Loudness - short term max?

[fork-media]

job came to me as a series for DVD. no specs at all- so I just treated it as film - altho since it was reworking of TV series from about 4 5 years ago when they just specified -10 dBFS ceiling decision was made to go in stereo since lots of the source was stereo.


now client has just sent specs - because they have *just* sold it on- so I will have to rework about 6 hours of program time.............

the specs say meet R128 -23LUFs target loudness level..

no mention of short term max =

never bothered me before - since I only had to do lots of dialog mixes to spec.

this however is something like Maxsteel- robots blasting etc...

searching around I find ( cos I asked them and they are being a bit dense about it - nothing new there 23 LUFS? more like FFS )

" The current target loudness level for broadcast material, as stipulated in EBU-R128, is an integrated loudness level of -23 LUFS (±0.5 LU) with a maximum permitted true peak level of -1 dBTP (or -2dBTP if being compressed to codec like mpeg layer 2) and a maximum short term loudness of -18 LUFS (≤ 0.5 +LU) "


obviously job is a lot easier if I can exceed -23 on short term max


actually they havent specified a +/- on the longterm

only stated "comply with EBU R128 -23LUFS "

most jobs I had before where +/- 2dB on 23LUFS


any advice? TIA

[fork-media]

I have asked them but no replies in the last week - everything will be last minute of course.

[quadraphonics]

It has been a while since I have done DVD work, but here goes.

The most important thing is to measure the DialNorm of the track and set it correctly in the AC-3 encoding process. When playing back, the decoded AC-3 will play back at -31LKFS/LUFS based on how you set the dialnorm. There are other things that happen around this setting as well.

The other thing you could do, which a number of us have done in the past, is just mix to whatever and set the DialNorm to -31 so no level change takes place. Things could get a bit odd with the RF and Line listening modes engaged.

Randall

Originally Posted by fork-media

I have asked them but no replies in the last week - everything will be last minute of course.

[fork-media]

yeh I'm just putting lots of stuff thru my monitors now - its a new temporary room I am in - just found my sound level meter A or C weighted.- its pretty basic - but that combine with lots of cross referencing has got me lots closer - running a WLM meter on the monitoring chain.


still not clear about how I can get away with being over 21LUFS for ( assuming spec is 23LUFS +/- 2LUFS ) on short term

I'm seeing the odd 18 or 19 LUFS ones on shows like Max steel ( ITV Hub - ) and dragons on Iplayer. ( similar type of shows )

[jeremiahmoore]

TL;DR: Using a setting of -31 = “no shift”


From Dolby Metadata Guide

The Scale
The scale used in the Dialogue Level setting ranges in 1 dB steps from –1 to –31 dB. Contrary to what you might assume at first, a setting of –31 represents no level shift in the consumer’s decoder, and –1 represents the maximum level shift. Here’s why:
Dolby Digital consumer decoders normalize the average output level—that is, the output level averaged over time using the equivalent loudness method, Leq(A)—
to –31 dBFS (31 dB below 0 dB full-scale digital output) by applying a shift in level based on the Dialogue Level parameter setting.

Note: The –31 dBFS Leq(A) should not be confused with the station reference level (often –18 or –20 dBFS). It is common to have different Leq(A) values for program material that has the same reference level. An average loudness level of –31 dBFS Leq(A) is quite compatible with facilities running at a variety of reference levels.
When a decoder receives an input signal with a Dialogue Level setting of –31, it applies no level shift to the signal because this indicates to the decoder that the signal already matches the target level and therefore requires no shift. In contrast, a louder program requires a shift to match the –31 dB standard. When the Dialogue Level parameter setting is –21, the decoder applies a 10 dB level shift to the signal. When the setting is –11, it applies a 20 dB level shift, and so on.


A Simple Rule:
31 + (dialogue level value) = Shift applied

Example:
31 + (–21) = 10 dB


The most important point to remember is that in setting the Dialogue Level parameter, you are providing your listener with an essential service. For your listeners, setting this level properly means:

• The volume level is consistent with other programs.
• The DRC profiles you make available to them work as you intend.
Once dialogue level is set, you can set up DRC profiles to further benefit the consumer.

[paulo m]

Hi there,

short term loudness of -18 LUFS, is only required for programmes up to 3 minutes lenght, which I believe is not your case. It makes part of the suplement 1, of the EBU loudness specs , in this case R128.

For programmes longer than that, -23 lufs integrated and maximum peak of -1 dbtp are the only requisites for broadcast material. If your client is asking that for a DVD release without mentioning short term, that´s the explanation.

You can find more info on this EBU doc here:

https://tech.ebu.ch/docs/r/r128s1.pdf

[fork-media]

thanks great Paulo thanks for that! its cleared things up a bit

- I have been recording various shows to take a look at their levels.. and on the whole they are showing up on WLM meter as 23/24 LUFS longterm and *occasionally* 18 LUFS here are there on some punches explosions

but my clients client seem to have added an additional requirement - max level of -9 dBFS. I think they are hedging bets and want one master for DVD and transmission.

I think its doable and perhaps means compatible with older TX systems perhaps?

I wonder if I should get a better meter system - VisLM2?

[paulo m]

Hummm, ok. If they´re asking for -9 dbfs limitation is because as you say, they want a mix that can also be broadcasted in a country that didn´t fully adhere to the R128 spec, probably because they´re either on a transition phase or because their TV transmission is still analogue.

It obviously prevents you from doing a more dynamic mix, as -9 dbfs is not part of R128 spec.

It´s still doable, you just need to compress less your louder sounds, as heavy compression increases the integrated value a lot. Just mix things more softly around - 10 dbfs and you´ll be fine. Set a true peak limiter to - 9dbtp on your final bus just to be sure that nothing goes up than that, but don´t squash things too much, otherwise you´ll end up with a reading much higher than -23 lufs.

I would suggest that you propose your clients to deliver two diferent mixes, one for DVD and the other for TV, but probably they don´t want to pay for that...

As meters, I like both Nugen VisLM and Izotope Insight.

I also use an Izotope plugin called RX Loudness Control that I use for offline measurement and ocasionally correction of delivery files. Nugen has a similar product.

https://www.izotope.com/en/products/...s-control.html

Originally Posted by fork-media

thanks great Paulo thanks for that! its cleared things up a bit

- I have been recording various shows to take a look at their levels.. and on the whole they are showing up on WLM meter as 23/24 LUFS longterm and *occasionally* 18 LUFS here are there on some punches explosions

but my clients client seem to have added an additional requirement - max level of -9 dBFS. I think they are hedging bets and want one master for DVD and transmission.

I think its doable and perhaps means compatible with older TX systems perhaps?

I wonder if I should get a better meter system - VisLM2?

[fork-media]

yes believe they are still analog and are requesting the R128 for the future when they do go digital - to be honest tho its the first time I have been asked for -9dBFS tho, -10 and -12 are more familiar flavours..

have downloaded NuVis LM 2.5 demo.. looks nice.

cant get the "timecode" mode to work - is it supported in PT10.39?

hadnt really seen any benefit for me upgrading really. 10 works for me otherwise

[paulo m]

Originally Posted by fork-media

have downloaded NuVis LM 2.5 demo.. looks nice.

cant get the "timecode" mode to work - is it supported in PT10.39?

hadnt really seen any benefit for me upgrading really. 10 works for me otherwise

Works fine on 11xx and 12xx versions. Never tried it on 10.