Standard mixing levels for movie theater, DVD, broadcast TV, commercials etc

[f1sound]

When doing location sound for a feature it seems that the trend can vary to setting a level of -20 to -12. I've been told that for HD media a level of -20 should be set because there is little head room before distortion occurs... In the past I've set my level to -20, and have been told my levels are "low" (clearly an audio signal though) so now sometimes I'll make the best judgement on the application (say sit down interviews) I would set level -12 (going to camera on a standard HD tape, MINI DV, P2, etc) because its pretty much a controlled enviorement; whereas for a film I prefer -20 for a maximum range for a scene that can have say whisper and shouting dialogue.

Is there a recommended standard or preference for the guys and girls in post; or does it go back to keeping it simple, "as long as it sounds good"?

[danijel]

Originally Posted by f1sound

When doing location sound for a feature it seems that the trend can vary to setting a level of -20 to -12. I've been told that for HD media a level of -20 should be set because there is little head room before distortion occurs... In the past I've set my level to -20, and have been told my levels are "low" (clearly an audio signal though) so now sometimes I'll make the best judgement on the application (say sit down interviews) I would set level -12 (going to camera on a standard HD tape, MINI DV, P2, etc) because its pretty much a controlled enviorement; whereas for a film I prefer -20 for a maximum range for a scene that can have say whisper and shouting dialogue.

Is there a recommended standard or preference for the guys and girls in post; or does it go back to keeping it simple, "as long as it sounds good"?

What do you mean by setting the level to -20 or -12? The line-up sine from the mixer to read those numbers on the recorder, or dialog peaks?

Anyway, recording level used to matter more because of S/N ratio, not because of the level the dialog is going to end up at (what this thread is all about). Now, with good 24-bit recorders, I see the trend is to ride the levels less in production, and record at safe low levels to prevent distortion. Production sound used to 'mix itself' much more, but now I have to apply 24+dB of gain on quiet parts of some scenes before I even start editing dialog.

[f1sound]

Oh what I meant was if I send a 1 Khz tone to 0 db on my field mixer and would calibrate my Hard Disk recorder to either, -20 or -12 db on the recorder at 24Bit 48 K/hz.

Thats interesting that you can boost whisper scenes +24 on gain; How does the S/N get affected on that, I imagine you would use some sort of noise reduction for the room ambiance of the scene after boosting up the level that much?

[danijel]

Originally Posted by f1sound

Thats interesting that you can boost whisper scenes +24 on gain; How does the S/N get affected on that, I imagine you would use some sort of noise reduction for the room ambiance of the scene after boosting up the level that much?

NR is probably used on almost every movie - in fact, I only did one short movie where I didn't have to use NR at all. In each quiet scene, there will be at least a few words which are so quiet, that you have to bring them up so much that some amount of noise is also brought up, no matter how good the shooting conditions were. And yes, often all the lines in a scene are NRed, but that has much more to do with voice VS the surrounding noises, then with the recording level at 24bit.

[HDJK]

You can find 'Recommendations for surround sound Production' here.

There is also a good section on mixing with/without bass management.

This is a real informative thread by the way

I mix theatrical spots in 5.1 and it's not that easy to find valuable information on the topic. Most soundengineers seem to shy away from anything with the term surround in it.

[rodGal]

Originally Posted by mbvoxx

I can offer this on Radio & TV mixes...our agency produces about a hundred spots per month, which are all sent out digitally to stations all over the country.
Most of the TV is Fast Channel delivery and radio is MP3 via email. Our TV spots go with a -10 to meet Fast Channel's requirements. Radio spots are mixed at the hottest level the mixer will allow, peaking at about +4. Plus we run the final mix thru a maximizer to get it as hot as possible. BTW: some radio stations have a 2mb limit on audio deliveries so we have to keep the rate at around 128. We have one station in smalltown Okla that has a 1mb limit, which requires the spot to be converted at a much lower rate...go figure...audio is their business and they prefer the worst quality they can live with just because they won't beef up their server....anyway....Once the radio spots get to the stations they are usally limited/compressed at least twice before they get to the transmitter. In the case of the last Clearchannel station I worked at we used an Optimod in the master engineering bay, plus the production studios had L/C's for processing all spots. So it got smashed twice before getting up the stick. There are some stations that use a final L/C at the transmitter too, which would process the audio 3 times before it gets to your radio...so it's important to produce at the agency level with as little L/C'ing as possible, otherewise, by the time it gets on the air, it's been squashed into oblivion...Fuel for thought....now you know why the music on FM radio sounds so horrible...

