To me that’s pretty much the same problem as with music consumption, although it really shouldn’t be. In other words we don’t have any control over playback levels in movie theaters so we should just ignore it. If the mix is done on a proper stage that’s all we can ask for and we should just make it sound good there. If individual theaters screw that up it’s on them.
I forget where I saw the chart but if I remember correctly someone had done a bunch of loudness tests on commercial releases and I think the average was lower. This was maybe 5-7 years ago. From what I recall the average was closer to -27LUFS. -24LUFS is still pretty hot and unless my math is wrong it would be close to 95dB average on a calibrated stage. That’s still pretty loud for a normal movie.
The times I’ve been “forced” to mix to -24LUFS for short film festivals I’ve found that to be a bit loud and the times I hit closer to -27LUFS it all sounded much more “cinematic”, for lack of a better word.
At least in the US the -10dBFS peak was/is really just a legacy number because of old analog gear. I remember I had to target that myself when laying back to analog or even Digibeta, but that’s like at least 10 years ago by now.
I don’t recall what my setup was calibrated to. I do almost only post for TV. When I’ve prepped for bigger systems and theatrical I’ve boosted monitoring by about 3-4dB though to get some more headroom and not be too loud later.
-24 LUFS will sound pretty low.
I got that information from Carsten from the DCP-O-Matic forum who is a cinema technician for a German movie theater group.
He’s seen and analyzed a lot of DCPs - and that is the actual standard in use.
The -27 LUFS you are talking about is the dialnorm for audio encoders for broadcast and home movie theater (DVD/Blu-ray).
Program material integrated loudness of -24 LUFS will sound pretty low @ level 7?
Standard practice maybe, not “a standard”.
And let’s think about this for a second;
As far as I recall Dolby pink noise for calibration comes out to -31 LUFS, and when the playback system is set to the proper #7 that equals 85dB SPL at the mix position. So an average of -24 LUFS would equal 92dB SPL, again average level.
Does that not seem quite loud?
No, I’m not talking about “Dialnorm”, I’m talking about dialog gated BS.1770-1.
If that playback level is the norm then making that the de facto “standard” is in part the problem. If people who work in that part of the industry start telling people that they play back features that low and that people need to mix hotter to compensate then all they are doing is promoting a loss in headroom and dynamic range. It becomes a loudness war driven by theater operators. It’s stupid and backwards. It would be like Spotify telling music producers to mix louder because Spotify keeps lowering the streamed playback level because music producers mix louder.
At “4” we’ve lost 10db. So whatever range we had above average is now crippled. Compression as a result. Why does loud dialog sound “squashed”? This will be why.
"the intended average playback level of all these movies, regardless the volume setting during mastering, is about -27 LUFS. This is approximately 79 dBA in the theatre. "
I find that the interesting part is the responses by those working in (self-assessed) “big budget movie situations”, where virtually all are monitoring at 7 / 85dB SPL, and their impression of loudness on a scale from 1 to 7 is in order 5, 4 then 6 (fig 17 p17).
Next take a look at Fig 23, p20…
I’ll just reiterate that I think it is wrong and dumb to advocate mixing to a playback level lower than 7 on a properly calibrated mix stage for several reasons:
To the extent that there are problems with people feeling the overall loudness is too high this clearly does not solve anything as loudness ends up being the same.
Because mixers will just mix louder to compensate we now lose effective dynamic range and we get less headroom leading to more compressed sounding mixes (apart from loudness itself). And,
Because of this limited range discerning dialog is in practice bound to be harder as people end up with this lesser range.
I know those articles, and have already read them years ago.
But reality is, that movie goers have complained about loudness and movie theaters have started to react to this by turning the volume down.
This would not have happend if directors and producers wouldn’t have had insisted to the audio mixers on making their movie sound louder, louder, and louder.
This could have been prevented by setting an enforced standard like -23 LUFS in EBU broadcast - maybe with a more relaxed ± tolerance.
Then movie theaters wouldn’t have got so many complaints and wouldn’t have had to react somehow.
So you can mix for -27 LUFS but the reality is - it will be too soft. At least in European movie theaters.
The other thing is bass/LFE - despite the movie theater technicians always telling me that everything gets calibrated by Dolby to spec - it is all over the place.
In the same theater testing the mix in different auditoriums can sound very different and drove me crazy because the director started to freak out.
Back to voice mixing - in the book “Surround Sound Up and Running” from Tomlinson Holman (developer of THX) I found this interesting:
Let’s say that a mixer puts a vocalist into center, and also into left and rigth down say 6 dB, called a “shouldered” or “divergence” mix. If you shut off center, you will hear the soloist as a phantom down 3 dB (with power addition as you are assumed to be in the reverberant-field dominated region; different complications arise if you are in the direct-field dominated space). Only 3 dB down, the phantom image has a different frequency response from the actual center loudspeaker. This is because of acoustical crosstalk from the left loudspeaker first striking the left ear, then about 200 µs later reaching the right ear, with diffraction about the head and interaction with the pinnae in both cases occurring. For a centered phantom the opposite path also occurs of course. The 200 µs delay between the adjacent and opposite side loudspeakers and acoustical summing causes a notch in the frequency response around 2 kHz, and ripples in the response above this frequency. Now when added to an actual center loudspeaker signal, and only 3 dB down, and with a different response, the result is to color the center channel sound, changing its timbre.
