Question about compression

[Blor007]

Hi,

I got a little question about compression:


I think i'm missing the point of compression, i'm kinda experimenting with it and everyone says it brings the sound level kinda more ''even'' and it pumps the sound up.
But when i use it i notice that the attack milliseconds are kinda working like a gate. What I mean is when i compress a snare like 10ms attack 150ms release 4:0 ratio, I get huge peaks of 10ms (where no compression is applied yet) and the rest is just so small that the peaks take lots of ''dB space'' in my mix but to let them really sound I really have to turn the rest of the mix down , leaving me with huge peaks.

Am I overcompressing, am I missing the point here ?

 

[SouthSIDE Glen]

You have the attack time set too slow to capture the initial transient of the snare. Try setting it as fast as it will go, and then dial it progressively slower and listen to how the different attack times change the sound and the wave.

If youre just beginning to learn compression, you may want to set your attack and release on auto (if your compressor has that) and concentrate first a bit on the threshold and ratio control pairing first. A 4:1 ratio - BTW, there is no 4:0 - in and of itself is meaningless without taking the threshold into account as well; 4:1 at a threshold of -20 can have a completly different effect and sound than 4:1 @ -12, for example.

I'm putting the finishing touches on "Compression Uncompressed", a free 42 page introductry hypertext treatise on compression and compressors. Keep your eyes open for more on this in the next few days.

G.

 

[JeffLancaster]

I'm putting the finishing touches on "Compression Uncompressed", a free 42 page introductry hypertext treatise on compression and compressors. Keep your eyes open for more on this in the next few days.

Hey Glen, I just ran across this thread while doing a search on "compression". Thing is, while I understand the basic principals of compression and use it in my projects, I would really like to get a more in-depth understanding. I was looking for sources to do some reading and this sounds great. Have you finished it yet?

 

[FALKEN]

anywayz, to the OP,

it sounds like you have a grasp on what's going on. as the other posters pointed out, if you want to grab that initial spike, you need a faster attack time...many compressors might not be fast enough to do this. the benefit of what you are doing now is to make the snare "cut through" more...i.e., you get the transient without all of the clunk. some people like the clunk though and want to get rid of the transient...you're going to need a different approach for that.

 

[SouthSIDE Glen]

Hey Glen, I just ran across this thread while doing a search on "compression". Thing is, while I understand the basic principals of compression and use it in my projects, I would really like to get a more in-depth understanding. I was looking for sources to do some reading and this sounds great. Have you finished it yet?

I'm || <- this close to done. I just have another couple of pages left that I want to give a final proofread to and then have an associate of mine beta test the whole thing on his browsers just to make sure I have the zip file set up correctly and that I didn't miss any major bugs in the DHTML. Barring any major problems with the beta (which I don't expect, I'm using mostly tried and true code that I've used before) I hope on releasing it via an announcement either in this forum or in the Recording Techniques forum either Monday night (5/22) or Tuesday (5/23) (US CDT).

I'm not sure what you mean by in-depth. This notebook is fairly beefy at about 42 pages and 35K words, but it's in-depth on a fairly basic level, if that makes any sense. Don't expect a whole lot of really advanced stuff here; it's mostly meant as an in-depth explanation and introdution to compression and basic technique, not as an advanced treatise or exposition of black belt tricks for professionals by any stretch of the imagination.

G.

 

[JeffLancaster]

I'm || <- this close to done. I just have another couple of pages left that I want to give a final proofread to and then have an associate of mine beta test the whole thing on his browsers just to make sure I have the zip file set up correctly and that I didn't miss any major bugs in the DHTML. Barring any major problems with the beta (which I don't expect, I'm using mostly tried and true code that I've used before) I hope on releasing it via an announcement either in this forum or in the Recording Techniques forum either Monday night (5/22) or Tuesday (5/23) (US CDT).

I'm not sure what you mean by in-depth. This notebook is fairly beefy at about 42 pages and 35K words, but it's in-depth on a fairly basic level, if that makes any sense. Don't expect a whole lot of really advanced stuff here; it's mostly meant as an in-depth explanation and introdution to compression and basic technique, not as an advanced treatise or exposition of black belt tricks for professionals by any stretch of the imagination.

G.

