Sea Lions

[Eric Gulbransen]

I posted here a while back asking what mics to get for recording wildlife. Big pill to swallow, but I finally managed to get some good mics. Thanks for the help. Thought maybe you guys would like knowing you made a difference.

One of my favorite places these days to practice recording audio is Moss Landing, just south of SantaCruz on the California coastline. There is sea life everywhere you turn. Of course that means it smells a little, but it also means the place abounds with sounds...

Here are two samples that go along with the following questions:
http://www.reelsense.net/Audio/Wildlife/SeaLions/

I'm having a few issues (that I know of).

1 - there's a highway nearby and my filtering skills are one notch above that of a four year old. My best solution so far has been simply re-recording (which I did to get these tracks). Is there a better way to filter certain background sounds than setting a noise print in Soundtrack Pro?

2 - I'm sure the locals hate these Sea Lions, but I could listen to them for hours. But when I listen to what I've recorded I find myself wishing I could "use Photoshop" on these clips. Maybe turn up the contrast, add a little saturation - maybe making it "richer" or more dramatic. Am I crazy or is this possible - and therefore yet another family of programs exist which I need (want) to learn as well?

[Seth Bloombaum]

For the highway, you might try recording a little before dawn. If the sea lions are awake and noisy at dawn, this might well be the best solution.

Generally, getting the mic closer to the sea lions and further away from the highway, of course.

Noise reduction is a distant third choice. Certainly worth trying. You also might try a high-pass eq, and move the frequency up and down while listening.

"Photoshopping" your recordings... well the first thing to do is to make sure you are listening on reference-level monitors. If you're not, you're photoshopping in the dark, or, let's say, photoshopping on a monitor that has too much red. You take the red out to make it look right, and everybody else in the world says "what happend to the red?"

From there, it is a matter of critical listening and thinking about potential improvements. Experimenting. Trying some mid-range emphasis, e.g. pulling up the frequencies about 1.2KHz by 6db. Roll that 1.2 up and down in frequency.

I'm not sure if you want or don't want compression. If you are anxious for more "zing" it is worth trying. Try a ratio of 4:1, then slowly bring the threshold down until you see the gain reduction display start to respond, continue bringing it down while you listen closely. This will also tend to make low level sounds like a distant highway more audible.

In assembling an effects chain, eq is usually before compression.

Do you have a stereo simulation plugin of some type? Worth playing with - the recording appears to be mono?

These comments aren't specific to soundtrack pro (I don't use it), but are generally applicable to any sound editing environment.

***edit***
Just for fun I did all of the above. For the stereo widening I used Izotope Ozone 3. Find some roughs here.

[Eric Gulbransen]

That's exactly what I'm after Seth. My recordings sound like the Sea Lions are in the next room while yours sound like they did in person - clear and crisp. This last attempt got me there just after dawn, to avoid traffic like you suggest, but next time I'll get there before dawn as there were still plenty of cars driving by at 7am.

I have to admit I am guilty of focusing my learning on the visual aspects of videography far more than the audio aspects - so far. But I do recognize that this is a typical ball to drop and I hope/plan to NOT drop it. Not even knowing what audio reference-level monitors are, I searched around a bit to find out. So far I've just plugged headphones into this mac to listen.

I need to learn about this, obviously. Are all those extra channels for monitoring the separate channels which you create in post? Or do you pros actually record sounds like these using eight mics? I am using only two - an MKH70 and an MKH30 in an MS configuration inside a blimp suspended on top of each other in tandem clips, on a pole, hung over a short cliff which put the mics about ten feet from the closest lion. What I put up here is straight from the mics, so you are right, they are two separate mono tracks - not stereo.

As far as compression, my understanding of that word so far is limited again to managing pixels - and making files smaller (also compressing blacks in color). But you suggest there's an entirely different meaning for compression considering audio?


Thanks for taking the time Seth, and also for explaining this in a user friendly language. Sorry for such basic questions. I have a hard time learning from books and I know no one local who shares the passion.

[Seth Bloombaum]

These are an M and S recording - M a long shotgun on ch. 1, S on Ch. 2? Straight recording with no processing?

I couldn't get them to decode properly to create a stereo image. I did have some fun setting up a decode with -S and +S delayed by .002 seconds.

I'm not too sure of the applicability of a long shotgun as a Mid mic. It makes pretty heavy use of phase differences to create directionality - and phase differences are essential to the M+S encode/decode.

Do you have access to a cardoid mic you could use for the M signal? That's the classic approach.

To decode an M+S recording (and you do need to decode it to create stereo):

The M signal is your cardoid (or shotgun, in your case). As a convention, record that to Ch. 1. The S signal is the bidirectional mic (MKH30), record that to Ch. 2.

Bring the recording into your editor, and ungroup the channels so you can work on them independently. Copy Ch. 2 and invert the phase.

You now have:
Track 1 - Ch. 1 - mid
Track 2 - Ch. 2 - side
Track 3 - Ch. 2 - inverted side

Pan Track 2 hard left, pan track 3 hard right, adjust the volume of Ch. 1 up and down - you should be hearing beautiful stereo. Wider image with Ch. 1 lower, narrower image with Ch. 1 higher.

The math used to describe this phase relationship:
L = M+S
R = M-S

When I used a stereo widening effect in the samples above, I sort of recreated the MS decode.

You certainly could record these beasts with a surround-sound mic array, but usually only worthwhile if you plan to release in surround.

Compression as related to audio can also refer to less data in a release format, but in this use we're talking about compression as a tool to reduce the dynamic range, the quieter sounds are louder and the louder sounds are quieter, relatively speaking... then the whole mix is turned up, and everything is perceived as "louder".

