image description


normalize audio??
  Forum Index » Studio One General Discussion 
Author
Message
Kahlbert
Presonificator

Joined: 27/11/2009 14:17:13
Messages: 4778
Location: Germany
Offline

Klypeman wrote:
Kahlbert wrote:

Yeah, of course you can't tell the difference, because turning the fader up is exactly the same, mathematical-wise, as normalizing. That's why I cunningly talked about any processing.

But sure, you're right, sometimes people tend to forget that anything you do in a DAW is some sort of processing and involves calculation - except changing track colors maybe ...


Sooo... We just stop mixing ?? Like right now ehh... ???
To prevent normalizing I mean... Oh God I suck at Math...

Sure. Just get all levels right where they belong while recording, and you can leave all faders at unity while mixing.



Just kidding, of course, don't worry ...
[WWW]
LMike
Baconator
[Avatar]

Joined: 19/02/2011 12:02:34
Messages: 8233
Offline

Like the Real Traps guy said when he kinda went "Yoda" on all that with a simple bit of subjective wisdom....

"If you can't hear it, it doesn't matter."

But .. (and there's always a "but" ) ... things like that usually have an unspoken baseline logical assumption built into them by default, to avoid unnecessary debate about the most clearly obvious logical exceptions, like it assuming that you actually are already in a position to be hearing what you should be hearing to begin with.
Studio One Tips For Newbs

"So.... Bitwig. I wonder what's queued up next in line on the drama list?"
hibidy
Presonoid

Joined: 01/02/2011 09:13:29
Messages: 2320
Offline

So we're saying that though the result should be a tasty meal, there is more than one way to fry a fish?

@home studio guys obsessing: Nah, get out! That never happens! What's sad in my case is that I still suck at mixing.
New computer:

2.6.2/win7x64 ultimate/SSD os and library/i7 4770/16GB ram

Focusrite saffire 14/Novation impulse 49/tons of VST goodies

Klypeman
Presonic
[Avatar]

Joined: 03/02/2011 13:47:00
Messages: 383
Location: Oslo, Norway
Offline

Kahlbert wrote:
Klypeman wrote:
Kahlbert wrote:

Yeah, of course you can't tell the difference, because turning the fader up is exactly the same, mathematical-wise, as normalizing. That's why I cunningly talked about any processing.

But sure, you're right, sometimes people tend to forget that anything you do in a DAW is some sort of processing and involves calculation - except changing track colors maybe ...


Sooo... We just stop mixing ?? Like right now ehh... ???
To prevent normalizing I mean... Oh God I suck at Math...

Sure. Just get all levels right where they belong while recording, and you can leave all faders at unity while mixing.



Just kidding, of course, don't worry ...

@Kahlbert
I Really like Your comments, they allways make me happy and sometimes Im just not funny when I try to be....


Im so fortunate that I have recorded and mixed some of Norways finest in Rock, Jazz, African Style and Underground PunkRock in the latest years and allways make quick decisions of where to do what and if something needs to be normalized other ways than via fader i do so with a keen ear in small portions, just as if I was mixing on my desk.
I think that our ears are the main tool to adjust sound in the mix and is using the K-14 System thingy (Bob Katz) to get where i want. I kind of get it.
Just not good at explanations. Let's say it works for me.
I try to stick to the EBU128 suggestions when it comes to LU, if possible in the mastering stage of things.
In the end there sometimes needs to be some small adjustments to get the CD's levels more aligned and then it's so.
I try to make things simple.
My 0.02 on this.


jemusic and others has better and longer explanations of the K-14 System than I.
Vocalpoint
Presonic

Joined: 17/10/2011 16:49:28
Messages: 192
Location: Calgary, AB
Offline

jemusic wrote:Tracks are often routed to buses and it is also very easy to get the correct K ref level on a buss too. The Masterbuss never really has to be moved from unity gain either. Once your buses are all sitting at the correct level then it is amazing how summing the buses to the Masterbuss seems to create the perfect mix and the final level on your Masterbuss is also sittiong at the K ref level.


A well balanced (and properly gain staged) layout should require little to no tweaking. I have never moved the master fader from unity in any DAW I have ever worked in and have never had to deal with "overs" or a hot mix either.

For my stuff - tracked via proper gain stage - I can practically move all my faders to unity and the thing practically mixes itself.

VP

This message was edited 2 times. Last update was at 15/05/2014 18:54:28

[WWW]
Vahevahe
Presonic
[Avatar]

Joined: 16/12/2011 09:46:31
Messages: 226
Location: Los Angeles
Offline



The normalizing feature should be removed from all daws
Studio One V2.6.2
Mountain Lion 10.9.2
Mac Pro 2.4 ghz dual quad core
16 gigs of ram
Apogee Symphony i/o
Lynx Aurora 16
Avedis Audio MA5 preamps
[WWW]
matthewgorman
Presonoid
[Avatar]

Joined: 24/02/2011 21:30:31
Messages: 3219
Location: Somewhere in the swamps of Jersey.
Offline

A well balanced (and properly gain staged) layout should require little to no tweaking. I have never moved the master fader from unity in any DAW I have ever worked in and have never had to deal with "overs" or a hot mix either.

For my stuff - tracked via proper gain stage - I can practically move all my faders to unity and the thing practically mixes itself.

