what does the attack shape of most compressors look like??
-
- KVRian
- 943 posts since 15 Mar, 2005
seeing as you rarely have control, im assuming its more generic than release shapes. so is it linear? 12-3 (clock face) exponential curve? or 9-12 curve?
and are there any comps that have a definable delay before the attack kiks in, or comps with control over attack shape?
finally, why is the release shape often implied to be more critical to a compressors character?
easy,
and are there any comps that have a definable delay before the attack kiks in, or comps with control over attack shape?
finally, why is the release shape often implied to be more critical to a compressors character?
easy,
Last edited by martian on Wed Jul 27, 2005 8:56 pm, edited 1 time in total.
-
- KVRAF
- 12235 posts since 18 Aug, 2003
I can't answer the other questions, but I'd guess that at least in part it has to do with release generally being longer, gives you more time to hear its effects.wm wrote:finally, why is the release shape often implied to be more critical to a compressors character?
-
- KVRian
- Topic Starter
- 943 posts since 15 Mar, 2005
yeh in normal situations attack is shorter, but for the effect im working on at the moment the attack has to be much longer than the release.shamann wrote:I can't answer the other questions, but I'd guess that at least in part it has to do with release generally being longer, gives you more time to hear its effects.
-
- KVRist
- 415 posts since 8 Dec, 2003 from NYC (born & raised in Philly)
most (if not all that i know of) employ an exponential attack, meaning it approaches full compression mostly in the first part of the attack phase, then more slowly towards the end.
if you were to visualize it on an axis, with time as X and envelope level as Y, it would look like the "9-12" arc of the clock face.
== chunk
if you were to visualize it on an axis, with time as X and envelope level as Y, it would look like the "9-12" arc of the clock face.
== chunk
- KVRAF
- 11381 posts since 3 Feb, 2003 from Finland, Espoo
Attack behaviour of a compressor is extremely important and is one of reason why some bad compressors are simply bad and unmusical (behringer cheapo comps, dbx 266, alesis 3236 or whatever that model was). Of course these can be abused to sound good in certain situations but that's not the point with my rant.
Good examples of pleasant attack 'shapes' are LA-2a, 1176 and 1178, tubetec cl1b and the non-opto one, l2b? dunno the models now.. another excellent one for most tasks is the distressor. An example of a high-end compressor I didn't much care for the attack 'shape' is the Millenia TLC-2 compressor which made is useless for doing serious mangling of tracks but then again it was designed with gentle mastering in mind.
A 'bad' attack shape to me is one that handles the incoming transient clumsily, letting trough a whole 'block' of audio and then sudenly squashing everything after this block without a gentle transition slope to compression. It's hard to explain but a nice compressor for vocals/drums/percussion/bass is usually one that goes "twwwaaaack!" or "flap!" instead of "CLACK!" or "Kack!".
Damn, this must read like a madmans ramblings.. oh well.
This is one of the biggest issues with digital compressors compared to analogue ones that can achieve extremely fast but still smooth attacks and one of the reasons that you see still the big guys at gearslutz and mastering forums complain that nobody has yet done digital compression right (some do swear by their weiss compressors but I found that unit very tame and boring..). As a sintefex owner I fully agree with all the statements. It does sound imho much better than most plugins but it still can't touch good analogue outboard, not even close.
Disclaimer: I talk about "huge" differences and such in my posts but note that to me a huge difference might be for some people just hard to hear so never mind my ramblings!
- bManic
Good examples of pleasant attack 'shapes' are LA-2a, 1176 and 1178, tubetec cl1b and the non-opto one, l2b? dunno the models now.. another excellent one for most tasks is the distressor. An example of a high-end compressor I didn't much care for the attack 'shape' is the Millenia TLC-2 compressor which made is useless for doing serious mangling of tracks but then again it was designed with gentle mastering in mind.
A 'bad' attack shape to me is one that handles the incoming transient clumsily, letting trough a whole 'block' of audio and then sudenly squashing everything after this block without a gentle transition slope to compression. It's hard to explain but a nice compressor for vocals/drums/percussion/bass is usually one that goes "twwwaaaack!" or "flap!" instead of "CLACK!" or "Kack!".
This is one of the biggest issues with digital compressors compared to analogue ones that can achieve extremely fast but still smooth attacks and one of the reasons that you see still the big guys at gearslutz and mastering forums complain that nobody has yet done digital compression right (some do swear by their weiss compressors but I found that unit very tame and boring..). As a sintefex owner I fully agree with all the statements. It does sound imho much better than most plugins but it still can't touch good analogue outboard, not even close.
Disclaimer: I talk about "huge" differences and such in my posts but note that to me a huge difference might be for some people just hard to hear so never mind my ramblings!
- bManic
-
- KVRAF
- 2028 posts since 18 Mar, 2004 from New York, N.Y.
The attack posture of the Northeastern White-Toothed Compressor gets its ditinctive shape by arching it's back, pointing its tail, and generally looking somewhat menacing like this:

