Refuting the external analog mixer theory

David Polich · Post by **David Polich** » Sun Jun 18, 2006 5:06 pm

chrispick wrote:
Splinter wrote:.

Still, for some pleasure's found in the nth degree. to that, I'm more apt to say "fix your song."

Man, that is so true.

Splinter · Post by **Splinter** » Sun Jun 18, 2006 5:12 pm

chrispick wrote:Why does it have to be considered hype? Some people like the sound better. Just expressing opinions, right? Certainly, analog devices add subtle aural aberrations and harmonics that many people prefer.

I'm not knocking anyone's opinion of what they like, it was just an observation of those in a "camp" or what is blatantly marketing hooey that tend to propagate the analog summing superiority myth. Most people can't tell the difference.

I participated in a blind listening test over at Gearslutz a while back that took an identical mix bounced in PT and a mix summed OTB through a Dangerous 2Bus (I think). Anyway, about 80% of the people got it wrong thinking the PT mix was the OTB mix which they preferred. Then some actually went back to change their posts to state they liked the other mix instead or tried to discredit the test. The pro-OTB crowd couldn't live with the fact that they actually preferred the PT bounce.

Timmy S · Post by **Timmy S** » Sun Jun 18, 2006 9:30 pm

At home I Use V72 preamps, a focusrite producers pack with an apogee trak 2 for A/D and a big Ben for master clock. Also a motu 1296 is my i/o but I don't use the coverters except for monitoring.

At hyde Street we couldn't get the big ben to work so we just used 4 pro tools interfaces with internal clock.

At home I use my Ramsa mixing board for 12 channels out of my computer. I started doing this once I tried routing different instruments to the different outputs and immediately noticed more depth. But it was nothing compared to the Neve.

I cannot be sure it is the analog summing bus but the mixes off of the Neve sound about twice as wide as both the ITB mix and the summed mix off my ramsa at home. You can take it all for what it's worth but the Neve makes the mix sound punchier right away. So whatever the reason is, due to coloration or summing bus, I will always use a neve to mix.

zandurian · Post by **zandurian** » Sun Jun 18, 2006 11:21 pm

It's fairly aggrivating for us creative people. We hear all the best stuff ever done in the entire history of recording and we want to do all that - and even one better (it's human nature). Sometimes the great results we hear were acheived with an entire team of people and a lot of money - and sometimes maybe with just one guitar and a stroke of genius.
The bottom line is that if you're not getting the results you want then stay open to the input of those who are getting it done. That is, if you are not finding sonic utopia on your own, find someone who has and learn to do what they do (whether it's gear or technique). If you hear something in your head that no one's ever achieved, then invent a way to to achieve it and if it's really really liked by a lot of people you can get rich and famous. If it's just a matter of not affording the right gear, then either work your but off to get it or try to block the whole thing out of your mind so you won't be miserable your whole life.
I agree with the poster who said to not forget to have fun.

philbrown · Post by **philbrown** » Mon Jun 19, 2006 3:48 am

A friend of mine has a Neve broadcast console. He had tracks up in Nuendo through an RME interface and we listened through his studio monitors (don't recall brand) at his nice but modest home studio. We listened to the ITB/Nuendo mix out of the RME then he ran a few submixes out of the same interface (I believe 3 or 4) to his recently delivered Neve. The difference was stunning. I could use lots of colorful adjectives, but one thing is certain - I know I could have told them apart blindfolded and so could you. It was NOT subtle. It was amazingly better, and it wasn't volume, either, it sounded better even when turned down lower than the ITB mix. A third friend who had just starting recording was there and his jaw was scraping the floor too. At that point were having a chuckle saying "So that's what all those analog plug-ins are trying to sound like!"

