When do you decide, "this track needs compression"?

[Mike Tholen]

Walters?

[Fredrik]

In tracking it´s not necessary @24bit, atleast thats my opinion. For mixing I would say any time its hard or impossible to find a good level in the mix for the sound, when it seems like it´s both to soft and to loud, reducing the dynamics will give the sound a more even level, hence it will be easier to find a perfect level for it in the mix.

I think it´s enough when it seems to correspond with the other sounds in the mix, or when the problem of finding the level is gone. Compare with the bypass switch to decide if the tone or the vibe of the performance is lost, if it is it´s to much or perhaps the wrong compressor for the job.

[3rd world order]

Originally Posted by Big 3rd

When do you know something needs compression at the tracking and at the mixing stage? (Talking strictly midi production)

when it has too much dynamic range... generally i go for about "half" of the setting i think will work in the mix because i can always compress more later... sometimes i'm feeling less cautious and go for what i think will work... sometimes i'll print a few different versions and pick in the mix... sometimes i'll use the different versions in different places in the song.

IDEALLY you want to get as close to the source as possible so you might wanna start with the envelope section of your MIDI device...

And, how do you know when it's a been enough compression?

when it has just enough or you reach a limitation of your gear.

you need to be much more specific if you want better responses.

[TonyBelmont]

I feel during tracking, vocals are needing compression more so than anything. Other than that, I use distressors, etc.. to get more flavor out of different instruments...

[beatzz]

Originally Posted by TonyBelmont

I feel during tracking, vocals are needing compression more so than anything. Other than that, I use distressors, etc.. to get more flavor out of different instruments...

Hey Tony, im curious......

If you are tracking vocals in 24 bit, why even bother compressing on the way in? Doesn't that just limit your options to what you can do afterwards?

[TonyBelmont]

Originally Posted by beatzz

Hey Tony, im curious......

If you are tracking vocals in 24 bit, why even bother compressing on the way in? Doesn't that just limit your options to what you can do afterwards?

No, not if you know what you're doing anyway...

It doesn't matter if you're recording at 24bit, you're limited by the headroom of your converter. Essentially proper use of compression on the way in allows the converter to see a hotter signal, that way you can take advantage of all those precious bits...

[Fredrik]

The dynamicrange at 24bit is 144db! My opinion is that it´s safe to use the last bit´s
for headroom. I would agree that compression could be used as an aesthetic
choise for the song or perhaps on highly dynamic stuff like percussion or drums.

When playing back a signal printed at 0dbfs it´s so hot you are at risk of cliping the inputs on the console, attenuating the levels in software will only result in loosing those bits anyway.

Isn´t it true also that most converters sound worse when they are driven hard, and that recomended peaklevels is -10dbfs?

[beatzz]

Originally Posted by TonyBelmont

No, not if you know what you're doing anyway...

It doesn't matter if you're recording at 24bit, you're limited by the headroom of your converter. Essentially proper use of compression on the way in allows the converter to see a hotter signal, that way you can take advantage of all those precious bits...

Sorry man i just don't buy that.

Most modern converters are at -100 db (give or take)...........i don't think their noise floor is an issue.

[TonyBelmont]

Originally Posted by Fredrik

The dynamicrange at 24bit is 144db! My opinion is that it´s safe to use the last bit´s
for headroom. I would agree that compression could be used as an aesthetic
choise for the song or perhaps on highly dynamic stuff like percussion or drums.

OK.... show me a converter that has that same dynamic range and I'll stop compressing on the way in.

[TonyBelmont]

Originally Posted by beatzz

Sorry man i just don't buy that.

Most modern converters are at -100 db (give or take)...........i don't think their noise floor is an issue.

Actually most modern converters have better specs than that... But, either way you guys are talking about dynamic range of software and signal to noise ratios. You aren't even taking into consideration the headroom of the analog circuitry of the converter.

Record the way you want, but I'm happy doing things the way I am right now... is it for you? guess not....

I also use EQ when I track vocals, too....

[Fredrik]

Since 1 bit = 6db a full 24bit signal = 144db I guess.

