Izotope Ozone Loudness Maximizer

I am using izotope to master some songs and when I use the Loudness Maximizer and really get the song loud at times, you can hear the compression really working and taking place, How can I boost the volume louder and not hear the compressor working?

The mixes are already really good, so I just want to essentially boost the volume and not hear any outside compression or "breathing".

I usually set the threshold HIGH and the Margin LOW, and I use Intelligent mode with "Smooth" Character. I do not change anything on the right panel.

Am I going about this the wrong way? what are some good settings for commercial quality loudness? thanks!

Use the "Fast and Loud?" (it's something to that effect) setting on the Limiter instead of the "Smooth" one. Leave the ceiling alone at -0.3dB and lower the Threshold until you get 6dB of reduction (max) in the loudest sections of the song.


HTH

ill try that out, thanks!

Make sure you're not overcompressing, rocker887. Turn on wave preview on the master bus and look at your output. My experience with Ozone has been that using the very settings you describe, I can totally squash a mix beyond all reason. And I almost never hear "breathing", even at very aggressive settings with a hot kick.

Follow Jose's advice about tweaking the threshold until you see only an occasional limiting of more than 6db. Go a little higher than that if you're after a sort of Green Day wall of continuous sound. But if you're after a cleaner metal sound, a la Dream Theater, stay under the 6db target.

If you find that the levels change drastically through the song, such that Ozone shows very little compression in one part and huge compression in another, then you might consider preceding Ozone with another compressor. But I'd avoid that if it's not really necessary. After all, you want most of your distortion to be intentional.

ORIGINAL: Jose7822
I think it's the best limiter out there (even better than the Waves stuff) IMHO.

Hehe. The Waves limiters are no competition for the Ozone limiter (or PSP Xenon, Oxford Limiter, Voxengo Elephant etc).

One comment though, I personally never like to hit a limiter with more than a couple of dBs of limiting. The trick to get things loud without too many artefacts is to get a little bit of the loudness from each processing step. A little bit from compression (1 or 2 compressors), a little bit from EQ, a little bit from harmonic distortion (analogue gear or stuff like the Oxford Inflator) a little bit from ADC clipping and a little bit from a limiter (or even two) etc...

The other thing to consider is that each mix has a "loudness potential". Some mixes can be turned up really loud without too many negative artefacts. Other mixes just can not be turned up loud without falling to bits. If your are mastering your own stuff, very often the solution is to go back to the mix and fix things there. Or... not make things as loud! :)

UnderTow

ORIGINAL: Jose7822

Undertow,

I've read about ADC clipping before in other forums and have always found it awkward (but interesting) since we're always trying to avoid clipping in all stages of the process. Could you please elaborate on when and how this process takes place? I'd like to know how you go about this. Thanks in advanced!

Well, as all things in trying to get ridiculous loud levels, it is a compromise. On some material, clipping converters (or clipping in the DAW) can give a cleaner result than limiting. The idea is that an extremely short burst of distortion (clipping) is undetectable. Even more so, this distortion can replace the energy lost by clipping those short transients.

The results depend on the converters and the material. Clipping a soundblaster won't give very good results. Clipping the converters on a Lavry Gold can be extremely transparent. More transparent than limiting in some cases. This is probably because the analogue stages of these converters are very clean and have plenty of headroom and when/if they saturate they do it in a pleasant way but most of the clipping is purely the digital stages.

In the same vain, clipping a violin solo will not sound good but if your mix has extremely short transients from, let's say the snare drum, that take up a bit of headroom, clipping those transients will probably go unnoticed. More so than limiting (which always has an attack and release time). In a sense it can be seen has brick wall limiting with no attack or release.

Most people don't have these kind of converters so instead you can clip digitally. I like to use Voxengo Elephant in clip mode and 4X oversampling. This means that allot of the clipping distortion (which is full bandwidth) gets filtered out when the plugin converts back down to the base sample rate. (Another way of looking at it is that the clipped waveform gets smoothed out by the anti-aliasing filters. Mathematically this is the same thing).

In the end, you have to try it out and see if it fits the material or not.

Back to Ozone: The limiter in Ozone gives you a compromise. You can set the release times so short that it is or nearly is clipping. With character set to something like 0.5, you are nearly clipping but not quite. Turn on the inter-sample peak protection and intelligent mode and your are also making sure that cheap converters won't clip on inter-sample peaks. (Many limiters do not take this into account).

Is this clear? I'm not entirely sober.

Edit: I just forgot one thing: This is usually the last or before last step in the chain. Often ME's will clip slightly their ADC's and then add maybe another dB of limiting in the box. Some won't limit after the clipping. (I usually do as it catches slightly different parts of the audio).

One other thing, you never want to clip the signal anywhere else in the music making process because clips during recording or mixing or whatever will be processed further down the line and you can not judge how that will sound (and the clipping might not stay on the peaks of the waveform but might end up anywhere in the waveform as the phase of the signal gets affected by the processing. This usually becomes much more audible). If you clip when mastering and can't hear it, you know it doesn't matter as that is the final product. There will be no more processing. (MP3 encoding, radio processing etc not withstanding).

UnderTow

ORIGINAL: Jose7822

OK, let's see if I got this right. Basically you're saying that ME's clip OTB by routing the signal into high quality converters and then back to digital where they might apply Limiting (similar to what you do when you use a summing Bus), is this correct?

Indeed.

I use a Fireface 400 which converters are obviously not in the same league as Lavry's Gold, so I guess it would be better for me to do this digitally with Ozone.

