Summing Shmumming? Analog vs digital summing test

In which I finally tackle the issue of analogue summing: is this the secret magic ingredient thats been missing from your ITB mixes? Or, are we being conned by charlatans and snake oil salesmen? Lets find out...
Music: dan-worrall.ba...
buy me a coffee :)
ko-fi.com/danw...
Affiliate links:
Waves Dorrough Stereo meters: waves.alzt.net...
Dangerous Music 2 BUS + Analog Summing Mixer: tidd.ly/3tRjj3f
NOTE: this is NOT the unit I tested in the video; I havent tried this unit, I'm not endorsing it; but if you're going to buy one anyway I would love you to use my affiliate link above ;)
If you like this type of content and you want to see it more often, consider signing up for Channel Membership: / @danworrall
Video edited with VEGAS Pro 17:
{affiliate link)
www.vegascreat...

[@acollectionofdaves]

Love mixing with my Dangerous 2bus+ and I certainly use less plugins. I would argue that it helps having 3 analog processors on it which imparts more character without using plugins. If I were smart, i'd follow the link Dan has posted to buy one in affiliate links :) I also enjoy mixing in digital - just different workflows :)

[@DanWorrall]

Here's that link again :) tidd.ly/3tRjj3f

[@PWMaarten]

@@DanWorrall unfortunately momentarily not in stock.

[@DanWorrall]

@@PWMaarten that would be a problem, if I was actually expecting anyone to use that link ;)

[@Mikey__R]

@@DanWorrall I'm going to use that link, but I'll be buying a microphone instead. Hopefully you'll still get your affiliate kickback. EDIT: I should add, eventually I'll be buying a new microphone. I need to buy acoustic treatment first, and I've got to move house before then.

[@petristorres8647]

Cool 💪

[@Tekkerue]

Dan is the perfect blend of engineer, instructor, voice actor, and internet troll. 😂

[@gladejustglade]

And scientist. His logical analysis is pretty spot on.

[@robshrock-shirakbari1862]

I believe he's walking the line of being a bit too cynical and snarky these days in the spirit of being more controversial and edgy for his channel. He has a hard-earned, vaulted position above the Internet fray that may eventually suffer if he trades in his clever, British common sense and steady-handedness for American-ized, antagonistic "flair." IMO & YMMV

[@Tekkerue]

​@@robshrock-shirakbari1862 Maybe this is his true form and he simply works hard to suppress his snarky troll-ish nature when he is making videos for other companies like FabFilter? 😂

[@hyperactivists9390]

Haha awesome comment dont you love his snarky voice?

[@RealHomeRecording]

@@robshrock-shirakbari1862 so in other words if he goes too far someone is going to call him on his bull snit and he will go down an internet flames? Agreed.

[@Whiteseastudio]

Wow! Watched the full video! I can now put my “summing mixer investment” on the “when you’ve got nothing else to spend money on anymore” list 😂

[@MonkeyBars1]

ie the Never List Though we sincerely hope you have infinite budget one day fine sir

[@SuperIZL]

So yeah, there goes your Neumann desk superior summing..... Don't you notice any difference when in summing analog? Thnx

[@made.online2149]

>starts a series called 'snake oil' for software with unread manuals >easily duped into craving multi-thousand dollar hardware

[@patrickalphenaar]

Well it seems i wasted 2k on my Neve Orbit 5057 hahahaha. Call me crazy as i do prefer to Sum with the Orbit 100%, but probably for its saturation and natural compression when driving it hard. Question is it worth 2k? Yep at least to me it is……….. more Analog is coming like The Fusion and Neve Tape Emulation 542, we all need to spend it all before we leave this planet hahahahaha. Cheers

[@marcelosuarez3027]

@@patrickalphenaar your Orbit has a LOT of saturation and coloration options, so not clean necessarily. But i do agree with Dan, I much prefer an analog mixbus that I run through over a summing mixer, as it's the actual electronics you go through that make a change, not the summing device. So keep your orbit if it does help you mix faster and better!

[@Purpose_Tortoise]

This kind of ,,myth busting" is just what the audiophile and music industry needs. So many people get emotionally invested into things like their gear, or the way they play, compose, mix... And they can end up leading others away from the path of reason and into marketing potholes, and ignorance ditches. Good work, Dan! Thanks for making those experiments, so that we don't have to (although I would encourage anyone with enough time and equipment to go for it) ! Cheers!

[@kristianTV1974]

"Bigger problems than a shitty mix" got a belly laugh from me. Planes falling out of the sky..

[@mooseymoose]

It’s impossibly bad in guitar circles, the cognitive bias is unstoppable. Yeah sure that 59 ‘burst is worth $350000 compared to the damn near perfect reproduction from another brand name…sure.

[@CitizenOfEverywhere]

@@mooseymoose Complete agree. Rock'n'Roll has become The Antiques Roadshow.

[@mika461983]

cmon! let the people make money) but I'd like to see a video about "vinyl vs digital"!!!

[@KozmykJ]

I'd say it's what the audio equipment consumers need more than the industry 😜.

[@marshy_moo]

If computers weren't good at adding up numbers, we'd have bigger problems than a shitty mix!! 😂 Worrall is a comedian!!

[@mmcb5199]

That got me

[@lovebirdsmusic]

nah he´s english.

[@lumpyfishgravy]

Believe it or not I do run into problems with single precision floats from time to time. But they're causal and deterministic so can be fixed.

[@DanWorrall]

When just summing signals?

[@mmcb5199]

@@lumpyfishgravy are you fun at parties

[@JanSandahl]

I null tested this video against several of Wytse's Snake Oil videos - and they all nulled 100%. This proves that this is a genuine Snake Oil video. Well done, Den!

[@artysanmobile]

When analog summing became the latest fad, I had cause to smile. In the mid 70s, I was an engineer at MCI, where we worked very hard to make great sounding, ultra clean, high headroom analog consoles. Common knowledge at the time was that the damned summing buss was the achilles heel ruining all our efforts.

[@nambams]

So funny how things go around. I worked at a studio that had an MXP which I believe is related to the MCI and had used it daily for several years. We eventually got an SSL for a second room. I still remember the first time hearing the SSL how much coloration was imparted to the mix and realizing just how clean the MXP was!

[@chipsnmydip]

Thats kind of funny. These days with DAW recordings I have to use all sorts of transformers, tubes, and soft clippers to get a pleasing sound. I actually work harder now that tape and consoles are gone because there all kinds of pokey transients, awkward dynamics, and dull lack of harmonics that need to be dealt with before it sounds pleasing to anyone.

[@jc.1191]

Back in Black was done on one of the MCI consoles. I'd say job well done. Yeah, summing all those outputs must be difficult. My board solves that by having vcas for fader automation that make that a bigger issue than it's summing😂.

[@jc.1191]

I personally think console Achilles heels are the switches and interconnects. Same with outboard gear. My brothers working on a neve vr now where they're replacing switches with relays as part of restoration.

[@djentlover]

​@@chipsnmydip Distorted sound is just what's familiar to you, that's why you chase it

[@kasomeno]

A man who actually knows how to conduct things scientifically. Bravo Dan.

[@nibblrrr7124]

00:24 - 01:28 *Possible desirable imperfections* · 03:32 04:37 *Comparison pitfalls* · 05:35 06:36 *Tests* · 07:43 · 09:29 10:44 *Reasons to disregard* · 11:13 · 13:07 · 13:48 - 00:00 -- *Intro* 00:24 -- Intro: Digital summing is mathematically perfect. 01:28 -- *Desirable imperfections:* The possibilities 01:51 -- Desirable imperfections: Frequency response 02:12 -- Desirable imperfections: Output non-linearities 02:40 -- Desirable imperfections: Input non-linearities 03:32 -- Desirable imperfections: Consider the Waves plugin. 04:37 -- *Comparison pitfalls:* Level matching, Pan laws, Gain calibration 05:35 -- Comparison pitfalls: Colouration not from summing 06:36 -- *Test:* Mystery prototype unit 07:43 -- *Test 1:* Setup 08:35 -- Test 1: Listening 08:47 -- Test 1: Null test 09:29 -- *Test 2:* Setup 09:59 -- Test 2: Null test 10:44 -- *Reasons to disregard:* Specific unit 11:13 -- Reasons to disregard: Bias & mixing process 13:07 -- Reasons to disregard: Not high-end enough 13:48 -- Reasons to disregard: High noise floor 14:38 -- Outro

[@gomesdiogo]

mvp

[@michaelbadaltv]

EXACTLY! hahaha

[@terryshaifer6831]

Love your content!!! Separating audio engineers from audiophiles is a dirty job, but someone has to do it!

[@MichaelW.1980]

How do you separate audio engineers (with a certain budget) from being audiophile though? I mean, in the sense of the word, audiophiles love to listen to “good sound”. Whatever that is supposed to be can be disregarded, because it’s already highly subjective. And in a way, Dan admitted to be subjective himself. And because of that, I found it quite fitting, when in another comment Dan was called an internet troll. And I guess, he is on a good track of being able to get away with it because of him admitting his bias. All of this being said, I need to put in my agreement with Dan on being skeptical about analogue summing as a disclaimer. 😏

[@terryshaifer6831]

@@MichaelW.1980 The difference is an audio engineer can determine through objective testing, what is happening to the audio signal that could possibly impact why one likes or dislikes what they are hearing. An audiophile usually does not really understand the details of what is going on underneath the hood, and needlessly spends money on gear because it looks nice in their listening space or because it was refered to them by a source that they trust. They do not have the ability nor inclination to objectively test to gain a deeper understanding.

[@terryshaifer6831]

@@MichaelW.1980 Engineering of any type, is a skill that requires tools, understanding and experience to perform. Anyone can be an audiophile.

[@ToniDelgadoAbellan]

​@@MichaelW.1980 engineers are technicians, audiophiles are enthusiasts

[@shaft9000]

@@MichaelW.1980 It's the same as driving sports cars vs being able to fix and/or design them. You don't have to understand how a Porsche works under the hood to drive one. Just pay up. You can even be sh*t at driving and be an enthusiast, just like someone with terrible taste in music can also be an audiophile.

[@PDJMDS]

Cognitive bias is one of the most fascinating and challenging aspects of training/working as a mix engineer. Placebo test... Imagination is powerful.

[@leopoldbluesky]

I've lost count of the number of times over the years when I've tweaked a knob on a plugin or channel strip, decided I like the result, then realised I'm on the wrong channel 🙃

[@PDJMDS]

@@leopoldbluesky i remember watching a drummer endlessly tweeking attack and release parameters on a compressor, and it was bypassed. I just kept quiet 🤫

[@harrisbeatsfrankou6304]

Walk away from the mix and then batch mix a few is my best way...if I get obsessed on one track... it gets worse.

[@leopoldbluesky]

@@PDJMDS 🤣

[@flipnap2112]

in humans in general. entire swaths of populations under full management and control through the ghost of cognitive bias.

[@Zet237yt]

Just throwing in that I'm up for more snake-oil-like videos with this level of investigation.

[@datajake1999]

I second this.

[@TASTELESSAUTO]

I also second this.

[@natus49]

It's like Christmas every time Dan puts up a new video, except without the awkward small talk - just pure nerdery

[@woutervandijk9961]

Combined with a healthy amount of cynicism ;)

[@crunchwrapsupremium1053]

Christmas never sounded so clear

[@Orrinton]

Yessss 😂

[@FrequencySummoner]

I do like a good null test! The most interesting take I've seen on nonlinear summing is the airwindows Console plugin series. There are a whole bunch of versions and the setup can be inconvenient, but taking the simplest one "PurestConsole" to illustrate the concept: Multiple channels go through sin(x) into a bus, and on that bus the nonlinearity is compensated with asin(x). So a single input will null, but mixtures do not. I don't think Chris based this on a deep electrical analysis, more of a "hmm, how can I make digital mix channels interact?"

[@eyeball226]

Hmm. That would still be nonlinear even on a single channel, since sin(pi) is indistinguishable from sin(0) making it impossible to undo. Maybe I'm misunderstanding something.

[@nickskywalker2568]

@@eyeball226 yes, digital amplitude is constrained between -1 and 1

[@FrequencySummoner]

@@nickskywalker2568 @eyeball226 Right, and in practice I think this effect uses a fraction of that. The sine is being used as a gentle saturator.

[@wesmoore1559]

yeah, iirc Chris talks about some electrical behavior of multiple AC sources being summed (I won't try to butcher the description, I can't go that deep in EE). No idea what his algos do, aside from the serial effect + anti-effect chain, but they do something I've never heard from any other effect. A bit subtle, but really interesting. And especially interesting that the more channels you add, the more pronounced the effect is, thus making it a unique tool that can't be replicated by a single 2-bus processor. Dare I say more interesting/useful than the type of "clean" analog summing Dan used (and it's free!)?

[@eyeball226]

@@nickskywalker2568 While -1 to 1 represents 0dB, in my experience of DAWs using floating point maths the signal can far exceed this and remain linear as long as it is attenuated back to that range before reaching the output. You may be right with regards to this plugin but not as a general statement about digital audio. However if the plugin hard clips at pi/2 and -pi/2 that would make sense.

[@noodlezeep5159]

"You can reassurey yourself that your specific unit does have some magic, right? You've heard the difference, after all. It's not just cognitive bias messing with you, right?!" The burn 😂😂

[@OneFingerSnap]

Thanks for the mythbusting video. I’d love to see a similar video focusing on real world differences between different quality AD/DA converters. Having heard a few test projects I really can’t hear a difference, but I’d love to know your take on it.

[@drrodopszin]

A big plus one on this. I think the differences are so minor that you can correct it with a less than -0.5db EQ move. So why would I spend 1000s of $ for a difference I can correct with the first move I will make?

[@flipnap2112]

how about the difference between the ultra cheapies and the ultra expensive?

[@RobertKgma]

I noticed a big step up going from a $700 roland interface to a $1500 UAD Apollo X Twin interface. Less noise, tighter low end and crisper highs… and thats just on the output, no da/ad loops.

[@PWMaarten]

@@RobertKgma have you done a null test?

[@RobertKgma]

@@PWMaarten I didnt, nor did I have to. The difference was obvious.

[@geemcd]

I'm happy to watch to the end despite never having had any interest in summing mixers, or even being aware that they were a thing, because your videos are always great and I always learn something. You're a diamond Dan x

[@NikolausBrocke]

Software engineers will agree, of course. Producers who have invested a lot of money in analog gear will never agree. I am glad that with this video another very competent and regarded specialist prooves this fact. Thanks for your time and effort! For me that means once more that one can recreate the analog sound in the digital domain. Huiii, that saves me a lot of money. But you need a lot of experience, the right tools and the right ears to know how to recreate that "analog" sound.

[@BenedictRoffMarsh]

Thanks Dan. I am heartened to see more people questioning (or straight calling-out) the BS that has all but swallowed the music game. Saturation does indeed help a mix sound nicer. Skulbuggery does not. Quite the opposite actually. As you pointed out those records made with stats we'd piss on work so well as everyone in the room was focused on the music, not the gear. It is the brain of the person that matters. As I often say: I'd hire Alan Parsons in a leaky cowshed over Fiverr Freddie at Abbey Road every time. :-)

[@intheory3130]

"I'd hire Alan Parsons in a leaky cowshed over Fiverr Freddie at Abbey Road every time" Now that's a quote I'm gonna use. Nice one. 👍😂

[@Orrinton]

skullbuggery 😂

[@burgerbeatz6293]

As always, very impressive and entertaining. Love your ability to self-reflect all while keeping up the snark.

[@leomatarazzo.musica]

I am pretty glad with my totally digital mixes, I've never felt the need for analog gear.

[@zwsh89]

Ya this sort of myth busting video had me on the edge of my seat, what a roller coaster! Good format!

[@thegroove2000]

ANALOG. Its not always about the numbers. THE VIBES BABY. Noise, the inaccuracies, and so forth.

[@PuRe_AdDicT]

I didn’t realise transformer-less/colour free analog summing was something people even wanted ^^ Seems the opposite of what I would want to achieve. I guess if you had other color boxes in each path then a clean sum at the end is ok (for a DAW free stereo tape print maybe), but as you’ve shown you can just do that particular task in a DAW easily

[@steamer2k319]

Maybe they should be called "multi-channel analog (pre-amp) re-amp" units. Maybe call 'em "re-pre" units for short. If that were the focus, you'd expect multi-channel returns to be commonplace (to facilitate further per-channel post-processing and 'perfect' digital summing). Perhaps some models would offer an array of switchable color options (e.g., circuits with different saturation curves) for each channel. FWIW, I have no idea what's available/common on the market or how it's presented since I'm a penny-pinching digital home hobbyist.

