The No.1 Website for Pro Audio
LtRt louder than 5.1?
Old 7th November 2013
  #1
Lives for gear
 
jahtao's Avatar
LtRt louder than 5.1?

Hello,

Can anyone explain why my LtRt down-mix is hotter on the LU meter than the 5.1 mix?

The latest BBC World Wide spec even says that its best to bring the LtRt down by a couple of dB approx to keep at the same loudness as the 5.1.

But I just can't get my head around why!

My Neyrinck LtRt settings
PL2
C -3dB
LsRs -6dB
LFE is turned off

My 5.1 mix had dialogue diverged across LCR by -6dB. Everything peaking at -10dBFS except the VO (routed to centre channel only) which peaks at -7dBFS so that when it goes thru the LtRt it peaks at -10dB.

I aim for the stereo mix to peak at -10dB but am having to bring that down a dB or so to get it at the -23LU required by the spec.

Any help much appreciated
Matt
Old 7th November 2013
  #2
Lives for gear
 

If you have dials in the L and R, when you LtRt encode, they don't get brought down to compensate, so effectively, you get the L and R, plus your C at -3 added on top. the maths doesn't work to make it encode at the correct level. Also you'll find that when you listen to your LtRt, the VO will be slightly buried, as that is C channel only in 5.1.

If everything was C channel only in terms of dial, then the encode would work perfectly.
Old 7th November 2013
  #3
Lives for gear
 
Henchman's Avatar
Another case where divergence of DX causes a problem.
Divergent for DX is a bad idea, and adds nothing .
Old 7th November 2013
  #4
Lives for gear
 

Quote:
Originally Posted by Henchman View Post
Another case where divergence of DX causes a problem.
Divergent for DX is a bad idea, and adds nothing .
+1

philp
Old 7th November 2013
  #5
Lives for gear
 
pethenis's Avatar
 

Quote:
Originally Posted by Henchman View Post
Another case where divergence of DX causes a problem.
Divergent for DX is a bad idea, and adds nothing .
Agreed, it sounds rubbish as well... But I just did a mix where it actually helped me: 1600's, VOC-ship caught in a storm, crew shouting, big Waves hitting the deck, wind howling, BIG musicscore, you get the picture.

On top of that a fairly calm VO.... Now how to mix that and retain both excitement and understandability (is that even a word?). I found -in this case- that using a little divergence got me closer to what I wanted. Not there, but a lot closer.
Old 7th November 2013
  #6
Lives for gear
 

I'd further that, it takes away clarity!
Old 7th November 2013
  #7
Lives for gear
 
jahtao's Avatar
Thanks for your responses so far.

Two things:

1. The network I'm mixing for (Sky) unfortunately demands diverged dialogue. They claim it makes the mixes more robust on poorly set-up home surround systems.

2. Surely, Mr Full Frequency, the maths DO add up when doubling up the centre channel (C -> L+R) and dropping by 3dB? I still don't understand where the extra LU's are coming from. Especially when I am dropping the surrounds by 6dB in the LtRt. I must be missing something... anyone know what it is?

Cheers,
Matt
Old 8th November 2013
  #8
Lives for gear
 
BIGBANGBUZZ's Avatar
 

Quote:
Originally Posted by Henchman View Post
Another case where divergence of DX causes a problem.
Divergent for DX is a bad idea, and adds nothing .
What about on dx verb, or do you play it straight up the centre too?

Sent from my Nexus 4
Old 8th November 2013
  #9
Lives for gear
 

Quote:
Originally Posted by jahtao View Post
Thanks for your responses so far.

Two things:

1. The network I'm mixing for (Sky) unfortunately demands diverged dialogue. They claim it makes the mixes more robust in poorly set up home surround systems.

2. Surely, Mr Full Frequency, the maths DO add up when doubling up the centre channel (C -> L+R) and dropping by 3dB? I still don't understand where the extra LU's are coming from. Especially when I am dropping the surrounds by 6dB in the LtRt. I must be missing something... anyone know what it is?

Cheers,
Matt
I've seen some weird specs, but (Sky) demanding diverged Dialogue is absurd and why wouldn't their argument apply to VO as well?

Anyway - If you Diverge you'll need to turn the LtRt down to match. The Dialog in the L/R channel combines with the C channel at the fold down but where as the C channel info hit the fold down -3 on each side, the L/R went straight through at 0 (It wasn't turned down like the Center). There's your Bad Math.

Related observations about Divergence:

Divergence is different than pan. In a pan as you turn the knob the energy is kept the same by attenuating the signal as it moves around. For example - take a Voice, Center pan it and take a LU score. Now, move the pan to 50L and take another score. You'll see it has the same LU.

