Surround & Downmixing help needed

[Stephen van Vuuren]

I'm finally getting to the point I want my mixes to be pro-quality but have some questions:

(1) When I've created a 5.1 mix, on previewing or rendering out a stereo version, I get clipping at the loudest points as if Vegas is just summing channels. The downmix function only affects monitoring (where I can hear the distortion) and Vegas lacks any sort of intelligence in creating a stereo file out of 5.1 project.

That makes it seem like you actually need to create a stereo mix in separate file and thus what's the point of a downmix function in Vegas? Don't home DVD players intelligent create a stereo playback and not just summing channels? The Vegas documentation says nothing about all of this.

(2) The AC-3 surround templates have all sorts of settings in the custom area for levels, dialog, center channel, surround mix level etc. Any need or advantage to change this stuff? Has anyone documented not just what the settings are but why they are available or useful?

(3) Film Style vs. Constant Power - any opinions on which panner is better for DVD? Any good tutorials on them?

thanks,

[Mike Kujbida]

I can't answer question #1 as I've never worked with 5.1 audio.
As far as home DVD players are concerned, there's usually a stereo/5.1 option for the viewer to choose from.

On questions 2 & 3, try the following settings.

Encode set to AC3
Click on custom tab
Dialog normalization: -31 dB.
Dynamic range compression: None
On the first tab set diag. norm to" -31"
On the last tab marked preprocessing
Set the Line Mode & RF mode profiles to "none"
Now save that as a preset.

There's a LOT more information on the Dolby site about AC3 & surround sound at http://www.dolby.com/
Use the "Select a Technology" drop-down menu at the bottom of this page to select various resources.

Mike

[Douglas Spotted Eagle]

Originally Posted by Stephen van Vuuren

(1) When I've created a 5.1 mix, on previewing or rendering out a stereo version, I get clipping at the loudest points as if Vegas is just summing channels. The downmix function only affects monitoring (where I can hear the distortion) and Vegas lacks any sort of intelligence in creating a stereo file out of 5.1 project.,

Vegas isn't summing, but you might have too much energy in one place or another in the 5.1 mix. Down mix is indeed only monitoring, but it is a very valuable tool as it helps you find phase issues and also know how well it will be collapsed in a stereo mix.

Originally Posted by Stephen van Vuuren

That makes it seem like you actually need to create a stereo mix in separate file and thus what's the point of a downmix function in Vegas? Don't home DVD players intelligent create a stereo playback and not just summing channels? The Vegas documentation says nothing about all of this.,

No, you don't need to. You certainly can, but these days, very few do. We still do, because you can't really count on all devices to correctly mux the 5.1 to a stereo. How are you monitoring all of this during authoring? That alone plays a big part in what you're experiencing.

Originally Posted by Stephen van Vuuren

(2) The AC-3 surround templates have all sorts of settings in the custom area for levels, dialog, center channel, surround mix level etc. Any need or advantage to change this stuff? Has anyone documented not just what the settings are but why they are available or useful?,

Set LMP to None. Set Film Mode to none. Set dialog norm to -31. Leave the rest alone. Read www.dolby.com if you want the gory details.

Originally Posted by Stephen van Vuuren

(3) Film Style vs. Constant Power - any opinions on which panner is better for DVD? Any good tutorials on them?,

Film mode is more intuitive for me to author in, but it's really dependent on your room, monitoring, etc.

[Stephen van Vuuren]

Originally Posted by Douglas Spotted Eagle

Vegas isn't summing, but you might have too much energy in one place or another in the 5.1 mix. Down mix is indeed only monitoring, but it is a very valuable tool as it helps you find phase issues and also know how well it will be collapsed in a stereo mix.

It must be partially summing (at least center + LR) - my 5.1 mix does not clip anywhere, sounds great in my monitor room and in HT playback. Stereo mix noticably clips at loudest sections.

No, you don't need to. You certainly can, but these days, very few do. We still do, because you can't really count on all devices to correctly mux the 5.1 to a stereo. How are you monitoring all of this during authoring? That alone plays a big part in what you're experiencing.

I have a 5.1 setup with Outlaw bass management. My 5.1 mixes are working great - it's the stereo down mixes I'm having trouble with.

Set LMP to None. Set Film Mode to none. Set dialog norm to -31. Leave the rest alone. Read www.dolby.com if you want the gory details.

Thanks to both of you. I assume dolby.com tells you the "why" of these settings?

Film mode is more intuitive for me to author in, but it's really dependent on your room, monitoring, etc.

My setup is all nearfield monitors very close to me (small studio). I've assumed contstant power is the more accurate way to monitor when speakers are close. However, I'm curious what you mean by "intuitive".

Thanks very much for the info though.

[Peter Jefferson]

hmm. well i wont ramble on, but the simple solution to a clear 5.1 mix has been answered.

Preferably, with dynamic range compressions and LMP , i set it to film light.. i also set my panner to film (like DSE mentioend, it IS intuitive) and its also the most accurate form of 5.1 mixing/speaker placement within the edit when it comes to "cinemtic" presentations.