This was really useful, thanks mbvoxx
I stumbled into this forum and this answer in particular while asking around about standard radio broadcast level (in particular for commercials). Some mixers said they all set it at -8, both TV and radio commercials. I went with the hottest level the mixer will allow, peaking at about +4 approach, worked just fine.

Now, I have a mix to deliver this week for Muzak systems - commercial again. Does anyone know if there is a standard level for this media?

[ergalthema]

Great sticky!

One question:
What would be the average level of a rocking music track used in a movie for DVD?

Just a general idea would be nice. My particular case is a mid-intensity film - not an action film, but not a quiet drama. And I'm referring to using an already maxed-out, mastered CD track. So the peaks wouldn't be much different from the average.

I'm guessing that the music would not stay at -0.1dB like most modern CDs.

[danijel]

Originally Posted by bradhebert

Great sticky!

One question:
What would be the average level of a rocking music track used in a movie for DVD?

Just a general idea would be nice. My particular case is a mid-intensity film - not an action film, but not a quiet drama. And I'm referring to using an already maxed-out, mastered CD track. So the peaks wouldn't be much different from the average.

I'm guessing that the music would not stay at -0.1dB like most modern CDs.

Thanks!

For super-freakin' loud: Fight Club's intro music goes all the way up to 0 in the center channel, while the RMS is around -10dB. The L and R are around -16dB RMS. This is the loudest/longest thing I have ever heard in a movie.

A 'general idea of rocking music' could be -15dB or -20dB RMS in C, and -20dB or -25dB RMS in L&R. But again, it all depends on the music, it's length, on what came before that scene, etc etc, so don't look at the meters, and trust your ears!!! If you are mixing in a less then perfect environment (like a small room), a trick that works for me is to close my eyes and imagine I'm in a theater. Suddenly, the music feels too soft, and I want to raise it by 10dB sometimes. And even after that, it can be too soft in the dub stage or theater.

[ergalthema]

Thank you.

Do all these levels refer to what the meters in something like Final Cut would read when a Pro Tools mix is brought back in to be attached to the video? (which should be the same as the master meters in Pro Tools, right?)

Sorry for the basic questions. I just want to make sure I don't make any stupid mistakes.

[ergalthema]

Originally Posted by danijel

A 'general idea of rocking music' could be -15dB or -20dB RMS in C, and -20dB or -25dB RMS in L&R.

Hmm. I would have thought a bit higher, in general. Since dialog tends to average around -22dB to -27dB, right?

[danijel]

Originally Posted by bradhebert

Thank you.

Do all these levels refer to what the meters in something like Final Cut would read when a Pro Tools mix is brought back in to be attached to the video? (which should be the same as the master meters in Pro Tools, right?)

Sorry for the basic questions. I just want to make sure I don't make any stupid mistakes.

I don't know Final Cut (KK, Georgia, anybody?), but I would guess it's peak, not RMS.

Originally Posted by bradhebert

Hmm. I would have thought a bit higher, in general. Since dialog tends to average around -22dB to -27dB, right?

Well, there is only so much we can discuss in terms of dB
The figures I gave you are from quick-checking some music cues in Little Miss Sunshine, Fight Club and other films I have in the timeline.

BTW, disregard the comment about imagining you're in a theater, I forgot you're doing a DVD.

[Jfriah]

Originally Posted by danijel

I've heard this before, that it's better not to compress the mix for radio play, and that music that is actually less compressed will sound louder, but then, I've also heard that's not true

Interestingly, on a road trip last weekend, I was playing "The Pretender" by Foo Fighters (fairly dynamic) via mp3 into my car stereo and it was tough to hear some of the quiet intro/bridge stuff over the road noise but the chorus kicked in pretty good. Then this week it was on the radio. The chorus was perceived as being quieter than the intro/bridge!

(broadcast) Compression...gotta love/hate it.