Then we took a (very mediocre) mix from the new guy on CD and ran it stereo out through the Neve and then bypassing the Neve. Again, an amazing difference, but obviously no summing going on. In this case the difference was much more the coloration (especially the Neve transformers I believe) rather than a ITB-OTB summing issue as it turned out. But if someone walked away saying "The Neve OTB mix blew away the ITB..." that would be inaccurate but the person might argue their point til next Tuesday (or post on rec.audio.pro) because "I heard it with my own ears! Neve smokes the ITB. You can't tell me otherwise!" But you could also say "When I run my mix through this Tube Screamer it colors the sound..." and be just as meaningful, but it wouldn't have anything to with ITB-OTB summing issues. However, the OTB signal chains often have coloration (hopefully good, high quality, desirable coloration) so the ITB-OTB summing issue gets obscured with another huge variable that is way bigger than just summing: analog coloration, whether it's a Neve or Soundcraft Ghost or Trident. Folks choose those boards specifically beacause the like that coloration. I'm not saying there are no differences between summing methods, just that coloration can be the overriding and often overlooked factor in these argu... discussions.

My $.02

Phil

Resonant Alien · Post by **Resonant Alien** » Mon Jun 19, 2006 7:19 am

I read the Roger Nichols article, and I think a lot of you are missing the point he was trying to make. In my reading of that article, I don't believe he is saying that ITB mixes and mixes done through an analog board SOUND identical. This would be silly, since most "classic" analog boards are not "straight wires with gain", so they obviously have some impact on the final sound - whether or not that makes a mix "better" or "worse" is totally subjective.

What Nichols is saying is that the summing portion of a DAW and the summing portion of an analog mixer are the same, if you use your faders the same way on both and pay attention to gain structure the same way on both. He is making a mathematical argument that does not say anything about the subjective sound of the mix.

Just because an analog board sums it's signals the same mathematical way that a DAW sums it's signals does not provide any insight or clear argument one way or the other regarding whether the two sound any different.

kwiz · Post by **kwiz** » Mon Jun 19, 2006 7:53 am

Has anyone investigated the different panning laws between Neve, SSL, and various DAWs? Could this explain the perception of width that is achieved when summing through analog circuitry?

Splinter · Post by **Splinter** » Mon Jun 19, 2006 8:34 am

philbrown wrote:...if someone walked away saying "The Neve OTB mix blew away the ITB..." that would be inaccurate but the person might argue their point til next Tuesday (or post on rec.audio.pro) because "I heard it with my own ears! Neve smokes the ITB. You can't tell me otherwise!" But you could also say "When I run my mix through this Tube Screamer it colors the sound..." and be just as meaningful, but it wouldn't have anything to with ITB-OTB summing issues. However, the OTB signal chains often have coloration (hopefully good, high quality, desirable coloration) so the ITB-OTB summing issue gets obscured with another huge variable that is way bigger than just summing: analog coloration, whether it's a Neve or Soundcraft Ghost or Trident. Folks choose those boards specifically beacause the like that coloration. I'm not saying there are no differences between summing methods, just that coloration can be the overriding and often overlooked factor in these argu... discussions.

My $.02

Phil

Yeah, what Phil said.

waxman · Post by **waxman** » Mon Jun 19, 2006 9:04 am

Timmy S.

You got it! The proof is in the pudding or in your case ears and body. Why do the itbers like waves or urs or NI instead of the other junkie plugs that are out there....

It just sounds better.

So T.S. even if your test is not "fair" it is true.

It just sounds better.
The way some plugs just sound better. The way some mics just sound better. So all you ITBers take the scopes, put down the dope and just listen! Is it better? then.... it's better! yogi bera

YouTube · Post by **James Steele** » Mon Jun 19, 2006 9:30 am

Hmm... what I'm taking away from this is that it really boils down to "coloration" as was pointed out... not necessarily purely summing. I think that makes sense. I also wonder if the people who think the coloration is more pleasing are people of my generation who grew up listening to vinyl records that were recorded on analog gear that not only contained coloration due to tape saturation but the analog mixer circuitry? Do you suppose younger ears that know only digital would prefer the sound that they're used to? I know that being part of the former group, I'd probably prefer the analog coloration, but perhaps this preference is not just subjective but can be traced to a certain age group that was raised on the sound of records made on Neves, etc.

Timeline · Post by **Timeline** » Mon Jun 19, 2006 9:32 am

This is a confusing thread. Has anyone tried this?

...take a big stereo track mix, drop the level at the DWA mixer input to -40 and buss through the master fader to an IO. Record the analog audio after making op the 40db with a GOOD pre-amp? Record it on a 1/2" two track.

OK then send the exact mix at unity gain out to the same source without gain makeup passing through the same gain makeup device at unity. Record it on the 1/2" two track. Compare...

You will find low level signals, because of bit loss=: lost clarity, lost low level signals like echo trails and some lo end punch etc.

When I mix in a DAW I accept this because a 32bit float will make up quite bit of the sound loss within the structure of a DAW but the lower the levels the more damage is done.

Even stems are marginalized. I would, if I could, send every channel out at the exact recorded level
if I could to separate IO's with NO PANS. Then to an analog mixer for comparison.

Think they did that?

Analog is infinitely better if done in this manner given the character of the console is of a type you like.

Consoles I would use in order of character I like: API, Trident A range, Neve 80 series. All discrete NO CHIP mixers.

This said I use ITB more often for ease of use but I don't have the consoles I used to own. Thinkin about building another one though. I just about have all the parts I need.

Discrete boards are simply awesome.

Splinter · Post by **Splinter** » Mon Jun 19, 2006 9:33 am

waxman wrote:Is it better? then.... it's better!

Sure, but, your conclusions about WHY it sounds better can be entirely false. That's the point of Nichol's article and this thread. It is easy, though, to miss the distinction between analog summing and OTB mixing. You are really talking about two different things. The point argued here is summing. I don't think anyone here would argue ITB and OTB mixing will yield different results, but as to whether "analog" is better than "digital," well, that's entirely subjective and I'm not going there.

Timeline · Post by **Timeline** » Mon Jun 19, 2006 9:41 am

Sorry our posts overlaped James. Your right on that but remember many top records today are done analog using a DAW front end. It's very in NOW not just yesterday.

Splinter · Post by **Splinter** » Mon Jun 19, 2006 9:50 am

Timeline wrote:You will find low level signals, because of bit loss=: lost clarity, lost low level signals like echo trails and some lo end punch etc.

When I mix in a DAW I accept this because a 32bit float will make up quite bit of the sound loss within the structure of a DAW but the lower the levels the more damage is done.

Sure, this could be an issue, but who works on their DAW at -40dBfs? And then we have the noise floor which essentially "wastes" the last 5 to 6 LSBs of 24 bit audio anyway. Sure technically analog is infinite, but it gets washed away in a sea of noise at -110 at best.

In your comparison, if you are raising the noise floor 40dB then of course you will have some maskings. Same would be true of an analog 24 track through a Neve.

Timeline · Post by **Timeline** » Mon Jun 19, 2006 9:59 am

Splinter wrote:
Timeline wrote:You will find low level signals, because of bit loss=: lost clarity, lost low level signals like echo trails and some lo end punch etc.

When I mix in a DAW I accept this because a 32bit float will make up quite bit of the sound loss within the structure of a DAW but the lower the levels the more damage is done.
Sure, this could be an issue, but who works on their DAW at -40dBfs? And then we have the noise floor which essentially "wastes" the last 5 to 6 LSBs of 24 bit audio anyway. Sure technically analog is infinite, but it gets washed away in a sea of noise at -110 at best.

In your comparison, if you are raising the noise floor 40dB then of course you will have some maskings. Same would be true of an analog 24 track through a Neve.

My point is that an analog board will not mess up low level signals. I disagree. Noise yes.. Digital mixers do.

Commonly in a mix, faders are not all at 0 level Splint.