Levelizer fun

[Grekim]

I am currently working on a 35 song mastering job:eek: To stay within budget, I'm supposed to spend around 20 minutes per song. It was mixed in a very small room, through a budget mixer, without compression and with NS-10 monitoring. So you could justify spending an hour on a tune.
The Levelizer has been great. I am using some nice outboard EQ/compression per request of client, but I have the Levelizer and Sonoris meter patched into an output track, with input-always-on-monitoring engaged so I can hear the combination of effects but not print the Levelizer just yet.

The Levelizer's normalize feature I have not gotten into yet and it sort of blows my mind that this feature exists. In the meantime, I figured out something very handy....when you set the peak limit to 50***37; you should be equivalently setting a threshold of -6.02 dBFS. What's nice is you can then safely increase the output gain by 6.0 dB and get within 0.02 dB of full scale. Similarly, a 25% peak limit should correspond to -12.04 dB threshold and you can safely bring the output up 12 dB. Of course, backing the output off by another 0.25 or 0.5 dB is a good idea.
So I left the Levelizer as a global setting (29% peak limit with a +10.5 dB output gain) for all tracks and adjusted the gain driving it. I used the SAW channel EQ to make some last minute fine tuning. When rendering a master via buildmix, all that was needed was to disengage mute for the channel of interest and tab to the region end to set boundaries.

When all is said and done I'll have 70 stereo tracks occupied. I'm halfway through and of course the CPU is not breaking a sweat at around 8%.

[Dave Labrecque]

FYI... you can set the Levelizer to read in dB or %.

[AudioAstronomer]

FYI... you can set the Levelizer to read in dB or %.

Yep. Click the IQS triangle and go for it ;)

[Grekim]

He he. Thanks! :o

The point was really that you can keep a lot of things static such as the Levelizer's settings and simply drive it via a channel fader, for example. Then you can effectively make use of things like the Sonoris meter (thanks Robert) applied to an output. So I have 70 tracks and only one Levelizer intance needed. No presets needed either. No dramatic new info here, just stuff for any Levelizer newbies and sharing the fun ;)

[Arco]

check out how bob uses the levelizer. always a good place to start:
http://www.sawstudio-media.com/video/levelizer.wmv

[Dave Labrecque]

He he. Thanks! :o

The point was really that you can keep a lot of things static such as the Levelizer's settings and simply drive it via a channel fader, for example. Then you can effectively make use of things like the Sonoris meter (thanks Robert) applied to an output. So I have 70 tracks and only one Levelizer intance needed. No presets needed either. No dramatic new info here, just stuff for any Levelizer newbies and sharing the fun ;)

Don't forget to listen. The Levelizer is very transparent as a brick-wall limiter up to around 5 dB of gain reduction, depending on program content of course. Above that the distortion becomes more noticeable in my experience.

Don't just set it and forget it. ;)

[Grekim]

Don't forget to listen. The Levelizer is very transparent as a brick-wall limiter up to around 5 dB of gain reduction, depending on program content of course. Above that the distortion becomes more noticeable in my experience.

Don't just set it and forget it. ;)

Good advice. I'm working under 5 dB of gain reduction through limiting on this project. And going with a K-14 avg loudness. It's absolutely transparent.

[Cary B. Cornett]

Good advice. I'm working under 5 dB of gain reduction through limiting on this project. And going with a K-14 avg loudness. It's absolutely transparent. I really like keeping track of my mix levels with K-14 metering. My only concern for your situation is whether the "loudness" result from using K-14 as a guide will satisfy your clients. Sadly, there is still way too much demand for "shred" mastering and average levels very close to 0 dBfs. Here's hoping your clients are more enlightened than that.

[Grekim]

I really like keeping track of my mix levels with K-14 metering. My only concern for your situation is whether the "loudness" result from using K-14 as a guide will satisfy your clients. Sadly, there is still way too much demand for "shred" mastering and average levels very close to 0 dBfs. Here's hoping your clients are more enlightened than that.

In most cases, yes K-14 is fine, but I'm mostly working with folk and acoustic rock. I mastered a punk demo once and that was fun:rolleyes:. I forget where I put the level, probably K-12, but they were happy so it must have been pretty darn loud. I did have a case where I had just mastered a folk album and one of the performers was concerned that when listening on her ipod that there was a significant level drop when her folky track was played between some Prince cuts. We did end up leaving the level where it was. I have sent projects off to several talented mastering engineers and there's nothing to stop the sound from going downhill once you reach a certain point. I think that point comes a lot sooner/lower threshold than people realize.