-Jeff

[danijel]

Originally Posted by Jfriah

Interestingly, on a road trip last weekend, I was playing "The Pretender" by Foo Fighters (fairly dynamic) via mp3 into my car stereo and it was tough to hear some of the quiet intro/bridge stuff over the road noise but the chorus kicked in pretty good. Then this week it was on the radio. The chorus was perceived as being quieter than the intro/bridge!

(broadcast) Compression...gotta love/hate it.

-Jeff

In the Radio section of the sticky here, I've recently added a link to the well-known Bob Orban / Frank Foti / Bob Katz article which clearly shows what happens on-air in regards to compression. That article has somehow escaped me thus far (although I have the Katz book right here on the shelf).... A must read for everyone interested in radio!

http://www.orban.com/support/orban/t..._Truth_1.3.pdf

[Eanan Patterson]

Doing a basic mix in Pro tools for a doc. What peak limit should my master out be set at? -3....-6 for TV broadcast?
thanks for the help guys!

[danijel]

Originally Posted by eanan

Doing a basic mix in Pro tools for a doc. What peak limit should my master out be set at? -3....-6 for TV broadcast?
thanks for the help guys!

Since you're in Ireland, I searched your national broadcaster Radio Telefís Éireann though google with this query: ppm site:rte.ie
It returned this document:
http://www.rte.ie/commissioning/docu...programmes.doc

which says:
Audio line up should be recorded at reference level -4VU, corresponding to 0dBu or PPM 4. Subsequent programme peaks must not exceed this reference level by more than 8dB (PPM 6). This corresponds to a line up level for digital audio (AES/EBU)
of – 18dBFS.


So, I suppose you could just limit at -10dBFS, and maybe check the mix with something like PPMulator in the end, to see if your PPM readings are OK. That will make your mix acceptable for all of European broadcast.

Of course, if there's someone from Ireland on-board, please speak up!

[Eanan Patterson]

thanks for getting back to me Danijel, so does this mean that my entire mix can't go past -10 on my levels??

[danijel]

Originally Posted by eanan

thanks for getting back to me Danijel, so does this mean that my entire mix can't go past -10 on my levels??

Probably, but I can't guarantee you that. If you can, get a PPM meter like PPMulator and make sure you don't cross the 6 mark.


BBC Type PPMs to BS 5428 are scaled in 4dB steps numbered from 1 to 7 with
increasing signal level. They are calibrated so that line-up Level will read PPM 4 and
thus Peak Programme Levels shall not read higher than PPM 6.
Digital “true” peak reading meters such as on VTRs and DAT recorders will typically
read 4dB higher than a BS 5428 PPM on programme material though they should
agree on steady tone. As it is unusual for digital level meters to match a BS 5428
PPM when measuring programme material, suppliers should never use “peak
reading” meters to assess programme levels accurately unless they are known to
meet BS 5428.

From this excerpt it looks like you'd be pretty safe at -10dBFS, because short peaks (as far as I understand it) could even go to -6dBFS, while the PPM wouldn't cross the 6 mark. But I'm just guessing here. If no-one from the PPM-land (UK, Ireland) chimes in here, start a new thread or try looking some more - I'm sure this has already been answered before, either here or at the DUC - try searching for "PPM", "peak", "BBC" etc.

[Deady11]

very nice work!!!

[gmarinov]

Originally Posted by danijel

From this excerpt it looks like you'd be pretty safe at -10dBFS, because short peaks (as far as I understand it) could even go to -6dBFS, while the PPM wouldn't cross the 6 mark. But I'm just guessing here. If no-one from the PPM-land (UK, Ireland) chimes in here, start a new thread or try looking some more - I'm sure this has already been answered before, either here or at the DUC - try searching for "PPM", "peak", "BBC" etc.

PPM6 is the maximum allowed. Although in theory it should correspond to -10dBFS, due to the nature's meter, peaks do go higher on a digital peak meter indeed. And then the sound ops typically don't make a fuss about an overshoot or two past PPM6, though according to the legend, the transmitter doesn't like it. PPMulator+ should help. Some channels have a minimum sound level requirement too.

TV sound - gotta love it...

[danijel]

Originally Posted by Deady11

very nice work!!!

Thnx!

