[LAD] Half-OT: Fader mapping - was - Ardour MIDI tracer

Ralf Mardorf ralf.mardorf at alice-dsl.net
Thu Aug 21 21:24:46 UTC 2014

On Thu, 2014-08-21 at 23:20 +0200, Ralf Mardorf wrote:
> On Thu, 2014-08-21 at 15:50 +0100, John Rigg wrote:
> > On Thu, Aug 21, 2014 at 12:39:05PM +0200, Ralf Mardorf wrote:
> > > IMO each channel by default
> > > should provide your (Fons') parametric EQ and post fader aux sends.
> > 
> > Those who usually try to record the right sound in the first place might find
> > this a nuisance. Unfortunately one size doesn't fit all.
> An EQ and a fader for each channel is a must have and never ever could
> be a nuisance. Control over volume and frequencies are needed tools for
> mixing music. There is nothing like "recording the right sound in the
> first place" regarding the mix of several tracks. Controlling
> frequencies can be used to get the "right sound" for an individual
> instrument and yes, you can record it that way in the first place and
> then you don't need the EQ for this task, but when mixing music the EQ
> is needed to control interaction of the sound from different tracks, too
> keep the sound you recorded right in the first place. Nowadays many
> people use compressors to do the work that actually is a task of an EQ
> and not the ask of a compressor. A transparent acoustic pattern and
> control over dynamic range by a good engineer mainly is done using EQs
> and faders and not by using compressors and fade ins and outs.
> > 
> > > If
> > > people start mixer automation for the fist time, a notification should
> > > pop up and mention, that:
> > > 
> > > > Fades are rare in music mixing
> > 
> > Some might find that patronising. I personally mix a lot of recordings with
> > fade outs, even if it's just to fade the end of the last note or a long reverb
> > tail.
> Fading out a reverb is disgusting, anyway, I'm mixing pop music myself,
> somebody perhaps sings ten words and you want a delay just on one of the
> words, this kind of mixing is done and some people might like fading out
> a reverb or to add reverb to the end of a guitar solo ... the claim is
> that fates are _rare_ and to add delay or reverb by a fade or to fade
> out reverb or to fade out a song and sometimes to adjust the volume is
> done, but it's rare and not that often as fades are done by amateurish
> computer music "engineers" using mixer automation.
> Another problem of bad mixing are dynamic effects for each track and
> different sorts of delay for each track. This usually isn't done with
                    ^^^^^ reverb, this should be reverb and not dely :D
> knowhow, it's usually done because of a lack of knowhow.
> On Thu, 2014-08-21 at 08:30 -0700, Len Ovens wrote:
> > Good I must be doing things right. The only fades I have used are fade
> > out at end of song and I have drawn that in. Often, to make it sound
> > right, I have to fade the lead (louder) tracks faster than the backing
> > tracks anyway.
> Pop music usually ends as classical music ends, with a composed end,
> only the around 3 minute radio versions need a fade out. Bad composers
> already make this part of their compositions, they don't compose an end,
> they repeat something and make the fade out part of their composition,
> usually the same kind of composer who don't shy away from repeating the
> first half of their song by transposing it two half steps for the second
> part of the song. There are different levels of simple pop music. I love
> simple pop music without too much cheap tricks. But within the last
> decades cheap tricks became more and more the business.
> However, a fade out for the radio version is ok, lead and backing
> usually are on separated subgroups, another feature that also a virtual
> mixer should provide by default, needed for the fade out you described
> and for the post fader aux sends I mentioned.
> The design of virtual mixers shouldn't be that modular as they are. Some
> people might not need this or that, but to learn how to mix and to
> decide when something is needed or when not, there's the need to have a
> good default mixer. Some people, including myself have the knowhow and
> could use a modular mixer, but it does cause extra work or at least a
> template.
> > Certainly adding individual reverb per track does not make sense as
> > having the whole "band" in one sound space would be the normal target.
> There are exceptions, such as gated drums, but indeed, it's good common
> practise to use one reverb. _But_ the reverb is not simply given to the
> clean stereo sum, it's controlled for each channel individually, that#s
> a good reason for using aux sends. The pad string like sound needs
> another volume of reverb, than the dominant pop music kick does need.
> On Thu, 2014-08-21 at 09:01 -0700, Len Ovens wrote:
> > Not having anyting other than level certainly ecourages one to record
> > the right sound. It has with me.
> You only can estimate what sound is needed for the end mix when you're
> much experienced and even then it can't be optimal. Faders and EQs are
> basics for doing a good mix. To act from necessity is ok, but stupid
> when good EQs are available for the mix. I will repeat it. The EQs for
> _mixing_ are less needed to give the recorded instrument a sound
> characteristic, this can be done for the recording. For _mixing_ the EQ
> is needed to _keep_ this recorded characteristic when mixing with other
> recorded instruments, when the frequencies start to interact. The EQ
> also is needed to control dynamic range of the mix and to control
> transparency of the mix.

More information about the Linux-audio-dev mailing list