I'm trying for an effect like Audrey is singing in a cathedral hall, so to speak.

And I'm doing this in a squashed room...

Reverb?

I would have to rank EQ, Compression and Reverb in the following order:

1) EQ - to get all the instruments working together in a mix you need to make sure they are EQ'd so that each of the instruments frequency "sweet spots" are represented, and those freqencies that have to overlap may have to be attenuated to eliminate a build up.  EQ helps to lay out the instruments horizontally in the Frequency spectrum.  The EQ strategy forms the basis of the mix - it is the ground floor.  It is essential to get this right.

2) Compression - It is a tool to control volumes and transients - vertically along the "amplitude spectrum."   A transient shaper is a type of compressor.  Compressors can be used to shape the sounds and add color.  I've found I've totally overused and abused compression in the past.  Too fast of attack on a snare drum, for example, can kill the snare, which was one of my mistakes.  I used to think that every instrument needs a compression strategy (just like EQ) but many times compression is not needed at all, or just a very slight compression is needed.

3) Reverb - helps define the room or space that the mix and gives depth.  It is important for sure, but you have to have the EQ working properly before you add reverb to anything.  Otherwise you just create mud.  And as Danny said, you need to EQ your reverb too.

Well, I think that the 'record it right' thing really means get it as close as you can naturally, then apply compression and EQ on the way in so that it's very close to the way it should sound as recorded, and will require minimal adjustment in the mix. It doesn't mean you can't process it on the way in. You could modify it to mean 'print it as it should sound', if that makes more sense.

Though this concept is dying fast these days, because fewer and fewer bands can really play, less and less money to give them the time to get it right as recorded, mroe and more bands having to record in studios without much outboard gear to print it processed or people running them with the extensive experience to do it anyway probably if they did, massive DIYer influx almost none of whom have the experience or gear to do it.
And of course a huge thing in the DIYer crowd is that they are very much more just about putting out songs, not about mastering any craft. So if it requires ten X more time editing it than playing it, then whatever, just delete the bad parts, copy and paste in from somewhere else, throw a hundred plugins on it, massively automate it, and get it out there.

droddey


Well, I think that the 'record it right' thing really means get it as close as you can naturally, then apply compression and EQ on the way in so that it's very close to the way it should sound as recorded, and will require minimal adjustment in the mix. It doesn't mean you can't process it on the way in. You could modify it to mean 'print it as it should sound', if that makes more sense.

I'm not sure I'm getting this Dean..if every instrument is recorded and 'printed' as it should sound, what happens if there is frequency masking? You've still got to do the eq work in the mix, right?

I know in the past I've come up with what I think is a great guitar tone in isolation, but I just can't get it sit right with other instruments in the mix without a whole lot of carving.

Am I misunderstanding your meaning?

How does one unmask two frequencies if they coincide?

Danny:  Ah might as well blame me...it's ok, I got big shoulders for a lil guy. :) 


LOL!! I was sitting here listening to an arrangement when all of a sudden BOOF that left hook hit me. 


+1 to what others said; go write that book and don't forget to autograph it ;)




Matt,


Sounds like you have it under control. I'm sure everything will turn out fine ;) 


Interesting concept about the white noise, I guess that would add dirt to the snare; emphasizing the actual snare 'wires' to help it cut thru the mix. 


So I guess the consensus is; start with a dry mix, get it sounding best as possible and 'then' if need be do the finer tweaks and processing on the tracks, be subtle, oh and watch out for those guns they pack a mean punch! LOL!! . 


Good luck with the album, wish you great success :)








.

trimph1


A question, I have an idea of recording my wife doing a version of Gaudette ( as in a church ) .... if one was to use a reverb which reverb would be a good one for that?

Do you have Perfect Space, Barry? Some nice church & cathedral impulses in the presets.

trimph1


How does one unmask two frequencies if they coincide?

If two instruments are sharing the same frequency range, and you've identified where they clash, then it's a matter of carving a 'notch' in one to reveal the other (simplistically speaking).

jamesg1213


I'm not sure I'm getting this Dean..if every instrument is recorded and 'printed' as it should sound, what happens if there is frequency masking? You've still got to do the eq work in the mix, right?

I know in the past I've come up with what I think is a great guitar tone in isolation, but I just can't get it sit right with other instruments in the mix without a whole lot of carving.

Am I misunderstanding your meaning?

But sounding like it should sound means how it should ultimately sound in the end, not that it sounds good in isolation. You should know what you want and get it like that up front, as close as you can. It's an experience thing obviously, and something that you can only continue to work forever to improve presumably. But every time at bat should be used to learn where you got it wrong and do better next time. I mean, ultimately, there are a reasonably small number of combinations of how bass and guitars fit together, for instance. Decide up front what type of mix you want and try to make it so as printed. If you get it wrong and have to adjust, figure out why and try not to make that mistake next time.
 
Recording it like it should sound is really equivalent to 'mixing it as you record', you are mixing it effectively on the way in, not after the fact (again, to the degree you can realistically do so, which is WAY more than most folks even attempt to.) I think that the biggest single problem is that you always want to give each individual instrument more frequency space than it can ultimately own, often way, way more. It makes it sound nice and fat and nice. Trying to learn what the appropriate frequency restriction needs to be for a given instrument in a given type of arrangement is the trick. If you do, then you can get it that way up front, even though it sounds a little wierd out of context. 
And the other big benfefit is that, since each part is going down as it should sound and balanced correctly before moving on, for each new part you can hear it in the context of what is effectively a pretty good mix. And this isn't a radical idea, given that it's exactly what an engineer would do if recording a real band, except that he can do it all once, not one track at a time. He's not sitting there making every instrument sound optimal on its own, he's making it sound good as a whole, then it's recorded.
 
I'm not saying any of this because I've mastered the art, not remotely. But I have figured out enough to know that this should be the goal to be working towards mastering, if you want to be a good recording engineer and I think that anyone recording themselves should be WAY more interested in that than in learning how to do heroic retroactive correction in the mix.
 

Ok, I get what you're saying, and I can see how working towards that goal would be useful.

It becomes much more difficult to do that though, if you're working with collaborative partners, as a lot of us are these days, and if you're using instruments outside the familiar guitar, bass & drums format. For example, in any given mix I can be working with mic-ed acoustic and mandolin , electric guitar recorded through pre-amp/amp sim, sampled percussion, loops, mic-ed ethnic percussion, plus electric bass, synths, real world samples..and the kitchen sink.

I realize I'm just talking about my own particular scenario, but I really don't *know* what kind of a mix I want until I have all the instruments tracked and laid out, therefore I don't know what any instrument *should* sound like in the end, until I start to mix.

I can see that the approach you're describing would work if I used all the raw tracks as scratch tracks/guides then re-tracked them as necessary, but that somehow seems to defeat the object. A mix to me is always like a blurry picture, slowly coming into focus.