Lots of the stuff has been discussed but in different ways of of course in general rather than from my personal preference perspectives.

Yes they could be - but it's not altogether sure that they actually have been, despite the claims made. The desire for 'analogue sound' has created a massive suite of myths that can be exploited

There is no doubt that pushing a basic digital system to its max is not the same as analogue - and of course some analogue being pushed is not the same as other analogue. The analogue field is awash with differences and dependencies that make it very touch and go and confusing - often sounding different from session to session.

Knowing what I do know about analogue (I grew up with it), I have to remain sceptical of people who claim to have emulated it in the digital domain - not because it's impossible, but because I would doubt that sufficient knowledge of the nuances had been researched or sufficient efforts had been expended to get them.. And anyway, under what particular conditions was the analogue gear under that the emulations refer to. Analogue stuff changes depending on all sorts of stuff, even the kind of sources and load connected to them..

Now this is an area I am heavily into moving into - this was indeed my plan before my resources dissolved. But I am not interested in emulating specific pieces of gear under specific circumstances (and claiming to have produced a clone without understanding why). I am much more interested in providing the ability to produce the wanted sonic artefacts and character of analogue gear in a way that's understandable, reliable and usable to the engineer doing creative stuff - not just a pretty GUI of some rusty old box - it's sound that interests me.. Possibly sounds akin to commonly found samples can be stored as set-ups etc.?
To me this kind of thing is one of the biggest artistic holes in the digital environment we have right now - and I'm passionate about it, because I know loads about what creates which sounds and nuances - I grew up with this stuff.


I honestly could not say. Though for claimed emulations of things I have designed in analogue in the past, I can obviously tell if the claims make any sense!! I have to be honest and say that it does not inspire confidence and adds weight to the idea that it's more psychology than technology..

Ok - from an entirely personal perspective (leaving out issues of control tactile control panels etc.), knowing what I know I would entirely steer clear of analogue as much as I possibly could. I am pretty confident that with the right efforts I can get pretty much all I want from stuff running digital - because I know what's in much of it from an insiders perspective and understand what its likely to do..

The business of making things powerful and loud without overwhelming distortion starts right at the beginning with the sound of the tracks themselves and how they can be made to gel in the mix - before - you hit it all with any maximisation. This is how I would go about it:

The very first thing I do is to bring the tracks down to reasonable level to give me some headroom freedom when I'm mixing (as discussed) weigh up the sounds and come up with a sonic plan in my head for it that I will aim for.

I then start working towards this 'plan' using EQ and rough mixing - EQ at this stage is to reduce or even take out overly loud resonances (either from instruments themselves or micing techniques). My EQs are almost all in cut at this time by small amounts in Hi Q at very finely specific freqs done by ear - and I have one on every channel always.

I then run around the balance adjusting the EQs gain and Q to get something like the basis of the plan. For natural stuff often at this point the mix sounds better than it ever will again - but this is never what the clients are expecting - so more has to be done.

For rock music I normally first start attacking the drums and bass in more detail - I put them on a buss so that I can reduce the rest of the mix by 6-10dB so I can hear the whole drum section in prominence with the song in the background. This is where I'll deal with spill from the overheads with dynamics (gentle gating) and more EQ notching to take out the unwanted stuff, basically to give me more freedom in attacking each drum in separation. I get the overheads and cymbal feeds sounding good entirely on their own using nothing but EQ.

After this I start applying dynamics to individual drums starting with the Kick and bass - not for compression, but to mold the attacks and dynamics so that they gel together. At this point I usually start bringing up the rest of the song bit by bit in rounds of percussion modification. I do the snare in a similar fashion and then the toms etc. I know I have done it when the rest of the song is back at normal level and all sounds roughly as my head plan for the song.

Now If I'm aiming for maximum volume (rather than quality) I need to have the majority of the impact and 'surprise' in the MF regions because this is where things sound loudest and produce the least peak values. So I will go round the rest of the instruments (excluding vocals) trying to build a texture that exercises the MF and lower HF regions without sounding harsh or intrusive. In goes more EQ often gentle stuff a few dB here and there at lower Q values to build a subtle sound palette that hangs together nicely..

