• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

How loud is loud, how to measure it? Is THX calibration bad for your health?

Soundmixer

Senior Member
Joined
Mar 8, 2021
Messages
433
Likes
296
Most recordings are mixed at around 80-85dB average SPL
Not movie soundtrack destined for the home. They are mastered at 75-79 dB. Music has no set mastering level - levels are all over the place. Calibration levels for THX commercial theaters are different from the calibration levels of THX in the home.
 

EEE272

Active Member
Joined
Jan 4, 2022
Messages
128
Likes
61
Not movie soundtrack destined for the home. They are mastered at 75-79 dB. Music has no set mastering level - levels are all over the place. Calibration levels for THX commercial theaters are different from the calibration levels of THX in the home.
Sorry that I get confused.
If we take a movie mastered at 75 for home use, it means that if I have my THX receiver at 0 dB, it would be 10dB too loud, right?
So the reference for home listening, if there is such a thing, would then be -10dB on the receiver.
Is that correct?

It would not surprise me because even at -20dB on the receiver, most of the time, it feels too loud to me.

One exception is The Dark Knight Rises on Netflix. I think it might have to do with the dialog constraint someone mentioned. Bane is very loud in comparison to others and if his voice is used to balance the rest, it would become very silent.
 

audio2920

Active Member
Joined
May 21, 2021
Messages
235
Likes
292
So the reference for home listening, if there is such a thing, would then be -10dB on the receiver.
Is that correct?
Exactly.

Well, almost. Another thing we haven't discussed here is the effect of Dial Norm on replay gain, which exists upstream of AVR calibration. I shan't go in to that (plenty of detail available on Internet search) but it's yet another variable that makes it impossible to say "reference is [x]dB"

Long story short, if I'd mixed something at 79, given I listen a few dB quieter at home than in the studio, I wouldn't be surprised if played that mix back anywhere between, say, -15 and -5 relative to THX ref, in a home environment.
 

TrevorLJ

New Member
Joined
Jan 7, 2022
Messages
4
Likes
15
Location
Fort Worth, TX
While watching movies I usually shoot for peaks of no more than 89 dBA with a meter. Anything more than that and it gets a bit uncomfortable. It's not only the peaks either as dialog can be unnaturally loud also. Depends on the mix where that 89 dBA peak lands volume wise. Super 8 is also mixed Super Hot...-22dB on my processor gets me just under 89 dBA during the clipping...I mean train wreck. Avengers Age of Ultron's intro battle needs -13dB to get there. Just gotta play it by ear! (Ba Dump Ba Tssss....:facepalm:)

Now...calibration wise I spend quite a bit more time in that 95-100 dBA (and ~105-115 dBC) area to ensure the overhead is there. I don't leave the room but I do sport double hearing protection (plugs and muffs) to keep my ears happy.
 

Soundmixer

Senior Member
Joined
Mar 8, 2021
Messages
433
Likes
296
Well, almost. Another thing we haven't discussed here is the effect of Dial Norm on replay gain, which exists upstream of AVR calibration. I shan't go in to that (plenty of detail available on Internet search) but it's yet another variable that makes it impossible to say "reference is [x]dB"
Dial Norm is not used with Atmos, and it was never used by DTS. It was mostly used with Dolby Digital and to a lesser extend TrueHD on disc. It is also not used for streaming, at least not on the titles I have encoded.
 
D

Deleted member 23982

Guest
who came up with this "standard" and whats the reason? if i watch movies the dynamic range is already too much with flat responding speakers, i mean litereally, i would guess that in mastering movies "normal conversations" are already at -10 to -20 db compared to "explosions/actions" and the THX standard suggest to basicly extend this dynamic range by 10db more??? wtf? i hope movies dont get mastered specially for "THX" use, it just seems like a crap standard to make movies more boomy on "casual" speakers which dont offer much dynamic range to begin with or something like that
 
Last edited by a moderator:

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,448
Likes
7,957
Location
Brussels, Belgium
Not movie soundtrack destined for the home. They are mastered at 75-79 dB. Music has no set mastering level - levels are all over the place. Calibration levels for THX commercial theaters are different from the calibration levels of THX in the home.
Sorry can you share with us the reference in more details?

Using a C-weighted microphone running pink noise (one speaker), what would 0 dBFS be while you’re mixing for a movie destined at home?
 

audio2920

Active Member
Joined
May 21, 2021
Messages
235
Likes
292
Dial Norm is not used with Atmos, and it was never used by DTS. It was mostly used with Dolby Digital and to a lesser extend TrueHD on disc. It is also not used for streaming, at least not on the titles I have encoded.
What's DTS? Hahaha... no, I get that it's a Dolby thing. That's exactly what I mean though, it's a discrepancy.

I'm surprised you encode your own mixes though, I've never had to do that. That happens at the authoring house or streaming platform's encoder whenever I've delivered. In the early days I had to set Dolby E metadata, or tell the authoring house what data to set on Dolby D, but that all got bumped downstream from me years ago.

Interesting point about Atmos / Dial Norm - I hadn't really clocked that. (Since I've almost never set Dial Norm, I didn't miss it when it wasn't in Atmos.) However, the DD part of a DD+JOC stream will have a value set for dial norm, so if that's discarded by the Atmos decoder and used by DD, there's probably a level discrepancy in required gain depending on which format you listen in.

[Edit: I do disagree with the comment about streaming not using Dial Norm though; if anyone's got a Dolby AVR which displays it, you can see the dial norm value changing with content. Maybe some platforms set-and-forget this value, which would be fair given the regulation/normalisation applied to mix levels we supply. But it's definitely still in play. My Atmos decoder doesn't show it - I guess proving your point that atmos doesn't use it! Anyway, I have no strong feelings about this as for the most part it's only introducing a few dB more discrepancy...]
 
Last edited:

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,448
Likes
7,957
Location
Brussels, Belgium
A very interesting read is the WHO-ITU for safe listening devices and systems, since the majority of the research shared here is strictly about noise and not music or active listening it might be of interest to all of us.


To summarize, they recommend ~80dBA (40 hours a week) for adults

It is also recommended that the device allow the user to select their reference exposure as
one of two operational modes:
Mode 1 WHO standard level for adults: this will apply 1.6 Pa 2h per 7 days as the
reference exposure (derived from 80 dBA for 40 hours a week)9.
Mode 2 WHO standard level for sensitive users (e.g. children): this will apply 0.51
Pa2h per 7 days as the reference exposure (derived from 75 dB for 40 hours a week).

1641643238813.png
 

audio2920

Active Member
Joined
May 21, 2021
Messages
235
Likes
292
Using a C-weighted microphone running pink noise (one speaker), what would 0 dBFS be while you’re mixing for a movie destined at home?

Don't know about soundmixer but my starting point is usually around 79 rel Dolby, so 0dBFS = 99dBC.

But this is only ballpark. I might turn the monitoring down even more if I want to deliberately restrict the dynamic range I use.

(Also peaks are limited lower than 0dBFS for home ent, typically maxing out at -2, so reducing peak required level to 97dBC in the example above.)
 

EEE272

Active Member
Joined
Jan 4, 2022
Messages
128
Likes
61
Don't know about soundmixer but my starting point is usually around 79 rel Dolby, so 0dBFS = 99dBC.

But this is only ballpark. I might turn the monitoring down even more if I want to deliberately restrict the dynamic range I use.

(Also peaks are limited lower than 0dBFS for home ent, typically maxing out at -2, so reducing peak required level to 97dBC in the example above.)
Sorry that I ask again but it is really confusing the way that loudness is handled...

When you say 0dbFS=99dbC then it means that you have already configured your system such that its max output leads to 99dBC out of your loudspeakers? In other words, you could set the amp to -5 then and use your input device up to full scale?
Then you are mixing (or creating) with the goal of around 79dB out of your loudspeakers?
So that means you are working towards -20dB output from your software, as we then meet this output out of the speakers.
This means that someone that is playing back your mix on a calibrated system at reference level would play it 6 dB too loud?
Because instead of 99dbC, the system would be calibrated to 105dbC.
Funny enough, any cinema receiving your content would probably think that all is good because the average is lifted to 85.
Is that what you meant? If I may ask, what type of content are you producing? For music, it seems the standards are basically non-defined...

Regarding Dialnorm, it gets even more confusing because Netflix does actually use it.
You can hear it quite clearly when comparing to the bluray version or Prime Video.
Fortunately, you can lookup the value in the sound info of the amplifier, which shows by how much the dialnorm shifted the output.
It's still quite a pain because the problem is that many of us use loudness compensation and that might yield a very wrong compensation, when things are not following the standard.
 

EEE272

Active Member
Joined
Jan 4, 2022
Messages
128
Likes
61
A very interesting read is the WHO-ITU for safe listening devices and systems, since the majority of the research shared here is strictly about noise and not music or active listening it might be of interest to all of us.


To summarize, they recommend ~80dBA (40 hours a week) for adults



View attachment 177532
I think someone mentioned this before but most of these standards are only meant to make sure that at the end of your work life, you are still able to follow a conversation properly. Your hearing might still be damaged. Seeing that the table stops at 40 hours, it seems the source might be work related.

In any case, the time doubles per 3 dB, which is reflecting the amount of energy you put into the speakers.
If you go for 74 dBA, you can basically listen the entire week... ;)

EDIT: I think I found it. Your document says it is based on: https://www.itu.int/rec/T-REC-H.870-201808-I/en
Page 47 describes the test of workers exposed to noise. At 83 dBA during the work week, they noticed a 10dB dip at 4KHz compared to the average person without additional noise exposure at the end of their worklife and around a 2-3dB dip after only two years.
A few pages before they describe the difference between music and work noise. While no long term study has been done, they observed the temporary shift and found out that music can be worse than work noise - with the funny exception of classical music and they also point out that the person who mentioned to dislike the music had the worst effect, which they do say might indicate that a subjective opinion can influence the outcome... In my opinion, it seems that this part is not well explored yet.
 
Last edited:

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,448
Likes
7,957
Location
Brussels, Belgium
I think someone mentioned this before but most of these standards are only meant to make sure that at the end of your work life, you are still able to follow a conversation properly. Your hearing might still be damaged. Seeing that the table stops at 40 hours, it seems the source might be work related.

In any case, the time doubles per 3 dB, which is reflecting the amount of energy you put into the speakers.
If you go for 74 dBA, you can basically listen the entire week... ;)
Work in production of music? because this is specifically about listening systems not random work related noise. That was kind of my point.
 

audio2920

Active Member
Joined
May 21, 2021
Messages
235
Likes
292
When you say 0dbFS=99dbC then it means that you have already configured your system such that its max output leads to 99dBC out of your loudspeakers? In other words, you could set the amp to -5 then and use your input device up to full scale?
Then you are mixing (or creating) with the goal of around 79dB out of your loudspeakers?
So that means you are working towards -20dB output from your software, as we then meet this output out of the speakers.
This means that someone that is playing back your mix on a calibrated system at reference level would play it 6 dB too loud?

Sooo.... firstly, yes, in this scenario our "amps"* are set so that 0dBFS = 99dBC if we set the monitor level to 79 in the studio.

This is because in studio monitoring (film & TV) we tend to use Dolby reference of -20 = 85dBC / 0 = 105dB as the starting point for calibration. We refer to this as "85" or "Fader 7". This is a fixed thing for mixing for cinema.

When mixing for home, we have to some extent to decide ourselves what level to use. If we turn out "amp" down 6dB, we refer to this level as 79.

But, it's a common misconception is that 79 (or 85, or whatever) is some kind of target while we mix. We are not directly targeting any particular room SPL when we mix. Mixers tend to have a dialogue level baked in to their brain that they're comfortable working at. So if you give a mixer a room set at 85 they'll end up recording the dialogue 6dB lower than if you set the room at 79. It's mostly done by ear, with perhaps a casual "comfort check" on meters.

Similarly -20dBFS is only a calibration level. It exists to allow gain structure to be checked (and historically it exists for analog tape calibration) such that we can know that what we're playing back is at the same level as it was recoded. Again, it has nothing to do with the mix itself, or any kind of loudness target. All these numbers are just for repeatability of replay gain.

To your last point, ignoring a whole bunch of other factors (e.g. any loudness spec normalisation which is sometimes needed after the mix, dial norm etc) then yes, playing back at "reference" level would mean that something mixed at 79 is playing 6dB louder than it was mixed.

Since you've no way of knowing what level it was mixed at, nor if any gain adjustment was made to normalise it then [sorry to keep repeating this but...] IMHO there's absolutely no such thing as reference level at home. You can only guess at it by ear. Reference level only exists in post-production and cinema (well, OK, maybe there's a few THX discs, or discs where they accidently used the cinema mix - yes that does happen!! - in that case "reference" would be the level it was mixed at - but again, there'd be no way to know for sure.)

*When I say "amps" I'm referring to the entire B-chain line up. We obviously don't go around changing gains on our amps, we have a monitoring controller, room calibration, console center section etc and we apply negative gain somewhere in the chain to achieve the same result.

Sorry that's very long winded, but hope that helps??
 
OP
sarumbear

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,324
Location
UK
Is there a reason why many of you use non-rounded numbers like 79dB or 99dB instead of 80dB or 100dB? Will a 1dBSPL difference an important or even noticeable?
 

audio2920

Active Member
Joined
May 21, 2021
Messages
235
Likes
292
Is there a reason why many of you use non-rounded numbers like 79dB or 99dB instead of 80dB or 100dB? Will a 1dBSPL difference an important or even noticeable?
Haha!! No not at all. It's just because Dolby use 85+20=105. Then, as people tend towards 3dB changes to be meaningful, monitoring levels are commonly found set at 85, 82, 79, 76 with peak being 20dB above that. But there's no law about it.
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,832
Anecdotally having seen Dune in a high end theatre I didn't find it that loud and I more often than not find movie theatres too quiet and not too loud though for my age I still have good hearing. Worked in audio for decades and will play my music at times wayyyy to loud, but I do that for a short period and get it out of my system. It's therapy. Being on the equipment / science side I don't have to experience regular volumes recording and mixing engineers will. I love live music especially local bands, but will use earplugs for some.

Thread started with a good premise but I think as others have stated the premise had flaws and actual experienced sound levels would rarely be at the level either in a theater nor at home of course where the listener has control.
 
OP
sarumbear

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,324
Location
UK
monitoring levels are commonly found set at 85, 82, 79, 76 with peak being 20dB above that. But there's no law about it.
Where are these set? By who? Who chose the 79dB for instance? Where is it written?
 
OP
sarumbear

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,324
Location
UK
Thread started with a good premise but I think as others have stated the premise had flaws and actual experienced sound levels would rarely be at the level either in a theater nor at home of course where the listener has control.
If you read enough ASR posts you will see that lots and lots of people talk about 100+dBSPL level listening levels. Yes they have the volume control but why do they talk about high levels. Not to mention my main premise for the post is the standard that defines home theatre listening is unhealthy. Again, many posters on ASR talk about having to reach the standard levels.

That is my premise.
 
Top Bottom