|
Post by EmRR on Jun 6, 2022 10:46:57 GMT -6
More on exact level matching - the higher the channel count, the higher the gain errors, obviously. You'll find most channels in a 16-32 channel DA to vary somewhat from each other, and all that adds up to a slightly different mix. 0.1dB doesn't seem like much, but it's enough to make or erase perceived differences. Add similar changes in a patchbay, connecting cables, precision of mixing resistors in a passive summing bus, etc, etc. Season with perception (deception?) bias....like so many things in audio we get impressions from long term varied experience, many no more conclusive than "something's different and I like this one", which is maybe as conclusive as it needs to be.
There was a time a 24 channel DA output summed through my Mackie 32*8 (pans, faders at unity) was usually preferable to ITB. A change to better converters undid that perception.
|
|
|
Post by drbill on Jun 6, 2022 11:15:21 GMT -6
I think everything matters, converters, wire type, electronics, etc.. All minute amounts. Maybe not "hearable" one at a time, but perhaps noticeable once you add em all up. Really, this is one of those "you gotta try it for yourself" situations to see if the differences are positive, or if they matter to you, and how much $$$ you want to spend to get that difference. It's a personal preference, and different for every combination of gear.
|
|
|
Post by jacobamerritt on Jun 6, 2022 12:37:15 GMT -6
Update from the OP: So I think for my next mix session this week, Im going to try printing the through my Vintech Dual 72s and Tonelux Equalux and see how much difference it makes... My thinking on using the Dual 72s is they are based on the Neve 1272, which was used for make up gain on the mix buss. I know the Vintech is a bit of a different thing. But it sounds fun to try
|
|
|
Post by Dan on Jun 6, 2022 13:00:47 GMT -6
More on exact level matching - the higher the channel count, the higher the gain errors, obviously. You'll find most channels in a 16-32 channel DA to vary somewhat from each other, and all that adds up to a slightly different mix. 0.1dB doesn't seem like much, but it's enough to make or erase perceived differences. Add similar changes in a patchbay, connecting cables, precision of mixing resistors in a passive summing bus, etc, etc. Season with perception (deception?) bias....like so many things in audio we get impressions from long term varied experience, many no more conclusive than "something's different and I like this one", which is maybe as conclusive as it needs to be. There was a time a 24 channel DA output summed through my Mackie 32*8 (pans, faders at unity) was usually preferable to ITB. A change to better converters undid that perception. Summing itself is simple math. Why buy an expensive box to do it?
Better DAs and the DAWs just got cleaner. 64-bit floating point pcm summing pretty much is perfect. There's no gradually rising as the frequency lowers distortion at higher sampling rates like with 32-bit float operations.
Digital itb has been fully capable with distortion since the mid 2000s and fully capable with less distortion than analog since the mid 2010s. There are still some crazy dynamic filters that haven't been realized in digital cleanly and of course analog gear has cool distortion, especially older gear with component degradation and drifting out of spec, but most people don't even know what they're super special weirdo dynamics smoothing filters are doing and most modern gear is just designed to sound cool, ie distort, rather than solve a problem. Form over function.
|
|
|
Post by Guitar on Jun 6, 2022 13:01:07 GMT -6
The other half of "everything matters," is "none of it matters." Too much emphasis on the first one, be reminded of the second one. And vice versa.
I think obsessive audio types often skew too far in the neurotic "detail oriented" direction, which can quickly result in psychosis of some sort. You can look at some of the thoughts people have been having here as an example.
I hesitate to discuss "summing" in general because once you've started to discuss it you're already far off in the weeds. The deeper you get, the more lost you are.
In my understanding, the light at the end of the tunnel is, "none of this really matters in any meaningful way." That can be your freedom.
|
|
|
Post by shakermaker on Jun 6, 2022 13:52:11 GMT -6
ITB summing is ofc perfectly fine in 2022 and I don't think there is much to gain from summing outputs into something v transparent. But the capi sumbus32 does something forrrr sure, that me likey. I should just start saying "big ass API sounding box that sounds rad af and is fun to use" instead of "summing mixer" haha. Two very diff things ... I would just sum in the box perfectly happy instead of a transparent sum mixer.
|
|
|
Post by christopher on Jun 6, 2022 13:55:51 GMT -6
Summing is dependent on the issue where all analog is subject to the laws of electromagnetic waveform physics, something that science is continually learning more and is really hard to understand. Transformers, inductors, etc exist because wire carrying signal also produces a magnetic field around the wire. That electricity is induced into conductors (copied over) within the field. The electromagnetic field of a signal on wire spreads like a magnet does but it’s invisible so it’s hard to understand the impact.
