Using a Loudness Meter

[scifunk]

Can anyone give a run down (or appropriate articles) on how to use something like Flower Audio's Loudness meter properly for dance music? Not for mastering though.

What is the target TOT?
Where should my peak be on the master fader?
If I'm aiming for 0 on the master fader do i then drop it to send for mastering?

RMS, VU, LKFS, FFS LOL, OMG?????

[scifunk]

HA HA no one knows

[SA Studio]

I see that, haha

I'd really like to hear Selig's take on this to clear it up as well. I know it's, for lack a better explanation, changed a little bit over the years as far as what's considered the global standard, I'm pretty sure.

A great question.

[riemac]

In mastering my trance tracks I aim for a RMS loudness between -4 and -5 RMS. The peak level I set to -0,3.
You should analyse the loudness of some other commercial tracks in your style.

[Last Alternative]

I've talked to a professional mastering engineer who told me they want NOTHING on the master channel and at least -.4dB of signal to work with.

[scifunk]

That goes without saying. But how do you use a loudness meter at mixdown to get the loudest mix? Or know your mix is loud enough?

[SA Studio]

Wow.

Shocked at no other responses. It's almost like we're being given enough rope to hang ourselves here, scifunk

Did you see the video I made? I try to give advice on how to read the Flower Loudness meter and what you should be looking for. In the video, I encourage around a -11RMS for the sake of Dynamic Range and fighting the loudness wars. Up above, when the guy says -4RMS, I'm betting he's talking about PEAK RMS, not average or total. In my video for example, my -11 Total RMS has a PEAK RMS of about -8RMS.

-4 Total RMS would be rather insanely loud. (It's totally cool that KOV is our resident vinyl guy - I'm sure some aspects to the finalization process before sending it to vinyl are different )

With that said, yes = I was schooled to measure RMS when Mastering. But I've also come to get familiar with the LKFS measurement as well.

In the video, you'll see how I use Adobe Audition's "Amplitude Statistics" feature that professionally scans the audio and gives you a very detailed description of your loudness stats. (Reading meters is fine, but scanning audio at the Mastering stage is empirical. Can't argue with it, and at the Mastering stage when you're finalizing a product for sale that needs to be within certain parameters, you scan that audio and finish it/examine it professionally.) I also cross-reference my stats from Audition with the readout I'm getting from the Flower Loudness meter in the video as well to try and make more sense of it all.

This here is a nice resource you should check out that explains some of the different meters out there: http://productionadvice.co.uk/lufs-dbfs-rms/

Here's my video:

[SA Studio]

This is also a great information to take in regarding YouTube's "Loudness Normalization": http://productionadvice.co.uk/youtube-loudness/

[selig]

I see that, haha

I'd really like to hear Selig's take on this to clear it up as well. I know it's, for lack a better explanation, changed a little bit over the years as far as what's considered the global standard, I'm pretty sure.

A great question.

My take is to submit what is requested by the powers that be. As an example, if they ask for peaks no greater than -6 dBFS, that's what I submit.

In my experience it's the broadcast (and possible film?) world that has promoted loudness standards for what should be obvious reasons (commercials have traditionally gone for maximum loudness making them intrusive compared to program material). By following these standards you are insured your material won't be further subjected to harsh processing. But there is also a trend for sites like Apple Music (iTunes) to lay down standards - see Apple's "Mastered for iTunes" specs for more info: https://www.apple.com/itunes/mastered-f ... itunes.pdf

Interestingly, and as far as I've been able to tell, there are absolutely no "loudness" standards at Apple. For example here's an excerpt from the above linked document:
"Whatever you decide—exquisitely overdriven and loud, or exquisitely nuanced and tasteful—we will be sure to encode it and reproduce it accurately. We only ask that you avoid clipping the signal."

So I'm not sure this was even a good example in this conversation, but it's a "standard" that has evolved in the digital music era that engineers should probably be aware of.

As for the original question, I don't use loudness meters, but would certainly need to use one if I was requested to target a specific level with my master. To date I've been perfectly happy to use the VU + Peak meter mode in Reason (or similar in Pro Tools) and have not had mixes/masters rejected to date. But I don't work in broadcast or TV/Film, so take my comments with a large grain of salt as always!

[selig]

This is also a great information to take in regarding YouTube's "Loudness Normalization": http://productionadvice.co.uk/youtube-loudness/

Hopefully Relevant Quote:
"And YouTube just took “loudness” out of the equation."

So they're saying youtube will make adjustments for you no matter what you submit, which to me means "make it sound good" and let youtube worry about loudness. That is where we are likely headed for all future internet audio IMO.

[8cros]

HA HA no one knows

Mixing in my old AKG headphones



Mixing in my old AKG headphones with the loudness meter.
Much louder.

[PhutureD]

That goes without saying. But how do you use a loudness meter at mixdown to get the loudest mix? Or know your mix is loud enough?

Why would you want the loudest mix? mastering engineer wants - 4db, give him that.
louder is not always better.
Just make sure the mix sounds good to you and measure the output on master bus Doesnt really matter the
exact number coz the guy the will will have a knob in his software which he can turn to increase , or decrese the gain, if too much or too little.
Last track i submitted was -10db. It gives him plenty room to work with.
.usual requirements are - no fx on master bus(reverb,delay,comp,limit,nothing), no stereo wideners on any channel , unless it is for creative purposes only.
So far every engineer who has mastered my work has not had any problem with this format.

[scifunk]

That goes without saying. But how do you use a loudness meter at mixdown to get the loudest mix? Or know your mix is loud enough?

Why would you want the loudest mix? mastering engineer wants - 4db, give him that.
louder is not always better.
Just make sure the mix sounds good to you and measure the output on master bus Doesnt really matter the
exact number coz the guy the will will have a knob in his software which he can turn to increase , or decrese the gain, if too much or too little.
Last track i submitted was -10db. It gives him plenty room to work with.
.usual requirements are - no fx on master bus(reverb,delay,comp,limit,nothing), no stereo wideners on any channel , unless it is for creative purposes only.
So far every engineer who has mastered my work has not had any problem with this format.

There's only so much a mastering engineer can do. Provide them with both a clear and loud mix has to be better for them to work on.

[ambeant]

In mastering my trance tracks I aim for a RMS loudness between -4 and -5 RMS. The peak level I set to -0,3.
You should analyse the loudness of some other commercial tracks in your style.

I agree, -6 RMS to -4 RMS is very comparable to commercial levels in that genre.

The Loudness War is inevitable

[MitchClark89]

sorry for to drag up the old thread but i have just bought the mastering bundle and also flower audio meter to see if i can do these things with -xx dB peak in the meter. to my ears my mix sounds quite nice (not perfect but i am happy with my progress from learning to mix down) with 4dyne into ozone into RE3Q on the master insert. it is reasonably loud although i do not necessarily want this, i would want to have a song with good dynamic range and appropriate volume as other songs i hear.

as you can see in my pic, the total rms is down but i cannot get the peaks down below 0dB. my problems is i dont know where to start but i have a few limiter and synapse eq in my track and some FET compressors on the drums and bass. on the SSL all levels are down at -10(dB?) and the master fader vol is also hover around that region.

would anybody just be able to give me a tip on where i might be able to look to reduce the peaks to get my project down to the apparent goal of -4 or -6dB peaK?

thanks to all reason people.

MC

[selig]

sorry for to drag up the old thread but i have just bought the mastering bundle and also flower audio meter to see if i can do these things with -xx dB peak in the meter. to my ears my mix sounds quite nice (not perfect but i am happy with my progress from learning to mix down) with 4dyne into ozone into RE3Q on the master insert. it is reasonably loud although i do not necessarily want this, i would want to have a song with good dynamic range and appropriate volume as other songs i hear.

as you can see in my pic, the total rms is down but i cannot get the peaks down below 0dB. my problems is i dont know where to start but i have a few limiter and synapse eq in my track and some FET compressors on the drums and bass. on the SSL all levels are down at -10(dB?) and the master fader vol is also hover around that region.

would anybody just be able to give me a tip on where i might be able to look to reduce the peaks to get my project down to the apparent goal of -4 or -6dB peaK?

thanks to all reason people.