VP


Honestly, between this, and subtractive eq, they do mix themselves. But there may be an additional reason for that. Folks that are paying careful attention to gain staging like this, are probably also taking more time when miking everything. The mix really starts as soon as the band sets their gear up. When you start with getting good tones from the equipment, tuned drums, etc, and add proper gain staging, the songs mix themselves. And when the songs mix themselves, there is very little to do in mastering. But it ALL starts on the front end, and not too many people pay attention to that.
Matt

Lenovo Thinkpad E520, Windows 7 64bt, 8 GB RAM, Intel i5 Processor
StarTech EC13942 34mm Expresscard
http://www.newegg.com/Product/Product.aspx?Item=N82E16839158010&Tpk=startech%20EC13942
S1Pro V2 (Special Dog Balls Edition), Melodyne Editor, Nomad Factory Studio Bundle, Waves Renaissance Bundle, Firestudio Tube, Faderport, Monitor Station, HP4 Headphone Amp Yamaha HS50 Monitors.
Shure Mics (57's, 58's, and a 1953 Unidyne), Various AKG, Various MXL Ribbon Mics.

1974 P-Bass, 1990 Jazz, 1985 Guild B302, Ampeg SVT with 4x10x15 cabinet

http://soundcloud.com/stars_apart
jemusic
Presonic
[Avatar]

Joined: 16/12/2010 14:14:20
Messages: 286
Location: Melbourne, Australia
Offline

Matt is very correct on this, it does all start right at the front end. K System goes a long way towards correct gain staging. Setting correct VU rms levels on tracks during tracking makes all the difference. It is so easy to do. I have real VU meters and they can be set to show levels coming in very easily. But even so Studio One allows you to monitor incoming levels so easily. All you do is insert the meter plugin on an input channel and set it for K system at the correct reference. (The meters on the inputs are not showing 0 dB VU for a given ref level. That is why you need a meter plug in on an input channel. However use the peak metering of the built in input meter combined with the rms meter.) I still find a VU VST such as the Klanghelm meter is very good because it looks and behaves very much like a real VU meter.

The VU's don't show fast transients but that is OK. Just make sure those sources are being monitored with a peak meter. The idea is to work with both forms of metering incoming. RMS and Peak. Once all the transient sounds eg drums are sent to a buss, normal VU rms metering works fine because by the time all the transient sounds add up there is enough rms energy to move a VU meter normally. One reason for this is the slight differences in timing between all the transients. They only have to be 150 ms apart and they will move a VU meter half way up the scale.

Virtual instruments can produce wildly varying levels and many of them need to be tamed and set correctly. Before you do any bouncing or converting VST's to audio, run the track in real time, solo it and check that the output is producing a nice level which just peaks your VU up to 0dB VU in the loudest sections. Some virtual instruments will need to be turned up, others way down etc..

Loops also can vary a lot in level. Many are either right on or low or way too loud. I fine tune the Event gain parameter handle works well. Grab the gain handle and just ease it either up or down to suit. You can cut a loop into sections and fine tune each part level wise. This is all better than using a compressor anywhere.

On mixdown once all track levels are tracked correctly it is easier to mix. In fact you can even partly mix using a quality hardware VU meter by eye. Engineers have done it before. You bring in the drums so they are just pushing -4 or so on the master. Then bring the bass in and the level goes up to about -3, -2 or so and by the time you bring the other elements in you are just hitting 0 dB VU on your master and the mix will be perfect. Master fader is never moved etc..

VU metering was all done before and it can be done now with modern DAW's. The problem is they dropped all the VU rms metering and just went peak only which is a bit dumb. Peak metering tells you nothing about rms levels that exist below the peaks. People are way too obsessed with keeping peak levels constant but have no idea about the wild varying rms levels underneath. (that is why they find it hard to mix because a peak meter might be showing the correct level but the rms level could be way too low hence making that track hard to get to the right level in a mix. One tends to reach for a compressor at this time and compress and use makeup gain. But why not track the rms part of the sound louder and use no dynamics at all)

K System is all about keeping your rms levels constant instead, right through the production. Just like in the old days, we did not worry about peaks too much because the built in headroom of the analog systems (eg mixers and tape machines) just took care of it. You can do the same in a modern DAW. The K System headroom takes care of peaks. You will never have a single clip light or over this way.

I like working at K -14 and one main reason is my Yamaha digital mixer is calibrated at that level. But K -20 is also a very cool level to work at. It only takes a few seconds to recalibrate the VU meters and away you go at -20. K -20 sounds very transient and just beautiful. There is no need to be pushing high levels in a digital medium, especially at 24 bit. The digital recording medium noise level is now way down at -144 dB or even -138 dB real world levels. So while you are even down at K -20 you still have a signal to noise ratio of 118 dB!

This message was edited 5 times. Last update was at 16/05/2014 05:09:24

[WWW]
jemusic
Presonic
[Avatar]

Joined: 16/12/2010 14:14:20
Messages: 286
Location: Melbourne, Australia
Offline

These -14 or -20 levels also only refer to pre mastered levels. It is good to do a dynamic range measurement on the pre mastered mix and also a LU FS reading. I find that LU FS readings are very close to K system reference levels.

Mastered levels can be higher in relation to a K System ref level. Such as aiming for average rms levels os say K-10 will still produce a dynamic range reading close to that. There is a sweet spot in mastering in between loudness and transient punchy sounds. Finding that 4 dB of extra rma gain is easy over the three mastering phases.

Normalising audio can be ineffective if there are very high transient levels throughout the waveform. I like opening tracks in an editor such as Adobe Audition or similar. What can work is taming very loud transients down to a lower level (limiting) than you can normalise the whole track but to a value lower than 0 dB FS such as -3 dB FS (peaks only to there) You will still end up with a much louder rms level overall and with leading edge transient peaks still healthy in comparison to rms levels for snap quality. These pre mix edits should be saved too under a different name so you can revert back to the original raw waveform if you need to.

[WWW]
 
Forum Index » Studio One General Discussion
Go to: