TranceAddict Forums (www.tranceaddict.com/forums)
- Production Studio
-- Cool panning thing I saw today

Cool panning thing I saw today

I was watching an engineer today and the mix was very busy and full. To find place for certain elements, he would split the signal into bands and pan those bands. Allowed him to place a certain sound in a space that fits while not loosing the frequency content that makes things harsh and brittle. I asked him about it, he said the principle is based around how stereo is perceived ie below 1700 or so , by time difference, and over 3000. by amplitude. But basically , he was able to have sounds fit without cutting everything all the time. Sort of along the same line of thinking as m/s EQ but with more control.

This genuinely interests me.

I was watching some of the Fabfilter videos the other day and they were splitting frequencies by channel, although not for panning purposes.

One day i'll have the time to go and do a proper engineering course that I keep promising myself.

i was impressed that he could just know in detail which where he could put a very specific frequency band. No spectrograms or visual aids, the guy must be approaching 50 so his hearing probably doesn't go past 10 but it was ridiculous watching him work. he mix we gave him was kinda shitty. He shook his head , and just went to work. I brought him a latte so he didn't mind me being there despite being in a bad mood. I didn't want to disturb him but this is pretty much what he was doing.

what i did notice was that he was using that frequency range in the mid lows - mid high hat has a limited stereo image and using that concept to fake perspective while finding room. It really was pretty cool.

If you ever get the chance to watch a pro, it is a learning experience. This mix was really busy. About 124 racks. He took a listen. Shook his head. He didn't bother with the group stems. Started from scratch. Was riding more than one fader at a time recording the automation in real time without making a mistake. I mean one pass. thats it. Like this guy was on speed.

And how he was able to organize things so fast. After the first listen, granted everything was labelled right using a colour scheme that many are used to, full orchestra was panned and grouped in 10 minutes. He did 4 passes and managed to get all the level riding done. I'm sure he did some final tweaks later on but man this guy was fucking sure of everything.

And it is a good feeling when your engineer appears to know what he is doing. Micromanaging is just a pain in the ass.

I asked about him and he used to record orchestras. Huge setups, MAin ORTF, omni flanks, house cardioids, ortf for winds, cardiods for harp percussion ..... And he would just invent he scheme in his head asking who the conductor was, and what orchestra and with he score. This guy could flip thru a conducor score and make decisions based on that . I mean most engineers can't reaad music, this guy could read a conductor score rather well.

now that is an engineer. I mean imagine watching Lang Lang on he piano and you wonder how the fuck. Well here are engineers like that. Virtuosic in every sense of the word. And that oldschool way of doing it. Where you just do it, and thats it. No double takes because your mic spacing is wrong or your mics are at the wrong height.

Great info man - you want to keep this guy as close as possible and suck out all his knowledge as much as you can (takes years to get to his level).

Cheers

oh we definatly aren't close enough to even be considered colleagues. I just happened to bring lattes because I had heard of his reputation. Everyone else was booted. And he was a fucking grump. Every time something in the tracks did not agree with him, he looked at me and gave me evil eyes like i had anything to do with it. Never said anything, Just looked at me with that sort of fuck you punk look. lol

I did not ask a question, I was a healthy distance away pretending to make adjustments to he score.

Bu ya, it was like watching a virtuosic piano player. No hesitation what so ever.

I would say the beast engineers usually have a classical background. That was where most advancements in sound quality happened. Recording an orchestra is very challenging. And when you do it every day as a living, you just develop these chops that make you go whuuaaat.

Crazy and fascinating - level of knowledge of some people is just incredible. It is so motivating and exciting to watch people like him doing his daily task, the way they think, they act and analyze the problem which for us could look/sound massive but for them simple and routine - damn it, one day (when I'm 70 or 80) somebody will look at me the same way I'm looking at people like him (at least I think my son may???? lol).

