Normalization...... and Reverb... and other things

[eeoo]

Also, I teach guitar and my number one priority is to never cause my students to get down on themselves, ever. And it's really not at all hard to do, way harder to go out of my way to make someone feel bad and then take their money. I don't get that.

[mojobone]

If you do everything in solo, you always optimize the sounds for things in solo. EDIT I mean to say: Instead of in solo, make your eq compression and ambience decisions in the full mix with everything playing - this can make the decisions matter a lot more. We don't get to listen to albums with tracks in solo. We listen to the full mix.

HTH

Andy hit the nail on the head. Especially with the the advice above. I preach this every time someone asks me about a mix they have done. It can't be overstated. No one listens to tracks in solo....except us cave dwellers. The other piece of advice I got from a mentor of mine was to get the sound in the tracking stage to be as close to the finished product as you possibly can. The more time you spend tracking, the less time it will take to mix. In the world of plug ins, we tend to rely on them to "fix" issues later. I still fall into this trap now and again, but not nearly as much as I did when I started out, and mixes sound much better.

Bravo. I'm a firm believer in getting the sound at the source; it's what's in front of the mic that's most important.

[burpo]

Even experts disagree. Education does not ensure consensus. Consensus does not equate to certain truth.

I'd better quit there. I'm sounding pretty snooty.

[mojobone]

Education ought to ensure accuracy; I don't think you'll find much agreement among experts that 'normalization is the opposite of compression'. Jes sayin'.

[burpo]

Not suggesting otherwise, just making a broader point.

I'm dippy that way.

(I guess I gots politics on m'mind.)

[TimWalter]

Thank you all so much for your comments and viewpoints. Sorry for my late response, havent been on here in a while.. too long of an interval. As Russell said in his interview, there is no reason not to log in daily, like most people do for Facebook. LOL.

I am printing off this discussion thread and digging into it deeply. Thanks again all so much.
Tim

[Kolstad]

When I hear the track I thought that the sax, which is an acoustic instrument, was blended in between electric instruments like bass and gtrs, and that the mix could be more balanced (depending on your vision of the role of the sax in the track). It's hard for an acoustic instrument to cut through between those high powered electrics.

So when I think of his comments from that perspective, I too wanted to hear the sax better relative to the gtrs (especially from a sax players perspective). And I'd say it's true that you might not want to compress the sax track too much here, because then you just get 3 in-your-face tracks trying to compete with each other.

The issue is then how to get to hear the sax better, and basically get a proper balance between your instruments in the mix.

I can understand the idea that normalizing the sax track is about making the waveform and volume of the sax bigger and louder without compressing it, but then there's an issue that you are loosing headroom, in stead of gaining it (as you would by compressing the track.

But there's another way to achieve that balance, and that would be to instead turn the louder elements in the mix down, and match the levels, and then either normalize (I know some prefer to do this, despite it's not being best mix practice, as you say) or compress the whole mix together (to "glue" it, and perhaps make some headroom for mastering/ limiting the mix up to levels).

So, I think your guy is right in what he wants to hear, but perhaps does not have the mix experience to elaborate proper advice for you, like Andy can (which is a much more elaborate way to proper mixing of the track).

Some good stuff here, too, but the succinct way of putting it is your guitars are too loud; my teacher would have said it sounds like a guitar player mixed it. In a live ensemble, when someone takes a solo, the other players adjust their volume to help highlight what's being featured in the arrangement. If this didn't happen on the floor, because...overdubs or whatnot, you can use a ducker, a special kind of compressor that's keyed by the sax track to turn the guitars bus down by a predetermined amount whenever the sax energy crosses a predetermined threshold, OR you could just nudge down those faders when the sax comes in, then print your fader move.

Hehe, yeah gtrs too loud, that's about it Mojo Good point

Anyway, I don't know if normalizing the sax actually would hurt anything in this example. It may be a proper tool to raise the volume of the sax track, without fiddling with the dynamic range, as you would with compression. Normalizing the Sax track may turn out to be a good starting point to balance the mix, fwiw, despite it is not exactly the opposite of compression. Sometimes hanging with the grammar police, may lead to miss a perfectly fine solution.