MC

Ozone MUST be last, pretty much all I can see at first glance that would cause what I think you are describing.

I don't understand what you mean by "getting your project down to - 4 to - 6 dB. If you are talking about the raw mix you do this by lowering levels or bringing down the master fader. You would not have any mastering at this point, if I'm understanding you correctly. Not sure what metering you're using on the Master meter either - too many variables to know for sure without looking at your song file. But start by putting Ozone LAST - to be sure it's last (and if you use dither) I'd put it between the Master Outputs and the Hardware Interface instead of the Master Insert. Just my 2 cents, hope I'm understanding the question correctly.

[MitchClark89]

sorry for to drag up the old thread but i have just bought the mastering bundle and also flower audio meter to see if i can do these things with -xx dB peak in the meter. to my ears my mix sounds quite nice (not perfect but i am happy with my progress from learning to mix down) with 4dyne into ozone into RE3Q on the master insert. it is reasonably loud although i do not necessarily want this, i would want to have a song with good dynamic range and appropriate volume as other songs i hear.

as you can see in my pic, the total rms is down but i cannot get the peaks down below 0dB. my problems is i dont know where to start but i have a few limiter and synapse eq in my track and some FET compressors on the drums and bass. on the SSL all levels are down at -10(dB?) and the master fader vol is also hover around that region.

would anybody just be able to give me a tip on where i might be able to look to reduce the peaks to get my project down to the apparent goal of -4 or -6dB peaK?

Ozone MUST be last, pretty much all I can see at first glance that would cause what I think you are describing.

I don't understand what you mean by "getting your project down to - 4 to - 6 dB. If you are talking about the raw mix you do this by lowering levels or bringing down the master fader. You would not have any mastering at this point, if I'm understanding you correctly. Not sure what metering you're using on the Master meter either - too many variables to know for sure without looking at your song file. But start by putting Ozone LAST - to be sure it's last (and if you use dither) I'd put it between the Master Outputs and the Hardware Interface instead of the Master Insert. Just my 2 cents, hope I'm understanding the question correctly.

hi to you selig thank you for the reply and for all your replies to all users on this board as i always find your answer and insight useful and educational

sorry my post - what i would want to have is, a mixed down track (mixed down by me) that i can then apply some basic mastering too with the 4dynes and ozone to have a "finished" (just for me and friends and online sharing) track and to have the output fall within the range of acceptable volume and dynamics. when i say this, i mean from what i see you and others talking to about the -10db peak for tracks so that there is room for mastering. my SSL meters are all in VU and the master is VU + peak. i do not know how to get dbFS in these meters. some of my volume meters in ssl are down so low they are between -10 to -20 VU.. i am scared to turn them down more. the gains are mostly set at 0.

i have moved the maximiser in the chain as you have suggested.

thank you again for all the helps on here. it is wonderful

[selig]

hi to you selig thank you for the reply and for all your replies to all users on this board as i always find your answer and insight useful and educational

sorry my post - what i would want to have is, a mixed down track (mixed down by me) that i can then apply some basic mastering too with the 4dynes and ozone to have a "finished" (just for me and friends and online sharing) track and to have the output fall within the range of acceptable volume and dynamics. when i say this, i mean from what i see you and others talking to about the -10db peak for tracks so that there is room for mastering. my SSL meters are all in VU and the master is VU + peak. i do not know how to get dbFS in these meters. some of my volume meters in ssl are down so low they are between -10 to -20 VU.. i am scared to turn them down more. the gains are mostly set at 0.

i have moved the maximiser in the chain as you have suggested.

thank you again for all the helps on here. it is wonderful

Much to unpack here. I'm not sure about the -10 dB Peak you mention. I have suggested to follow the Props suggestion to have all individual tracks PEAK at around -12 dBFS. Many mastering engineers request mixes that peak between -6 and -3 dBFS - this would be with no mastering plugins "on". This should then be what you see when you bypass mastering plugins - make sense so far?

Next, ALL meters show dBFS in Reason. That just means that 0 dB is the clipping point, and values below that are in negative numbers. What is IMPORTANT is whether the meters are showing RMS (average) or Peak.

