Mixing with the attack principle

This is a neat little technique for drawing the ear's attention to things in the mix that you're having trouble finding the right level for - hardly mind blowing, but interesting nonetheless:
 

.

Yes, it is a curious phenomenon of audio perception. We've all had this experience: a part isn't standing out enough, so we give it a little boost and then we're happy. Sometimes the boost only needs to be 1db to make us happy. But ask yourself if a casual listener would ever notice a 1db difference. The answer is no. So why did it satisfy our own ears? Because we were actively listening for the difference, and focused on the part. To achieve the same effect for the casual listener, you have to dramatically force their focus onto it.
 
Another example. I had a minor tambourine part in a song. A friend listened to the track and commented the tambourine was too loud. As soon as he said that, I was aware that it was indeed to loud and too piercing. From then on I could not listen to the song without being annoyed by that damn tambourine. But previously, I'd listened to the song a hundred times and had actually forgotten that it had a tambourine in it at all! 
 
It's just how our auditory sense works. It's how we can hold a conversation in a noisy room. Or how, when hunting for supper, our distant ancestors could pick out the footsteps of their prey among the other sounds of the forest.
 
It's also the same mechanism that keeps us from going insane from the constant barrage of extraneous sounds in our environment. I live a mile from the freeway, but I usually don't hear it. At least, not until I think about it - then suddenly it's very loud and very annoying. So I try not to think about it. Oh, dang it, too late - now I'll be hearing it all day!

Buzz buzz buzz ...
 
I've said this before but people need to have their blood pressure checked too.  Tinnitus (oops, said it!) can be aggravated by that condition.  Not hearing is bad enough - no heart beat is a bit more immediate of a problem.
 
@

Cool method, I can see where this would be useful at times.  Thanks for sharing.

Forgive me for posting here, but can anyone tell me where to find the button with which to start a new thread? I know I'm ADD and all, but this is always seems to be a great mystery on every forum I join.
 
Thanks,
John

It has to be said that a lot of these types of things should optimally be built into the composition and performance. It's not like players are limited to a single volume. If the composition is well worked out, then given instruments should naturally come up between vocal phrases or trade off with other instruments just naturally as played. And it's a lot easier and more natural (more like music than data processing) than doing lots of automation curves.
 

Whenever I learn about a useful technique like this, I always want to go back and remix everything I've ever done.
 

If you are running a compressor with a slow attack you are already doing it.....No Mike this is not the same thing at all. A compressor even with a very slow attack may allow a sound to rise up in level in its attack portion for only a relatively small time. What we are talking about here does involve the word attack but in this context we are talking about several seconds of a part being louder before bringing it back down. ie a much longer attack. 'Attack' is not a good word to use in this situation. More like the start of any given part.
 
In fact for this technique to be effective it would have to involve the start of the part being louder for at least a few seconds. The sort of attack Mike is talking about would not even be noticeable in this situation. Effective yes for increasing the short term attack of a sound (initial rise time) for sure but not for this technique I don't think.
 
Besides the video does not mention a compressor anywhere it talks about editing automation, a very different thing. What if you did not want the attack portion of a sound to be louder at all or remain completely unaltered but just wanted the first few seconds to be louder in general. The compressor won't do it in that situation. But editing automation will. It is not the sort of effect that can be created by slapping a compressor over and expect to solve your problems. It is something that needs to be done by hand. And I can imagine that there would be no set time to return a part back to its normal level either. That would need to be evaluated for each given situation and adjusted accordingly to be most effective. In one case it might only be 2 seconds but in another it might be less or much longer.

You're both right, but the OP is describing longer timeframes than we'd normally set a compressor's attack time to. A slow attack setting on a compressor might mean 100-200ms. To draw focus to a musical passage we'd usually be talking 10 or 20 times that.

You guys will all kill me for what I'm about to say, but it's just my personal opinion so don't be too hard on me. :)
 
I wouldn't waste a second doing something like this. The reasons being? If you don't know the part is there, neither will anyone else. Seriously...we could put a ton of work into this little thing only to have it fall on deaf ears. Does it make enough of a difference texture wise to even waste the time doing the attack thing? In my personal opinion, definitely not. Reason 2...
 
The sound is jacked. If you want a sound to stand out more (even as a textured instrument that is a backer...not something that will be focal) you choose the right sound for starters. Next you eq the sound to fit the mix. Next, you compress it just right and you don't even have to worry about it and it will be audible as a texture. I could barely hear that sound at all with the attack method. Not because it was too low....but because it was not eq'd correctly for that mix.
 