[TimWalter]

Guys
Not to belabor a point..I did get a return on this this morning, as I expected after the excellent and on point discussions on this thread. The screener pointed out the parts weren't in the same room, so that fits with suggestions on here (and its a good sign to me).

with that said, if anyone can suggest helpful tips on how to use glue reverb? I have the section on Mike Seniors book Section 16.3 on Mixing WIth Reverb (pages 235-242).. Reverb for Blend, but I still find myself wanting/needing more help. In this section he discourages using plate or spring reverbs for blend purposes, so i have been trying to use UAD Realverb (which came free with my Apollo). The settings on this GUI dont have Predelay setting, instead they have a number of other attributes that are a bit foreign to me. I know there are other methods, like using delay with no feedback with setting at 125 to 165 ms( those are values I have heard suggested). Also, some folks suggest having multiple delays and reverbs set up (3 delays and 3 reverbs, each varying in degrees of predelay, with front to back positioning being close, medium and far), but that seems to me to be counterintuitive. If I want all my tracks to seem to be in the same room, shouldnt I use the same effect (s) (perhaps to varying degrees) on all of them? Just trying to grope my way through.

I guess my questions would be:
1) what reverbs do you more experienced guys use for Blending,
2) if anyone has experience with UAD reverbs (more sepcifically Realverb) can you suggest any tutorial videos or other resources? I found one on youtube, but it was in spanish, and it didnt help me at all. Or have I chosen the wrong reverb plugin for this purpose (UAD or otherwise?)

Any suggestions very much appreciated.

Thanks in advance to any of you who spend your valuable time answering.

Thanks and regards.
Tim

[andygabrys]

I guess my questions would be:
1) what reverbs do you more experienced guys use for Blending,

This is all relative. For years people have been using real chambers which made their reverbs.

e.g. - reflective concrete chambers, stairways, bathrooms, leading to Plate reverbs etc.

Basically the signal is broadcast into this space via a loudspeaker and the result is picked up with a microphone.
Sampled / convolution reverbs end up with much the same result.
Digital reverbs often try to approximate the same thing.

Plate reverbs tend to have a pretty dense effect with a lot of high frequency content - often they are used for vocals more than anything
Spring reverbs even more so - and unless its a guitar or organ, are typically used for vocals where you want some retro character (except for the new AKG BX 20 model that UAD came out with which is a very "smooth" sounding Spring reverb if I read correctly - whatever, it sounds pretty cool).

Convolution reverbs of Halls, Rooms, Cathedrals etc. tend to be fairly neutral sounding for a lot of blending uses.
Digital / algorithmic reverbs with Room, Hall, Chamber, or Cathedral settings like the Lexicon 224 or 480 or PCM96 or Bricasti (and the digital models of all these hardware units) can be used for similar uses

So a plate reverb might still work well for blending, if you tailor it (maybe eq radically) to keep the parts you like, and get rid of the parts that are either clouding up the low end or midrange, or causing too much splash up high.

2) if anyone has experience with UAD reverbs (more sepcifically Realverb) can you suggest any tutorial videos or other resources? I found one on youtube, but it was in spanish, and it didnt help me at all. Or have I chosen the wrong reverb plugin for this purpose (UAD or otherwise?)

Its not the wrong reverb, but it might take some time to first learn the controls (read the manual man!) and figure out what those relate to in Mike Seniors specific book language, and more importantly what is a good texture to use for which instrument.

Look at the interface. There are a number of things that equate to other reverbs.

Shape - more comers in the room will lead to a more neutral space
Material - harder material should generally have a colder, brighter, more reflective sound - and increased thickness should effect in the same way

Size - an office sound vs. a gymnasium

Resonance - basically an eq to tighten up various regions.
Timing - length of decay - two materials so two timing graphs

Levels in and out - if the red lights go on, turn down the input.