They are not Lavry Gold's but RME converters are not bad. Try clipping them to see how it sounds on your material. It might work. :)

I usually enable those same settings you're talking about in Ozone's Limiter (for the same reasons you've described) but I'll experiment with pushing the Limiter a bit more next time. Thanks a bunch!

You are welcome. :)

Take care!

You too Jose. :)

P.S. Yeah, you're always drinking aren't you?

Actually I rarely drink at all! But thess last two weeks have been beautiful and I've been indulging much more than usual. :) I've spent the afternoon and evening on a terrace with mates. Very enjoyable. :) Also, drinking doesn't help me make any music so I'm left with time over to spend on the net. :)

Just one (redundant) reminder to everyone reading this: These kind of techniques are ways to achieve competitive loud levels in the this silly loudness war. You can always aim for lower levels. It can only do the music more justice!

UnderTow

ORIGINAL: Jose7822


There's nothing wrong with that as long as you do it consciously, specially if you get to spend more time around here (your knowledge is always appreciated ). Glad to hear you're having fun relaxing with your mates, that's good.

Have a great weekend my friend!

Are you saying that me drinking is a public service? Maybe I can tax deduct the beers! Lol.

Thanks for the kind words.

UnderTow

Try using the multiband as a subtle but effective parallel compressor to the whole mix - and then limiting with whatever (elephant is amazing). You get all of the bright & punch, with none of the squash - and it's limiter ready. I read this in Bob Katz's book - easy tip, and it works.

Actually, he goes into some detail re: the advantages of using mc's - but with the proviso, as you note, that a crappy mix is a crappy mix, basically. I find that with kick heavy, "dirty" tracks (house music), a multi is a cleaner way to tidy up the track, before limiting. The cool part was how to use upward compression and upward expansion - together. No mud. THAT made a big difference, at least for me.


BTW - "Mastering Audio - the art and the science, 2nd edition" was one of the best music purchases I've made in a while. It's also a great read.

ORIGINAL: UnderTow

ORIGINAL: vespesian

Try using the multiband as a subtle but effective parallel compressor to the whole mix - and then limiting with whatever (elephant is amazing). You get all of the bright & punch, with none of the squash - and it's limiter ready. I read this in Bob Katz's book - easy tip, and it works.

But don't forget what Bob writes on Multi-band compression: "I largely use multiband compression (and expansion) to fix bad mixes that could not be remixed, for one key to a great master is to start with a great mix!".

When mastering your own mixes you should usually be able to fix things in the mix if need be.

His opinion might have changed in the second edition of the book. I still need to get it so I don't know. (I really should as he asked if he could quote me in the 2nd edition! )

UnderTow

Undertow - I really appreciate your contributions on this subject. It is a complex and interesting subject since we're talking about mastering tools, yet there doesn't seem to be too many actual mastering engineers on the forum. And you seem to have a lot of experience in this area.

I'm getting what I feel are very musical results from Ozone3. My material is heavier kindof like the dynamics of Nickleback. However, I can't get my mix to sound as loud as the Nickleback CD's. I can get them to sound as loud as some good material from the early 90's though, so I'm pretty satisfied. But It would be cool to get it louder without squashing the heck out of it.

I'm using the intelligent mode with character at 4.0. My material before going into Ozone probably averages -8 to -6 with peaks from snare sometimes going over -6. My threshold on the maximizer is -8.5, margin -0.3. I wasn't using the prevent inter-sample clips. If I go below -8.5 with the threshold the result starts getting unmusical. My input and output are both at 0. Can you see anything obviously wrong with these settings?

Thanks

ORIGINAL: UnderTow

ORIGINAL: guitartrek

Undertow - I really appreciate your contributions on this subject. It is a complex and interesting subject since we're talking about mastering tools, yet there doesn't seem to be too many actual mastering engineers on the forum. And you seem to have a lot of experience in this area.

Thanks. :)

I'm using the intelligent mode with character at 4.0. My material before going into Ozone probably averages -8 to -6 with peaks from snare sometimes going over -6. My threshold on the maximizer is -8.5, margin -0.3. I wasn't using the prevent inter-sample clips. If I go below -8.5 with the threshold the result starts getting unmusical. My input and output are both at 0. Can you see anything obviously wrong with these settings?

It is impossible to say without having the material. There really is no "mastering by numbers". The only thing I can suggest is to try lowering the character setting.

Are you using anything else besides limiting? A typical approach is to have a couple of compressors before the limiter. One that has a very slow attack (150 ms) and medium release, low ratio (1.3:1) and quite a low threshold. This compressor might catch something like a dB and maybe a bit more on the loudest parts of the song. The second compressor would have a bit faster attack but not actually fast. (Maybe 50 ms or something like that depending on what the material needs) a faster release and a slightly higher ratio (maybe 2:1 or something like that) that catches a bit more.

If the compressors are pumping on the low-end, try using a side-chain compressor and filter out some of the low-end in the side-chain. (70-120Hz depending on material).

Something else you might want to check out is adding harmonic distortion. I've already mentioned Sonnox Inflator. Voxengo VariSaturator is also worth checking out.

But it all really depends on what the mix is doing. It is every step from tracking to mastering that will determine the final possible loudness.

UnderTow
(All numbers in this post are very vague indications and should not be taken as absolutes).

Thank you.

Those settings look to be right in the ballpark. Might want to consider bumping the margin up to -.2 or -.3 though, just to be on the safe side. The difference in level is almost nothing.