This is the thought experiment that got me: - a 30Hz waveform at +22dB is going to have 30 positive and negative moments. - a 15khz waveform that’s 40dB quieter is going to have 15,000 positive and negative moments.
If they are both going at the same time, then those +22dB peaks at 30 Hz will be a massive ocean of positive electrons during its peak.. and cause the 15kHz quieter, much smaller negative moments to change its sign from negative to positive. We know that if you put a negative charge in a bigger positive charge it will align with the surrounding bigger charge.
And so really summing ITB shouldn’t be the same, the math is exact. The effect that analog electromagnetic I’d sort of expect is to act to side chain de-ess and blend things into one during extreme peaks of low end.
Eric Valentine does have a video on a cheap DIY summer based off neve, sounds pretty nice as well
|
|
|
Post by Dan on Jun 6, 2022 14:33:50 GMT -6
I think everything matters, converters, wire type, electronics, etc.. All minute amounts. Maybe not "hearable" one at a time, but perhaps noticeable once you add em all up. Really, this is one of those "you gotta try it for yourself" situations to see if the differences are positive, or if they matter to you, and how much $$$ you want to spend to get that difference. It's a personal preference, and different for every combination of gear. Also HEAT! and different interfaces and multichannel converters respond differently. Especially those that split the signal of a multichannel dac chip meant to run in parallel and sum and average them for more accuracy in stereo. that's a big cheap out to keep the price down and most of the big name brands do it, thus the "mastering" converters and such that everyone is offering. The multichannel AK chips are gone so almost everyone has now copied Apogee in splitting up "channels" of the ESS chips like surround receivers but they're not selling 32 channels for 10k, it's cheaped out even more in your cheaper brand box.
so an analog summer for whatever analog distortion you want can lead to distortion from the typical cheap multichannel converters (yes the ones not insane money are all cheap and cut down vs the two channel offerings) that you don't want. they're not a dangerous convert 8 which is 4 convert 2s on the same board or a lynx aurora with more channels which just has more cirrus converter chips.
and this is just from the chips themselves, not the analog distortion from heat, circuitry, the older parts used in most summers, etc.
|
|
|
Post by jmoose on Jun 6, 2022 15:38:15 GMT -6
Even if stems are being separated they are still coming back in and being heard through a stereo output hitting 2 channels of conversion. So taking it out of the box in a summing mixer I don't understand how thats any different from taking it out of the box through lets say a hardware insert. It's a pretty confusing topic... ~~~ Workflow and ergo's is a completely different topic though. Knobs and gear is fun. Workflow & ergonomics are completely intertwined. IMO impossible to separate. With the option for both here sonics aren't much of a debate. We can stay ITB or go outboard through the little SSL sidecar. Both are cool. But my approach with each is completely different, and since they're made up of completely different tool kits then the resulting mixes? One isn't better then the other. But... maybe the workflow for a given project makes one a better choice then the other? Deadlines & such can sometimes pull rank over chasing tone and "the cool factor" but that's life. One of those methods stays inboard and doesn't use a single patch cable. The other takes 16 outputs and hits a 20 input desk with, very likely a couple dozen pieces of outboard hung of various points. Obviously different methods towards achieving the end result. Ultimately any similarities between an outboard & inboard mix boil down to the constants... and there's only a couple of those. One is the room. Monitors are the same and yeah, center section of the desk controls all those so that's the same too. And my converters are the same, only its 2 channels not 16. But the only other thing between those methods that's the same? Me. Very likely that any similarities are going to boil down to how I'm hearing things and my reactions. Elements like vocal placement against the drums... the kind of reverb/delays & general depth of field... how I hear that stuff doesn't change from ITB to outboard. I'm still me. Wherever I go there I am..? So while like everyone else I have preferences... for the crowd that's asking "If it makes a difference" - I'm not sure I'd go thinking about a summing box being a magic wand that "poof" is going to be an instant shot of horsepower. If you aren't happy with your mixes in general? Ehhhh... right. But if you might already have a collection of outboard, and want to use some of it while mixing... and it encourages a different way of thinking & working? And maybe that provides a superior, or at least radically different result? Now maybe 'ya got something.