Positioning - Mix - generally if you are using reverb as a send to save CPU and blend a bunch of instruments together you will set this reverb to 100% wet.

Distance - effectively the same as pre-delay. Pre-delay refers to the amount of time after a specific transient that the reverb starts. It can simulate the sound of banging a drum in the center of a big room far away from the walls if the pre-delay time is long. Or if the pre-delay is short or nonexistent this can simulate banging a drum right next to a wall - you hear the reverb immediately.

In orchestral work - generally pre-delay is longest for the strings, and then gets shorter for the woodwinds and brass, and then is the shortest for the percussion section. Theoretically this should mean the strings are still ambient but have more definition, and the percussion can really be drowned in the verb to make it sound big, and clarity of the percussion is less important (or at least it is what it is if you are trying to replicate an actual orchestral seating arrangement).

in pop mixing - pre delay is used not necessarily to give distance in space, as it is to delay the onset of reverb and make things clear and audible. Like when using a plate reverb, a tape slap delay would often be used to delay the input signal into the plate. So the singer says "superstition" and 125 ms (30 IPS) or 250 ms (15 IPS) later the signal would hit the reverb. Its an effect. It might work.

Now this reverb has some great sounds in it (and I use it too) but it takes some work to find out what the best uses are, or some tweaking.

regarding your other questions:

I know there are other methods, like using delay with no feedback with setting at 125 to 165 ms( those are values I have heard suggested). Also, some folks suggest having multiple delays and reverbs set up (3 delays and 3 reverbs, each varying in degrees of predelay, with front to back positioning being close, medium and far), but that seems to me to be counterintuitive. If I want all my tracks to seem to be in the same room, shouldnt I use the same effect (s) (perhaps to varying degrees) on all of them? Just trying to grope my way through.

for blend, sure try one nice medium room reverb first. Add other reverbs as you need to get front and back depth. One reverb applied at low levels can make things sound like they are part of the same band in the same room just as Mike Senior says.

But beyond just making them sound less discrete (make them blended) you need to also consider a front to back sound stage. Stuff that is closer might have subliminal level short reverbs, stuff that is very distant might have less high end, and more long big reverb.

Delays, I usually am using tempo synced delays because they make things sound big without the delays being as audible - like 1/4, 1/8, 1/16 note lengths.

the no repeat delay trick does this: it makes you feel like you are in a room, a space. Imagine singing in a bathroom that is 8' x 20' long and all tiled. There is going to be a "sound" of that space. Using delay makes you hear or feel the room. It can make things stand out a bit (like vocals - put it at a subliminal level for more modern styles, put it at a loud level to make it sound like Lenny Kravitz's album Mama Said - the vintage retro thing).

This can be an asset because after you have blended everything, the mix might feel a little too much the same - so now how do you give things a little separation. Same as a live studio session with the whole band - in years past you didn't need to blend, you had so much room mic sound and bleed that you were looking for ways to make things clear and stand out - compression and eq. Now as is said in the book, with stuff close mic'd and overdubbed, you need to find ways to make stuff feel like the same band to start.

over to you!

[andygabrys]

Ok further on the above:

maybe its time to check out some demos of other reverbs to see what you are missing.

just on UAD you have the following:

EMT 140 (Plate)
EMT 250 (digital reverb)
Lexicon 224 (digital reverb)
UAD Precision reflection engine (algorithmic / digital)
AKG BX20 (big spring)
RealVerb
DreamVerb
Ocean Way Studios - models the rooms at Ocean way. very cool and incredibly DSP hungry

so you can try all those for free for 14 days.

Then there are COUNTLESS other reverbs out there, some that are less (like the Valhalla reverb series which are very inexpensive and well regarded) and some that are more (the NI series that include models of the Lexicon 224 and 480, and the Softube TSAR-1 and the Lexicon series of native reverbs)

plus all the ones that came in your DAW (you are on studio one these days right? - if you have the pro version you have a few different flavors of reverb for sure).

I would start with the UAD stuff you can demo, and the stuff in Studio One.