Cool - i am really looking forward to it. I've done some looking around and noticed there's a real lack of really good information available on the subject right now. It's easy to find the basic stuff, like explanations of terms like ratio and threshold and stuff like that, but rarely does anyone go much beyond that. I have several books on recording, etc. and most of them talk about how to select your gear, setup your studio, place your mics, etc, but they always seem to cut the subject of compression a little short. For example, there was a thread on this forum recently where a guy was asking why it sounded like his compression ratio settings were affecting his high frequency more than his lows - it would be helpful to have a more solid understanding of the principals behind those sorts of things.

 

[SouthSIDE Glen]

It's easy to find the basic stuff, like explanations of terms like ratio and threshold and stuff like that, but rarely does anyone go much beyond that.
...
For example, there was a thread on this forum recently where a guy was asking why it sounded like his compression ratio settings were affecting his high frequency more than his lows - it would be helpful to have a more solid understanding of the principals behind those sorts of things.

Yeah, the reason I decided to write this was becasue of the propensity of first-year-compression questions that pop up on these forums and the apparent lack of really good comprensive articles available on the Interent. There are a few decent partial articles here and there, but nothing that I was happy with referringto someone looking to really cover the basics of properly USING a compressor.

My "notebook" does also explain all the controls, since it does start at the very beginning. The good news is you can always skip that chapter if you wish It does touch on the subject of frequency vs. compression at a couple of points in the text, including it's relation to signal chain ordering and compressing for mastering, not to mention a full chapter on (*shudder*) multi-band compression.

I'm sure there will be a whole lot of questions that I don't directly address in the text, not to mention a few in which my answers may not foster universal agreement (you can't please everybody .) That's what followup questions and arguments in the forum can be for .

G.

 

[DrJones]

Yeah, the reason I decided to write this was becasue of the propensity of first-year-compression questions that pop up on these forums and the apparent lack of really good comprensive articles available on the Interent. There are a few decent partial articles here and there, but nothing that I was happy with referringto someone looking to really cover the basics of properly USING a compressor.

My "notebook" does also explain all the controls, since it does start at the very beginning. The good news is you can always skip that chapter if you wish It does touch on the subject of frequency vs. compression at a couple of points in the text, including it's relation to signal chain ordering and compressing for mastering, not to mention a full chapter on (*shudder*) multi-band compression.

I'm sure there will be a whole lot of questions that I don't directly address in the text, not to mention a few in which my answers may not foster universal agreement (you can't please everybody .) That's what followup questions and arguments in the forum can be for .

G.

I am looking forward to dowloading this file. I have some of the same questions. You are right about finding good tutorials online. Thanks

 

[Blor007]

My biggest problem is that I get a result that sounds good , but when I check it out visually it eats a HUGE amount of dB in the peaks

So I end up limiting those peaks, thus eliminating the attack ; guess thats all the wrong way

Looking forward to your tutorial Glen , nice initiative

 

[SouthSIDE Glen]

My biggest problem is that I get a result that sounds good , but when I check it out visually it eats a HUGE amount of dB in the peaks

So I end up limiting those peaks, thus eliminating the attack ; guess thats all the wrong way

Looking forward to your tutorial Glen , nice initiative

Regarding that and your original post, Blor, the attack is basically the response time of the compressor, how fast it responds to signal peaks. When you have the attack set to 10ms, that means that transients faster/shorter than 10ms will "get past" the compressor before it clamps down. If you need to get those faster transients, then decrease the attack time/increase the attack speed.

Alternately, if the excesive peaks are caused by the summing of two smaller peaks from different tracks, you might try knocking down one of the individual peaks in mixing before you mix it all together.

And there's more than one way to knock down peaks, whether in mixing or mastering. Often times it can be more transparent to manually knock down a small handful pf peaks with manual or automated volume control in your editor in the digital domain or via manual control of a dB pad button or fader on your mixer if you're working in the analog domain.

If you're getting a result that sounds good, and limiting the peaks and removing those dynamics makes it sound worse, then don't worry about it and leave those peaks there . Volume at the cost of sound quality is no bargain. One can always turn up the volume on playback, but there is no "Quality" knob that can be turned up. And if it sounds bad, no amount of volume is going to make it sound better.

G.