When you say your channel meters in the mixer are all down at -10 to -20 dB this is normal because they are in RMS mode. Peak levels can be as much as 10 to 20 dB or MORE above RMS levels, so those channels could ACTUALLY be peaking ABOVE the -12 dBFS levels that are recommended! In other words, though they look low, they COULD actually be high because they are not showing you peak levels.

I would not be "afraid" to turn levels down since Reason is a floating point system you have 1500 dB to work with. You can turn channels down, or up, by MANY dB before you'd have problems. The problem with levels being extremely low (or extremely high) inside of Reason has more to do with getting them to hit a compressor at a level that will produce any results - and the fact you'll have to eventually get the low levels back up or the high levels back down before you can get a good mix out of Reason. Meaning it's possible to have a level so low that the threshold on a compressor can't go low enough to get any compression, or high enough you can't NOT get compression! This is extreme to be sure, but it's POSSIBLE.

This is all to say, it's best IMO to keep ALL leaves peaking around -12 dB on individual channels, starting at the instrument or audio track and continuing on through to the fader. Any processing you add, especially compression, should be compensated to keep levels constant - this makes comparing A/B before/after processing more accurate as you won't be fooling yourself by simply adding gain.

Sub-mixes (can be hotter, and the final mix can also be even hotter (-6 to -3 dBFS). Hopefully this is not too much information and I'm making sense here!



Sent from my iPad using Tapatalk

[MitchClark89]

Much to unpack here. I'm not sure about the -10 dB Peak you mention. I have suggested to follow the Props suggestion to have all individual tracks PEAK at around -12 dBFS. Many mastering engineers request mixes that peak between -6 and -3 dBFS - this would be with no mastering plugins "on". This should then be what you see when you bypass mastering plugins - make sense so far?

Next, ALL meters show dBFS in Reason. That just means that 0 dB is the clipping point, and values below that are in negative numbers. What is IMPORTANT is whether the meters are showing RMS (average) or Peak.

When you say your channel meters in the mixer are all down at -10 to -20 dB this is normal because they are in RMS mode. Peak levels can be as much as 10 to 20 dB or MORE above RMS levels, so those channels could ACTUALLY be peaking ABOVE the -12 dBFS levels that are recommended! In other words, though they look low, they COULD actually be high because they are not showing you peak levels.

I would not be "afraid" to turn levels down since Reason is a floating point system you have 1500 dB to work with. You can turn channels down, or up, by MANY dB before you'd have problems. The problem with levels being extremely low (or extremely high) inside of Reason has more to do with getting them to hit a compressor at a level that will produce any results - and the fact you'll have to eventually get the low levels back up or the high levels back down before you can get a good mix out of Reason. Meaning it's possible to have a level so low that the threshold on a compressor can't go low enough to get any compression, or high enough you can't NOT get compression! This is extreme to be sure, but it's POSSIBLE.

This is all to say, it's best IMO to keep ALL leaves peaking around -12 dB on individual channels, starting at the instrument or audio track and continuing on through to the fader. Any processing you add, especially compression, should be compensated to keep levels constant - this makes comparing A/B before/after processing more accurate as you won't be fooling yourself by simply adding gain.

Sub-mixes (can be hotter, and the final mix can also be even hotter (-6 to -3 dBFS). Hopefully this is not too much information and I'm making sense here!

selig your reply has been a revelation to me! i did not know this about the SSL track levels showing RMS not peak! wow. i know how to get the master level to change modes to show peak but as i understand it the only way to check the peak levels on each individual track besides doing solo of each one is to insert a RE meter (selig gain for example ) on each one at the end of its chain (after compression fx etc) and then as you have said, work through each one to get the desired level.

ps i am proud to own your REs now which came with the bundle. i hope you still get the full entitelment from the sale. i could not buy them without being in the bundle i am sorry

[selig]

selig your reply has been a revelation to me! i did not know this about the SSL track levels showing RMS not peak! wow. i know how to get the master level to change modes to show peak but as i understand it the only way to check the peak levels on each individual track besides doing solo of each one is to insert a RE meter (selig gain for example ) on each one at the end of its chain (after compression fx etc) and then as you have said, work through each one to get the desired level.

ps i am proud to own your REs now which came with the bundle. i hope you still get the full entitelment from the sale. i could not buy them without being in the bundle i am sorry

One of the motivations for Selig Gain was to see Peak levels anywhere in the signal path, especially the SSL Channels which have only RMS metering.

