YouLoud LUFS discrepancy

[RICHARD HUTCHINS]

Hi all,

This was to an extent covered in my earlier post but I started a new thread because the earlier post covered lots of other things, and got a little side-tracked ( a good side-track though, loads of interesting and useful comments, many well over my technical knowledge but lots of stuff I learnt along the way).

Anyhow I bought YouLoud to try and equalise the volume on my backing tracks for gigs.

Its a great app but I've hit a few issues perhaps someone could enlighten me?

Okay, this is how I use it; I mix the backing track stems in Cakewalk and then I open YouLoud in the Master, and adjust the gain to aim for around 14LUFS integrated.

I export this as an MP3.

But when I then drag the MP3 from desktop into the plugin ( still open in CW as a master plugin) the LUFS (integrated i.e. average for the track) are now in the 10s not 14s. Its got louder?

Does the MP3 format change the way the loudness is set as opposed to the tracks in Cakewalk before they are exported?

 

Am I doing it all wrong?

 

I have till Friday!

[Glenn Stanton]

so, export the file to MP3 with the plugin on the master showing (after fully playing it) an integrated LUFS of -14db.

close the project (after saving it of course).

then open the standalone version of YouLean and drag the MP3 file onto it, and it reads?

in my projects, if i use Expose 2 or YouLean, they're usually within about 1/10's LUFS for integrated level based on my project settings of -14LUFS. the short term, dynamic range, etc values can be somewhat different because the tools may be using variants of the "standards" in their application (for example, using AES in both results in slightly different values, but again usually are in 1/10's of differences).

 

Edited January 10 by Glenn Stanton

[John Vere]

The drag and drop reading is the most accurate and in my experience is very accurate. I have a bunch of other analyzing tools and they all seem to agree on the loudness but You Lean will catch true peaks none of the others catch. 
I first thought it was just being inaccurate but I read the info on his web site which explains why it catches those peak at a per sample level. The others are just averaging it out. 
They  would never be audible but I found it was great way to test brickwall limiters. The Loud Max and Boost11 won the shoot out of 15 free BW limiters. 
Any way back to real time VS drag and drop Real-time readings give you only a ballpark and you have to play the whole song. 

And as I said in the other thread- make sure the YL meter is the very last thing in the Master bus effect bin. The master fader has to be at unity. 
Mute the master to double check nothing is bypassing it. 
I would test the integrity of MP3 by exporting as a Wave file. Then compare it to the MP3 version. 
 

[RICHARD HUTCHINS]

Thanks. I managed to install the plug in , using this path; c:\program files\steinberg\vstplugins\Youlean Loudness Meter 2.dll

 

But how do I use a standalone version on my desktop? Cant seem to find an exe file for the shortcut, just a dll

[John Vere]

You open its location in the VST folder and right click then choose more options ( w11 ) and then use create shortcut 

 

 

 But what you can do as well is keep the project open and after the export open the location dialogue from the blue toast and then open You Lean in Cakewalk and drag and drop from the export folder. I do this on my Laptop but on my main computer I have the Desktop shortcut on my second monitor and the system tray.  

Edited January 10 by John Vere

[Glenn Stanton]

make a shortcut to "C:\Program Files\Youlean\Youlean Loudness Meter 2\Youlean Loudness Meter 2.exe" and put it on your desktop

Edited January 10 by Glenn Stanton

[RICHARD HUTCHINS]

Thanks all will look at this later tonight

[RICHARD HUTCHINS]

I'm getting there. Shortcut sorted.

Now trying to get the LUFS readings roughly the same for each song. Checked a WAV against MP3 John and that's ok.

 

[gustabo]

3 minutes ago, RICHARD HUTCHINS said:

I'm getting there. Shortcut sorted.

Now trying to get the LUFS readings roughly the same for each song. Checked a WAV against MP3 John and that's ok.

 

Does LUFS
https://www.hornetplugins.com/plugins/hornet-thenormalizer/

[RICHARD HUTCHINS]

Hi all,

I hope this is the right place to add my comment, its an extension of the above so I guess so.