Now do a similar test using Divergence and you'll see the energy is not "automatically" kept the same but rather it is additive. Set your Divergence to 50% and you will see your overall LU rise a couple dB's over the signal set to 0% Divergence. If you Diverge the signal 100% it will be 3dB louder than when it was center panned with 0% divergence. I think the reason why those who say they think it sounds better when they Diverge the voice are either sitting to close to their speakers are more exactly, they like it better because it's louder.
Old 8th November 2013
  #10
Lives for gear
 
Henchman's Avatar
Quote:
Originally Posted by BIGBANGBUZZ View Post
What about on dx verb, or do you play it straight up the centre too?

Sent from my Nexus 4
If I am using it to match ADR to production, then yes, mono up the center.
If I'm using it for effect, surround.
Old 8th November 2013
  #11
Lives for gear
 

Quote:
Originally Posted by jahtao View Post
Thanks for your responses so far.

Two things:

1. The network I'm mixing for (Sky) unfortunately demands diverged dialogue. They claim it makes the mixes more robust on poorly set-up home surround systems.

2. Surely, Mr Full Frequency, the maths DO add up when doubling up the centre channel (C -> L+R) and dropping by 3dB? I still don't understand where the extra LU's are coming from. Especially when I am dropping the surrounds by 6dB in the LtRt. I must be missing something... anyone know what it is?

Cheers,
Matt
Listen to the downmix.... It clearly doesn't add up. Play with the dialogue divergence whilst monitoring the downmix, go from full centre to 50/50. You can hear the downmix level jump up the more it's diverged. The left and right don't get dropped at all in an ltrt encode. The centre does. This means that the more centre percentage of dialogue you have in your mix, the lower the dialogue after encode (and closer to your 5.1 level).
Especially check through a VO to dialogue transition on the downmix, this is where the bad maths really translates to viewer experience as where the level would sound even on the 5.1, the downmix VO level would suddenly suck away, and become slightly unintelligible by comparison.
Old 8th November 2013
  #12
Lives for gear
 
pethenis's Avatar
 

Quote:
Originally Posted by nathand View Post
I think the reason why those who say they think it sounds better when they Diverge the voice are either sitting to close to their speakers are more exactly, they like it better because it's louder.
Could be, but louder divided over three speakers is different than the same boost in 1. That's probably why I thought it worked in that case.
Old 8th November 2013
  #13
Lives for gear
 
soundboy's Avatar
I hate it when people make across the board statements about this crap. Divergence on Dx can be useful. Let's say you have a doc with really noisy dx. After cleaning, you still have noise, or your dx has been cleaned to the point of being underwater. Rather than centering that noise in just one channel, and really pinpointing it, a little divergence spreads it out, and softens it somewhat. Unless you are going to print a track of that noise and use it as a BG across the front that's one way to deal with it. This was from a recommendation by Georgia years ago. She may not stand by it now, but I've heard it work. It's a tool. As far as the LtRt, sometimes you have to go in and tweak the settings on the encoder, and listen to the encode and decode.
Old 10th November 2013
  #14
Quote:
Originally Posted by Henchman View Post
Another case where divergence of DX causes a problem.
Divergent for DX is a bad idea, and adds nothing .
That settles that. I guess I screwed up the last 120 shows I mixed.
Old 11th November 2013
  #15
Lives for gear
 
Henchman's Avatar
Quote:
Originally Posted by soundboy View Post
I hate it when people make across the board statements about this crap. Divergence on Dx can be useful. Let's say you have a doc with really noisy dx. After cleaning, you still have noise, or your dx has been cleaned to the point of being underwater. Rather than centering that noise in just one channel, and really pinpointing it, a little divergence spreads it out, and softens it somewhat. Unless you are going to print a track of that noise and use it as a BG across the front that's one way to deal with it. This was from a recommendation by Georgia years ago. She may not stand by it now, but I've heard it work. It's a tool. As far as the LtRt, sometimes you have to go in and tweak the settings on the encoder, and listen to the encode and decode.
If I need to cover the noise, I add mono center fill. Or back off on huge noise reduction if it's doing more harm than good.
And tweaking the encoder and listen the the encode decode isn't an option. We have to deliver it right, without problems, from the stage. We barely get enough time to do a mix.
That's why I never mess with this type of stuff, as I have seen people lose gigs because something like this screwed up on broadcast. And it was their head that was on the chopping block.
Old 2 weeks ago
  #16
Gear Nut
 

Just curious as to what settings you guys use for the Neyrinck 5.1 to LtRt downmixer.

I had previously used what I liked in the default downmixer that comes with ProTools (LR 0, C -3, LsRs -6, LFE -inf), but have found that the dialogue really comes out loud in the LtRt downmix, and have found that putting the Center down to -5db better represents my surround mixes.