[@harrisbeatsfrankou6304]

I know right? I use Kush transformers 458 and TWK usually and the sound is immeasurably better...I'm even getting my 8 track going into the daw for the pres eq's and tape sat. Then I can have 10 in and sent them to my Daw.

[@biasrocks]

Bottom line. I want color from my summing chain. The 2Bus+ I have here provides it and more. Wouldn't do a mix without it. I think a more valid test would be to mix into a summing chain and then try to recreate that ITB.

[@maedaeburning]

Right? Tf would I step into the analog realm for other than saturation (or workflow if you have that cash monie I guess)

[@kensmechanicalaffair]

@@maedaeburning It's good to have a few pieces.

[@production_facts]

I was on the side of analog, but this video has given me clarity and will influence my workflow. I can now rejoice in printing my stems back into my DAW, after summing, instead of using my hardware in realtime. Recalls will be a breeze! Thanks, Dan.

[@Mikey__R]

I enjoyed this! You put together a well defined test to exercise the hypothesis that linear analogue summing is different to linear digital summing, and to my satisfaction at least you proved they are the same. This is definitely the first video I've watched where the summing is tested rather than the preamp before the ADC.

[@lunalangton5776]

Wow, I'm amazed you can do a null test with analog gear and get that close to silence. Thanks again, your videos are always impressive.

[@MrJemabaris]

That's what I thought!

[@NamelessSmile]

I was surprised as well! But then I thought about how we're all so used to analog things in software being used to colour things so much, when real pro audio hi end analog is meant to be as noiseless and distortionless as possible.

[@hyperactivists9390]

yeah that rocked my world too

[@Alex_Martz]

Awesome! and brutal honesty for debunking analog snake oil and also for saying "I just made the video for view counts and revenue" 😄 Thank you Dan!

[@pittjpb8]

Thank you for your patience and thoroughness as always Dan! Thoroughness of the test itself, and also of your thinking and arguments. This confirms my suspicion that what I like about my summed vs ITB mixes is really just all the little differences between converter channels and resistor values, plus mostly the preamp following the passive summer. Still love the results, but this eases my mind for the times I need to stay in the box. Grateful you’re here and making such excellent videos sir!

[@OwenSome1]

Love this. It's a breath of fresh air. This conversation is usually dominated by those who have invested in loads of analogue equipment and need to justify it.

[@donnas8408]

And you think you’re happy because you can’t afford high end amazing gear to put hands on ha ha

[@Lutzifer31337]

i always wondered WHY people would want to SUM with an analog box and nothing i found on the internet convinced me, apart from the feeling that it might just add the right kind of saturation. But then i could just use saturation really. So thanx for making me feel smart for the first time, watching one of your videos for figuring something out myself. Still, i m amazed, as always, in the way you present weirdly niche topics in such an interesting and eloquent way. Love listening to your audio expertise!

[@paukin9344]

The most useful element of summing boxes us if you are running a hybrid setup and need analog inserts without going through another round of conversion which will add latency between your hardware and software

[@saftpackerl]

@@paukin9344 There is latency compensation in your DAW. If you use your analog outboard gear and print it in series, sum it in the box, there wont be any latency from ad/da conversion.

[@prodigalretrod]

I had concluded that analogue summing wasn't better _before_ I bought my summing mixer, but I did it anyway because that way I get to run separate analogue outboard on each bus before hitting the analogue 2-bus, so for me it was a workflow choice more than anything (though having a single roundtrip D/A/D conversion at mixdown is also preferable).

[@Cautionary-Tales-Band]

The main advantage I can see of something like a Cranborne 500 summing mixer is this - being able to buss everything to your sub mixes, process with whatever units you want, and sum to stereo without additional conversion.

[@artale4404]

I never thought about using analog summing before and I'm glad I didn't. I like the way my mixes sound without too much complication. Thanks Dan!

[@ejmikk]

I would love to see a video from you on one or some of the many current resonance suppressors / automatic mix balance plugins! For example Gullfoss, Soothe 2 or Smooth Operator.

[@RafaelBernatto]

Yes, please!

[@brailyndsummers]

Dan Worrall has a P.H.D. in throwing shade Don't know how in the hell I've made it this far in life without subbing to your channel, mate. You've always got great advice. You never fail to bring the spice; and for that, I subscribe. Good day, Sir.

[@LAxemann]

Dan: Gives a well-thought-through, honest, empiric and replicable proof and guides people through all the steps Random dude: Bullshit lol

[@LPChip]

Reading the text in the clickbait thumbnail, referring to White Sea Studios, I not only knew I wanted to watch this sometime, but I knew I had to watch it like instantly! :) Well done.

[@mudsh4rk]

What I'd like to see is null tests between Pro Tools 8 or 9 and the versions of a few DAWs that were current at that time. I've ogt a hunch that part of the "different DAWs have different summing sounds" myth comes from the period when most other DAWs had been internally floating point for a few years but Pro Tools was still fixed point. When every pan fader move, insert effect, or any other track processing in an entire mix is adding a generation of dithering (or truncation, I honestly wouldn't be surprised if at least older versions of Pro Tools used simple truncation) before summing even happens, it will probably sound a little worse than an otherwise identical mix created in a fully floating point DAW, and the actual summing math itself wouldn't even enter into it.

[@MikaelNitscheDrums]

This would be really interesting!

[@kelvinfunkner]

the final null was a surprise to me and one of the reasons why I just subscribed to your channel...actually...it was just the "nail in the coffin" so to speak...I've just watched the one about the Harrison 32C plugin and a few others on cramping too...this one just brought it to my attention that I hadn't yet hit the subscribe button...which equally surprised me...looking forward to binge watching more of your content 🙂

[@Hostile_Design]

I obviously know way less about any of this than you or probably most people here in the comments, but I'm mostly relieved by this video. I saw a video with Glenn Fricker where he pointed out the trick he found for fat guitars was ultimately to sum two sources analog before going into the DAW. I immediately felt that sounded like an odd statement, my basic understanding telling me the signals should just get mathed together regardless. So unless I'm missing something that would be specific to whatever gear he used, the only difference would be saturation, something I can add myself with better control via any good VST, and if so, fantastic. Less random variables and extra trinkets to worry about :).

[@nibblrrr7124]

It seems to be partly about the color of specific low-end Behringer preamps that has nothing to do with summing, but, *most importantly: workflow.* Thing is, while the analog summing sticks out as the most unintuitive bit, it's just one ingredient of a complete recipe that some seasoned engineers swear by because it just gets great a metal guitar tone quickly & reliably: Two mics (SM57 in a bright position, MD421 or other tom mic dark), both super close. Room reflections deadened by blanket fortress/gobos/vocal booth. Both mics go into an older Behringer Eurorack mixer - which one of the engineers who taught Glenn the recipe picked specifically its "creamy" preamps, over all the other, much more expensive ones he tried for metal guitar. Glenn says to watch the gain staging and not drive it to hard, judged by the indicator LEDs - not sure how much saturation you'll get anyway, but I suspect it's mostly about frequency response in the mids? Also note that while Glenn often does A/B comparisons, he doesn't bother here, but instead emphasises the workflow aspect in the end: "It's so easy to just dial in the mics and get a great tone and then not worry about it after. One of the added bonuses of this method is that *it forces you to make a decision right at the source and not be left constantly tweaking after the fact.* It's quite liberating to be honest." (BTW, the video in question is "The Greatest METAL GUITAR recording trick I ever learned" by SMG Studios.)

[@Hostile_Design]

@@nibblrrr7124 This made me realize that quite some time has passed since I actually saw this video, and when I did, I knew even less than I currently do :D. Thinking more closely, the thing I'm referencing was an answer from another video, responding to a comment based on the one with the trick, where he specifically points toward the summing inside an analog box as the important part where the commenter was opposed to that idea. But I can't swear I remember this properly and I should probably go and re-watch the content. Also, I think it was clear but just to be sure. I of course didn't mean to throw any shade towards Glenn. I've learned a lot of good things from his channel :).

[@MrJC1]

@@Hostile_Design glenn doesn't always talk sense.