Head off to dolby.com and download their pdfs on this matter.. tehre are HEAPS of white papers and tech specs which go over all this in specific detail

[Stephen van Vuuren]

Originally Posted by Peter Jefferson

hmm. well i wont ramble on, but the simple solution to a clear 5.1 mix has been answered.

Preferably, with dynamic range compressions and LMP , i set it to film light.. i also set my panner to film (like DSE mentioend, it IS intuitive) and its also the most accurate form of 5.1 mixing/speaker placement within the edit when it comes to "cinemtic" presentations.

Head off to dolby.com and download their pdfs on this matter.. tehre are HEAPS of white papers and tech specs which go over all this in specific detail

Thanks - I have downloaded a bunch of them for reading. But I still wish I knew why my stereo renders from 5.1 projects are clipping.

[Dale Paterson]

Hi Stephen,

I have also had this problem in the past but eventually it came down to workflow and the original stereo track.

For what it's worth here is my workflow (at present):

Either (if making no pan changes):

Start with Stereo Project
Ensure that stereo audio track has correct levels and normalize if necessary
Render PCM Stereo track (already been shot down for this)
Change Project Properties to 5.1 Surround
Render to AC3 Stereo
Render to 5.1 Surround

Or:

Start with 5.1 Surround Project
Ensure that audio tracks have correct levels and normalize if necessary
Pan changes
Render to AC3 Stereo
Render to 5.1 Surround
Change Project Properties to Stereo
Render PCM Stereo Track

I always include PCM Stereo, AC3 Stereo, and 5.1 Surround (AC3) on my final product.

Basically when rendering to PCM Stereo from a 5.1 Surround Project I have noticed that if you do not work as described above the PCM Stereo track levels are 'boosted' almost as if the channels were 'summed' as you describe.

Also be aware that when nesting Vegas Projects within 5.1 Surround Projects the audio of the imported projects will be loaded into Vegas downmixed to stereo.

I have also noticed that if you monitor a 5.1 Surround Project using a stereo only output device you also get distorted sound i.e. this is obviously not the same as monitoring or previewing the stereo downmix of a 5.1 Surround Project on 5.1 Surround capable device in stereo if you know what I mean.

Hope this is helpful.

Regards,

Dale.

[Stephen van Vuuren]

Originally Posted by Dale Paterson

Or:

Start with 5.1 Surround Project
Ensure that audio tracks have correct levels and normalize if necessary
Pan changes
Render to AC3 Stereo
Render to 5.1 Surround
Change Project Properties to Stereo
Render PCM Stereo Track

Hope this is helpful.

Regards,

Dale.

It was helpful - thanks. Since all these projects started and final mix down in 5.1, I used your second steps. However, a couple of notes. First, I after changing project properties to "Stereo", I need to mute LFE tracks. The couple of spots were a still a little hot (just around 1 db or so) but easy to fix.

Thanks again.

The only problem with this issue is that if you make changes to your project you have two files to keep in sync. I would really like to see Vegas support a nice mixdown render to stereo from a 5.1 project.

[Peter Jefferson]

"But I still wish I knew why my stereo renders from 5.1 projects are clipping"

well whats happening here is that your 5.1 levels at approx 0db or -12 (depending on your preferance) are being mixed form 6 channels into 2. from here, consider that each of these channels requires a base amount of "headroom"

Now in a 5.1 mix at anything between -12 to 0 db, once u ADD the surround, LFE and centre channel to the downmixed stereo track, what your pretty much doing is going over the Headspace. So your trying to cram in data which was intended for 6 channels into 2 channels. U cna see where im going with this..

What u can do (and this DOES work) is do ur final 5.1 mix at anythgn between -12 to 0 db.
Once u have that ac3 file, save the project and rename add teh subname of "stereo".
Once u have this second 5.1 projct file, all u need to do is drop all your levels down to anythign between -18db to -24db. Just go to you track header and drop levels from there... very simple stuff..

Now, once ur levels are basically half of what your original 5.1 was, go to properties and change ur audio from 5.1 to stereo.

From here, each track should now be in stereo, and u can kill off or or run an effect on the channels which were allocated for the rears. When i do mxdowns, i usualy run a slight reverb to the rear speaker channels. This will bode well for prologicII and standard stereo surround playback, as the decoder will recognise the frequencies and alloate the source to the appropriate channel based on the frequencies and delayed response of teh reverb.
For the centre, i run it mono. Even though its a stereo track and in my 5.1 mix it was allocated to the centre channel, in stereo, u still want the vox to be centred. One thing to note that with ur stereo mix, u might want to thin out your compression and EQ settings so as not to blow out and overdrive themselves.