The trick here is to somehow allow the drums and bass to occupy a believable place in the song without creating massive peaks that will reduce your overall average levels. I quite often use the inflator on the drums to flatten the peaks without total loss of their sounds (remembering to leave the cymbals and overheads out of it cos they will alias) - a 3dB overall improvement can most often be had without any apparent loss in the percussion sound.. I will sometimes introduce and inflator on the kick on it's own and deliberately overdrive it to produce some 'smack' if it sounds too soft. I sometimes EQ very slightly again to reduce stuff that causes level but does not feature much in the song artistically. All this is of course entirely subjective done by ear and nothing much to help you.. Remember that at all times I am surgically taking out what I don't want - rather than going into a merry-go-round of trying to boost what I do want. The latter always ends in a hyperbole mixing disaster IMLE :-(

I then do the normal stuff with the vocals fretting about the sound, dynamics and atmosphere they produce. There's normally very little you can do to boost the relative loudness of the voices apart from placing them as much as possible in the mid ranges. (The DSM pre-set boy band does this as an illustration). This works for some material and not at all for others. Getting it right is everything IMLE.. So I spend ages on them most of the time.. Dynamic compression with sibilance control is absolutely crucial to every vocal track I have ever encountered. There is advantage in using an application that caters for both dynamic and sibilance control naturally because they can achieve so much more natural results than a blind de-esser followed by a blind compressor IMO. Before I had the DSM I often used to boost the vocals in the HF with EQ - compress them - then undo the EQ afterwards...

At this point I have something like demo mix and my mix buss is normally peaking at around -10 to -6dB. I will start to put in the buss processing I want to use - and start 'mixing into it' - the idea being to produce something that will exercise the buss compressor optimally, because I must produce a result with an overall controlled dynamic range. With normal compressors I just set them to soft knee at highest ratio and the appropriate attack and release times to match musical content ( I am not limiting at this point). If I am using the DSM it's great because I can either dial up a pre-set that has something like the appropriate sound on it - or I can grab a capture of the song itself and use that.

The job at this point is to sylise the production, bringing up a bit of this and that at these or those moments to enhance the expression of the work and 'feature' important aspects of the performance all along the song. This is pure fader and automation work - keeping an eye and ear on the effects this has on the buss compression so as to avoid doing anything that will over compress and lose level in the end product compression/limiting. The DSM again is great here because it will compress freqs in isolation - so there is much more freedom than in normal compressors and you can get away with loads more mixing manipulations without the nasties occurring.. It's just so much easier to do :-)

This is where I most miss physical faders, because I was used to 'playing' the faders intuitively on the fly a bit like an instrument at this point in a mix. On a screen based W/S doing this a part at a time on one track at a time requires you to somehow segment the effects of the manipulations in your head and hold them in your brain long enough to do associated stuff on another track on the next run - despite the fact that the sound of the unfinished (and broken) combination is rattling around the control room! :-( It's far too easy to find yourself prodding and poking this and that ad-infinitum using automation in bits and pieces with increasing frustration, to do what you would have just done on the fly intuitively on a good control surface. And of course it's easy to lose the plot while you're fiddling with a screen selecting this or that for editing or worrying about how best to actually do it. But there's a price to be paid for being able to afford to do the thing in the first place - it's still better than nothing at all - which is what you would get for the same price in a HW digital console...

Ok now at this point (when I think the mix is done and sounding OK even though it's probably overly loud and impactful for what would most suit the song) this is what I would send to the mastering engineer, controlled but not limited, loud but not excruciating - he has still something to work with.

However from the client's point of view this is where outright loudness comes into play (if only to impress the A&R people). You end up having to compare this with some other topical sounding CD the client has brought in - making sure that yours is as loud as it (or you are history) :-(

This is where I pop in the limiter (either the DSM limiter section or the Oxford limiter) and start to increase level (increasing thresholds and or gain) to strike a balance between loudness and distortion. I run round the gain threshold and time constants trying to produce the best balance between lost or squashed peaks, distortion and loudness until you reach the loudness of the other CD. If I have the DSM I may try other captures or pre-sets to see if one gets further than another.

But in the end, when I am finished I could always get more loudness (that the client would even accept as sounding great) than I am prepared to finish on. This is because (even though I could) I do not want to produce more loudness than a mastering engineer himself would settle for - because not only will he have your clean mix, he will also likely get your loud mix as well as a kind of carrot to aim for the clients can hit him with. And all this would do is reduce his options to the extent that he could not do what you pay him for... Your loud mix only has to match other popular productions, not beat them :-)

This seems to have developed into a rambling post - sorry, it may have gone off topic :-(

But the point is that in the above there is absolutely no step that I wouldn't have done in the days of analogue (except manipulating physical faders), nothing anywhere is clipping or distorting unnecessarily, there is no outboard kit or unnecessary DACs/ADCs etc. The only thing I would like extra in this process is some more artistic applications that were properly under my control rather than the phase of the moon :-) And this is something I have always wanted - so no change there :-)

IMVHO I am pretty sure I can produce as much loudness and punch in the box than any combination of outboard stuff - even when dealing with someone else's finished mix, where all the above ramble would not even apply.