[@matthewgray713]

preaching to the choir here. I always had suspicions that it was the non-linearities and component tolerance differences in consoles or summing boxes that imparted additional depth and width but then I would also argue that you could get similar results summing digitally and then mixing into or mastering through analog hardware. Thanks for proving the point!

[@robertlofgren6834]

Dan, I’m happy to learn that you have an x32. Given your mixing skills it would be fascinating to hear what you could accomplish mixing only on the x32 alone (or the Wing for that matter), without resorting to external resources. Cheers, /-robert (Mr.X32)

[@ViktorNova]

I was happy to hear this too =)

[@threepe0]

So glad I came across this. I was thinking about your channel the other day, kicking myself for not having subscribed after the few videos I had seen. This morning I was going to try and find the channel in order to sub, and this popped up.

[@ViktorNova]

Rad video as always - though you didn't answer the question that I think would be the most relevant to the most amount of people (who would never consider buying a summing mixer): What about the digital summing "console" emulation plugins? I've been mixing for years and only recently have been exposed to this. Would love to hear your in-depth take on pugins like Airwindows Console8/Console8 Buss (free), Acustica buss plugins (as included in the Sand suite) meant to go on every channel, etc. for the purpose of "emulating analog summing in the box"

[@razhmazh]

I know a couple people that would need to see this. Saturation does, indeed, sound good!

[@ashleykoukou]

Hi, Dan, I've done a BIG summing test last year. I use 7 pairs of stereo stems（14 outputs）+ 36 different ways including 4 large format consoles : SSL 4000G+ / SSL 9000J / API 1608 / Custom Series 75 Console（Neve）. In digital summing, I insert analog emulation plug-ins like bx_SSL 4000G / Waves NLS Channel...etc. (with no EQ or compression, only input and output saturation) to avoid the situation you mentioned : the saturation make analog ones sound better and richer. All channels are carefully calibrated with test tone first, and tweaked the gain settings so that they were returning to the DAW at the same level, same as you did. All summed tracks are matched to the same LUFS level (-20.1 LUFS) , so no one sounds louder than another theoretically. When I listen, I use HOFA 4U+ Blind Test Plug-in to avoid any bias or analog myth to mislead me. Hear are the 36 players in my summing test： 【Player List】 （Player number does not match the list oder below） 【Team Software】 《DAW》 Ableton Live 10 Cubase 8 Pro Tools 2020.12 Pro Tools 10 Reaper 6 Logic Pro 10 《Plug-in（Pre-amp only, no EQ / Comp）》 PA bx_console_Focusrite SC PA bx_console_Lindell 50 Channel（API） PA bx_console_Lindell 80 Channel（Neve） PA bx_console_SSL 4000G PA bx_console_SSL 9000J Slate VMR Virtual Channel Ψ（Trident 80B） UAD AVALON 737sp UAD Manley Vox Box UAD Neve 1081 Waves TG Mastering Chain（TG12410） Waves NLS Channel_MIKE（TG12345） Waves NLS Channel_Nevo（Neve 5516） 【Team Hardware】 《Summing Mixer / Summing Box / Small Format Console》 Dangerous 2Bus+ Folcrom Passive Summing Mixer KAHAYAN EPSILON 32-500_Output Preamp = Clean KAHAYAN EPSILON 32-500_Output Preamp = Brent Avrill 1272 KAHAYAN EPSILON 32-500_Output Preamp = Burl B1D KAHAYAN EPSILON 32-500_Output Preamp = SSL 4000 Clone Manley 16x2 Rupert Neve 5059 SSL Six SSL X-Rack SSL XL Desk STUDER 961 《interface / AD / DA》 Metric Halo ULN-2 DA → Metric Halo ULN-2 AD Metric Halo ULN-2 DD → RME ADI-2 FS R DD → Metric Halo ULN-2 《Large Format Console》 API 1608 Neve Custom Series 75 SSL 4000G+ SSL 9000J -- Here is the summed wave 48kHz / 24bit files link if you are interested（the song is from my album released in 2013, Taiwan)： drive.google.com/drive/folders/1iiJmmG2ZOQQm_qPyHk0qLEGSvF8ER51g?usp=sharing -- Please message me for the answer : facebook.com/MixBetter

[@Alleroc]

That's the same conclusion I came to. Got to borrow a summing mixer, did the same type of test as you, and got the same results.

[@ledheavy26]

As some others have mentioned it would be interesting to see your take on Studio One's mix engine FX, the console shaper plugins specifically. They seem to be unique in their way of working, being integrated at the mix engine level "behind the scenes". Whether or not it makes a bunch of difference compared to having something like Waves NLS on every channel, I'm not equipped well enough to know.

[@kilkaserco]

The Mix engine is really interesting because besides stuff like saturation it does channel bleed which adds a really 'analogue' factor. I don't necessarily mean in a sound way, but in a way that it presents you with issues you aren't used to dealing with in digital. It's fun, and I guess that has to be a big part of it all for us.

[@olivergretz]

Thank you so much! I’m also convinced that the differences one hears with analog summing mainly stem from the tolerances of the components and the imperfections of the input/output stages, not from the analog summing itself.

[@AlexCornier]

I would love to hear your take (and by that I mean one of your masterclasses) on the difference between lower and higher end converters.

[@DanWorrall]

I bought a TV for the living room not long ago, and belatedly realised it didn't have a proper line output. So I bought a cheap £10 DAC to plug into the spdif outs. Honestly, it sounds great! We're living in a golden age of affordable excellent recording gear :)

[@digidope]

@@DanWorrall Everytime one thinks of buying new shiny converters, one should watch good ol' Monty explaining D/A with (now 20 year old) usb soundcard. Spoiler: converters were already then that good that he can't get the sound to change when passing trough AD-DA :D kzbin.info/www/bejne/mXq0anyOiLqtq68

[@jonoelke7944]

Dan has a great relaxing voice. I love the presentation of the information. Now I am so calm I will sleep well tonight.

[@jamesbannon1057]

I've always been sceptical about the claims made for analogue summing, once signals are in the box, especially nowadays when digital signal processors are available.

[@sub-jec-tiv]

This was so helpful! I feel like we have such a great choice these days. Between roughing up sources with hardware before they go into the box, to just running out a source for re-amping or what have you. You can get such a great ITB mix just being careful about how you print or process your individual channels (or busses). I do think that on desks/mixers there *can* be other odd things happening like transformer coloration and channel crosstalk. So while the actual summing question seems solved to me, it does seem like it is within the realm of possibility that different analog mixers *can* impart pleasing nonlinearities. Case-by-case of course. I would guess that if you ran 22 channels out of ProTools into the desk that used to be at AIR Montserrat, you would end up with a beautiful sounding end result. But I absolutely don;’t think that’s due to ‘analog summing being better.’ It’s likely more like the transformers in each channel, etc.

[@maximtheofanides7961]

I'd love to see Dan take a look at the S1 MixFx to see what he thinks. That's a per channel basis with cross talk, harmonics etc.

[@rautshsale1948]

was thinking the same

[@superlozen]

Hey Dan, thanks for yet another awesome video! I recently got a new interface and took out the passive summing mixer i built a few years back out of the closet for a new test in hopes that I may start using it again, as I have even more io to use now. My first test was to see if the claim that a passive summer, which is basically a load of resistors, does give a non linear response and check the more headroom claim - can you believe it, no matter how hot i went out of my interface, what i got was exactly the same sine wave, regardless of the frequency, no harmonics, absolutely no change, except the noise floor offcourse, which because the unit is balanced was incredibly low. Next i wanted to see if the summing on a voltage level will indeed change the sound stage and pan perception - initially I was fooled to think that yes it does - as the unit I made is a simple 4 channel unit - 2 mono ins and 1 stereo pair - very simple, very much made to make some fun with the pan right, well actually as with your tests, other than the slight differences between the 2 channels in level - 0.1db - which i accomidated for digitally post summing, I found that it does not actually alter the sound stage at all. So I would completely agree with your test that, a 2 bus processor might be more handy than a passive summer going into transparent preamps. I also took out an old passive summer with 2 transformers on the outputs, which i made during the same time period, and that did saturate nicely with mostly odd, but also some even harmonics. So the magic appears to all be within the transofrers and active circuitry and not in the summing at all. All the best, MBY

[@G_handle]