What you can do is insert Selig Gain on an instrument (Thor, Kong, or a Combinator) and adjust the Volume on the instrument to hit around -12 dBFS peaks, then shift/drag Gain to the next instrument and adjust volume etc. Instruments aren't likely to change much even if you play a different part so you won't have to have a bunch of Gains all over your rack if you don't want them there.

That ensures your instruments are all entering the SSL mixer at around the same peak level, which aids in two ways - any compressors/gates/saturation/amps will all react similarly with similar settings, and your mix won't likely clip (if you ALSO follow the next step).

The next step and "best practice" is to be sure that when you do any processing such as compression, EQ, etc, that you're not drastically changing the overall level, which can fool you into thinking things are sounding better when they're mostly just louder. This will also keep your final mix level from getting out of control as you turn things up across the board.

In the Mixer you can check the final peak level by putting Selig Gain in the Insert (making sure it's still at the default routing position of post DYN post EQ (Check the diagram at the top of the channel). At this setting the insert is the last thing before the fader, and the last place where you want to check to be sure you're still hitting around -12 dBFS peaks on all channels, adjusting the Gain fader if need be. If you have insert FX already in place, put Gain LAST in the chain.

At this point you've checked levels at the beginning and the end of your channel signal path meaning that you're not adding or taking away any overall level with any processing in between. This means if you bypass processing you should STILL see the same level, making an A/B comparison more accurate and "fair". This should also keep your mix levels reasonable without having to chase the master fader etc.

[ClassickHitz]

selig your reply has been a revelation to me! i did not know this about the SSL track levels showing RMS not peak! wow. i know how to get the master level to change modes to show peak but as i understand it the only way to check the peak levels on each individual track besides doing solo of each one is to insert a RE meter (selig gain for example ) on each one at the end of its chain (after compression fx etc) and then as you have said, work through each one to get the desired level.

ps i am proud to own your REs now which came with the bundle. i hope you still get the full entitelment from the sale. i could not buy them without being in the bundle i am sorry

One of the motivations for Selig Gain was to see Peak levels anywhere in the signal path, especially the SSL Channels which have only RMS metering.

What you can do is insert Selig Gain on an instrument (Thor, Kong, or a Combinator) and adjust the Volume on the instrument to hit around -12 dBFS peaks, then shift/drag Gain to the next instrument and adjust volume etc. Instruments aren't likely to change much even if you play a different part so you won't have to have a bunch of Gains all over your rack if you don't want them there.

That ensures your instruments are all entering the SSL mixer at around the same peak level, which aids in two ways - any compressors/gates/saturation/amps will all react similarly with similar settings, and your mix won't likely clip (if you ALSO follow the next step).

The next step and "best practice" is to be sure that when you do any processing such as compression, EQ, etc, that you're not drastically changing the overall level, which can fool you into thinking things are sounding better when they're mostly just louder. This will also keep your final mix level from getting out of control as you turn things up across the board.

In the Mixer you can check the final peak level by putting Selig Gain in the Insert (making sure it's still at the default routing position of post DYN post EQ (Check the diagram at the top of the channel). At this setting the insert is the last thing before the fader, and the last place where you want to check to be sure you're still hitting around -12 dBFS peaks on all channels, adjusting the Gain fader if need be. If you have insert FX already in place, put Gain LAST in the chain.

At this point you've checked levels at the beginning and the end of your channel signal path meaning that you're not adding or taking away any overall level with any processing in between. This means if you bypass processing you should STILL see the same level, making an A/B comparison more accurate and "fair". This should also keep your mix levels reasonable without having to chase the master fader etc.

