-
Microfiles: Univox DF-14B, A Smooth (Looking) Operator
“Compare its quality…its good looks…” -
Top Concerns Faced By Churches When Purchasing…
-
In the Studio: Timing Your Effects To The Song
-
The Eight Prime Factors That Determine The Sonic…
How And Why Unity Mixing Can Make All The Difference In The World
You can tell the difference just by listening? Yes! Good unity mixes sound open, alive, immediate and unrestrained
Page 2
Back to first page.
Recording, Audio, Console, Mixing, Sound, System, Engineer, LiveOver time, I gradually changed the way I set the preamp gain, evolving to soloing up the channel and adjusting the knob until I saw what I believed I should see on the console meter. Yet my paradigm was still about adjusting by sight - until being confronted by Delwyn Brooks.
Delwyn came up assisting at Little Mountain Sound and has a page full of credits to his name. We met when I was the recording engineer on a session at Imperative Studios and he was coming in to mix the songs. I showed him the system and we brought up the first song.
The first thing that happened was not expected. Delwyn yelled at me. He hurt my feelings. What an unappreciative jerk! So what if I ignorantly made the job of mixing more difficult for him because of my poor preamp gain setting technique!
At Little Mountain, junior engineers were given the night recording gigs (usually for the local bar bands wanting the Little Mountain name for half price). There was one huge stipulation placed on them: they could use nothing other than the gain knob and console EQ and they absolutely could not use any outboard gear.
When the senior engineer arrived in the morning, all of the console faders had to be at unity. and when he pressed play on the multitrack, there had to be a solid mix already happening. Not a perfect mix, but one where all the elements were in their appropriate places. If the junior engineer couldn’t learn to get this right, he didn’t continue to work there.
When Delwyn explained this to me, it was like fireworks going off. Wow! This was the answer I had been seeking, and the problem all along was that I did not know how to ask the question!
Finally, I could see and understand a much better way to set gain. My brain did a major paradigm shift, and fortunately, Delwyn eventually forgave me.
The proper way for setting the channel gain on the console is not focusing on each channel itself, but rather, listening to where that channel fits in the mix.
Ever seen the house console photos in Live Sound magazine taken before the soundcheck, where you can see all of the console faders parked at unity? Now I know what these guys know. Dialing in a mix is so much easier when you’ve properly gain matched at the front-end, and it sounds best too.
This is not to say that you mix the whole show from the gain knobs - once they’re in the right position, mixing is properly done at the faders. A side benefit is that those mixing monitors from the main console aux will find it easier to get a quality monitor mix.
So if your mixes look more like Picture A than Picture B, it might be time to re-think your gain structure approach. Put yourself through the exercise of “What if I can’t move the faders from unity?” I think you’ll like the results.
Author’s Note: My thanks to Ted Brown and the Martin Woldson Theater at the Fox.
Previous articles on PSW by James Cadwallader:
Yes, Virginia, System Gain Structure Matters - Here’s Why
No Slave to Gear: Maximizing What You Get Out Of What You Have
Viewed 23951 times.
Topics:
Live Sound, Feature, Consoles, Education, Engineer, Mixer, Sound Reinforcement, System, Technician,
View all topics
Comments (31) Most recent displayed first | All comments in chronological order
First off, I totally disagree with the 'faders at unity' concept.
Almost every 'pro' sound tech (guys who do sound on 'known' bands tours) that I've worked along side do this whole 'fader at unity' technique. I ask them all why they do it and the only answer I've heard is, "it's how I was taught". Then every single one of them who have to run monitors from FOH never have enough gain to drive their aux's for monitors. . . THEN the performers get upset when their monitor mixes change constantly as the mixing engineer is tweaking his gains.
After taking Electronics Engineering and having to build A to D converters at school I understand how they work. As a preliminary statement, let me say that I also am a recording engineer and would like to point out that there is a reason people record at high bit depths - it's to achieve a more 'analog' (smooth) waveform. See photo of analog (or high bit depth and sample rate) vs low bit depth/low sample rate/low gained digital waveform here:
http://media.soundonsound.com/sos/sep05/images/forgotten1aquantisation.l.jpg
Quoting the webpage from where I got that picture... "With a higher bit rate, there are more possible amplitude values". Well technically speaking, the larger the waveform going into a A to D converter, the truer the representation of the waveform. If the preamp is turned way down (because you want that channel on your mixing console quiet in the mix), then you may end up with a block like digital square wave (sounding like junk) vs a truer to analog type waveform with a properly gained input to your converter.
To touch back onto my earlier statement about recording at higher bit depths, it's mostly heard in things like reverb tails where near the end of the tail are very small signals... which in the digital realm end up being square waves.
Having said all of that, your theory is flawed... big time. . . at least on digital consoles.
For analog consoles, it's all about signal to noise ratio. EVERY single electronic component introduces noise. So technically you SHOULD keep your signal to noise ratio as awesome as possible... in other words keep your signal as large as possible.
Lastly, if your having troubles with clipping summing stages (like you say in your article), fix yourself. haha Keep an eye on everything, that's your job. If you clip after a compressor, turn your makeup gain down! If you're clipping the stereo buss input, slide all your faders (or group faders) down! Making sure you're not clipping is the first step to mixing!
Sadly, the people who mix with their faders at unity probably will always mix that way. . . and I'll continue to laugh inside at them haha
Julia (if you are real):
Please read Matthew 6:5-7.
Jesus doesn't care about the prayers in the megachurches. Jesus doesn't need a huge PA system. Jesus thinks your flashy pastors and preachers are all hypocrites.
Read your book and tell me what I say is wrong.
Hey...this is a wonderful website buddy and an informative post!!! i am new here and i found this site very interesting and informative ,, you are a professional blogger i think i have a great interest in such things...thank you for the post buddy and keep on posting nice stuff like this :) i am a student.
+ View all comments on this article
Commenting is not available in this weblog entry.
Product Showcase
Advertising
Learn more about product showcase listings
Sponsored Links
Some comments. You've "designed" ADCs, so you should have some familiarity with sampling theory. As such, you'd know full well that the picture to which you linked, which shows a stair-step approximation of the digitized waveform, is completely _WRONG_. Hint: a sample represents the amplitude of the signal at the sampling instant. There is no amplitude value between the samples. It could like like this: http://cnx.org/content/m0009/latest/cosine.png
To get back to a continuous-time signal, you need to apply a reconstruction filter, which is what fills in the time between discrete samples. And this is _independent_ of sampling frequency or sample word length ("bit depth" is the wrong term).
Second, the reason for using longer word lengths is to increase resolution at the bottom end. A 24-bit word pushes the quantization noise down below the noise floor of the analog electronics. So turning down the preamp and digitizing a cold signal won't give you that "block like" signal you describe. Reverb tails are _NOT_ square waves. As noted, that's NOT how sampling works.
For BOTH analog and digital consoles, you make a trade-off, which is summing headroom vs signal-to-noise ratio. In both cases, if you trim the inputs hot, in order to mix without overloading a mix bus, you have to pull down the input faders. You can get away with hot input trims when doing a solo-acoustic act on a big console, because of how the console's summing attenuates each input, but when you run all of the inputs hot, you're in trouble.
Please, go back and study your textbooks about sampling theory and come back to the conversation when you actually know what you're talking about. It's clear that Cadwallader has no idea what he's talking about -- don't be like him.