Two plus Two is Four. Every time. 1) The fixation on the actual “summing” was always misplaced. I think most serious players realized that at the start of the “Analog Summing Cult” era. In defense of Waves (a sentence I’ve never uttered before. Waves business model seems to be “how can we be less defensible?”) (( No F…ing way! As I typed that, a KZbin notification just flashed my iPad with “What’s new in Waves V14”. They really are the worst!)). Wow. Okay….. …in..defense…of…Waves, when they released the waves NLS system, I read and watched videos from them basically stating that In their testing they found that the “magic” was Not in the actual summing, but in the subtle Non-Linearities within different analog circuits, And the minor non-identical differences between each of the channel strips in an analog console, before being summed together, by yet another non-linear analog circuit. That’s why the system included 3 different console models (SSL 4000G, EMI TG12345, and Neve 5116) each modeling 32 Different channels a piece, and capturing their frequency response curves, saturation / harmonic distortion, and the noise profiles separately. And again for the summing circuits in each console. 2) To my understanding and experience, the irony has always been that the big analog console manufacturers themselves spent decades chasing after essentially perfectly flat frequency responses, zero distortion, bottomless noise floors, no crosstalk, and perfect summing. And once digital summing gave perfect to us, both in Digital Consoles and in DAW Mixers, we supposedly wanted the “more than perfect” analog summing back again. Go figure. 3) There’s always been a divide within the audio industry between the engineers advancing the technology, and the marketing departments selling us the products. Some of which is definitely Snake Oil. But every good lie has some truth to it and everyone that mixed in analog, really did hear a big and undesirable difference making the transition to digital. What was causing that difference, or ”Problem”, and what to do about it became both a Challenge for the engineers, and an Opportunity for the snake oil salesmen. The “Magic Box” was born. And the more that box cost, thousands of dollars, and the bigger the name brand logo was printed on the front, the more it must have been infusing your mix with unicorn urine and dragon dung. 4) Facetiousness aside, and the thing I wanted to include into this discussion, is that all of this decades long debate over “Analog Summing Magic” wasn’t for nothing. I think Dan even mentioned in this video, that it forced those of us resisting this cult like following to think deeper about what really Was happening within those Analog circuits. Today I think there’s much more discussion and understanding of distortion, saturation, odd & even harmonics, etc by the Audio Engineers when that used to be mostly in the purview of the Electrical Engineers. The guys who use the stuff think more like the guys who make and fix the stuff. Many engineers now reach for saturation Before or instead of reaching for EQ. This isn’t all in the name of debunking analog summing, but as part of a whole, the old “you don’t know what you’ve got till it’s gone” lead many to investigate the summing part of the analog mixer, which spurred many other investigations, and ultimately deeper understandings. 5) So the Waves NLS system (with unfortunately Summing being in its name, but ultimately being more about the NL part) for me fits into a package of “products” that all essentially attempt to make the “Perfect” Digital Mixers in every DAW, behave more like the “More than Perfect” Analog Mixers in the magazine photos from our rose colored past. DAW even build some of this in. Pro Tools has H.E.A.T., which started life as Phoenix. Studio One has Mix Engine FX. I’d also argue that the PA Brainworks “tmt” (tolerance modeling technology) behaves very much on the same principle as NLS. And I guess the most obvious of all of the products in this category, are the subject of DW’s other recent takedown videos: Channel Strip Plugins. 6) I have Most of the channel strips from Waves, Plugin Alliance, Universal Audio, etc, as well as Softube’s Console 1 and SSL’s Native Channel Strip & Bus Compressor systems. I’m way deep into the “turn this DAW mixer into a Board” camp. I also have FOUR actual Analog consoles, that I absolutely love for working Out Of the Box. Each has an entirely different sonic flavor, which is transferred into whichever project I deem appropriate for each piece of machinery. And then, like everything analog, I can patch anything into anything else, tubes here, transformers over there, transistors, DOAs….. Add to that, and don’t tell anybody, I actually Built 4 little Summing Boxes (DIYRE they’re like $50 bucks, try it) that are completely Passive, have zero controls, no power, nothing. Just two 8-channel DB25 connectors feeding two L&R XLRs, 16 summing into 2. To my knowledge, there is no cleaner Analog Summing than these little boxes. Obviously I don’t know what Dan was testing, nor have I done That exhaustive of testing myself, but I’d be willing to suspect either the same or even lower noise Nulls from these units. So if it’s exactly the same summing digital or analog, what’s the point? If anybody here is actually interested I’ll write up what the point is for Me, but for now let me just say that: What Dan just did for me was Confirm what I had hoped was the case. He encouraged rather than discouraged me into using my Analog Summing. If the summing itself is indeed irrelevant, then I can choose freely whether or not to use it based upon the million Other, far more significant factors in deciding when and whether to work digital or analog, in or out of the box. Or both!

[@ts4gv]

Quick turnaround for such a lengthy comment lol Good writeup, interesting read. Thanks

[@LightWthoutTheStatic]

I've always wondered about this. Fascinating. It was just about color and distortion, not the summing itself.

[@jmalmsten]

It is interesting to me that the reasoning of audiophiles so often resembles argumentative tactics used by religious apologists. Logic, math and evidence say one thing. They say the same thing, even. But subjective experience and pre-held beliefs always trumps reality for the true believer. I say that, as I await the delivery of my first analog Moog setup. But I try to be honest with myself. I want to try it out not because it's the best way to produce the sounds. Part of the appeal is the irregularities of analog components not quite exactly lining up. And I just enjoy twiddling with knobs and cables overall. I liken it to discussions with sword enthusiasts. The Katana may objectively speaking be a pretty bad sword design born out of less than ideal metallurgy circumstances and a isolationalist and elitist culture. I just find them more appealing.

[@lennartrauhe]

Nice one! :) It's very nice to finally dispel all doubts about the actual summing part itself. To me the coloration that it adds sounds absolutely beautiful! It seems to sing within the song, the null test against the digital version was exactly what emphasises my emotional response music! That was awesome to find out! :) To me, that would be worth a summing mixer, especially when I can tailor that by the individual input levels. It is not an early investment tho.

[@lennartrauhe]

I'm still trying to figure out, why the top-end transients on the hats and stuff felt so much more pleasant on the analogue, since there was no high-frequency content in the difference.. must be about the added density in the midrange, maybe that balances out the tops somehow 🤔

[@joewohlitz5616]

Hey Dan. Thanks for your testing. In the first run of the test you stated you ran the stereo digital stereo summed mix through channels 1+2 of the summing mixer. Can you tell us if that would null with a digital summed version that stays completely in the box(bounced itb and never leaves the computer)? In testing I've found that even running a stereo pair out of converters and into two channels on a summing mixer is beneficial to the mix. To be fair, I have not tried this with a passive/transformer-less/colorless summing mixer. In the second test, it seems that you ran everything out solo'd in pairs and then had those summed digitally vs the 8 pairs straight into the summing mixer. Those nulled which I am not completely surprised about because if I understood correctly they went out of the converters and then into the transparent summing mixer that has no transformer/harmonics and then back into the DAW. I would like to know if a straight itb summed bounce(bounded itb and then never leaves the computer) would null here as well? I am also biased because I do believe converters, summing, and transformers positively impact audio and mixes from my years of use =).

[@DanWorrall]

I haven't actually tested those specific converters, but I would expect some phase shift from the converters which would break the null.

[@joewohlitz5616]

@@DanWorrall Thanks for the reply. Any way you can let us know what the summing mixer is when it is released commercially? Would be curious to check it out. We do have a Dangerous 2 Bus + in the studio being routed to a console and hope people will use your link because that thing is great!

[@unifinful]

@@DanWorrallI'd also be interested. Since you wanted to eliminate the saturation, didn't you also force it trough a summing stage? If you solo and record master, it went through summing. If you use separate aux, these have a summing stage as well. I think direct out wouldn't pass a summing stage, but probably wouldn't count, as it only passes through the preamp, but not the other coloring parts, like eq or comp.

[@waterknot1]

In the early 2000s, I had the opportunity to mix an album and sum it in 3 different ways...In the box, on a custom API console, and on a Trident 3B. We carefully calibrated each situation. In the end, when it came to choosing versions for mastering, they chose about 1/3 of the songs from each of the summing methods. So ultimately, the choices had more to do with mix tweaks than it did with the summing method. From then on I haven't bothered with external summing. It just doesn't make enough difference to warrant the effort.

