10/10 James for matching these mixes. You are doing a fabulous job at this now. Interesting thing is this is first time I managed to get some sort of null thing happening here with this one. Nowhere a full null by any means but I can get the low end to drop out now when I null these which means the bottom end of the mixes are matching for some reason a little more accurately.
 
As for sound still very very similar, very hard for me to tell them apart. Still no big OMG moment here! I am still finding the ITB mix to be pretty amazing.
 
One thing I am doing now that I did not do before is I am limiting how much low end I am letting into the room with my monitors.  Before they went down to about 35Hz but I am now only allowing them to go down to about 41 Hz or so.  After reading some articles about the benefit of not letting too much low end into a smaller control which is what I have.  I am liking my bottom end more as a result.
 
I have also got a pair of Yamaha HS7's (which don't go down real low anyway) Mixes sound extremely similar on those to me as well.
 
So if the bass end on one of these (maybe the pres) is little heavier I am not hearing it.  Maybe James you can clarify here if that is happening. 
 
Just done some more listening and maybe the top end a hair louder the pres but it is real close though. Nothing that some +1dB shelving at around 8K on the ITB mix would not easily match though. 
 

Thanks for staying with me Jeff! :)
 
I am actually really stunned at how similar this mix is to the the ITB box.
 
For the mix bypassing the pres, I was using 4 stereo channels on the board but those are only mono for mic pres so I moved those 8 channels to 8 mono channels. They might behave a bit different from the stereo channels. In fact, when calibrating I noticed the center detents on the stereo channels aren't exactly where they should be.
 
There are only two things I can't explain right now:
 
1. At 53 seconds the big guitars come in. The are very low in volume in the ITB mix but big and present in the analog mix. I have to believe I have a routing issue and I'm going to try and track that down.
 
2. The various reverb aux buses are all louder in the analog mixes. I don't have a clue why this is so because I've calibrated those two channels on the Midas multiple times. Of course, you could increase the volume ITB to match, but I'd like to discover exactly why this is so.
 
Interesting that you mentioned the low end. I have a friend that listened to this on his big system and he says there is some mud around 40 hz. I'm using Adam F7 monitors and my room is well treated, but when I look at the Sonarworks plot for my monitors I see there is a big drop off at 80 and an even bigger drop off around 50 or so. I have ordered a sub woofer that will be here by the weekend; I really want to hear what he's hearing to be able to clean up that low end!
 
I want to do one more mix and then I think I'll be satisfied. I'd like to bump all 16 channel outs by 3db and calibrate the master fader on the Midas down to compensate. I'm told analog summing has much more headroom, so I'd like to push a bit into the Midas pres to see if that brings out some interesting character.
 
It will have to wow me though - at this time I don't see any reason to sum in the analog domain based on the results of my tests...

James,
 
Thanks for taking the resources and time to do this experiment.
 
I think many of us (especially "long timers" - read old ) suspected the results you're finding.
But... all these years later... it's still good to see/hear those suspicions confirmed.
 
There's no doubt that prime pieces of analog gear have that "something special".
I picked up a Neve channel-strip a while back.  Everything sounds better recorded thru it.
DI electric bass, vocals, you name it...
 

Thanks Jim!
 
This is a journey I just had to take and I'm glad I did. I love doing this stuff. :)  In the end I still have a great desk with tons of quality Midas mi-pres and 16 good converters in the UAD Apollo 16 that I wanted to have for recording anyway...
 
I am eager to try the Stam Neve pres and the Stam SSL4k bus comp when they get here. I don't suspect I'll want to use the pres on the mix bus but you never know. I do suspect I'll make the SSL a staple on the mix bus but I won't bother with analog summing - just shoot l/r out, through the comp, and back in.
 
But of course only trying it will tell me for sure! :)

Thanks James it has been a pleasure and fun.  I have started to really get to know that track too!  There is nothing wrong with external mixing once you start using the EQ's for example on the mixer.  When I was mixing externally on a Tascam console I really liked the EQ's on the channels.  They were fast and enabled me to get the sound about 80% right very quickly.  Coupled with an EQ plugin leaving the DAW I was able to get sounds very easily.  Using the faders on the mixer is rather nice too. Nice to be able to just grab them and move them etc.. Having physical sends to reverbs and things is also rather nice.  But these things are other factors rather than summing engines alone.
 
Not sure pushing pres by 3dB will make things a lot different. I would have thought higher values e.g. 6 dB or more.  Pres tend to add less of sound when their gains are lower though so it might even make things closer!  
 
There is plenty of headroom mixing digitally especially if you work with something like the K system at a ref level of -20 dB FS = 0 dB VU. There is 20 dB of headroom in that situation which is only really available in the finest of analog consoles.
 
The biggest thing for me is the recall ability of digital though.  That in itself is amazing.  When I was producing a rather large project back in 2007 on a large format analog mixer I did have to recreate mixes like 2 to 3 weeks later because the client decided to change something.  Even after writing down every detail about the settings of everything on the mixer and the outboard I never got the mix exactly the same.  Analog works fine but it is best to put it to bed there and then if you can.
 

OK, last test here. I pushed the 16 outputs up 6db each and leveled the Midas master to bring it back down. I still found I had to adjust the new track -1 db to level match so it's not an exact science.
 
ITB:
https://www.dropbox.com/s/ie57qcr1nwufyfk/ITB%2002.wav?dl=0
 
+db into Midas pres (remember to drop 1db when you listen):
https://www.dropbox.com/s/wiyzjolg9yhk7w6/Analogplus6.wav?dl=0
 
As far as I'm concerned, there's almost no difference.
 
I believe my experiment is done! Any differences in the analog summing just don't justify (to me) the work involved. In fact, short of the guitar and reverb aux issues (which I think I may know what happened) I don't think I could pick out what was what in blind test. I even decided to try to add some "transformer magic" by putting the new UAD SSL E channel just for the xformer sound on all 16 tracks and mixed ITB. I'm not posting the results because again they just weren't noticeable.
 
I'm calling this one done: MYTH BUSTED! LOL
 
My crossover showed up this morning so now on to hooking up and calibrating a sub! :)

I was reading about summing mixers on the web, curious about them I guess and whether I could use a small Yamaha mixer as an experiment.  'Let me check the Cakewalk Forums,' thought I.
 
Thank you so much for this!

I should add as an update that I spent some more time experimenting and calibrating and I absolutely hear a difference. It is small but there, but when I started adding external gear it really made a difference. I am currently using the summing mixer and waiting on arrival of a Stam Audio SA4000+ SSL comp clone with transformer mod. My intent is to come out of the board into the SSL clone and then back to the PC...