Okay, so I'm progressing with using YouLean to adjust the LUFS on my backing tracks, with varying degrees of success. But my problem is , on some tracks, they are very quiet to start with, they downloaded like that I reckon, and need to be louder; so using the master gain I lift the volume, aiming for round -14 LUFS integrated. trouble is, this sometimes ends up on full gain, all the way to the right, and it starts distorting with clipping in the VUs.

Is there a solution to this/ Am I doing something wrong?

[David Baay]

4 hours ago, RICHARD HUTCHINS said:

using the master gain I lift the volume

Depending on the musical content, it may not be possible to get it to -14LUFS without applying some compression to raise the RMS level while keeping the peaks from clipping.

[OutrageProductions]

17 minutes ago, David Baay said:

Depending on the musical content, it may not be possible to get it to -14LUFS without applying some compression to raise the RMS level while keeping the peaks from clipping.

Or a brick wall limiter set lightly.

Mixing well is both a science and an art. Practice makes improvements. 😀

Edited February 21 by OutrageProductions

[John Vere]

I noticed that Band in  a Box uses very low levels in its midi files. So I can see it’s easy to end up there. 
My 2 favourite brick wall limiters after testing over a dozen freebies are LoudMax and the Boost 11. 
Make sure and turn the make up gain all the way down . Then set the peak level to 1.0 or 1.2 which every you like. This goes on the master at the top and the YouLean underneath. 
Here’s my other trick. 
On the master. 
Turn on the pro channel EQ. Now when you over load the master the on off button will turn red. 
So you back off until it doesn’t. You can turn the pro channel off when done but it really shouldn’t matter . 
The Loud Max will catch overs but it’s best they are not steady.
I don’t care what anyone says that it doesn’t matter in digital, I’ve read otherwise from trustworthy sources. 

If that project still is peaking but has low LUFS readings then it will have something with high transient peaks. Possibly a snare of piano. 
Use Span or the pro channel EQ on the master to see if any particular frequency is peaking.  Dang it gets complicated fast sorry ! 
it might be also that the bass isn’t loud enough .
Bass is the most noticeable part of a backing track. That is why I always use only the AmpleP bass set at exactly the same velocity and output level of -4.0 db. 

Edited February 21 by John Vere

[RICHARD HUTCHINS]

Okay I'll try those fixes thanks. Its a lot of hard work much harder than I thought, I prefer a band rehearsal to be honest! But until I find another Paul McCartney that's not looking likely!

[hsmusic]

19 hours ago, RICHARD HUTCHINS said:

Hi all,

I hope this is the right place to add my comment, its an extension of the above so I guess so.

Okay, so I'm progressing with using YouLean to adjust the LUFS on my backing tracks, with varying degrees of success. But my problem is , on some tracks, they are very quiet to start with, they downloaded like that I reckon, and need to be louder; so using the master gain I lift the volume, aiming for round -14 LUFS integrated. trouble is, this sometimes ends up on full gain, all the way to the right, and it starts distorting with clipping in the VUs.

Is there a solution to this/ Am I doing something wrong?

Two songs having the same LUFS doesn't necessarily mean they will sound the same level. E.G. a soft ballad might have a higher LUFS than a rock track in a mastered album for continuity. (or the other way around)

[John Vere]

32 minutes ago, hsmusic said:

Two songs having the same LUFS doesn't necessarily mean they will sound the same level. E.G. a soft ballad might have a higher LUFS than a rock track in a mastered album for continuity. (or the other way around)

Absolutely. But backing tracks tend to be just like a real band after sound check. And in my case the only difference between regular dance tunes and a waltz will be instead of -14 LUFS that song will be expected to be probably-15 or -16. Mostly due to the drums being more subdued. But I keep the bad the same in all songs just like you would with a real bass player. 
 
 I also use backing tracks for our folk/ bluegrass band and those don’t  use loud drums And the other guys make me turn any keyboards just about off. So the bass is the driving force. We run that through a bass amp. The drums and keyboards only go to the floor wedges and in ears as a glorified click track. 
I only make sure the bass is always the same level. 

Edited February 21 by John Vere