|
|
|
Post by johneppstein on Jun 6, 2022 17:04:55 GMT -6
I've been wondering if I should invest in a summing mixer... If so, why? Here is a basic rundown on my rig: Manley Core UA 4-710 Revive- Vintage Audio M12 (Handmade API 312 clones) Iron Age Audio QPP Vintech Dual72 Langevin DVC Lindell 18xs Weight Tank WT-72 Weight Tank Comp Lindell 17xs JDK R22 KT-2A x 2 Drawmer 1978 Tonelux Equalux MOTU 16A Interface Most of the usual plugs from SoundToys, FabFilter, Valhalla, PSP etc So, lots of flavors for pres, EQ and compression. Live room treated well with GiK, legit mic locker. I've done some reading on summing and its hard to really nail down what is a quantifiable benefit. Like... should I just get better at mixing, or can summing actually make a difference with headroom, making a mix feel 'big', adding punch to drums, etc?? Or should I just do stuff like print drums through the Drawmer and the mix the Tonelux EQs for a little flavor? Edit - Also am I just trying to do something a good mastering engineer is going to do anyway? No, don't. They sound appealing but they really don't do anything a DAW can't DO. If you really want analog summing(as I do) buy a friggin' console! Not much more expensive unless you insist on an API, Neve, etc, but there are many that will do the job of the redundant (for me) summimng mixer.
I have a Soundcraft DC2020... 32 channels, cost me $6000. (Much more when new...maybe 50-60 grand.)
You mix - into the summing stage - which does its job, it sums. Master buss processing is done before this stage. It can be ITHB or in your rack - same thing, same result, all things being equal (which is another discussion).
Note that I'm NOT knocking the Silver Bullet. It's just that if I had one I wouldn't need the redundant (for me) summing features. OHOH, I have a whole rack full of gear, and then some....
|
|
|
Post by [email protected] on Jun 6, 2022 17:28:28 GMT -6
Things to consider.
Number of channels being summed is going to greatly impact the effect because of the makeup gain being asked from the summing amp.
Small differences between channels are going to introduce variance that’s hard to replicate. A few tenths of a db and so.
Pushing a summing amp hard is going to be similar to pushing a mic pre hard. If you do or don’t appreciate or hear the latter you will or wont the former.
Noise is increased with analog summing, right wrong or indifferent.
THD is increased with analog summing, right wrong or indifferent.
|
|
|
Post by notneeson on Jun 7, 2022 15:13:58 GMT -6
I've been wondering if I should invest in a summing mixer... If so, why? Here is a basic rundown on my rig: Manley Core UA 4-710 Revive- Vintage Audio M12 (Handmade API 312 clones) Iron Age Audio QPP Vintech Dual72 Langevin DVC Lindell 18xs Weight Tank WT-72 Weight Tank Comp Lindell 17xs JDK R22 KT-2A x 2 Drawmer 1978 Tonelux Equalux MOTU 16A Interface Most of the usual plugs from SoundToys, FabFilter, Valhalla, PSP etc So, lots of flavors for pres, EQ and compression. Live room treated well with GiK, legit mic locker. I've done some reading on summing and its hard to really nail down what is a quantifiable benefit. Like... should I just get better at mixing, or can summing actually make a difference with headroom, making a mix feel 'big', adding punch to drums, etc?? Or should I just do stuff like print drums through the Drawmer and the mix the Tonelux EQs for a little flavor? Edit - Also am I just trying to do something a good mastering engineer is going to do anyway? No, don't. They sound appealing but they really don't do anything a DAW can't DO. If you really want analog summing(as I do) buy a friggin' console! Not much more expensive unless you insist on an API, Neve, etc, but there are many that will do the job of the redundant (for me) summimng mixer.
I have a Soundcraft DC2020... 32 channels, cost me $6000. (Much more when new...maybe 50-60 grand.)
You mix - into the summing stage - which does its job, it sums. Master buss processing is done before this stage. It can be ITHB or in your rack - same thing, same result, all things being equal (which is another discussion).
Note that I'm NOT knocking the Silver Bullet. It's just that if I had one I wouldn't need the redundant (for me) summing features. OHOH, I have a whole rack full of gear, and then some....