Has anybody else found this to be the case? Am I crazy? haha
Old 2 weeks ago
  #17
Lives for gear
 

Also, stereo music with a lot of mono content that is pan-spread or diverged will affect LU.

My experience is that shows with a lot of commercial music suffers the most, actual scores a lot less. Probably also somewhat related to the compression of conmercail music releases.
Old 2 weeks ago
  #18
Lives for gear
 

Quote:
Originally Posted by eighttwelve View Post
Just curious as to what settings you guys use for the Neyrinck 5.1 to LtRt downmixer.

I had previously used what I liked in the default downmixer that comes with ProTools (LR 0, C -3, LsRs -6, LFE -inf), but have found that the dialogue really comes out loud in the LtRt downmix, and have found that putting the Center down to -5db better represents my surround mixes.

Has anybody else found this to be the case? Am I crazy? haha
Not crazy, but I'm quite surprised that someone is actually downmixing to LtRt in 2020.
Old 2 weeks ago
  #19
Lives for gear
 

In LKFS land, the surrounds are not measured as hot as the LRC when coming up with a measurement. If the 5.1 material has a good deal of surround information and is down mixed to stereo (LtRt, LoRo) the resulting measurement can be higher than the 5.1 source.

The ITU down mix co-efficients are a good place to start. Some people add a bit (-10) LFe back into their Stereo downmix depending on how the LFE was generated.

Randall


Quote:
Originally Posted by ErikG View Post
Also, stereo music with a lot of mono content that is pan-spread or diverged will affect LU.

My experience is that shows with a lot of commercial music suffers the most, actual scores a lot less. Probably also somewhat related to the compression of conmercail music releases.
Old 2 weeks ago
  #20
Gear Nut
 

Quote:
Originally Posted by kosmokrator View Post
Not crazy, but I'm quite surprised that someone is actually downmixing to LtRt in 2020.
Haha, I'm just doing what the deliverables from the distributors' spec sheets require, and often times they require that. Are you mostly delivering regular stereo downmixes, or something else?
Old 2 weeks ago
  #21
Gear Addict
 
Leverson's Avatar
Quote:
Originally Posted by eighttwelve View Post
Haha, I'm just doing what the deliverables from the distributors' spec sheets require, and often times they require that. Are you mostly delivering regular stereo downmixes, or something else?
Hardly ever LtRt's any more. Even when they are listed in the spec, these days I like to follow up and ask if they specifically mean Dolby Pro Logic or if stereo LoRo is okay. The vast majority of the time these days I get told LoRo is okay, if not preferred. I swear some spec sheets it's only still listed because the people who wrote it thought it meant 'stereo'. Every once in a blue moon I still get asked to deliver an LtRt but it's getting quite rare. And thankfully too, I never liked how they sounded.
Old 2 weeks ago
  #22
Lives for gear
 

Quote:
Originally Posted by eighttwelve View Post
Haha, I'm just doing what the deliverables from the distributors' spec sheets require, and often times they require that. Are you mostly delivering regular stereo downmixes, or something else?
LoRo. It just sounds better. Always IMO.

I wonder if there is a single customer still somewhere with an analogue surround system using ProLogic (II) decoding?
Old 2 weeks ago
  #23
Lives for gear
 

I also only do LoRo. I did one LTRT a few years ago for TV. Not since.
Old 2 weeks ago
  #24
Lives for gear
 

Pro Logic II encodes can pull the stereo image really badly if one is not careful.

Randall

Quote:
Originally Posted by kosmokrator View Post
LoRo. It just sounds better. Always IMO.

I wonder if there is a single customer still somewhere with an analogue surround system using ProLogic (II) decoding?
Old 2 weeks ago
  #25
Gear Maniac
Quote:
Originally Posted by eighttwelve View Post
I had previously used what I liked in the default downmixer that comes with ProTools (LR 0, C -3, LsRs -6, LFE -inf), but have found that the dialogue really comes out loud in the LtRt downmix, and have found that putting the Center down to -5db better represents my surround mixes.
The Avid Downmixer that comes with Pro Tools is not an LtRt encoder. It only creates LoRo.

I also always ask if LoRo is accepted when a spec says LtRt. 99% of the time they say LoRo is fine.
Old 1 week ago
  #26
Gear Maniac
 

When did this change? I only see LTRT on spec sheets... never LoRo.
Old 1 week ago
  #27
Lives for gear
 

We have been asking specifically for LoRo mixes for at least 6 - 7 years. We noticed that on many surround heavy 5.1 mixes that were folded down to PLii, that the stereo image would pull hard to either the left or the right (program dependant). We ran a bunch of tests and found that LoRo was the most fool-proof.

Randall


Quote:
Originally Posted by Travelers View Post
When did this change? I only see LTRT on spec sheets... never LoRo.
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump