-16 turns out to be pretty close to what you get if you follow K-14. I've tested a number of commercial references using the Tonebooster EBULoudness meter, and almost none of them come in that low. My own material does, because I master to K-14. 
 
I wonder how automatic volume adjustment adapts to very dynamic music that might start out at -16 but then rise to -9 before it's through.

That's a really good question bit...

I would think some "look ahead" processing would have to transpire...
Best case scenario would be complete analysis before processing...

Some more info: http://www.tcelectronic.com/loudness/loudness-explained/
 
 
Gonna do a test and use :
MNoisegenerator and check what levels it will be in RMS to LUFS readings at a few points.

Gonna use : TT DR Meter and  klangfreund Lufsmeter for the metering.

Test is pretty coarse and average metering.

pink noise more variation in the rms metering due to more dynamic changes.
-10db rms  peaks at +0.6db = LUFS at -10.3 (S) -10.3 ( I )
-12db rms  peaks at -1.8db  = LUFS at -12.3 (S) -12.3 ( I )
-14db rms  peaks at -3.3db  = LUFS at -14.4 (S) -14.4 ( I )
-16db rms  peaks at -6db     = LUFS at -16.3 (S) -16.3 ( I )


White Noise
-10db rms  and peaks at -8.3db    =  -6.9LUFS (S) -6.9 ( I )
-12db rms  and peaks at -10.2db  =  -8.9 LUFS (S)  -8.9 LUFS ( I )
-14db rms  and peaks at -12.2db  =  -11 Lufs (S) and -11 LUFS ( I )
-16db rms  and peaks at -14.3db  =  -12.9 LUFS (S) and -13 LUFS ( I )

Note: now that was unexpected, i would expect 3db less reading in the LUFS.
there was a bigger change in the White noise between meters than with the Pink noise with more low end.

my own track testrack : (EDIT : Deleted ....... )

TT DR metering            : rms reading from -9.6db to -11.6db as a coarse reading, peak at -0.1db
klangfreund Lufsmeter :  -14LUFS ( I ) and -13.2LUFS (S) to -14.4LUFS (S) reading.

EDIT: gonna lower the track to -23LUFS as they want sound to be, and do a new RMS reading of it.

Heres the new RMS data to the -23LUFS ( I ) settings:

TT DR Meter :  -16.6db rms and peaks at -9.56db ( i find this a bit to low for my liking's at least as a master )
Maybe it is ok for a pre master !

I personally would go for a average of -21LUFS ( I ) as a standard for music
that would give about RMS of -15db.

I guess -23LUFS is needed for Cinema and DVD, cause they need a huge dynamic range. 

this concludes my quick research for now.

Jan

Anyone want to do a similar reading, maybe with other meters
or want to add something that i not have discovered yet 

Jan

Guys,
 
First, thanks for all of the great links! I am fascinated by this topic, and I really want to grok it but I'm struggling. A few questions. Sorry if these are basic, I feel like I'm missing a few key pieces of understanding here.
 
1. I have a project that is already recorded, so I put Hornet VU on with auto-gain and reference of -18. After playing the entire project, each track has now had a trim applied. Has this set the trim so that the max peak of each track hits -18, or that the average (RMS?) hits -18? Is this a good way to start?
 
2. Now that the tracks are trimmed, should I be checking the levels at each stage? For example, I have the Klanghelm VUMT set to -18. Again, is this peak at -18 or RMS at -18? Should I be making my tracks average -18 on this meter, or should they be averaging at 0 (is 0 actually -18 when this is set to -18 for reference)? What about when mixing busses?
 
3. When mastering (using Ozone 5 or Slate fgx), should you not be pushing up to 0 but instead peeking at -18, or again RMS at -18?
 
3. I used to just mix each track the best I could keeping the gain staging level through compressors, etc. I liked to output each track as loud as possible, each feeding a buss. I then use the busses to actually mix levels with. If everything got too loud for the 2bus, I could just use the gain to cut it back as long as nothing was distorting. I never had a problem, but now I'm thinking this might not be good...

Thanks for sharing your test results, Soundblend.
 
It's going to be hard to correlate any EBU meter to other methods, because what it does is different from anything that came before. I think of LUFS as average RMS, even though that's not technically accurate. While LUFS is a conventional reading, it's weighted to try and more closely reflect how we hear things. Consequently, it will never match up exactly with RMS calculations unless you're measuring pink noise.
 
The TTS DR meter is even further removed, as it has its own scheme. It ignores the statistically outlying data in an attempt to concentrate on the meat of the song. Extremely dynamic music confuses it. Extremely dynamic music is likely to confuse most methods of loudness metering.
 
One problem is that there are so many different goals one might shoot for. You've got technical requirements for broadcast, technical concerns for compressed formats, automatic gain adjustments in players, film standards, and what sounds best on a CD. Compound those with genre-specific expectations and you've got an awful lot of conflicting end goals.
 
For myself, the first priority is consistency. It starts with placing speakers at their optimal location and then always mixing at a consistent volume. It also entails listening to commercial recordings in that same environment, so I'll know when I'm drifting outside the norm even without looking at any meters.

Anyone want to run this one through it's paces and test for accuracy?

it's comes in at $363.99

http://bedsp.net/audio-lo...ue-peak-meter-alm-5-1/

cclarry
Anyone want to run this one through it's paces and test for accuracy?

it's comes in at $363.99

http://bedsp.net/audio-lo...ue-peak-meter-alm-5-1/

Only if it includes vintage analog hiss.

Oh, my word.  Has anyone priced a pair of high quality hardware VUs?   They are proud of those rascals, evidently.
 

Just wanted to post a follow-up since this topic really spurred some interest and activity from me.
 
1. I bought the Klanghelm meters. Based on a post here, I changed the default settings on rise and fall (200/400 I believe).
2. I used the Hornet Gain meter to auto-set the gain on all of my recorded tracks so that the faders at 0 support the -18RMS that the meters are set to.
3. For each plugin I inserted on the track, I put a Klanghelm in front and behind, and made sure the adjust the plugin so that the gain stayed fairly consistent coming out of the plugin. This gave me different results that matching peek metering in and out and matching by ear. (still using my ears, of course).
4. I used the same approach when routing multiple tracks to busses. Although, I have to admit I found that the busses were almost pre-mixed in levels when feeding them tracks that were metered like this.
 
I did this on a project that has been killing me - just haven't been able to get it mixed well. The results were FAR AND ABOVE what I would have expected - a much much better mix. Adding and moving all these meters around as I mix is a pain, but the results are great.
 
Not sure if this is a known technique or doesn't even work and I'm experiencing a placebo effect, but man I have a better mix! :)