Perhaps I´m missing something?

[TonyBelmont]

Originally Posted by Fredrik

Since 1 bit = 6db a full 24bit signal = 144db I guess.

Perhaps I´m missing something?

144 db is a theoretical max.

You're not taking into account the analog circuitry which is part of every A/D converter, etc... Look at something like an Apogee AD8000 it's dynamic range is only 112db... get the idea?

[Fredrik]

I get your point. But there are converters with better specs than the ad8000 in terms of dynamicrange.

[Disco D]

i compress on the way in, in, and usually on the way out too

[Disco D]

actually you know what compressor is surprisingly good? the internal compressor on the mpc 1000.

[superburtm]

compressing during tracking has alot to do with the singer. Most singers I come across need the help to smooth out their performance dynamically. And I usually track with a touch of eq as well.

[blayz2002]

Originally Posted by Big 3rd

When do you know something needs compression at the tracking and at the mixing stage? (Talking strictly midi production)
And, how do you know when it's a been enough compression?

It's down to instinct and experience. Once you start to get to know your compresors and what effect they have on different signal types, and also when your after certain types of effect..i.e....I know I want a LA2A when I need more transparency and maybe 1176 when I want fatness or pumping or Waves Ren comp when I want colour (warmth/distortion)

I usually compress Midi bass a bit a least on the way in, and maybe keyboards and pads and leads... to thicken em up a bit (i've got a TLA FAT MAN Fat 1, not high end but adds nice valve harmonics to bass and stuff, whilst smoothing it out) I cant really achieve the same in the box.

Like the other guys said when somethin won't sit in the mix at your prefered volume, but the flavours right, that's when I reach for the compressor..or sometimes could just be a eq issue.

In Logic also I do use the dyanamic parameters sometimes to dampen stuff down depending, this would be on the way in aslo

[3rd world order]

24 bit and compressing vocals? funny... I compress vocals because it sounds better that way, vocals have to be GLUED these days.

i don't care about the bits.. no bits.. whatever.

you compress to achieve a certain sound.. or to push up the breath.. or so every thrid word isn't drowned out.. or to give a vocal a bit of GRAB..

bits NEVER enter my mind. should they?

[Igotsoul4u]

i don't care about the bits.. no bits.. whatever.

bits NEVER enter my mind. should they? [/QUOTE]


I agree. I have never thought of bit rate and compression as some sort of yin and yang. Since when does bit rate determine if compression is necessary? It sounds like some sort of scientific explanation or something. I personally use compression because it makes things sound better. That's the reason I compress to tape. Am I concerned with my input level? Yeah of course I am, but I am not going to let it control my sound. It's just a very unmusical approach in my opinion.

"In tracking it´s not necessary @24bit"

What a weird thing to say!!!

[Fredrik]

Originally Posted by Igotsoul4u

What a weird thing to say!!!

My point is, if you track in 16 bit and have a material with lots of peaks, you might have to record at such a low level not to have overs. The low level wont use all 16 bits available, hence protecting overs with compression makes for a higher resolution recording. The same goes for tape but there it´s more a fact of masking noise levels of the medium itself.

In 24 bit you can afford to lose 1-2 bits so what, that was my point.

Ofcourse aplying compression can be an aesthetic choise aswell, but the decision can be left till later it isn´t 100% necesary to do it in tracking.

[bloodstone]

I used to compress a lot when tracking and mixing. I mainly do rock, pop, and roots-oriented stuff. I've gotten away from compressing when tracking. I've started using compression at mix only when a signal isn't "behaving" itself and quick fader moves can't solve the problem. I also use compression when a sound doesn't have enough presence in the mix.

Bass guitar, kick, acoustic guitar & voice almost always get compressed at mix in my world.

I guess you could say I'm a recovering compression addict. I like a light touch on the 2-buss, and sometimes will create sub groups and compress the drum sub.

My ultimate rule of thumb: if it makes it sound better to your ears or the artists ears, compress. If it doesn't don't. "Whether 'tis nobler..."