Hey Selig,

I would love to see what you just explained in a video tutorial. I'm more of a visual person myself and I also just like to archive helpful information for future reference to refer back to when needed.

[Galaxy]

Especially since Selig's long explanation never ends on the fact that once you've "re-unity-ed", if that's even a word, your mix at -12, that's when you're ready to start balancing levels.

Point = a lot of high end sounds, will sound way to loud if they peak at the same level as your mid and even worse low end sounds. You're either turning your bass drum up, or your hi hat down.

Selig, you never mention this part of the process clearly, leaving ppl to think a mix should have all sound peak at -12.

[selig]

Especially since Selig's long explanation never ends on the fact that once you've "re-unity-ed", if that's even a word, your mix at -12, that's when you're ready to start balancing levels.

Point = a lot of high end sounds, will sound way to loud if they peak at the same level as your mid and even worse low end sounds. You're either turning your bass drum up, or your hi hat down.

Selig, you never mention this part of the process clearly, leaving ppl to think a mix should have all sound peak at -12.

All I spoke about was setting levels. The question I was answering wasn't about how to mix.

So do I REALLY need to tell folks to move faders to achieve the desired balances?

[MitchClark89]

One of the motivations for Selig Gain was to see Peak levels anywhere in the signal path, especially the SSL Channels which have only RMS metering.

What you can do is insert Selig Gain on an instrument (Thor, Kong, or a Combinator) and adjust the Volume on the instrument to hit around -12 dBFS peaks, then shift/drag Gain to the next instrument and adjust volume etc. Instruments aren't likely to change much even if you play a different part so you won't have to have a bunch of Gains all over your rack if you don't want them there.

That ensures your instruments are all entering the SSL mixer at around the same peak level, which aids in two ways - any compressors/gates/saturation/amps will all react similarly with similar settings, and your mix won't likely clip (if you ALSO follow the next step).

The next step and "best practice" is to be sure that when you do any processing such as compression, EQ, etc, that you're not drastically changing the overall level, which can fool you into thinking things are sounding better when they're mostly just louder. This will also keep your final mix level from getting out of control as you turn things up across the board.

In the Mixer you can check the final peak level by putting Selig Gain in the Insert (making sure it's still at the default routing position of post DYN post EQ (Check the diagram at the top of the channel). At this setting the insert is the last thing before the fader, and the last place where you want to check to be sure you're still hitting around -12 dBFS peaks on all channels, adjusting the Gain fader if need be. If you have insert FX already in place, put Gain LAST in the chain.

At this point you've checked levels at the beginning and the end of your channel signal path meaning that you're not adding or taking away any overall level with any processing in between. This means if you bypass processing you should STILL see the same level, making an A/B comparison more accurate and "fair". This should also keep your mix levels reasonable without having to chase the master fader etc.

i really want to thank you for taking time and effort to make this explanation to me. i found old thread from last year where you had explained something close to this but the new information has provided perfect clarity. i am very excited now to go back to my mix and set all the levels first to get consistent peak and then start the mix down process - until now i was guilty of driving eq and compression too hard to get wht i thought was good mix but was just ear bashing loudness after the 4th or 5th listen. im am grateful to you for this help

edit: i thought to have a cool idea would be (takes a little time) but set up a whole section of rack containing the selig gain meter for each instrument/player etc and label them accordingly. this would be a help to me until i got confident to just have the one meter and drag it across each instrument as you mentioned above. i saw another user said they have a selig gain after each eq/fx/comp in a instrument chain to keep watchful eye on each stage of the signal!

[MitchClark89]

Especially since Selig's long explanation never ends on the fact that once you've "re-unity-ed", if that's even a word, your mix at -12, that's when you're ready to start balancing levels.

Point = a lot of high end sounds, will sound way to loud if they peak at the same level as your mid and even worse low end sounds. You're either turning your bass drum up, or your hi hat down.

Selig, you never mention this part of the process clearly, leaving ppl to think a mix should have all sound peak at -12.

please dont come in to the thread and upset the apple cart. there is good information being shared (by selig) and learned (by me). i do fully understand that gain staging is separate and pre requisite to starting mix down. i appreciate that you thought this was a help to me though.