Sorry, I couldn't help myself...

Sorry, I couldn't help myself...
- KVRAF
- 2750 posts since 2 Feb, 2005 from Raincoast of Grayland
Thanks, jplanet, this shows why those analog dogs can sound better, but unless kept on a short lease, we DAW cats don't like them hanging around our bedroom studios.
perception: the stuff reality is made of.
-
- KVRist
- 415 posts since 8 Dec, 2003 from NYC (born & raised in Philly)
one step at a time. and shhhhhhh.....bmanic wrote:Speaking of which.. citizen chunk, where is your compressor?? You know, the one after vanilla?
- bManic
== chunk
-
- KVRAF
- 10597 posts since 13 Jun, 2004 from Alberto Balsam
i dont know at all for real, but to me it usually sounds like the release is exponential and the attack is the opposite (forget what thats called). Heres a quick drawing i made to explain what i mean:


-
- KVRian
- Topic Starter
- 943 posts since 15 Mar, 2005
great picture.
so anyone want to confirm, is that correct?
you see, i really need a comp that does the opposite. An attack which starts slow and clamps down towards the end, then a release that does the same. i'm using a compressor to act on the off beat of drums
A further bonus would be for it to 'look ahead' and ignore gaps (by not activating the comp) between transients under a certain specified time, this would stop the comp acting on drums rolls, and complex fills.
ive started building something in SE but pretty stuck as i cant make anything beyond a simple compressor of the type i dont want
can anyone put me out of my misery or help build it?
-
- KVRist
- 262 posts since 8 Jun, 2005
Hmm, I suggest you to use a sidechain - with this you can let any compressor act how and when you want it...wm wrote:![]()
great picture.
so anyone want to confirm, is that correct?
you see, i really need a comp that does the opposite. An attack which starts slow and clamps down towards the end, then a release that does the same. i'm using a compressor to act on the off beat of drumsi also need a 'delay' parameter to delay the start of the attack after a trasient has activated the comp.
A further bonus would be for it to 'look ahead' and ignore gaps (by not activating the comp) between transients under a certain specified time, this would stop the comp acting on drums rolls, and complex fills.
ive started building something in SE but pretty stuck as i cant make anything beyond a simple compressor of the type i dont want
can anyone put me out of my misery or help build it?
I often do this on drum loops to give them a specific groove. Just insert a compressor on the drum loop track and feed it's sidechain with a different signal (which needs not to be heard in the mix), e.g. a simple synth or whatever. You then can 'play' the pumping compressor groove in evrey matter you dreamed of...
-
- KVRian
- Topic Starter
- 943 posts since 15 Mar, 2005
yes, would make sense wouldnt it.
but i'm using sx and was trying to achieve something a lot less long winded and more automatic than setting up the various work around side chain methods.
but sure, thats going to give the flexiblity.
but i'm using sx and was trying to achieve something a lot less long winded and more automatic than setting up the various work around side chain methods.
but sure, thats going to give the flexiblity.
- KVRist
- 231 posts since 20 Jul, 2002 from Connecticut
Oh man...I'm sorry to pop in here with not much to say, but I just wanted to say that that picture really made me laugh...Chase wrote:i dont know at all for real, but to me it usually sounds like the release is exponential and the attack is the opposite (forget what thats called). Heres a quick drawing i made to explain what i mean:
As far as the attack shape goes, I guess there are different ways to get such a shape - logarithmic, upwards shift of a negative exponential, etc. Or you could just call the attack concave down and the release concave up.
The sesame seeds on the hamburger bun...genius...
[url=http://www.dangries.com]
