Voxengo needs a freeware compressor.

[bduffy]

Yeah, brok, I just don't see anything compelling in what you're saying; people don't record at -0db on every track, nor do you need to in order to use limiting...

[brok landers]

?? kingston, i'll not start a discussion about what you are up to here, but i strongly suggest you to read what i stated.
of course you dont clip when lowering the fader, you won't clip even up to +6db on 32 bit hosts ... but in the end it's going to be on a cd.
i won't argue all the stuff now, but it seems that you completeley missed my point ...

btw, it's one thing to state your opinion here, but be careful with simply wrong facts.
every analog gear has units, and all these units have their limits. when overdriven they slightly sound different ... ALLWAYS in analog, no matter how good the equipment is. i said that it's subltle, but these subtle things happen _everywhere_ on analog gear, also on highend, or how you said, "linear" gear, and they're greatly summed up at the end.
i don't want to come like an ass, but i do my job since close to 20 years, i worked a long time with tapes, analog consoles, and i did a lot of mixes and mastering by the time on absoluteley highend gear and on the cheapest crap.
what i say is true.
and about rms:
have a snare at 0db.
compress it, while leaving the attack fairlyopen, to get the snap.
you cannot raise the signal as it is allready on 0db.
now, what you did is you left the attack pass the compressor unaffected. it still hits the same level as it did before 0db.
but the rest of the time-based signal, that comes after the attack, the body, is quieter now. the audible "feeling" of level is quieter now.
thats rms ... the average loudness is quieter.
to get the same rms that you had on the snare before compressing, you have to raise the signal again, but not exeeding the peak level, which allready was on 0db.
when raising, the signal is indicated as clipping in the channel.
i know it doesnt clip yet (32bit actually leaves headroom) but why should i raise it up over the clip sign?
just to know that at the end i have to lower the masterfader in order to get a balanced mix again??
and loosing my reference point for mixing?
and loosing resolution on my masterfader?
if you mix carefully/correctly, you end up with the masterfader around 0db, no clipping.
thats what the faders are for.
sorry, kingston, my english maybe suffers, but not my knowledge.
what i say is absolutely correct.

[jens]

and in the digital domaim, because of the strict 0db border, the only way to achieve the same loudness in the end, a limiter after the compression stage in an individual channel is absoluteley necessary.

there is no 0bd border on the track-stage of the mixing engine - only at the master-stage...

[brok landers]

jens, yes, i know that.
but at the end you have the summed signal which mustn't exeed 0db. thats the whole point.

[Kingston]

if you mix carefully/correctly, you end up with the masterfader around 0db, no clipping.

and the point of this is? Why can't I use the master fader? There's 32bits of headroom available. Say if I clipped the stage of even +20db? There's still more than 130dB of usable headroom left in the 32bits!

Let's not bring the analog headroom/distortion talk to this. When I say linear, I mean most equipment will work with practically no distortion on the whole of their (large) dynamic range (THD below 0.01% most of the time). This means that you will *not* distort them in any practical situation, only when abused, which is rarely done.

So what if you compress a snare and the overall RMS goes down! stick a limiter in the masterbus and/or bring down the fader of everything else. You have more than 150dB of headroom to play with which means you will need to worry about the overshoots *less* than on the analog side.

Who said you have to mix with every track set to 0dB all the time, with no peaking? It's not possible to clip the 32bit headroom provided you take the advantage of it, and don't always mix right at the 0dB point.

jens, yes, i know that.
but at the end you have the summed signal which mustn't exeed 0db. thats the whole point.

This is just an arbitrary treshold. Do not fear the master fader. It has more 150dB of playroom just for this.

[brok landers]

Yeah, brok, I just don't see anything compelling in what you're saying; people don't record at -0db on every track, nor do you need to in order to use limiting...

no, of course not.
what i meant is, that the loudest signal cannot pass 0db in the summary stage.
let's say you have a drum hit, which has a lot of snappyness. it hits 0db.
the snappyness is the transient. if the signal is compared to the _same_ signal without or less snapyness, the signal with less snapyness sounds louder at 0db. the average level along the timeline is louder, the rms level.
the snapiness actually comes from the fact that everything _after_ the initial transient is actually less loud. like if you shape a signal with a fast envelope-decay.the signal afterwards has to be lower (the sustain, if you will), otherwise the decay is not affecting the signal at all. a cheap comparsion, but it works.
on an analog mixing console you just don't care too much, as you simply raise the gain of the channel. as there is no digital cutting of the transient, but a way more natural, nonlinear shaping, this mostly sounds good (depending on the signal, of course), and you just push the signal into that "driving behaviour".
this results in a louder sound, as the snappyness gets saturated, cos it is the part of the signal which first hits the driving stage.
you cannot do this in digital without a saturation stage at the output of the compressor.
if you do so (just raising the fader), you just keep the transient in its relation to the rest of the signal, until, when rising too much, it either digitally clips at the masterstage, or you lower the masterfader with that ammount, resulting in a less loud mix, remember, it is the loudest signal in the mix.

