What's under the hood?

[RMorgan]

I would suggest not. All of us here apparently discern a certain 'something' out of MB - warmth/softness-or-something.
Unless it's mass audio-hypnosis.

Well, it's good to know that!

Anyway, it would be nice if the guys from Harrison could jump in and elaborate a bit more about this topic.

In fact, in my opinion, they should actually make a video or article explaining why exactly Mixbus is superior sound-wise in relation to the other available console emulation software out there, what's the real advantages of it being a DAW instead of a plugin, etc...

They should really talk more about this kind of stuff. It's the heart of their software. Otherwise, a lot of people might dismiss Mixbus thinking it's just a nice GUI with some cool built-in effects.

I, for one, though Mixbus was pure placebo until I took the courage to buy it. Since they don't offer a demo, a lot of people might be thinking along the same lines.

[Ian Ballard]

A real console has each channel plugged into the busses individually and each input has its own circuit path that adds something, however subtle. A plugin cannot do this. It can only take the linear-summed signal and add non-linearities to it to "simulate" a channel-into-bus sound. Harrison can actually emulate the signal path and add non-linearities to each channel, like a real analog desk. Ardour makes this easier, being open-source, plus Harrison has more experience getting digital boards to sound analog than Waves, Slate or whomever.

[joegiampaoli]

This is like asking Col. Sanders for the "Secret Recipe"...

[RMorgan]

This is like asking Col. Sanders for the "Secret Recipe"...

Not necessarily, my friend.

Explaining what Mixbus does to achieve such peculiar results is very different from explaining how it does it.

[joegiampaoli]

This is like asking Col. Sanders for the "Secret Recipe"...

Not necessarily, my friend.

Explaining what Mixbus does to achieve such peculiar results is very different from explaining how it does it.

Once you answer one of the 5 w's (what, why, where, when and how) you can descifrate the rest....

[RMorgan]

This is like asking Col. Sanders for the "Secret Recipe"...

Not necessarily, my friend.

Explaining what Mixbus does to achieve such peculiar results is very different from explaining how it does it.

Once you answer one of the 5 w's (what, why, where, when and how) you can descifrate the rest....

Well, this may be true, but only sometimes...

I could imagine several situations where someone could explain a complex process with satisfactory accuracy without actually exposing its secrets...

As, an example, one could explain a lot about how an atomic bomb works without actually telling anything about how to build one...Or one could outline how a car engine works without actually teaching you anything about how to build one...An structural engineer could give you a fairly solid explanation of how a complex bridge works in five minutes, without getting into any of the hardcore physics behind it.

It's totally feasible for the people behind Mixbus to deliver a deeper explanation about its essence without showing a single line of the code.

[Ian Ballard]

I think in a nutshell you can say that there is no disconnect between the DAW and the DSP, unlike other workstations. From emulating the inputs of a channel, through the EQ, through the compression, through the outputs and inputs into the mix buses etc. etc. Everything has the mojo and it emulates the actual circuit path rather than simulating it with plug-ins.

[Ben@Harrison]

Hi guys,

We have explained our techniques in lots of different forums, interviews and such. But it's a bit technical, and not terribly interesting, so I think it goes unnoticed. Ian is closest; typical plugin signal-flow means that each plugin has to be treated separately. Whereas with a "mixer" you have more control over the signal as it progresses through the strip.

Here's the official story:

Harrison has a unique history. In the mid­80’s, we developed fully automated, digitally ­controlled ​analog mixer technology which was adopted by premiere film ­mixing facilities around the world. The analog processing racks were in a back room, separate from the control surface which was in the studio.

When the digital revolution came, we were asked to convert the analog processing racks into a digital processor, while leaving the control surface, already installed at the facilities, unchanged.

The mixers wanted the new digital mixer to work ­and sound­ exactly like the analog mixer they were using for previous projects. This required us to develop a digital audio engine that operated and sounded exactly like the analog mixer they were using for previous projects. (because the control surface remained exactly the same, you see?) This transition was not undertaken by any other company.

We do not add any distortion, crosstalk, or other "analog" effects. Instead our goal is to avoid digital artifacts; and in so doing, make it sound more "analog". This involves well-known techniques such as dithering and parameter ramping, but taken to the 9th degree through our experience with several iterations of ultra-high-end digital mixers. (when someone pays nearly $1M for a console, they put it through much more rigorous testing and listening than a typical DAW mixer)

That's pretty much it.... !

-Ben

[James Greenlee]

I'm guessing that the really low-level noise that I'm seeing on the channel and bus faders is dither then?

It's quite obvious that the tape saturation produces a third order harmonic (and only a third) and seems to also do something to transients.

Any insight on that? Why not a more complex harmonic sequence in the distortion?

---

Also, if you have any links to where you discuss this, I'd be grateful. There was mention about some details on Gearslutz, but I've searched without much luck.

[Ben@Harrison]

Hi James

Sending a tone through the saturation stage, and driving it into heavy distortion, is pedantically interesting but not really an indication of what the effect is doing. As you noted, it is a dynamic effect that depends greatly on the input signal. The clipping you are seeing is only applied to loud transients ( the lever mode, for example, has a slow attack that lets these transients through to the saturation stage where they are soft-clipped ). If you drive it into sustained clipping, you're not measuring anything interesting.

The tape effect was derived from analyzing several tape machines and their effect on the workflow of traditional 'pop' music mixing.

-Ben