The Silver Bullet has no summing features.
|
|
|
Post by jmoose on Jun 7, 2022 15:37:17 GMT -6
Things to consider. Number of channels being summed is going to greatly impact the effect because of the makeup gain being asked from the summing amp. Small differences between channels are going to introduce variance that’s hard to replicate. A few tenths of a db and so. Pushing a summing amp hard is going to be similar to pushing a mic pre hard. If you do or don’t appreciate or hear the latter you will or wont the former. Noise is increased with analog summing, right wrong or indifferent. THD is increased with analog summing, right wrong or indifferent. All true. At least from a quantum theory sotra level. In practice though? I dunno... some of the digital tools can have far more THD & noise then anything in the analog domain short of a RAT pedal. Thinking of things like tape & console emulations... various saturation plugs that resemble an effect more then anything else. Like if a tape machine or console was saturating that much in real life? We'd probably say it was broken. Plopping the 2 mix through a set of mic pres? Yeah I did the 1272 on the output of a smackie 8 buss thing a very long time ago... was an attempt to get some "tone" back into the picture. And if you like what that does its totally valid... but ultimately its an effect. Very obvious from the way it flat lines transients & crushes up when hit hard. That's kind of a different thing then running a real desk/box which, even when loaded down is quite a bit more linear. Unless you go for the intentional overload and run flat out, which most people won't do because it doesn't sound all that great. Not all the time anyway. Anyway... I find that typically at the intersection of practice & theory... noise floor? Between the inboard & outboard mixes... probably due to noise emulations in plugins there really isn't much of a difference. There's always hiss and other junk with either method. Digital is never 100% silent.
|
|
|
Post by Guitar on Jun 7, 2022 15:53:37 GMT -6
plugins don't have anything to do with summing.
%THD/N is totally under your control in digital, unlike "analog."
one person's practice is another person's "does not apply to me."
|
|
|
Post by drbill on Jun 7, 2022 16:39:28 GMT -6
Note that I'm NOT knocking the Silver Bullet. It's just that if I had one I wouldn't need the redundant (for me) summing features. OHOH, I have a whole rack full of gear, and then some....
Note : backing up what notneeson mentioned - the Silver Bullet is NOT a summing box. For your application, it would GREATLY augment your mixing setup - with a Neve and/or API and/or SSL master buss inserted into your Soundcraft. I've used the 2020, and it would significantly give you more options should you try one. I used to use one all the time on my D&R's which were notoriously "transparent" when I wanted some classic color.
|
|
|
Post by drbill on Jun 7, 2022 16:41:42 GMT -6
plugins don't have anything to do with summing. Haha! Indeed. At this point, we must conclude that the words "summing" and "stems" are hugely misunderstood by many if not most on the internet. 
|
|
|
Post by notneeson on Jun 7, 2022 18:59:22 GMT -6
plugins don't have anything to do with summing. Haha! Indeed. At this point, we must conclude that the words "summing" and "stems" are hugely misunderstood by many if not most on the internet.  Add reamping to the list.
|
|
|
Post by yewtreemagic on Jun 8, 2022 9:24:21 GMT -6
plugins don't have anything to do with summing. The Airwindows Console series is an exception, since it adds a calibrated amount of distortion to each channel, and then 'subtracts' the same amount of distortion at the point of summing via a separate buss plug-in, but in a way that aims to mimic the non-perfect action of the virtual earth mixing circuits found in the analogue world. At the risk of starting a forum bush fire, all I can say is that to my ears, they add glue and spacial depth to mixes, and a sense of space around each sound, and I now use them in every project. Here are some tests I did a few years ago - see if you can hear the difference soundcloud.com/yew-tree-magic/sets/airwindows-console5Martin
|
|
|
Post by johneppstein on Jun 8, 2022 10:03:16 GMT -6
The Silver Bullet has no summing features. Oh, good! I was misled by some of the other posts in the thread.
But to reiterate, I really don't see any reason for a seperate summing mixer. Exccept to seperate the unwary from cash that might be better spent on microphones, EQs, compressors, or (for that matter) burgers and whiskey!
|
|
|
Post by johneppstein on Jun 8, 2022 10:06:32 GMT -6
plugins don't have anything to do with summing. %THD/N is totally under your control in digital, unlike "analog." one person's practice is another person's "does not apply to me." #1 - of course!
#2 - I dunno about "totally under your control".
#3 - Coitantly!
|
|
|
Post by johneppstein on Jun 8, 2022 10:17:22 GMT -6
Note that I'm NOT knocking the Silver Bullet. It's just that if I had one I wouldn't need the redundant (for me) summing features. OHOH, I have a whole rack full of gear, and then some....