Cheers

quote:

Originally posted by Looney4Clooney i was impressed that he could just know in detail which where he could put a very specific frequency band. No spectrograms or visual aids, the guy must be approaching 50 so his hearing probably doesn't go past 10 but it was ridiculous watching him work. he mix we gave him was kinda shitty. He shook his head , and just went to work. I brought him a latte so he didn't mind me being there despite being in a bad mood. I didn't want to disturb him but this is pretty much what he was doing. what i did notice was that he was using that frequency range in the mid lows - mid high hat has a limited stereo image and using that concept to fake perspective while finding room. It really was pretty cool. If you ever get the chance to watch a pro, it is a learning experience. This mix was really busy. About 124 racks. He took a listen. Shook his head. He didn't bother with the group stems. Started from scratch. Was riding more than one fader at a time recording the automation in real time without making a mistake. I mean one pass. thats it. Like this guy was on speed. And how he was able to organize things so fast. After the first listen, granted everything was labelled right using a colour scheme that many are used to, full orchestra was panned and grouped in 10 minutes. He did 4 passes and managed to get all the level riding done. I'm sure he did some final tweaks later on but man this guy was fucking sure of everything. And it is a good feeling when your engineer appears to know what he is doing. Micromanaging is just a pain in the ass. I asked about him and he used to record orchestras. Huge setups, MAin ORTF, omni flanks, house cardioids, ortf for winds, cardiods for harp percussion ..... And he would just invent he scheme in his head asking who the conductor was, and what orchestra and with he score. This guy could flip thru a conducor score and make decisions based on that . I mean most engineers can't reaad music, this guy could read a conductor score rather well. now that is an engineer. I mean imagine watching Lang Lang on he piano and you wonder how the fuck. Well here are engineers like that. Virtuosic in every sense of the word. And that oldschool way of doing it. Where you just do it, and thats it. No double takes because your mic spacing is wrong or your mics are at the wrong height.

Fuck this sounds exactly like one of the main engineers at work.

I mean I think it IS him.

Was his name Alan by any chance?

I can't tell you how much i learned in the first month just sitting there with my mouth shut just watching them. It was stuff I couldn't even understand from a technical point, even though I could see the processes and had been an engineer for close to 10 years.

To be honest, all but one of the truly great engineers I've been lucky enough to work with have that exact demeanor. Gruff, borderline pissed with everyone, like to work in silence but so fucking damn good at what looks like instinctive on the fly decision making....

....but the truth is, that in most situations, it's simply the vast experience of having done it a thousand times before.

Going back to the orchestra aspect: all of the great score mixing engineers I know are active and current live orchestra recording engineers.

Now if you put that in to context; that they create the mic list and mic arrangement based on the composition, orchestra and conductor, then they have to mix that on the other end when the recording is done, it becomes second nature to know where everything is.

In fact, I know one major score engineer that asks to be the recording engineer whenever he's asked to mix a film project for that very reason - it's the control of recording it exactly the way you want to mix it and takes 99% of the guess work out of how to mix it all.

Finally, on the actual subject of panning, what you're talking about is frequency based stereo image separation. Score engineers often do this based on a LCRLsRsLFE (i.e. 5.1 setup) then just bounce those down to stereo with a preset mix ration for the channels involved.

Therefore with a 5.1 setup you're panning different things automatically based on their frequency band, and ultimately, this comes from knowing the instruments invloved.

So for instance we were working on a large 2008 super hero blockbuster
film. We were recording cello overdubs, which doesn't go low enough as an instrument (under normal conditions) to be in the LFE channel, so we set up four mics: L, R, Ls & Rs, two in front, two above and behind.

You now know you have the channels to be panned. From here you can place the balance of those channels anywhere you want in the image field, but (and there's the whole secret to it) the way they do it is based on frequency as you'll also have many other instruments all using 4 mics, 3 mics, 2 mics or 1 mono mic.

The basic principles goes that the higher the frequency the wider you can go with your panning but your can essentially "trick" the listener in to hearing a wider space by exponentially (rather than linear) increasing the width as you go up the spectrum.

The real pros do this however based on the score itself, not just the potential range of a given instrument.

It's actually incredible how much more you can do when you have this down. In that same film, we had one cue that maxed out at over 500 tracks, from a 48 piece orchestra and hours of studio overdubs and the engineer found space for all of them. He was riding 8 faders at once for most of the mix, on a 264 channel (96 physical) and that was linked, controlling another 264 channel desk in the other mix room.

