Summing

Track Rat

Just Your Average Sized Member
Spent the day doing some experimentation with the process of getting a multitrack into a stereo pair. Traditionally for the last 20 years or so I've been mixing on my hybrid system which was computer>MOTU2408mkII>Alesis HD24>Tascam M-3500. I really like this system as I can use plugins and hardware simultaneously and I feel I got a really good mix. BUT, as it was mixed on a board, the mixes were not repeatable. The last few years I tried to work exclusively "In The Box" with Reaper. Decent mixes and repeatable. But I still don't feel they sound as good as my mixes on the M-3500. So today I pulled out a passive summing network I built some time back but never got around to using. The computer still drives the MOTU>HD24 but the sixteen outs patch into sixteen passive inputs (eight per side) summing thru a resister network into a left/right output. Make up gain provided by mic preamps. Lots of possibilities. Spent the day making mixes thru the board, thru the summing network and rendered mixes. The jury is still out. So, anybody else doing anything "different" mix wise?
 
... i.m.h.o. :
The digtal notwithstanding the high resolutions achieved (192khz)
still cannot take all the nuances "details" (air) of sound , as the "quality" analog devices still can do...

... overall the digital that do not have dedicated DSP cards like Pro Tools ... ,

it is true that with increasingly powerful computers the digital data "compression"is always smaller...

but as for a digital photo when you zoom on it suddenly appears the "square" of the max resolution pixel

( a.k.a. : Linit ? .... )

same as you set max zoom on the digital audio wave ,

the analog do not have that "square limit" ,

so now multiply that "slight" difference by the number of tracks of the song .... , and the "winner" is ?

.... ok , the analog have other "issues", like noise , gain limit before distorsion , etc..
(..no copy-paste , and other editing options , .... include the perfect total recall )

but isn't it true that it is just the right distortion of the harmonics to often makes the sound more interesting ?

besides the fact that the analog every time it plays ,
little or a lot is never the same as the previous one ?

unlike sampled sounds that sound practically always the same,
and that even if they are made well, after a while they generate boredom ?
(...psychoacoustics ? )

digital distortion apart , ...still "annoying" though ,
despite of the results achieved in elecric guitar digital simulators of amplifiers and distorsion effects ...

and another aspect of the matter to be considered , is the depth of sound ,
in particular the one related to reverberation fx ,
determined by the "tails" of the lower levels ,

and again , the digital have the "squares"... (large or small, depending on the quality of the resolution) ,
while analog is sinusoidal... ,

..square and sine waves ,
... the endless saga of sound ? .. or the crucial difference ?

which is why, for over 20 years, there has been an overwhelming return to quality valves devices ? (Avalon , Manley , Thermionic Culture , & co. )
... which seems to be the only way to restore roundness to the squares....

hoping it will help ,
cheers

Ps :
Ever tried to compare the same tracks recorded in the 70s and 80s , also 90s , recorded , mixed , mastered in hi pro quality analog domain (a.k.a. "AAA" ) ,
between the original vinyl version, the subsequent (pretty cold) first reissue in Compact Disc made with the first digital technologies ...
and the latest remastered with quality valve devices ?

and hear details that were not audible in the previous ones ?

apart from an additional sound quality always enhanced by the quality of "those" valve devices...

And :
If originally instead of being made with the professional analog technology of the time ,
had been made with digital ,
would they sound like this ?
 
Last edited:
... also, be really careful when you go out on your boat. If you go too far, you'll fall off the edge of the world! 🚢

There are some fundamental errors in your assessment of digital audio, most prominent is the "stairstep/square wave" comment. If you look at any proper digital to analog converter on a scope, the output is decidedly NOT squared off or stairstepped unless you try to go over 0dB. That is a convenience in trying to express the concept of sampling. A more proper way to visualize it would be to place dots at the point of sampling, at which point the converter will interpolate the waveform between those points. Now, what if you want to go louder than 0dB? That's the same as saying, what if you wanted to get 50 watts out of your 5 watt tube. It's not going to happen. If you need think you need more resolution, then move to 32bit floating point. That gives you a theoretical 1550 dB range. I can guarantee that there is NOT an electrical device made that can come anywhere close to that potential. The highest shockwave possible on the earth is somewhere in the 200dB range. Combined with 192kHz sample rate, and you are now at two octaves higher than anything a human can hear.

I would challenge you to do a very basic test. Take a GOOD analog source (something with real dynamic range and frequency response) and do a proper digitization of it. Now do a very careful comparison the two. I'll bet you can't tell the difference statistically in a blind test. Now take a purely digital source and put it on the analog device of your choice (I don't care if it's vinyl, cassette, 1/4 inch tape). I'll bet I can pick out the analog copy every time. That tells me that the analog copy is the less accurate medium.

This has nothing to do with whether you like the sound of even order harmonics to make your sound "better", "fatter" or "warmer". That's the same as having a preference for putting a soft filter on your camera lens when doing portraits with your Kodachrome 25 Nikon. It tends to hide all those blackheads, big pores and skin flaws. Oh yeah, you can do the same with with your digital Nikon with a soft focus filter.
 
Wow. I haven't heard some of those arguments against digital in over a decade.

BTW, the harshness of some of the early CD's was due to them being made from the masters used for making vinyl with the RIAA filter applied. Stupid mistakes like that were made. Also, the early 80's was all about the clarity of the high end, so it was fashionable for everything to be really bright. Of course, if you mixed something so that it would be really bright on tape of vinyl, since CDs didn't soften the highs, it ended up being too much.

As in the video above, there is no stair-stepping. There never was.
 
Some people might also not be aware that the Red Book standard also allows for an artificially boosted high end using a Pre-emphasis flag bit. If that flag is not properly read, the CD will be much brighter, akin to recording a cassette with Dolby B and playing back without the Dolby engaged. I think the boost is about 9 or 10 dB at 20kHz. That's a LOT!

Pre-emphasis seems to be primarily used on a few older Japanese pressings. If you Google CD Pre-emphasis, you'll find some discussions and lists of this. ExactAudioCopy can show the Cue sheet for a CD and will list if the pre-emphasis flag has been set. Computer CD players and many ripping programs do not check for the boost. I've only found 1 CD with PE and that was a Japanese import copy of DSOTM a friend of mine had. None of my Japanese pressings have PE (Hendrix remastered discs), Mannheim Steamroller, etc.)

I don't know of anybody using PE anymore.
 
I’ll just place this video here



...technically no "moles"..
but i.m.h.o.
ears are much different than measuring instruments...

and the fact that professional ADA converters like Apogee , Prism Sound , UAudio , Grace Design , Antelope audio , & all the other "snack mates"... ,
do not sound the same one from the other , can only be its confirmation ,

... same for "the other side of the moon" , like Motu , M Audio , RME , etc... ,

and for sure the output sine wave of each one will be perfect ..
 
Back
Top