[Lagrange]

In this case I'd rather pay.. I like to throw my financial support behind companies like voxengo...

[jens]

http://mdsp.smartelectronix.com/classic ... ressor.php
is a top end compressor, try it ! Beats most commercial ones actually

yes - not very flexible though - but very light on the cpu as well...

[Kingston]

brok, you're confusing mixing and mastering big time. when you do the final mixdown, you can safely leave it more than 10dB below the digital zero and export, because you have more than 150dB of headroom. worry about the RMS in mastering.

need the saturation as well? well add one. You can safely assume a good saturation plugin at 96khz in mastering will sound better than the average allen&heath or similar when distorted. If you really *want* the crappy clipped overdriven transistor saturation of the mixer summing bus, go ahead. Oh and remember, your average or even expensive mixer won't have more than 100db (on a good day) of headroom. below is noise, and above the cheap saturated transistors. Nothing glorious about that.

Thanks, but no thanks, I'll keep my mixes safely in the digital realm where I have the choice of type of distortion, and far superior headroom.

Loudness has nothing to do with this. It is but an arbitrary threshold in the digital realm. Yes it's easy to see the 0dB point we're always aiming at, but it's far easier to forget the +150dB of dynamic range below it. That's a whole friggin lot in case this is news to you.

and excuse me everyone for the thread hi-jack.

[brok landers]

so what if you compress a snare and the overall RMS goes down!... [snip] ...or bring down the fader of everything else.

and resulting in a less loud mix again ...
man, you don't seem to understand me.

Who said you have to mix with every track set to 0dB all the time, with no peaking?

nobody, nor did i ...
i said that the loudest signal in a mix must not go over 0db in the summary stage.
thus it depends on how loud the loudest signal in the mix actually is. if you can actually raise the rms level of the loudest signal (still not exeeding 0dp peak), you can push up the rest of the mix in relation, which leads to a louder sum.
the louder the sum is, the less you have to compress at the mastering stage.
thats what i meant.

This is just an arbitrary treshold. Do not fear the master fader. It has more 150dB of playroom just for this.

i know that ... i just don't want to waste the "fader way" up to the point i have 4 mm to adjust the master, just because i pushed all trackfader up ... understand?

[Lee702]

different strokes

[Kingston]

so what if you compress a snare and the overall RMS goes down!... [snip] ...or bring down the fader of everything else.

and resulting in a less loud mix again ...
man, you don't seem to understand me.

yes I do. You have confused mixing and mastering.

Who said you have to mix with every track set to 0dB all the time, with no peaking?

nobody, nor did i ...
i said that the loudest signal in a mix must not go over 0db in the summary stage.
thus it depends on how loud the loudest signal in the mix actually is. if you can actually raise the rms level of the loudest signal (still not exeeding 0dp peak), you can push up the rest of the mix in relation, which leads to a louder sum.
the louder the sum is, the less you have to compress at the mastering stage.
thats what i meant.

No it just means you have saved more dynamic range for the mastering, actually giving you more choice.

This is just an arbitrary treshold. Do not fear the master fader. It has more 150dB of playroom just for this.

i know that ... i just don't want to waste the "fader way" up to the point i have 4 mm to adjust the master, just because i pushed all trackfader up ... understand?

because it's also illegal to use volume and gain plugins, right?

Know that none of this affects the exported mixdown in any meaningful way. You can take care of the loudness in mastering. No you don't have to compress any more this way. The loudness you think you achieve with these placebo brickwallers have negligible impact on the mastered result. Look up gain staging in some audio dictionary.

*sigh*

[brok landers]

kingston, let's forget it.
i know all your points, you just don't understand me ...
just see this:
the 150 db headroom (downwards, as you want to achieve 0db never theless, and want a fairly loud mix, too) are not of matter in this case.
the louder you get the rms signal of the loudest signal in the mix, the more you can rise the rest of the tracks in relation.
this leads to a louder rms-sum, which leads to the fact that you don't have to compress the summary in the masterstage so heavy. the more you let the transients pass in a mix, the less loud you'll get that mix without overdoing the mastercompressor.
the mastercompressor works on the summary of all signals, triggered by the loudest signal in the mix, but affecting the summary signal.
if my snare is very snappy and affects the mastering limiter , which again affects the whole summ, why not saturate the transient in that snarechannel, so that it doesn't even trigger the mastering compressor so hard?
i can the rise the whole tracks (or the summary bus that goes into the mastering chain) in relation and i am louder then, and the mastering compressor doesn't kill my master.

[bool]

Then why some like to drive converters hot even when tracking?

[Kingston]

brok,

I see what you're trying to achieve. It's just that I have never even heard of a mastering project where a spurious snappy transient here and there couldn't be taken care of. It could be clipped or limited before mastering compressor.

All this has negligible impact on the final result.

Then why some like to drive converters hot even when tracking?

loudness race and unschooled engineers don't go together well. ie. why do my transients sound like shite?