[@nj1255]

3:16 I can only repeat that sentiment: why spend money on a summing mixer when you can buy gear that actually does something useful? If you only want the crosstalk (something that is largely unwanted in analog gear) you can easily replicate it ITB for free.

[@yanguimedia]

Scientific method, a rigorous analysis and a critical mind accepting results whatever they may be. It's so refreshing in an audio world full of myths and legends ... 👍👍👍

[@greenlobsterrecordingstudi3033]

Interesting! Although by presenting an especially "clean" design you are maybe not taking into consideration that f.ex. the active summing network of classic consoles, with the feedback shunt through a big row of resistors (yes, noise) and into whatever the output stage before that resistor consists of will have interactions between the summing amp and the output stages that a "modern and clean" summing box which is connected to the modern and immission resistant outputs of a converter - probably buffered by yet another extremely low Z opamp stage - does not have. Consoles have factors like crosstalk, power limitations, all kinds of complex behavior a summing box will not have. You could call it "dirty", I call that charming and I've not heard that simulated convincingly by plugins. Also, one of the very important reasons to work with external summing is that the signals stay in the analog domain until they're converted back, after summing amp and master bus compressor/EQ etc... (coloring, usually). So I'm saving two stages of AD or DA conversion, and yes you can hear (and null-test) converter differences. Also, with your test setup, each channel separately routed through the full device "stack" including the amp, will still experience whatever difference the circuit makes. Of course, the "perfect" part of the whole thing, the summing itself, should then make no more difference. Everything else would be illogical. That point is proven, thank you for that ;) Anyway, as usual, great content! Cheers from a hybrid studio with all kinds of summing, digital or analog, which we combine for different reasons (e.g. latency) and regularly blind-test against confirmation bias 😁🤘

[@G_handle]

I 2nd!

[@pyratellamarecordingstudio1062]

This is not a video about consoles. It’s about transparent summing mixers and how they don’t do anything. I believe he made that pretty clear. Of course any mixer that has any kind of saturation is going to yield different results. Its the master buss compressor and other analog gear that make a difference. For example you could sum everything itb and run 2 channels through your outboard and get, exactly the same results. There can be some real magic in a summing mixer that has transformers on every input channel and mix out. But then again that’s not what this video is about.

[@greenlobsterrecordingstudi3033]

​@@pyratellamarecordingstudio1062 I know what it's about, but I tend to disagree with the idea that every summing mixer SUMS the same way: Depending on the circuit, the coloration comes maybe not only from the signal chain before/after summing. Physically "imperfect" summing theoretically will introduce nonlinearities and interactions BETWEEn the sources. It would need more examination on the topic, but if you understand a summing circuit including all potential factors, like parasitic capacitance affecting the negative feedback loop, channel crosstalk etc, there will be something going on which OF COURSE doesn't happen in a physically perfect summing. I still agree that most products on the market will not make anything better :D 64bit float digital summing is, of course, "perfect". (Btw: People who've worked with DAWs 1995ff will remember a time when this was CLEARLY not the case....)

[@ToddBeal]

Dan, I trust your analysis. I love the work you put into your final conclusions. Keep it up.

[@wazzpqazzza]

Okay, amazing. It's always so satisfying to see you work with such precision and I am genuinely impressed by the result. I had assumed that intermodulation would play a passive role in either case just through the copper wire but it appears not. What are your thoughts about digital summing clones that actually apply to tracks, busses and outs such as AirWindows Console series? I understand now that they work by applying saturation to every instance and then the effect on the master buss apply desaturation which gives the intermod effect of older consoles and that besides that there would be no difference. Never used them because I'd never had a job that would improve it.

[@ts4gv]

I love how you ended the video, from 10:44 onward. reminds me of the final battle in 8 Mile where Em lists off everything wrong with his life in his verse, so his opponent has nothing to attack him on. (Corny reference, I know)

[@TrippyTechnologyMusic]

Dan is indeed, The man

[@connorharris1900]

I really appreciate your contributions dan you bring clarity and truth

[@BojanBojovic]

Gear religiosity destroyer! :) Dan, you are a genius. :)

[@descendinguniverse]

Oh I remember exactly how I was obsessed with that analog summing and different hardware units but stopped in time

[@patrickalphenaar]

Well Analog Summing is indeed very Subtle. For sure you can do all in the box! However i still love my Neve Orbit 5057 for it saturation and coloring especially when driving it hard its saturates very nicely, it gives me depth and Air and separation, altho cabling is very important! My whole Studio is hooked up with “Mogami” and it makes a huge difference in sounding very natural which i cant get in the box myself. For sure other people can and even do it better but i still prefer analog summing for sure. Its just a personal preference i guess. Yea The Orbit stays were it is and be used for all mixing and Mastering i will do……… digital is kust too perfect for me sometimes. Nice vid and loads more to be debunked! Hahahaha Cheers

[@Sebastianandthedeepbluemusic]

got one of those too, and it sounds more better for everything I have used it on. I tried a null test and I found the side signal to have larger differences and middle channel had some smaller differences

[@patrickalphenaar]

@@Sebastianandthedeepbluemusic yea agree, it gives me some sound signature i cant find in the box……..

[@Sebastianandthedeepbluemusic]

@@patrickalphenaar yeah well, I mean, it is in a box haha. like a neve box. but still a box. and it is in there. JK JK yeah I still like it too. and even if you can get the sound to null with a digital mix hitting the transformer. well either way you have the analog transformer. anyway I didnt sell it yet. prob will keep it ha

[@saftpackerl]

Im not sure you got the point of this video.There is no difference in just linear summing. analog or digital. Nobody questions the nonlinearities your box adds to the signal. BUT: you could buy two channels of that aswell and Run your stems through one after the other and sum digitally with the same effect, and less money spent... It may be more time consuming, but 2 channels should be a lot less expensive than 16/24 channels! plus converters!

[@Sebastianandthedeepbluemusic]

@@saftpackerl I did get it I think

[@MaxCarola]

And to add a few little bits about level matching. I had once a seller to run aneat test of the newest interface that was to be a far better replacement of my old avid I/0. The run the test and the new one indeed sounded slightly better. Then I asked to match the levels and, guess what, they NULLED perfectly. Strange world is the one we leave in.

[@kylembagley]

Jokes on you, I had my fingers in my ears the whole time

[@Mnr.222]

Use to mix ITB, then bought a neve 5059. It didn’t take long to see how unnecessary it was, so I sold it. Back to ITB now and life is much easier. Still use outboard compression/eq for individual tracks, premaster or sub groups but I’m so happy I got rid of that thing.

[@TransistorLSD]

Well you can push a Neve for harmonics, but... You can also push Neve plugins on your busses and get 99% of the same effect. Maybe 100% with oversampling.

[@zwsh89]

Great experiment and analysis! One angle you didn’t explore is the argument that what is special about analog summing isn’t the pure clean summing, but instead how when tubes and transformers are integrated into the actual summing process, not pre or post, but are used in the summing stage, that the analog summing units would behave in unique and unpredictable ways that digital summing, or summing clean and running the signal through the same unit for that matter, can’t replicate. Even with the best emulations of the same exact components, maybe summing is still such a complicated and delicate process that digital emulations can’t quite pull off the intricate behavior of more colorful units. And if this special characteristic behavior is reliant on the intermodulation, as you speculated (which I suspect is correct), running the signal through the unit after clean summing of any kind, digital or analog, wouldn’t suffice. Yes, computers add numbers well, but when the process gets complicated, analog units don’t care how many zeros and ones a complex audio signal would require it to crunch. They just pass voltage from point A to point B reliably. This could lead one to conclude, or at least hold out hope that the behavior of certain analog units, when the colorful components are engaged, truly do produce unique, superior results. Testing for this would be harder, but not impossible. I’m totally speculating, I too am just genuinely curious if there is an angle to analog summing that truly justifies it, or if we’re just letting various companies’ marketing departments get us all turned around over what is really just a simple concept in fancy clothes.