It was too dark and was distorted in a way that went against the music...it didn't compliment it. You don't add in a driven guitar that you can't hear that is also the wrong sound choice for the song. You have to be especially careful in pop/commercial music. If you use a driven guitar for "a little bit of buzz" it has to be the right buzz. I'm talking almost analog synth type buzz and it has to be low in the mix. In my opinion, this particular tone in the example was a blanket. It had no personality in terms of complimenting.
 
Don't get me wrong, I'm not saying you need to add in a razor sharp metal tone...but think along the lines of "Danger Zone"...or something Steve Lukather may use in a controlled, pop atmosphere.
 
This is also where a sound that sizzles can compliment IF you choose the right sizzle frequency. The worst choice (in my opinion) that one can make is to use a dark sounding guitar as a support sound in a song like this. Warm up your REAL guitars all you want in the song that play throughout. But when you need that little extra support on a chorus section or a bridge...by all means you will have a little more impact when you have a bit more sizzle in the tone, you compress it and then keep it at a level in the mix that enhances...not a level that makes it lash out "I am a driven guitar, I'm here to rip the flesh off your bones!"
 
So though the principal to this whole thing makes sense and it can be a tool to use in certain situations *possibly*, it's not something I use nor would it be be something I would teach my students. One thng I'm noticing these days is...people continuously try to cut corners in this field. There is no substitute for doing something right. Somewhat off topic, but a quick example....
 
Side chaining. To me, if you know how to eq and compress your instruments, it's rare that you'd need to use something like this. How many years have we had incredible recordings without it? Seriously. There's always a trade-off...and in a good mix, nothing needs to really play second fiddle and it shouldn't if it's mixed correctly. The most over-use of this is people that side chain kicks and bass guitars usually because they:
 
1. Have not been fortunate making a kick and bass work as a team
2. Are not sure how to use compression, which compressor to use or how to go about it
3. Have major problems with frequency masking due to:
 
a) not knowing how to deal with it
b) don't know what to listen for
c) no sub or not having their monitors and room tuned to know what is too much/too little
 
Granted, some engineers use SC as a tool and know how to do the things I've mentioned. But trust me, there is a huge percentage that are totally clueless that need to learn how to crawl before they start to sprint. The only thing I *may* use sidechaining on is a guitar solo where the lead and rhythm guitars will share the same compressor. The "trade-off" is not as noticable as a kick and bass and the rhythm guitars NEVER drop out of the mix totally. The end result is...when the lead solo is done, the rhythm guitars return to where they were before the solo started which doesn't leave you with that gap/drop out in sound when an instrument starts or stops playing. But that's a conversation for another time.
 
This attack thing to me is an attempt at a short cut to make something sort of fit in a mix that doesn't fit at all. If you didn't know it was there...you wouldn't know it was there. If you want an instrument to be there, learn how to eq and compress it along with a little automation or a bit of verb to make it less up front if need be when you need something to be more of a texture. To just envelope a transient does nothing in my book. Keep telling yourself...if you didn't know that sound was there like he mentioned....none of it would matter. It didn't add a texture to me...it added a blanket. Ok, you may want that sometime in your mix....me personally, that wouldn't be the mix where I'd want something like that. But hey, whatever works for you. This is just my take. :)
 
-Danny

Good points, and I definitely agree about side chaining the bass with the kick. I remember back when I first started and watching a tutorial about compression on Lynda.com where the guy was talking about using this technique to give the kick space, and my immediate thought as a totally clueless newbie was "why in the hell would you want to do that to your bass line"? 

sharke
Good points, and I definitely agree about side chaining the bass with the kick. I remember back when I first started and watching a tutorial about compression on Lynda.com where the guy was talking about using this technique to give the kick space, and my immediate thought as a totally clueless newbie was "why in the hell would you want to do that to your bass line"? 

Yeah I'm sure we're not alone in feeling that way. I gotta say this though....for dance music/club music, it's becoming acceptable to have that trade-off. As long as that kick drum is going boom boom boom boomboomboom boom boom with enough force to make you think there's a bass instrument of sorts...you win. LOL! :)
 
For we that do music where a bass needs to have a bit more character other than a low note that resonates someone's bile ducts, you need a totally different approach as you know. :) I think these techniques are cool...but it should be mandatory that a person isn't allowed to view them until they learn how to do things the right way FIRST. Hahaha! :)
 
-Danny

I didn't say it was necessarily bad. I just said that it seems to that often younger folks who came into it post-2000 don't understand (or appreciate) the value of fewer, bigger parts, and how much easier it can be to create good compositions with far less folding, spindling, and mutilating involved if you don't try to pack too much stuff into it. Obviously, sometimes, any technique might be useful to create a particular effect, like pumping the mix with a side chain or something like that.

bile ducts?
LOL......