Its hard to explain without spenind hours explaining how it all works. Not to mention the licensing requiremenst if u intend on using this format as a marketing tool to help in your sales. Be aware that to use the technogy its ok, but to advertise the use of it, u must be licensed. WHich means your work is scrutinised and tested for quaity assurance and mixing quality in general. Basically Dolby wont put their name on any product which is not commensurate with a base standard of quaity set by them.
There are specific requirements to adhere to, however as your just learning this stuff, i would stronlgy recomend u read up as much as u can. Also if u can head off to a studio which already does this sort of post produciotn work, it will give u a hands on and a real life representation of how its all done.

[Stephen van Vuuren]

Originally Posted by Peter Jefferson

" So your trying to cram in data which was intended for 6 channels into 2 channels. U cna see where im going with this..

I understand - I was just expecting Vegas to offer this feature rather than half to do it manually. That's what I initialy thought the "downmix" function in the app was. Clearly, it's not and you have to do it manually.

Not to mention the licensing requiremenst if u intend on using this format as a marketing tool to help in your sales. Be aware that to use the technogy its ok, but to advertise the use of it, u must be licensed. WHich means your work is scrutinised and tested for quaity assurance and mixing quality in general.

Does the license that comes with Vegas and DVD Architect not cover this? I was under the impression it did. I did not understand that ever film using Dolby tech had to be inspected by Dolby themselves. Is there any documented case of Dolby ever go after someone for a bad mix that displayed a Dolby logo

Also if u can head off to a studio which already does this sort of post produciotn work, it will give u a hands on and a real life representation of how its all done.

Would love to but in Greensboro, NC, that just does not exist. But I am going to slowly work through this over the summer before I do my first surround mix that will play in theaters...

[Dan Euritt]

can i just say, thanks to everyone for contributing to this great thread!

[Emre Safak]

Does anyone here use a Dolby Pro Logic II or Circle Surround II encoder for their stereo tracks?

[Peter Jefferson]

"Stephen van Vuuren]I understand - I was just expecting Vegas to offer this feature rather than half to do it manually. That's what I initialy thought the "downmix" function in the app was. Clearly, it's not and you have to do it manually."

heheh wish it was that easy... no, u see when u encode to a 5.1 track, you are given options within the extended bitstream to offer "suggestions" for stereo downmix. Basically inject some metadata to tell the amp how to play the stereo track from a 5.1 source. this is VERY restricted as it DOSENT represent the way in which the piece was originally intended.. this is a keypoint to the way in which Dolby Technologies work. Basically as its a compression tool, theyre trying to to have it F*^$#ed with too much. I guess thats how i see it and from my communications with my own contacts in the UK.
The misdown tool is just that.. a misdown tool, however the assumption is that your using dynamic compressions, which is a standard for the format anyway (this helps with mixdown as well as "levelling out" your actual levels within the final encode

Does the license that comes with Vegas and DVD Architect not cover this? I was under the impression it did. I did not understand that ever film using Dolby tech had to be inspected by Dolby themselves. Is there any documented case of Dolby ever go after someone for a bad mix that displayed a Dolby logo

((No, the license u get with DVDA (u dont actually get the Dolby encoder until u install DVDA), DVDA is the actual program which is licensed... i guess this is a cost thing and a way to sell more copies of DVDA... as well as implmenting Native AC3 as opposed to mpg audio.) One thing abotu Sony, is taht theyre VERY strict on licensing and standards, and MPG audio is NOT a standard for dvd delivery...
anyways, the license you have is to basicaly instal and use the software... like any other software package.. be it a game, or a spreadsheet..
the Dolby license on the other hand, is a trademark and standards agreement which polices the quality of products marked and branded with the Dolby name (and artwork)
EVERY DVD production house must be licensed if they intend on using trademarks or the technology itself. No ifs or buts about it.
To get this license, one must ofer a series of works for scrutiny. From there, if u pass, your given the TSA and u jsut go from there. Occasionally youre asked to send work across for checking, other times, if you have given a good number of various projects, these can be used as a precedent for your work.
IF HOWEVER you change the way you work, you must re-send your new work, for scrutiny, else theres no guarantee that what u mixed 6 months ago is still at that same standard. Basically this applies if you change your encoding software, change your NLE, change your mixing technique etc etc

From what i know theyre policing the use of trademarks and going hard to those not licensed. I dont blame them as Dolby Digital has the potential to be a great selling tool for any business, but if its not done right, then Dolby themselves have their reputation tainted.

Persoanlly on a production side of things, and being one of the very few who are licensed here in Aus (theres a handful of us.. count em.. lol ) , i find it appalling how companies are blatantly using these trademarks and technologies without the licenses to do so, and more importantly, without the KNOWLEDGE AND SKILL to be able to actually use the technology effectively. Not only does it bring the format down a notch, but it also puts this foramt in bad light if the mix is inferior to the standards set by Dolby Labs themselves

[Peter Jefferson]

Originally Posted by Emre Safak

Does anyone here use a Dolby Pro Logic II or Circle Surround II encoder for their stereo tracks?

all the time

but what do u mean by "circle surround"

[Emre Safak]

Originally Posted by Peter Jefferson

...what do u mean by "circle surround"

SRS Labs surround sound solutions.