In your second question, what i usually do is when i finish mixing i bounce it and switch all plug ins off, archive and hide all the tracks and what will be left is 1 track ready for mastering, then click " save as" . Start mastering. One reason for exporting might be to ease the load of your cpu. If you do what youre saying in your question plus add more plug ins in your master bus you might experience BSOD or  your cpu might stop working.

2 major arguments against saving up to buy that expensive analogue summing gear just came up (thanks, Jeff, well said): 

first, "great composition or song followed by great performances" - think about the songs that moved you most and listen closely; you'll hear some strange mixes, sometimes even dubious quality, yet, those songs get to you ... most are probably summed the analogue way, but as good as these songs are they'd move you even if they'd be totally digi-sterile ...

next, "mastering" - if you want analogue summation without splashing out huge amounts, give mix stems to your mastering engineering who owns all that gear; i'm pretty sure that would make a huge difference to the overall sound, but more likely due to another professional with a different approach and fresh ears working on your material rather than just analogue summation ...

It is funny because people seem to think that all the great music and songs were produced using analog technology and mixed using analog summing. But the real truth is that is all they had, then.

What about the huge amount of amazing music and songs that have been produced all digitally and summed that way too ITB. I do think it is important to have the precision on the mix buss though but the good news is that most DAW's provide it. Even on 32 bit systems with double precision operation where it counts. A DAW does not have to be 64 bit end to end to produce an amazing result, but it is generally known that on the mix buss it is needed. 

Summing systems vary too. Neve summing box is full of transformers, input and output. That is what you may be hearing. (one review of the Neve box said even sending a stereo signal to and from that box changed the sound) Others have valve stages built in.  But at the other end of the scale, how about pure resistive summing followed by transformerless pristine gain stages. Might not sound so different then. 

Quite a while ago a forum member built a resistive summing box followed by two Mic Pre amps for the makeup gain. He posted two mixes and did not tell which one was which. After careful listening I deduced which was which and the digital version was superior, not the analog summing version. People were seduced by the analog summing at first but in fact it was inferior. Turned out the Mic pres had transformers in them and they were actually altering the integrity of the mix for the worse.

Mastering does not have to come back to analog to be good or effective. One can stay in an all digital medium and master very well. Some aspects of mastering are better performed digitally like limiting for example. It is up to the mastering engineer, not the analog gear he may have outboard.

When people are searching or obsessed with the tools that make the music it could mean they are looking in the wrong places. It might be time to go right back to the music instead. When one is very happy with the music the rest does not seem to matter so much. 

Here is a great quote I found on the net re analog summing:

Analog summing will not make your bad mixes better. More mixing will help make your bad mixes better.

I have done some research too and some boxes like the Dangerous are purely resistive followed by active Op amps. You could build the same thing for a fraction of the cost. I can understand valve stages and decent transformers costing a bit but not pure resistive summing followed by a quality gain stage such as the Dangerous.

We are straying away from the OP's questions - and to answer the last - try it.  Mix your song, bounce it to a master (at high resolution, of course), and then play that out through your TL comp and record it back into the DAW.  If you like it enough to make the extra step part of your system, do it.  When I do that, I send it out through a TC Konnekt and use the on-board DSP comp/EQ and an analog comp/limiter.  Generally, serial compression sounds more natural than one comp slamming, as you add a couple of dBs at a time.  You can do the same in the DAW on the stereo track.

Most of the qualities we are talking about are sublte.  The better the room and monitors, the more you can pick up the differences.  The more experienced the ear, the more you can hear them.  And the smarter you are, the more you integrate your experiences into your own system.  I know several professionals who have gone in-the-box.  They feel that digital is close enough (usually mentioning UAD) but mostly it is the fact they can recall settings when whomever is paying the bills wants "just a small change."  Resetting a bunch of outboard is a tedious, time-consuming process, and time is money.  Given their druthers, most would stick w/ hardware.  That should tell you enough about your own home-brewed mixes.

Where analog hardware does rule is in capturing your sound.  You need a room that doesn't have terrible flaws that you will record.  A good mic - or several, of various kinds.  An LCD for vocals and leads, a pair of SCDs for stereo stuff and acoustics, dynamics for drums, and a ribbon.  I recently picked up a cheap mxl ribbon that works better my mid-range LCD on guitar.  When placed right, it rounds the sound nicely.  A good preamp(s) is nice to have, for the flexiblity provided by more gain and the fact it is harder to crap them out when the guitarist/vocalist suddenly goes to 11.  Not to mention the sound they can impart.  A good compressor can help w/ keeping the level even.  Eq going in can help w/ that also, narrowing the energy.  All those settings should be minimal, unless you are sure about what you are getting rid of since you might need it for mixing.  Think of it as pre-shaping the signal.