Originally Posted by gmarinov

PPM6 is the maximum allowed. Although in theory it should correspond to -10dBFS, due to the nature's meter, peaks do go higher on a digital peak meter indeed. And then the sound ops typically don't make a fuss about an overshoot or two past PPM6, though according to the legend, the transmitter doesn't like it. PPMulator+ should help. Some channels have a minimum sound level requirement too.

TV sound - gotta love it...

Thanks! It's good to know our continental mixes are good to go in the UK

[BVS]

In New Zealand the specs for TV3 transmission are as follows( Audio only):


TV3 PROGRAMME DELIVERY – 0092
( Audio Specs Only )

SD
Digital Betacam with full stereo mix on tracks 1 and 2, music and effects on 3 and 4

HD
High Definition should be 1080i50. Our preference is for HDCam-SR with audio tracks to SMPTE standards, that is:
1. Left
2. Right
3. Centre
4. Sub
5. Left Surround
6. Right Surround
7. Left Total (LT)
8. Right Total (RT)
with music and effects on tracks 9 and 10 if available. If 5.1 audio is not available then audio should be as for SD delivery: tracks 1 and 2 full stereo mix, 3 and 4 music and effects. Please note we cannot accept Dolby E.

If HDCam-SR is not an option then we will accept HDCam with audio as per SD delivery.

We will only accept HDCam-SR, HDCam or Digital Betacam for master delivery.



Audio
Programmes should be delivered in Stereo format

Left audio shall be present on the A Leg or Channel 1
Right audio shall be present on the B Leg or Channel 2

If Mono is to be supplied, by agreement, it shall be dual mono format with identical and coherent audio on both left and right channels so that it may be used amongst stereo programmes.

Reference audio shall be -18dB below digital peak that is to correspond with PPM mark 4 on a standard Peak Programme Meter (PPM). The maximum or peak programme level indicated with a PPM shall never exceed PPM mark 6 or 8dB above this reference level.

Line up Tones serve to identify individual signal channels and to provide Reference Levels to indicate that without adjustment the programme transmitted will be within the signal level limits specified. At least 1 minute of reference tone (PPM4) between 800 and 1000Hz is to be recorded on Audio track 1 (left) and Audio track 2 (Right) so that the phase is identical on both tracks to allow replat phase checks.This tone shall be accompanied by visual Colour bars and be present on each track carrying programme sound. All tracks to be mute for 2 seconds preceeding programme start.

The relative timing of sound to vision shall not exhibit any perceptible error. Sound shall not lead vision by more than 20ms or lag by more than 40ms. (Note: 40ms=1 frame)

AC-3 audio specifications are available on request only if programme is to be delivered on HD Cam SR tape with all audio channels as discreet (No Dolby E encoded accepted)


.

[kheftel]

Thanks for this thread, it has been very informative. Lots of good links to lots of good reading.

[danijel]

Just found this great research about European broadcast levels:
http://loudness.hku.nl/Peak_and_LKFS_-_Grimm_ea.pdf

They have actually measured 50 TV stations peak, PPM and LKFS, and the results are very interesting. There are differences in loudness of up to 16dB from the loudest to the softest station Some stations limit their program so that their dynamic range is sqashed to 9dB (between LKFS and true peak), while some leave it so that it sometimes reads 20dB of difference. And I actually thought Europe was 'OK' compared to North America.... If you don't want to read the entire article, scroll to the bottom, there are some color graphs

They recommend peak level of -5dBFS, and loudness of -21LKFS, because that would require the least amount of change in most brodcasters. On the other hand, the official EBU P/LOUD group suggests -1dBFS /-24LKFS:
Main Page - EBU Wiki - Technical - P/LOUD

[so loud]

Ok, So I've done quite a bit of broadcast mixing for TV Ads in the US, but have limited experience mixing for cinema. I'm taking a spot that we mixed for tv which was limited at -8db and delivering a stereo mix for cinema. The room that we're working with is not calibrated so room levels are not applicable. Is there a safe limiting level that anyone would suggest for a stereo cinema mix? I would assume -.3db to make it as hot as possible since we're only using two channels.... but you tell me. Thanks!!!

