Analog summing - Can you hear the difference?

Hi all!

Last week I set up a an eight channel analog summing box like the based on New-York-Daves schematics. I used a couple of Telefunken V672 as amps.
With a friend I made a blind test to see if there was a difference between doing different mixes using the mix bus of my computer system or sending different groups of the mix into the summing box. We also tried doing analog summing with a small Mackie 1404 mixer.

The computersystem is Mac G5 dual 1.8Ghz with Logic Pro 7.1.1 and MOTU 2408 mk II PCI424 based hardware.

...and the result: more noise in the analog summing version, but no difference apart from that :shock: :shock:

The noise was better from the Mackie but no sound difference that would make it worth the trouble.

It should be noted that this is a subjective test. We haven't made any scientific measurements but have been listening for things like wider stereo image, intrument seperation, bass, air, headroom etc. things we strive for when doing a good mix

The mixes we did had between 24 and 60 tracks....next step will be my friends Pro Tools HD192 system.

Well that is my experience so far - I would love to hear about your experience?? :razz:

Morten

I hear a difference a big difference you will notice it with the slow tools highly disfunctional. Your bouce to disk will loose high end and the stereo field wil get more narrow.

I just ought to mention, I really don't stake out a position in this whole in-the-box vs. outta-the-box mixing debate. I only started posting designs for mixing circuits because a lot of people asked about them, and I was annoyed by some of the poorly-designed circuits on the web. I never set out to become some sort of "advocate" of passive mixing.

One reason I don't have a strong opinion on the matter is because outside of doing some simple mono reductions of stereo files in Cool Edit Pro, I've never done any in-the-box mixing at all. When I went from analog to digital recording, the only thing that changed was my recorder. I still use an analog mixer and analog outboard like I always have. So I can't pass judgement on something I've never really tried.

But having said that, I think a better test would be to take a recording of eight tracks and send the tracks out directly through individual D/A outputs. In the test you describe, you're already doing a significant amount of pre-mixing inside the DAW. It's like doing a taste test between a bar of chocolate, an apple and an orange--when the apple and orange are already covered in chocolate. Mmmmm, I'm making myself hungry...

(By the way: 60 tracks? My God. My multitrack has eight tracks and sometimes I don't even use all of them).

Hi NY Dave!

I think you're right that you are going to hear a difference (advantage analog) if you give all the tracks separate output and channels on the
mixer.

My reason for testing this setup was that I've heard that also here you would hear a big difference to the better compared to the computer-only summing. So of course I wanted (and still wants) to do a test since I'm always looking for ways to improve my studio :grin:

The projects we do in my studio calls for the flexibility of the computer since we have to go from recording for one client to mixing for another etc. very fast. The only analog mixer I've worked with that can do some of that is the Euphonix and I'm still saving for one of those :grin:

...and yes 60 tracks..big production with band, strings and harmonyvocals in bucketloads :grin:


Morten

A-B comparisons are very fickle things, in my experience. Often there appears to be no difference at all when you switch from one thing to another. i find that the best way to define the character of a piece of gear is to mix through it (at unity gain, no eq etc,) and when the mix is done, then remove the item from the signal chain. THEN you can hear what it isn't doing, and you willl instantly feel your mix to be "comfortable" again when you re-insert it. I am surprised to hear that you can't tell the difference between V672, Mackie and digital, but i do accept that it will depend very much on what you are mixing. FWIW, as an experiment, recently I ran off a computer-only digital mix as 8 stereo pairs out of logic, (drums, bass, acc gtr, lecky gtr, Keys, Strings, Bv's and LdVox.). I then re-combined them in Logic, and did the same in Protools, since I always had the notion that they both had a different character. Surprise Surprise! They ARE different! And I was always lead to believe that they would be identical in the digital domain! I am a habitual Logic user, and hence it is no surprise that I prefered the Logic Mix. MAybe it is a question of familiarity, but the Protools mix sounded slightly colder and more sterile to me.

Ho HuM


ANdyP

ps I use a passive mixer network with my Helioses, even with a F@@kusrite mic-pre, and There is a massive difference in my experience! BUt it is not always a desirable effect.

i think the difference comes apparent mainly if you drive some channels like drum and bass channels into the red

dave has probably the most direct point here, its not really very suprising that you didnt hear a difference bussing 60 tracks down to 8 for mixing out of the box. that isnt mixing out of the box, thats mixing in the box and bussing 8 stems out. I personally wouldnt expect to hear much of a difference at all. I also wouldnt pick a bright line amp to use for makeup gain, if you use a 1272 you are gonna hear huge differences compared to the v-whatever. The more realistic test is 8 tracks out, 8 back in. If you do that you can judge what the difference is and figure out, if you realistically NEED to track 60 tracks then perhaps you can buy more interfaces for OTB mixing if you decide there is a benefit. I think you need to provide a fair test before coming to a conclusion either way though.

dave

There will be more noise because there is another big gain stage...