Note : backing up what notneeson mentioned - the Silver Bullet is NOT a summing box. For your application, it would GREATLY augment your mixing setup - with a Neve and/or API and/or SSL master buss inserted into your Soundcraft. I've used the 2020, and it would significantly give you more options should you try one. I used to use one all the time on my D&R's which were notoriously "transparent" when I wanted some classic color. I would LOVE one - but can't afford $2700 right now...
|
|
|
Post by jacobamerritt on Jun 8, 2022 12:14:16 GMT -6
OP here. Fwiw- I don't think a summing box is necessary for me- I have enough decent analog outboard I track through with decent mics, in a good sounding room. I do think experimenting with printing my final mixes through some outboard on the 2 bus is worth the effort. Ive tried things occasionally but never with much intention, and earlier in my path of recording with much less knowledge. Also- started some mixes last night for a band who did a (very mediocre) job recording themselves. Printed all the tracks through outboard after doing some editing/comping and corrective EQ. Drums - UA 4-710 Guitars and Vocals - IAA QPP from [email protected] (RE0887 input 4x N DOAs 4x Cinemag output) and KT-2As Bass - Lindell 17xs and WT-COMP from Locomotive AudioMade a significant improvement! This is something Ive done almost every time I mix something someone else tracked. It had been awhile since Ive done that, but was reminded its definitely worth the time.
|
|
|
Post by Guitar on Jun 8, 2022 12:33:23 GMT -6
plugins don't have anything to do with summing. The Airwindows Console series is an exception, since it adds a calibrated amount of distortion to each channel, and then 'subtracts' the same amount of distortion at the point of summing via a separate buss plug-in, but in a way that aims to mimic the non-perfect action of the virtual earth mixing circuits found in the analogue world. At the risk of starting a forum bush fire, all I can say is that to my ears, they add glue and spacial depth to mixes, and a sense of space around each sound, and I now use them in every project. Here are some tests I did a few years ago - see if you can hear the difference soundcloud.com/yew-tree-magic/sets/airwindows-console5Martin I can hear a difference. The difference "I" can hear, is the Airwindows mixes are way worse! They neuter the punch of the kick and bass sounds, and basically add nothing postive in return. To each your own. And I'm going to continue to be a stickler for technical details. The Airwindows plugins work "around" summing in the same way that Waves NLS, BX Console and others do with separate "channel" and "bus" plugs. But these are really distortion tricks, and in fact are "NOT" summing! They work around summing. These plugins don't sum in any way. Summing is basically a whole bunch of wires converging down to one or two wires. With whatever passive or active electronics and/or algorithmic DSP is required to make this happen correctly. The opposite of splitting or multing. That's it. It's not going to make you sound like Britney Spears and it's not going to get you a dinner date with Paul McCartney or Sting.
|
|
|
Post by yewtreemagic on Jun 8, 2022 13:16:31 GMT -6
It's a good job we musicians are all different - the world would be a boring place otherwise "I" hear more space and detail around those percussive vocals, while the sustained ones end up deeper into the mix. I've tried focusing my hearing on the kick and bass sounds, and (thus far anyway) I can't hear any change in them between the various versions, but I'll persevere. And while I totally understand what you're saying about the technical details (AFAIK, the majority of channel/buss plugins just add distortion at both channel and buss stages), I'd maintain that at least Chris Johnson is attempting to model his take on a flawed summing process, by 'removing' that distortion at his 'summing' stage to leave only the interactions between channel signals, which to my mind does at least mimic what an imperfect virtual earth mixer would do when it's not a perfect 'virtual earth' (i.e. not quite zero ohms), but instead changes small details in the combined buss signal in what many people consider a pleasing way. I like what it does anyway, but thanks for having a listen monkeyxx
|
|
|
Post by ShadowK on Jun 10, 2022 16:30:06 GMT -6
The other half of "everything matters," is "none of it matters." Too much emphasis on the first one, be reminded of the second one. And vice versa. I think obsessive audio types often skew too far in the neurotic "detail oriented" direction, which can quickly result in psychosis of some sort. You can look at some of the thoughts people have been having here as an example. I fall into that camp, fortunately I came to the simple conclusion that if something doesn't make an immediately noticeable dramatic difference it's not worth entertaining. It's amazing how one small thought changed my perception on everything audio related..
|
|