[@anteshell]

"One angle you didn’t explore is the argument that what is special about analog summing isn’t the pure clean summing." Didn't you watch the video. He did explore that argument at 05:35. He didn't go very deep into it, since analyzing analog summing was not the topic. Just analog summing compared to digital, for which the explanation was adequate. "They just pass voltage from point A to point B reliably." This does nothing more than give analog music essentially infinite sample points as opposed to lower sample rate of digital music. This doesn't give the music any audible coloration. It just gives digital audio slightly higher noise floor. The effect is basically digital noise that comes from lowering the bit-rate, but any consumer lever gear is adequate enough to keep it so low that it's completely inaudible. The coloration you talk about after that quote is completely separate thing that comes from the exact opposite than what you're thinking about; the unreliability of the components passing the voltage through them.

[@NealMiskinMusic]

Yeah my thoughts exactly. Summing in the box is just fine. If you're going to invest in something to sum your audio in the analogue domain you might as well get yourself a mixer. With a mixer at least you get EQ and faders, plus some extra routing options, and you can adjust your gain structure if you want some saturation on specific elements in your mix. But the summing buss itself is basically just a buss bar and some resistors.

[@KingGrio]

I work at an audio company and I'm a lowly tester. I'll ask my egghead colleagues doing the designs and DSP, but honestly without having heard their answer I'm expecting to hear from them that analog summing has no magic sauce for the reasons you explained in the intro. If there was any magic sauce, you'd get it from non linearities entirely independent from the summing operations and you could just model them as a standalone saturation/distortion/coloring unit, there's no such thing as colour intrinsic to summing. This video reminds me of Adam Neely's video about non-equal-temperament tuning systems where the internet was going crazy over A=432Hz and then the relationship of all the other notes to A depended on which crackpot you asked. It mostly all sounded like crap, but to Adam's credit, he did his best to make some of the tunings sound better by playing melodies targeting the better intervals the tuning offered. He concluded though that even then, it restricted what you could play. I'd be lying if I said I never believed whacky theories even for a while, but I'm amazed by some of the myths that gain traction on the internet.

[@simongunkel7457]

A=432Hz is a 12TET, though. Non-TET systems have been around far longer than TET and you can make the case for period-appropriate tuning systems for older music. And you can also make a case for a variety of non-TET microtonal systems. Any tuning system is a compromise, you always have to figure out what to do abouth the Pythagorean comma. 12TET isn't the best solution in every case. If what you took away from Adams video was that 12TET is somehow the optimal thing, you should rewatch it (I know that fairly recently Adam had some music with 24TET in it).

[@KingGrio]

@@simongunkel7457 These days I think we use A=440Hz and equal temperament, but yes, you can define A=432Hz and then build the other notes in relationship to that A with equal temperament as well. And yes, I didn't mean to say equal temperament was the one and only tuning. But as you mentioned tunings are about compromise, and equal temperament is a really good one. I wasn't saying microtonal music is to be dismissed. I'm also pretty confident (though not certain), that when a guitar player bends a note to a certain pitch, he might be doing it by ear to what sounds right and if you measured it, it may occasionally be slightly outside of equal temperament, intentionally. What I was criticizing though, were the whacky "shakras" tuning that just sounded like your piano was broken. Unless you're going for a lo-fi, broken piano vibe intentionally, which I guess can work in a specific context. What came out to me though, is that equal temperament is a very strong system, and all that hype on the internet at the time about other tuning systems was just baloney, for the most part.

[@simongunkel7457]

@@KingGrio Well, 12TET at A=432Hz is baloney, but things that are not 12TET aren't. Guitar players from the blues tradition don't occassionally bend notes outside of 12TET, they do it pretty much always. Jazz singers tend to hit notes a bit flat as part of that singing style. 12TET really was built to ensure one thing: You could freely modulate and transpose between keys without changing the harmony. That evolved in tandem with the musical style of 18th century orchestral music, which made this type of modulation central to its structure - sonata form demands it. If you are writing modal music like in the Renaissance, it's not a good system. It's kinda baked into MIDI as well, which can be an issue, if you want to explore alternatives, because you're looking to do something it wasn't designed for and want to optimize the tuning system for your music.

[@Anteflop]

Appreciate the straightforward, no BS video as always. Summing mixers by themselves aren't anything to write home about. Companies try to compare them to consoles and that's just laughable. The summing of a high number of unique channels, with all their quirks and non-linearities is what helps give console recordings that sound. An analog thru-box is going to yield nullable results like you show here. I have a clean summing box and use mine as a utility for putting choice hardware pieces on groups. The mix goes out of the various stereo groups and comes back in summed. That way I don't have to mess with returns for those groups or a patch bay etc. Ultimately, it's an expensive way of putting some hair on a track. I would definitely say, even with top of the line hardware behind one - a summing mixer is not the secret to a good mix. It's of course all in the ears. Keep em comin' Dan!

[@TheZappazapper]

I think the idea of analog summing comes from the same place as the resistance to amp modelling - it's the kind of thing that was conceived in the earlier days of digital, and it might have had some benefits then beyond the inherent saturation of analog gear, but just like amp modelling is much better than it was in the year 2000, digital summing algorithms would most definitely have to have been improved over the last 20 or so years, and just like some guitarists will never let go of the "amp modelling sounds like crap" position that they took when they heard year 2000-era amp modelling and never allowed modern modelling to change their minds, I'm sure there are those that will always use summing mixers because back in 2000 when they went digital, there WAS a difference and now the gear is in their studio, paid for and all hooked up the way they like and it's now part of their workflow, whether it's a necessary piece of gear in 2022 or not.

[@nambams]

Digital summing is really just math, literally addition. So I think the sonic improvements with digital come more from increased bit resolution (16

[@TheZappazapper]

@@nambams re: POD 2.0... the question is whether "good" means the POD can do accurate modeling, or whether you can use it to get a sound that you like. I don't think anybody thinks the Fender Bassman model in the 2.0 is particularly accurate, but everybody should be able to do something with it that they like. So it will always be a useful piece of gear, just like my POD X3L is still a useful piece of gear, and just like my Digitech RP2000 is still a useful piece of gear. The ART Pro Channel that I bought the same day I bought my first interface, when the best EQ and compressor I had were the ones that came with Cubase VST32, and I wanted at least ONE of my 16 channels to sound good, not so much. Old amp modelers at least offer some unique sounds over modern ones, whether they sound "good" or not, but a late-90s budget channel strip really doesn't offer anything that makes it make sense to use over plugins. It's still in my rack, though, just in case I need an emergency 12AX7 😂

[@saftpackerl]

Early 2000s digital summing just doesnt sound that well anymore since they’ve reinvented math! 😄

[@rayrecordings]

It’s always good to keep all this in mind. I always compare my ITB>analog 2 bus to Full summing mixes, ending up choosing one or the other depending on the song. There is one important factor though: We don’t make sound to only satisfy scientific outcomes, and reach perfection, but to feel good, have fun, and analog sound, with all its imperfections, gives you something else to play with…a tactile, human, imperfect bliss. Ooompf !

[@1176hambone]

I prefer the workflow of sending 4-6 Stereo digital summed group outs to corresponding analog compressors (aliasing free compression and saturation) with analog Wes Audio automation, then summed through an Orbit for additional aliasing free saturation through an inserted mix buss compressor/limiter and discrete transformer shelving EQ. This is for my preferred bussing scheme workflow and aliasing mitigated saturation with screen free compression metering and tactile controls. Expensive yet fun with heat, metal, knobs, switches, buttons, and lights! I find screens depressing and boring. The less, the better, for me.

[@hyperactivists9390]

there is actual merit to this approach as analog compressors probably “sand down” any last bits of digital stepping or aliasing. or at least this was a sure thing a few years ago. converters have improved, or my hearing has declined, as I no longer worry as much at 48k or above lol

[@DanWorrall]

Digital stair stepping is a myth, and aliasing can't be fixed after the fact. But using analogue gear is certainly a valid way to avoid the aliasing in the first place.

[@thewolfcries8670]

@@hyperactivists9390 digital stepping?

[@whistletom]

There's no stair stepping when you use an anti-imaging filter. And anti-imaging filters (or reconstruction filters) have been in DACs pretty much since the beginning.

[@1176hambone]

@@whistletom stair stepping doesn't exist at all. It's a visual misrepresentation of PCM audio. When visual depicted in animation one can see that an accurately constructed waveform results. Aliasing is reflection of frequencies beyond Nyquist that are spewed back down enharmonicly. This is different than noise. Aliasing destroys timbre and pitch when it is above the noise floor. This happens quite easily far to often when engineers try to drive, compress, and saturate a through multiple plugin instances in a single signal path. Accumulation of Aliasing is the culprit of digital overprocessing and poor plugin design.