The main thing that bugs me about guys like this is the lack of second guessing and them just fucking knowing they get it right first time.

i did a quick search and he has worked with one of the sound heads for that project you worked on, the big one. don't want to name him but , hmmm his name is also a stage at he radford. The guy he has worked with that you know is known as the King. Unless you are on a first name basis. lol shit i will just pm you.

"what i did notice was that he was using that frequency range in the mid lows - mid high hat has a limited stereo image and using that concept to fake perspective while finding room. It really was pretty cool."

Could you describe in more detail what this technique is and how it works? I can't fully grasp it yet..

For instance I don't get how stereo can be percieved over 3000 hz by amplitude alone, cause the haas effect etc still majorly applies.. I don't quite get the concept I think.. :/

Very interesting. Also check your "t" key. It's appears to be a bit shy.

quote:

Originally posted by lenieNt Force For instance I don't get how stereo can be percieved over 3000 hz by amplitude alone, cause the haas effect etc still majorly applies.. I don't quite get the concept I think.. :/

3000 is about the fundamental tone of he highest G on a 88 piano. What you think is 3000 hz is harmonic material based on fundamental that is much lower. It is a matter of science. Your ear cannot use inter aural time difference for hz above 1500 which is the maximum value generally given for ITD

i should add that i am simplifying the matter. Although this is true , the brain does seem to able to use spectral grouping and come to think of it, i'm starting to doubt whether ITD and IAD matters when you have a stereo output. The Hass effect makes 2 signals seem like 1 if they are lower than 20 ms in delay. So although the brain cannot use ITD from a single sound source, a stereo source could. But I think this is more related to perhaps how first order reflections can impact the localization. To be honest, i am not quite so sure now.

ps: was curious so i just did a pure tone at 3000 hz. added a delay to the right speaker of 30 ms, couldn't tell the difference. But as you lower it , at about 1500 you start to be able to notice the emerging tone. Did the same for 10 ms, at 3000 no change in localization, as you lower , the sound in the left speaker seems to get louder.

I imagine it gets a little more complicated for complex noises.

I wish we had a dedicated thread on here where we talked about nothing but engineering or mixing techniques.

Its threads like this that make me realize how absurdly small my "engineering toolbox" is. I mean fuck I discovered Von Doorns 5 tuts only of yesterday and about 90% of the shit he did were things I've never even thought of doing before.

Are there books that specifically discuss engineering tactics? I really need to start thinking outside the box but when you don't even actually have a box thats a hard thing to do. I mean its people like me that would read shit like this and abuse it terribly. If you pan a sound exponentially would it be smarter to use more than 3 regions of the spectrum? Like instead of low/mid/hi doing low low/mid low/hi low/low mid/mid mid/mid hi/low hi/mid hi/hi? Should produce a more dramatic image right?

I have a hard enough time mixing 10-15 tracks never mind anything over 100. I don't even get how that is possible unless they aren't even real sounds but just blips of sounds. It took me 9 years just to learn something as stupid as putting tremolo on noise I can't keep learning at this slow a pace I feel like I'm wasting tons of time.

the technique described is not one you would typically do. Not like that. I could see how one might use M/S EQ but actually panning left or right is rare.

How many mixes have you finished ?
You need to read, both the practical and science stuff
and you need to just do it.

it isn't really that hard. Almost every advanced technique is just simple technique but used in another way. The core comes down to knowing how to listen, spotting the problems and knowing how to fix it .

you learn this by mixing and listening. Reading gives you the language to describe some issues and some general context and tips.

As far as how you get 100 tracks.

The orchestra was not real. So right there,

about 60 tracks. 30 tracks of percussion. misc stuff like synths guitars drum kit bass.....


it adds up quick. The only way to get a realistic sound from samples is to blend. So violin 1 can have 6 stereo tracks. Now if you account for all the instruments of an orchestra roughly 35. It adds up. 100 is actually rather light.

Re: Cool panning thing I saw today

quote:

Originally posted by Looney4Clooney he was able to have sounds fit without cutting everything all the time.

A good engineer once said that an awful lot of mixing is about using the levelling faders correctly. I sorta took no notice and thought to myself that what he said was obvious surely, so I didn't take on board the critical importance of his comment.