Some mixers prefer the sound of a phantom image to that of an actual center loudspeaker. This is due to long practice in stereo. For instance, microphones are routinely chosen by recording vocalists and then panning them to the center of a 2-channel stereo monitor rig, that suffers from the 2 kHz dip. In a kind of audio Darwinism survival of the fittest, microphones with 2 kHz range presence peaks just happen to sell very well. Why? Because they are overcoming a problem in stereo. When the same mic is evaluated over a 5.1-channel system and panned to center, it sounds peaky, because the stereo problem is no longer there.
It is a danger to have much in-phase content accross all three front channels, as the inevitable result, even when the monitor system is properly aligned, is to produce noticeable and degrading timbre changes.
Arguably yes. Or movie goers and reviewers could have punished the film makers by simply boycotting the movies. That would send a clear message to creators.
Not really. -23 LUFS integrated still allows for balancing out the movie with very quiet and very loud parts. people would still complain. Or conversely you could end up with a movie you want to be quiet that is now too loud.
If anything, what I think would make sense moving forward is still mixing at 7 and then just analyze the mix and dump that as metadata in a sidecar-file next to the DCP. Then you take those metrics and the cinema can tell the audience how loud the movie will play back.
You could for example decide to play back at 7, look at a loudness histogram and calculate how much time is spent at SPL XdB in the middle of the theater. People can then judge for themselves if that’s what they are ok with. Or they can run some polls with the average audiences and translate perception of loudness to some scale and tie that to histogram etc.
Then European theater operators are idiots. I mean that. If they care about a loudness war they shouldn’t contribute to it.
“Mix your movie louder because I’m going to play it back softer.”
“I’m going to play back your movie softer because you mix it louder”
You see how dumb that is, surely. (?) It is idiotic.
The absolute best thing theater operators can do now is to tell re-recording engineers and directors to always mix at 7. That’s the best thing they can do. That way levels remain low, dynamic range and headroom high.
How is this even debatable?
To me that’s basically the same problem as we see with music: us engineers have zero control over how things are played back once we let go of them. Since we can’t know how they mess it up we can’t proactively deal with it either.
Only if creators would get the reason as feedback. Otherwise they would blame everyone else.
I agree - I just don’t know how to deal with that other then playing the game. There are no big studios with the influence to change things in my country.
You can imagine the fun sitting through the test screening with the director and his rising anxiety that there is too little bass/LFE and afterwards talking him out of raising the LFE to crazy levels.
Perhaps what is needed is an advocacy group coming out of AES/EBU and others, made up of re-recording engineers, who can advocate for better theater operation as well as informing the public.
One of the ways I work - and this is with music, not film mixes so might be a totally different kettle of ball games - is to abandon all these ‘flavour of the Month’ meters and simply stick to VU meters calibrated to 0dB VU = -18dB FS. For me, this works and works well - if it ain’t broken, why try to ‘fix’ it as this ends up in a nasty, vicious circle.
But as I said at the start of this post, I’m not working with film although I suspect the same rules should apply - but I know very little about film work so I could well be talking nonsense. The one thing that bugs me are the insane number of ‘standards’ we seem to have these days, and I honestly wish someone could explain to me why we ever went away from VU meters in the first place.
Indeed, regs can be put in place, yet even if the horse’s mouth is forced into the water; only said horse can take the responsibility to drink. We’ve got speed limits posted everywhere and the 2 second rule, yet very, very few following the guidelines
Let’s be clear, film and music recording are completely different arenas, so there is that
nevertheless, I want my productions to sound good and translate well, yet if we don’t have a consistent baseline infrastructure, we are all just guessing
My goodness do I appreciate this discussion, thanks to everyone taking the matter seriously…not my thread, but I appreciate all the advice
I don’t believe anyone strayed away from VU levels but in-terms of monitoring audio signal accurately, it’s just lacks too much depth in giving you the absolute TRUTH (about the entire picture). I believe over time (for the past 30 years) audio engineers brains have evolved and quickly adapted the idea/notion that “as long as it’s not clipping then it’s okay” mentality and that’s where I believe the VU meter demonstrates it’s purpose.
Other than that, the spectragraph/spectragraph is the most accurate way to visually view the details of audio. I just hope for the next version of Spectralayers users get a live monitoring mode so users can always monitor live incoming audio for everything.
I actually think the amount of standards is really reasonable, it’s just the implementation of them that varies. And as for VU meters the loudness standard is much better in estimating perceived loudness, especially over different periods of duration.
Though I don’t care for the ballistics (technically) of Supervision’s short-term loudness measurement it really does offer a really large amount of great tools in general. The way I have it set up I can get so much more useful information about what I’m working on than a VU meter could possibly give me. It makes the work done much more consistent in "quality’.
This is true. My personal preference is to display a VU meter alongside loudness metering. VU metering is still quite good for judging the real-time ‘average’ level of the audio and could be considered a more intuitive tool in some circumstances.
In connection with this, Supervision is indeed good for viewing multiple meters but unfortunately we don’t have it directly in SpectraLayers, unless we are using it as an ARA extension. I mention the latter only because we are in the SpectraLayers forum. Might be helpful if we did have something like Supervision or at least loudness meters in SpectraLayers. AFAIK we have only a basic digital level meter (unless I missed something). But this minimalist approach does have advantages.