[georgia]

This ain't TV... you need to mix in a calibrated room. period. but if you inisist on cutting your own ( and your clients ) throat....

you need yo get the dialogue to around -12 for yelling with a bit higher for the occasional scream, and down to -22ish -20 for speaking, and down to -23 for whispers.
big crashes / hits/ explosions can tap 0. quiet ambience can get all the way down to -28 ish or lower.

then mix everything else around that. you can go to 0 if you are doing digital, but if you are going optical there are more issues. you also need to mix while monitoring thru the x-curve or film curve, because the low end is rolled off in the cinema and the high end and low end is rolled off in the cinema. Serious lows need to be bussed to the LFE track to the theater subs...

also a Cinema ad in the US... MUST meet a couple specs for levels ( maybe some one can tell you, I don't recall off hand and I don't have time to chase it down right now sorry )

do a test and go screen it it see how far off you are, so you can correct.

good luck... you'll need luck to make this work well.

cheers
geo

[marty lester]

Originally Posted by georgia

then mix everything else around that. you can go to 0 if you are doing digital, but if you are going optical there are more issues. you also need to mix while monitoring thru the x-curve or film curve, because the low end is rolled off in the cinema and the high end and low end is rolled off in the cinema. Serious lows need to be bussed to the LFE track to the theater subs...
geo

Georgia...i know you were in a hurry. would you mind clarifying your point about "low end is rolled off in the cinema..."

that sentence tripped me up...sorry if i'm mis-reading it.

thanks a ton! i'm about to finish a documentary that'll likely air twice in a theater and then-onward on DVD, so it seems like i may be making two mixes. i'm mixing to "regular" stereo (which i've been warned is somewhat dangerous in a theater) by the way.

thanks,
marty.

[georgia]

just that the x-curve ( film curve ) rolls off the highs and the lows. so if you want serious boom you need to route to the LFE track. do a search for the eq curve there are lots of posts all about this and mix levels. I'd try to help more, but i'm in crunch mode before I screen my new show tomorrow in NYC....

cheers
geo

[danijel]

Originally Posted by georgia

also a Cinema ad in the US... MUST meet a couple specs for levels ( maybe some one can tell you, I don't recall off hand and I don't have time to chase it down right now sorry )

Here it is (and not just for the US):

Originally Posted by danijel

However, there is a maximum loudness level for theatrical trailers and commercials which is measured with the Dolby Model 737 Soundtrack Loudness Meter.
Trailer loudness should not exceed 85 dB Leq(m), as regulated by TASA.
Commercial loudness should not exceed 82 dB Leq(m), as regulated by SAWA.

If you are going to film tape, the easiest safe way would be to send your mix to someone with the aforementioned meter who will turn your mix up or down to bring it to 82, and bounce the mix back to you before you send it to the lab. If the commercial will play off of a DVD, noone can help you - I'd personally go for as loud as possible, or at least with the TV mix.....

[paulo m]

Just reading through the threads here and would like to introduce another topic on the debate, if you don´t mind:

regarding trailer and commercials voice overs for cinema presentation, what are your experiences regarding channel assignment and level? Are you using compression or other effects to bring the voice over to a really in your face kind of sound or not? Where do you usually pan, to the center only or try to have divergence to front left and right also? I recently read a book from T. Holman were he refers that in some movie, they placed the voice over on the main three front channels to give it a closer and more intimate sound that could easily be diferentiated in timbre to the film dialogues.

What average level do you mix to, considering for example the meter bridge scale on the Dolby DMU? In TV, I usually set my voice over levels around -20 dbfs, as I start to add music and effects the meters would easily go above -10 dbfs.Then I mix by hear, just taking control of mix elements by compressing individual elements if needed and then applying a gentle compression on main bus. This way I can achieve a good degree of loudness without smashing the signal and during the years, this has proved more than adequate.

But in TV, dynamic range is short, compared to the cinema, so I really want to take advantage of that, in order to provide a cleaner sound and definition.

I come from TV business but starting doing trailer and add sound for the cinema at the moment, so any experiences welcome.Thanks

[danijel]

Originally Posted by paulo m

Just reading through the threads here and would like to introduce another topic on the debate, if you don´t mind:

That's an interesting topic on its own - I suggest you open up a new thread and copy the text from your post over there