Basically, recording and mixing is art.  There are rules that it is good to know and understand, and it is ok to break them once you do.  There are reasons people spend big bucks racking old neves for vocals and guitars or first send the bass through an UAD or use the same mic for the drum room.  It works most of the time.  Figure out how to use what you have so you can plug-in new stuff in the holes.  Finances might keep you from having the kind of analog set up you'd prefer, but digital should suffice once you figure out how you like to use the tools.


@

Thanks guys. 
A very interesting discussion.
Just the fact that so many experienced mixers disagree tells me something.
At my level and the fact that I am only doing my own stuff and don't expect, or care, to have a hit at my age, it's not cost effective to spend a lot for subtle improvements. Bang for the Buck concerns.
I just thought I'd use this borrowed analog Compressor to see if it made a large difference to my ears.
*******************
Still not sure about the meaning of "Stems"?.
Would it not be the same as Exporting a mix but instead, sending "what you hear" thru the analog stereo Compressor and recording it?
Or is it better to do it dry first?
Or are these all just things to experiment with?

M_Glenn_M


Still not sure about the meaning of "Stems"?.
Would it not be the same as Exporting a mix but instead, sending "what you hear" thru the analog stereo Compressor and recording it?
Or is it better to do it dry first?
Or are these all just things to experiment with?

Generally speaking when folks say "stems" they are referring to similar tracks exported as single files or outputs (usually based around common bussing techniques). So, for example, when I output stems into my SSL I'm doing something like sending all the vox into st1, all the drums into st2, all the send fx into st3, bass into 4, guitar into 5 and so on. I don't know of any real rules for stems...for example, I've seen lead vox stems and backing vox stems, and then I've seen vox stems that include both.


Essentially think of stems as logical groups of tracks that are exported or sent to another device.


Dean

Certainly there are those (including respected professionals like Eric Serafin, AKA Mixerman) who believe analog summing is better, and certainly there are hardware vendors out there cashing in on that belief with overpriced "summing boxes" (read: passive mixers). 

But I have yet to find ANY rational technical explanation as to why that might be. If the effect is really that obvious, then surely it can be measured, analyzed and explained. AFAIK, it never has. 

Audio can be a very complex subject, but there is no mystical component. Every aspect of audio can be described in objective, measurable terms. I reject the concept of "mojo". If you can't tell me why something should sound better, and if most people can't hear the difference, then the default presumption must be that you're imagining it.

Digital summing is transparent.

Analog summing is, at best, less transparent. If one prefers analog, and they aren't just imagining things, it's because they find the coloration added to their signal to be pleasing. There is nothing wrong with this, assuming transparency isn't the goal.

drewfx1


Digital summing is transparent.

Analog summing is, at best, less transparent. If one prefers analog, and they aren't just imagining things, it's because they find the coloration added to their signal to be pleasing. There is nothing wrong with this, assuming transparency isn't the goal.

BINGO!


Summing audio signals in the analog and digital domains is different. It is two different processes. And I think it sounds different, but others may argue that it doesn't. And it may not, to them.

As someone pointed out, it's the last 2%, and if everything else in your project is rockin, you probably don't need it. And if everything else in your project sucks, well, analog summing isn't going to fix it<G>!

I think it is worth trying, and I think everyone that has the ability to do so should, if for no other reason than it'll make you a little better informed. That's difficult to argue with<G>!

There are areas in recording/mixing where I probably go overboard... I almost always set up pedals, amplifiers, and microphones to track guitar... I think it sounds better, and I'm happy to give up the ability to try 22 different chorus effects, both tempo-sync'd and free-running, to do so.

But I 'grew up' during a period where 8 tracks was humongous, and if you had more than two channels of compressors you were living large, so I learned to make a LOT of choices very early in the process. So it doesn't bother me to limit my options, that's how/what I learned.

For someone that has never had limits imposed maybe it's a different story?

Anyway, it is easy enough to build a simple passive summing box, and if you don't know which end of the soldering iron hurts you can probably find someone who does. So give it a try... it'll be fun, and you'll know your personal answer.

Summing audio signals in the analog and digital domains is different.

The difference, however, is not in the summing itself. The difference is that analog summing requires extra D/A and A/D conversions (with added quantization noise), noise from amplifiers in the interface, plus any noise picked up by cabling, connectors, patch bays and internal components in the summing box. 

The degradation is probably insignificant, but I can't imagine how that could possibly enhance the signal in any way.