[Cary B. Cornett]

In most cases, yes K-14 is fine, but I'm mostly working with folk and acoustic rock. I would expect K-14 to work very well with those genres, especially with current expectations.

Working with some live musical-theater stuff was a real eye-opener for me. In that context, and with "traditional" choirs (not backed by rock rhythm sections), I learned to expect a dynamic range of about 30 dB on average. I still use K-14 metering on that stuff, but I expect average reading levels to be a ways below "0" reference in that context.

[Carl G.]

Don't forget to listen. The Levelizer is very transparent as a brick-wall limiter up to around 5 dB of gain reduction, depending on program content of course. Above that the distortion becomes more noticeable in my experience.

Don't just set it and forget it. ;)

5% is a HECK of a lot of gain reduction (starting from the top of the most of the material (instead of just transient peaks). That's excellent transparency!
I tend to use it more in the 1-2% mode at most just to allow more dynamics - so this is the cleanest transparent limiter I've ever known.
Now if you want "effect" limiting/compression - the world is full of those!

[Grekim]

5% is a HECK of a lot of gain reduction (starting from the top of the most of the material (instead of just transient peaks). That's excellent transparency!
I tend to use it more in the 1-2% mode at most just to allow more dynamics - so this is the cleanest transparent limiter I've ever known.
Now if you want "effect" limiting/compression - the world is full of those!


Carl you lost me. If you have a full scale input and reduce the level by 50%, that's 6.02 dB of gain reduction.

[Iain Westland]

if i read Carl right hes talking about the general body of music, you wont notice distortion on most of the peaks, but a 5% reduction in body mass is huge and liable to distortion, i think its also relative to how and what your recording.

i believe i left my coat over here

Iain

[Grekim]

if i read Carl right hes talking about the general body of music, you wont notice distortion on most of the peaks, but a 5% reduction in body mass is huge and liable to distortion, i think its also relative to how and what your recording.

i believe i left my coat over here

Iain

Okay. I've just never applied a limiter deeper than anything but the transients and don't think it would be a good idea at all. And 6-8 dB of reduction is about all I dare to do with other limiter plug-ins on percussive material.

[Dave Labrecque]

5% is a HECK of a lot of gain reduction (starting from the top of the most of the material (instead of just transient peaks). That's excellent transparency!
I tend to use it more in the 1-2% mode at most just to allow more dynamics - so this is the cleanest transparent limiter I've ever known.
Now if you want "effect" limiting/compression - the world is full of those!

5%? I said 5 dB. Which do you mean?

Also -- I'm talking about max gain reduction of a given signal, which means on the transients.

[Carl G.]

5***37;? I said 5 dB. Which do you mean?

Also -- I'm talking about max gain reduction of a given signal, which means on the transients.
Thanks Dave.... I kept pressing % when I meant dB.
.... and I'm talking about main body of material (as Ian pointed out).
I reference main body of audio simply because when using a compressor prior to the Levelizer, transients are whatever we want to make them in the compressor stage - dependent upon what attack time you like to use.

Actually, I view each piece of audio ais its own work of art.... and applying set numerical figures (important for reference as they are) to a broad spectrum of audio files seems counter productive to why we're in business :) (so I mean this whole discussion on my part simply as an observation I've noticed).

[Grekim]

Actually, I view each piece of audio ais its own work of art.... and applying set numerical figures (important for reference as they are) to a broad spectrum of audio files seems counter productive to why we're in business :) (so I mean this whole discussion on my part simply as an observation I've noticed).

Hey Carl, I agree, but the project was budgeted for 20 minutes of time per song. So although each track got its own EQ, I was comfortable with the compressors and limiter having the same setting for each song and simply varying the source input level.

[Dave Labrecque]

Thanks Dave.... I kept pressing % when I meant dB.
.... and I'm talking about main body of material (as Ian pointed out).
I reference main body of audio simply because when using a compressor prior to the Levelizer, transients are whatever we want to make them in the compressor stage - dependent upon what attack time you like to use.

Actually, I view each piece of audio ais its own work of art.... and applying set numerical figures (important for reference as they are) to a broad spectrum of audio files seems counter productive to why we're in business :) (so I mean this whole discussion on my part simply as an observation I've noticed).

Gotcha. Thanks. :)

[Carl G.]

Hey Carl, I agree, but the project was budgeted for 20 minutes of time per song. So although each track got its own EQ, I was comfortable with the compressors and limiter having the same setting for each song and simply varying the source input level.

I've done the same before on a huge project one time - used Hammer in S.F. batch mode. 2 minutes later - several hours of songs were 'mastered' :) :)
... (uh, its own 'work of art' you might say:eek:)
... but Levelizer sounds way better.