-L2@yahoogroups.com
Sent: Monday, October 30, 2017 10:08 PM
Subject: Re: [Avid-L2] Today's QC cornerI've never run Audition other than once to try and process an aaf for protools on a premier project. I don't see audition on my system although I had an icon in the dock, weird. I'll have to look into this tomorrow. It sounds like audition will run an analysis. Would it do that on avid media as that is what I was given by the network? I also have the mxf mezzanine format file they gave me. Would audition handle that?
Am I correct that you are saying that Audition will run an analysis of the original source files I was given?
---In Avid-L2@yahoogroups.com, <bogdan_grigorescu@...> wrote :As I said, run the 5.1 through audition(make sure you choose BS.1770-1 algorithm) and see what the result is.Target should be -23/-24LKFS.BG
From: "bigfish@... [Avid-L2]" <Avid-L2@yahoogroups.com>
To: Avid-L2@yahoogroups.com
Sent: Monday, October 30, 2017 5:47 PM
Subject: Re: [Avid-L2] Today's QC cornerHere is all we were given:Measured Loudness (LKFS, BS.1770-1) - English 5.1Out of Range
---In Avid-L2@yahoogroups.com, <bigfish@...> wrote :Correct loudness is not the same as levels but if they point out gain structure then tone is a simple way to prove that is correct. I've not run Audition much and they gave us no specific spec that was out just a generic loudness was greater.
Is there some feature in audition that would allow a direct comparison of two files to see if levels match?
---In Avid-L2@yahoogroups.com, <bogdan_grigorescu@...> wrote :loudness is not the same as levels.run your stems through Audition and see what it comes up with.make sure to not include commercial breaks, lineup or textless.BG
From: "John Moore bigfish@... [Avid-L2]" <Avid-L2@yahoogroups.com>
To: Yahoogroups <avid-l2@yahoogroups.com>; Yahoogroups <editing-list@yahoogroups.com>
Sent: Monday, October 30, 2017 4:43 PM
Subject: [Avid-L2] Today's QC cornerOver the course of several weeks we've been delivering test Jpeg2K mxf files exported from Avid to a cable network. The source is avid media they supplied of program master with a bin. They also gave us an mxf jpeg2k file for reference. The claim they are making is the loudness of the 5.1 stems are higher than they should be but the stereo tracks are okay. I know loudness isn't the same as peak and it's something I assume related to a dialogue norm type analysis.As a correction I was able to link to the mxf file they supplied and it included bars and tone. The supplied Avid media had not bars and tone. The bars came in correct as did the tone at -20 on the Avid meters. So I made audio mixdowns of those stems and used them for the test. Still they say it did not pass. They insist there is something wrong with our gain structure.Their suggested solution is to have us deliver ProResHQ instead. Well audio is audio has anyone heard of a reason Avid would not have unity audio gain when exporting an interleaved jpeg2K with 12 channels of audio?When I ama back to the jpeg2K files I've made the audio appears identical and in the case of tone it is also right at +20 on Avid meters. I have no idea other than their source material was off to begin with.What amazes me in this process is at no point did they suggest sending the bars and tone. If my bars and tone are correct there is no gain structure problem and all my internal tests, including showing the material to our protools mixer, have come up correct. How can a major network suggest a gain structure problem and not request a sample of tone? And if we have such a problem what possible difference is it delivering ProResHQ? It really feels like they just want to make the mezzanine format themselves. That I could understand.So has anybody had issue with audio levels when exporting a jpeg2k mxf with 12 tracks of audio from Avid?John Moore Barking Trout Productions Studio City, CA bigfish@...
Posted by: bigfish@pacbell.net
Reply via web post | • | Reply to sender | • | Reply to group | • | Start a New Topic | • | Messages in this topic (17) |
No comments:
Post a Comment