The whole point of these mix out of the box solutions is to be able to use all of our outboard gear! Use anything you can find and blend it into your in the box mix. Choose the baddest makeup gain you can handle and have some fun with all your toys.

:guinness:

I thought the whole point was avoiding digital summing with the ability to use outboard gear the cherry on top. Most folks building little summing boxes probably dont have mountains of outboard anyhow.

I bet this comes across really grumpy, I dont intend it to be.

dave

I think the difference shrinks as the quality of the mixer goes up.

[quote author="jrmintz"]I think the difference shrinks as the quality of the mixer goes up.[/quote]
I take that as better sticking to mixing inside the box ! :wink:













(MX8000)

No, Peter, it's a wrong conclusion...
I used a mackie lm3204 for a long time to sum up my synths (and tried to drive it very clean on the single signals...)
The difference to record the sum of the synths out of the mixer or summing up the synths digitally (with the same AD track by track with the exactly same signals and best possible tracking quality) is more than significant, even with a mackie compared to the most usual daws (cub*se and son*r...).
I did it similarly to compare the internal digital mix of my korg EMX with same results.
However, noise level is low with my synths, so it's only about the sound itself (esp. higher mids are more transparent, stereo image is better, basses are much less muffled *audiophoolish mode*).

Kind regards

Martin

[quote author="soundguy"] its not really very suprising that you didnt hear a difference bussing 60 tracks down to 8 for mixing out of the box. that isnt mixing out of the box, thats mixing in the box and bussing 8 stems out.
dave[/quote]

FWIW, I have heard a lot of people swear that this method causes a major improvement. I knew a guy who swore up and down that a single D2Buss summing prototools stems on a large project made a night and day difference. I haven't ever done any inthebox/outtathebox listening myself, so I can't comment.

What I can comment on, is that I am about ready to throw my DAW out of the window and start again with my old setup of a Mackie 8-buss and a TEAC 80-8. Damn, I miss a real analog mixer, and not for sonic reasons. I have come to the conclusion that little DAW control surfaces are pretty much reatarded. But that's another thread.

I recently built a 8 stereo channel, passive, balanced mix buss that uses mic pres for gain and I can assure you the difference is noticeable. However, I am pretty sure that the sonic differences are caused by the outboard preamp used for makeup and less by the act of passive, analog summing. The favorite mic pre for makeup has been a pair of SCA N72's. When I first built the device, we did some pretty extensive, blind AB's and here is what we found:

The analog mixes were noticeably wider. This is pretty noticeable and after tracking a whole project while listening to the PT out and then switching to analog; it's quite refreshing to hear.

The top end sounded better (to us).

The low end was tighter and more focused, but did not extend as low as the straight digital mixes, maybe due to the transformer rolloffs at low frequencies?

Anyway, I just figured I should share my experiences because I we definately do hear a difference. To listen to two records that have recently been mixed through my summing box, check out

Days Like These - Inventure
http://www.daysliketheseband.com

The Coral Sea - S/T
http://www.thecoralsea.net

Ian

[quote author="smallbutfine"]No, Peter, it's a wrong conclusion...
I used a mackie lm3204 for a long time to sum up my synths (and tried to drive it very clean on the single signals...)
The difference to record the sum of the synths out of the mixer or summing up the synths digitally (with the same AD track by track with the exactly same signals and best possible tracking quality) is more than significant, even with a mackie compared to the most usual daws (cub*se and son*r...).
I did it similarly to compare the internal digital mix of my korg EMX with same results.
However, noise level is low with my synths, so it's only about the sound itself (esp. higher mids are more transparent, stereo image is better, basses are much less muffled *audiophoolish mode*).

Kind regards

Martin[/quote]
I was just kind of kidding :wink: it's encouraging though. :thumb:

I figure when one's mixing in- or outside, his/her behaviour is obviously not the same since you're hearing different stuff and you'll be doing things differently according to that. So a 'real' comparison would be hard to make I think.

Bye,

Peter

internal or external they are different beasts and require differnt things. Sending out stems to an analog summing box offers a differnt sonic quality not necessarly better for the song at hand. think of it as another tool in the shead at which to use and work with.
Digidesign has a very interesting write on their 48bit mixer engine and how it does the number crunching. I think it's all suspect myself especially with some of the numbers they say are in theory possible. lord knows if thoses number existed I would never overload the stereo buss internally.
(making a thread with the new link.)