I've only recently begun to recognise the role of correct levelling and as such am now trying to put a lot more care into the volume of each instrument.

I tended to subconsciously maximise the level of everything and ended up with that brittle wall of sound mush, in fact I made just this mistake on a recent track (Endtimes).

I remember reading the same thing about the importance of the volume fader a while ago and it's really so obvious when you think about it.

When I started making music with a DAW again a few years back I probably spent 20 hours a week just reading stuff, and then created perhaps 5-10 practise projects every week, most of which were little more than 8-16 bar loops just to investigate what i'd been reading.

The importance of the fader hit me when I redid a small project that I thought sounded ok at the time. I set every channel back to zero and relevelled it from scratch, keeping every sound as quiet as I could get away with. When I finished I thought it sounded pretty close to what I had originally done. I lol'zd when I compared them.

I'm probably just starting to get to the stage where I feel I can realistically call some of the stuff I do as good (technically at least) but L4C is correct when he talks about numbers of projects and just getting on with it. There really is no secret to it, in fact it's really very simple. Just takes time and experience.

I've never finished a track, but I've read plenty about it...

Can you guys remember where you read this about the volume faders? Would be an interesting read I'm sure! Really happy if any of you can find it!

quote:

Originally posted by Looney4Clooney ps: was curious so i just did a pure tone at 3000 hz. added a delay to the right speaker of 30 ms, couldn't tell the difference. But as you lower it , at about 1500 you start to be able to notice the emerging tone. Did the same for 10 ms, at 3000 no change in localization, as you lower , the sound in the left speaker seems to get louder. I imagine it gets a little more complicated for complex noises.

Ah interesting! I will do some testing myself when I get home! But I think it gets more complicated myself yes with complex noises, cause I often a time delay one channel i.e. to get my snares (highend above 3000hz) more stereo, and always have to lower one channel in volume to compensate for Haas..

quote:

Originally posted by DJ Robby Rox I wish we had a dedicated thread on here where we talked about nothing but engineering or mixing techniques. Its threads like this that make me realize how absurdly small my "engineering toolbox" is. I mean fuck I discovered Von Doorns 5 tuts only of yesterday and about 90% of the shit he did were things I've never even thought of doing before. Are there books that specifically discuss engineering tactics? I really need to start thinking outside the box but when you don't even actually have a box thats a hard thing to do. I mean its people like me that would read shit like this and abuse it terribly. If you pan a sound exponentially would it be smarter to use more than 3 regions of the spectrum? Like instead of low/mid/hi doing low low/mid low/hi low/low mid/mid mid/mid hi/low hi/mid hi/hi? Should produce a more dramatic image right? I have a hard enough time mixing 10-15 tracks never mind anything over 100. I don't even get how that is possible unless they aren't even real sounds but just blips of sounds. It took me 9 years just to learn something as stupid as putting tremolo on noise I can't keep learning at this slow a pace I feel like I'm wasting tons of time.

Don't get disheartened Robby - we've all had (or have) those moments where we just seem so out of our depths you think "what's the fucking point".

I remember when I first moved out to LA, I thought I was a pretty decent engineer, had about 10 years of working audio experience, having graduated a good engineering school and had worked with some pretty big name studios in London.

The last studio job was a complete eye-opener and I probably learned more in the first few month there than I had in the preceding 10 years. What L4C said about watching and working with real pros is absolutely true. Sitting there by yourself will be way slower than discussing with, or watching other people better than you.

Another point: you may want threads just about enigeering, but honestly this place echos studio life quite well: the sharp humor, the acidic wit and even harsh handling of noobs/dumb questions is like a pale version of studio life. If you can't hack it here, then lifespan in a real studio would be less than a tse tse fly.

Why? because the intense pressure and monotony, combined with brains overflowing with knowledge (not that I'm accusing this place of those attributes) need a release and a punchbag.

Back on topic, I would not use this technique until you've got just simple level balancing and frequency separation via EQ truly down.

I'm going to post a bunch of shit about this theroy and technique but I'm knackered so I'll post is tomorrow.

Powered by: vBulletin
Copyright © 2000-2021, Jelsoft Enterprises Ltd.