[@ardkok9483]

Hahah, oh my lord. This video was the most nerdy disstracks i have ever seen. So amazing. Love your videos, your knowledge, and love your attitude.

[@tobitweaks]

Not bad, tho it feels like even if you got a hit on the analog summing as the selling point wich I also think is misleading people into buying stuff they might not need, I still feel that summing boxes still stand after this video as useful function generators. Like having input stages impart saturation on your channels or groups without intermodulation. While this is not particularly the effect of summing, it might still be a desirable effect no? All the best, Tobi

[@TransistorLSD]

So you should probably call them "saturation units" and not "summing boxes"... Probably...

[@acdnrg]

@@TransistorLSD Probably depends on whether it is a saturation unit that also sums or a summing unit that also saturates :P

[@saftpackerl]

Yes, but you dont need 16 or 24 channels plus converters simultaneously. 2 would suffice…

[@tobitweaks]

@@saftpackerl well the concept of multichannel saturation to avoid intermodulation sadly incorporates multiple channels

[@saftpackerl]

@@tobitweaks …but not in prallel. 😉

[@ZanyPro]

Great test Dan, I've always preferred a mixbus EQ and Compressor over analog summing box. Less cables, less i/o and less hassle 😅. Glad the NULL test proves there's little or no difference compared to the cost of setting it up.

[@LukeBass1000]

I'm still trying to get my kick and bass to sit well together nevermind worry about analog summing 😂

[@lordberly]

Me too, bud. Me too but I watch this anyways 😂

[@FabvL]

Love the video. My only counter would be that clipping certain summing mixers can be beneficial, I've found pushing past 0 in my Neve Orbit can be pleasing. Am I missing something here?

[@DanWorrall]

That'd valid, but it's clipping not sunning.

[@Robangledorf]

I’d personally never spend money on a summing mixer but this test doesn’t really make sense because the product doesn’t really make sense. The whole point of a summing mixer is that inconsistencies in multiple channels of multimono saturation and the crosstalk between those channels would sum up to a more pleasing saturation than saturating them all individually without crosstalk. Having a “totally transparent summing mixer that imparts no distortion” wouldn’t then do the job. A distortionless summing mixer is essentially just a patch bay that we would expect to do nothing. This really doesn’t show us anything about summing mixers hahaha.

[@PrecisionProductions]

Exactly what I was thinking. I get what Dan is " going for " but if you want no sonic qualities imparted then just use a patch bay. Summing mixers were designed to bring the magic of a console in a smaller form without the size, expense and maintenance of a big format console, therefore creating a hybrid environment with your computer. So correct me if i'm wrong, based on this video and argument , this means that a big format SSL console ' SUMMING" does nothing different to ITB (Pro tools or logic ) summing, forgetting all the other parts and only judging the " summing ." To me you can't separate one from the other, they all come together.

[@MrJC1]

@@PrecisionProductions but what you are both talking about is NOT summing. Lmao. I think you both missed the point. Smh.

[@Robangledorf]

Please explain what summing is then. Because based on what Dan is saying there’s no such thing as summing.

[@MrJC1]

@@Robangledorf Huh? Uhhh no Rob. No. Lmao. You missed the point. Edit: okay... summing is just the addition of two or more waveforms. Analog or digital. It doesn't matter. There is nothing else to it. These summing devices that claim the "magic" is in the summing are misleading. Summing doesn't colour or anything like that. All that stuff is everything else in the chain. Transformers etc. But you could have all those and use digital summing for the same results.

[@Robangledorf]

Tell me you’ve never heard of crosstalk without telling me you’ve never heard of crosstalk. A summing mixer isn’t just “adding the signals together better”. That makes no sense and nobody I know is claiming that. Summing mixers are saturation boxes, and they are not the same as individually saturating elements with other gear because the crosstalk between channels changes what’s being fed to each saturating circuit. If your drum channel is hitting pretty hard and the vocals, bass, and music bus are all bleeding through to the drum channel at -65db each, all of those other elements in the song are going to be contributing to the harmonic structure that’s imparted when the kick hits. That won’t happen by just saturating each element individually. Some people like this effect, typically because it adds a perceived cohesion. By testing a summing mixer that doesn’t saturate, all Dan has showed us is that he and his friend both don’t understand this concept yet. I say this as someone that has learned boatloads from Dan. If he reads this comment then I’m hyped I get to return the favor. Anyways, best of luck on your audio journey.

[@BrianBiscione]

No me voy a cansar de comentar en cada video lo enorme que sos Dan! Gracias.

[@scshepherd]

In my situation I definitely hear a difference. I mean a big positive difference. So no testing is necessary to prove it. If you like one better then just go with it

[@Dr-Curious]

"to prove it." Well the problem is, humans sense things that aren't real based on their cognitive biases. That's a fact proven by science. So, when you can hear a certain thing it means other people without that bias are not going to hear it. Itr's my opinion that all audio classes should teach this, but too many have too much to lose. Like sales.

[@Dr-Curious]

Oh, and if you don't believe how utterly real you think EVERYTHING you hear is - look up the "rubber hand and hammer" experiments. Then you can ponder upon how much you imagine is happening.

[@moksja667]

Of course you can hear it, you are super special.

[@seedmole]

I think it's a matter of people wanting multiple incompatible things from a single device. They want a summing device that somehow both perfectly passes the incoming signals, but also adds some magic. It can do either one or the other of those, but asking something that is transparent to somehow affect a signal in a positive way is worthy of being its own multipurpose idiom.

[@connermusic]

As someone who would otherwise share the same cognitive bias as you’ve declared on this test (and doesn’t own nor plans on owning an analog summing mixer), I am curious to see if analog summing with some boxes may have some kind of crosstalk between channels is also at play here - like some kind of nonlinearities determined by any adjacent input signals. Whether it be significant or not I also think that “null” isn’t really the correct term here - even though it’s the same type of test - your signals are cancelling enough to reach or be below the noise floor but because that floor is still present, they aren’t necessarily “nulling” which is pedantic but important

[@MrJC1]

Its just about how accurate his gain matching is.

[@DavidDavis-FA-photog]

I laughed my ass off when I saw the canceling out. Great video should save lots of people lots of money. As an FY I saw another video that describes how some famous mixers use consoles and others mix ITB. It's about 50/50. They all had incredible mixes. But what was interesting was the common denominator, "the individual". The mixing engineer is the one making great mixes not the console or computer.

[@MixMeMcGee]

😂 that thumbnail!!

[@ProgressiveSoundAudio]

I love your delivery style Dan. Great video and keep up the great content.

[@emilbirkedal8430]

Dont tell wytse

[@scacchomattho]

Come on, he has a console and a 24 channel tape machine, he doesn't need summing mixers at all, he just love to fiddle with knobs and faders (and get that sweet tape saturation he loves)

[@emilbirkedal8430]

@@scacchomattho True that :)

[@balladeerstudio]

This made me so happy. Ever since I heard of analog summing it has made no sense to me, for exactly the reasons you listed at the beginning. It's nice to know I wasn't overlooking something in ignorance. I love your humor. I laughed out loud at least five times.

[@MotokoKusanagi]

It made me somehow very happy that your approach was so rigorous, honest..

[@dannymaher664]

Dan, I love all your videos my friend, I find them really informative. The thing about them is that they are always to the point and without any bullshit. I always find that you leave a satisfying argument for people to ponder over whilst always remaining intuitive with your explanation as to why you have come to your conclusion. Great stuff!! Thank you for all you do for music. Dan UK

[@marcan42]

That was a great demo! To play devil's advocate and hypothesize how analog summing could add "magic" that is hard to pinpoint (and null like this), I think the obvious case would be nonlinearities on both the input and output. NL->(+)->NL is not equivalent to doing the sum before or after, and without some way of snooping on the mix bus directly, there's no way to use such a summing mixer in single-channel mode to get the same result, and it's tricky (though by no means impossible) to analyze it in order to replicate it in software. This is basically equivalent to having the sum operation itself be not quite a linear sum. Still no real magic, of course; functionally this is no different from adding saturation in each channel and to the final mix and still doing the summing normally. It would just be harder to replicate the exact saturation of that specific hardware.