Clear as mud?

Here's a good article from Sound On Sound Magazine.

The Wikipedia entry is, well, encyclopedic.

[Eric Gulbransen]

Funny story Seth;

I'm at 20 fathoms into your response tonight, with my wife talking to me from the kitchen. Apparently she asked me a few questions which I neglected to respond to (or even hear). I think I was at R=M-S when I realized her disgust in me. "I"m telling you, when you get on that computer I could walk around naked and you wouldn't even notice me.." (trust me, I would notice her)

So for once I got clever, "OK Queen Cranky Pants, I tell you what. How about I read Seth's response to you and then YOU make sense of it!"

Again I got to R=M-S when she stopped me for the second time, only this time without the attitude. "OK I take it back" she says. "Because I can't even understand the first sentence.."

"I tell you what" I responded, "Come do a few laps around the room naked and we'll call it even"



Can't thank you enough Seth. While I may not understand all of this tonight, I now get where I need to focus the learning energy. Very kind of you to take the time. Pieces are already coming together in my confusing maze of a mind.

For what it's worth, I shoot a lot of telephoto while chasing wildlife (still scared of people). This is why I got the MKH70. While I do understand it can't reach as far as a 600mm lens can, I was hoping any bit would help. Of course in this particular case I managed to get pretty close to the subjects so I do understand why maybe a cardoid instead of a shotgun would have been better. I don't understand what "phase differences" are, but I'll read up on that. Unfortunately I only have the MKH70, 30, and an ME66. What would you suggest I get to replace the 70 for these cases? I plan to sell the 66.

Thanks again Seth. I'll have an education in this if it kills me. And hey, I hear footsteps - I might even get lucky tonight...

[Eric Gulbransen]

OK Seth you are right. I loaded the original track, originally, into Soundtrack Pro just long enough to screw it up (inadvertently) because I was completely clueless (but had the best intentions?).

The REAL original recording is here http://reelsense.net/Audio/Wildlife/...sGrowl_RAW.wav

I only opened it in stp this time to export a few seconds of the 40 minute original. I did none of the things you suggested above - yet. I'm anxious to try that after work tonight.

In reading about phase differences last night and this morning I am warming up (slightly) to what this means. I must not be even close to comprehension yet though, because I don't understand what is going on in the 70 compared to the 30, to create this difference. Tons more reading (and re-reading) to go.

Thanks again (and no, I did not get lucky :- )

[Seth Bloombaum]

Ah, that's much better.

I've posted a straight MS decode of the raw wav file, nothing else done to it, now we've got some nice stereo!!!

Have a look at the track layout below. It's all I did to do the decode. It would probably be improved by some eq & compression, as in yesterday's samples.

[Eric Gulbransen]

Huge Seth. Simply huge.

Thank you for helping me so much. These lessons are going a long long way.

I ordered a new camera recently which won't be here until November. I'm left to capture audio only for almost six months now. Ultimately I would like to complement great images with great sound. This is a great beginning

[Bill Davis]

Originally Posted by Eric Gulbransen

Huge Seth. Simply huge.

Thank you for helping me so much. These lessons are going a long long way.

I ordered a new camera recently which won't be here until November. I'm left to capture audio only for almost six months now. Ultimately I would like to complement great images with great sound. This is a great beginning

Eric,

If I could, I'd like to add a little "concept" information that might be helpful.

It's both tempting, easy, and WRONG to think that by altering the pickup pattern on any microphone you're somehow altering the "REACH" of that microphone. If you think about it, you'll realize that this is IMPOSSIBLE.

Given 10 microphones (or a thousand) at position X relative to a sound source at Y - THE SOUND ARRIVES AT ALL OF THEM EXACTLY THE SAME. This is the stuff of physics and the inverse square principal and is UNAVOIDABLE.

So given that the sound reaching the the transducer will always be a constant - the ONLY thing any microphone can do to sound different from another is to alter the sensitivity of the diaphram(s), alter the physical layout of the structure the capsule is housed in (interference tube designs, etc), and/or employ superior filtering and amplification electronics.

Essentially, we're talking about is boosting the results AFTER something has been SUBTRACTED from the sound.

"Directional" microphones do exactly that. They try to REJECT sounds arriving from the sides or rear. Filtering out those from the overall recording. Then they attempt to boost the remaining signal in a way that doesn't alter the sound or add noise.

People new to sound need to wrap their brains around the concept that a microphone "pattern" is always a SUBTRACTIVE proposition.

in other words, a "shotgun" type microphone doesn't ever "reach" for anything - and the mic itself can't truly make sounds in front LOUDER, because it's IMPOSSIBLE to make sounds louder than they actually are. All any "directional" mic can really do is remove part of the signal and then boost the rest.

This means that the quality of the electronics in any directional microphone - the design of the interference tubes or other electronic methods of filtering out the directional sound information you don't want - plus the amplification required to boost what's left will always cost something.

Pristine "close to noiseless" amplification and careful mic design engineering cost money.

The inexpensive mics everyone seems to be in love with in this age of cheap digital recording gear - even the ones with a decent reputation - are generally ones used in work where "good enough" is just that. Good enough.

So for a handycam shooter working with direct to camera recorded DV, mixing and monitoring on small studio monitors in an acoustically untreated production space and outputting an audio mix that will largely be heard on consumer TVs - well again, good enough, is good enough. And there's nothing wrong with good enough for that kind of work.

However, when the goal is to generate clean audio suitable to drive a state of the art modern theater cinema sound system in Dolby Digital - just good enough generally isn't.

To truly be "big screen ready" you've got to record VERY clean - preserving the kind of signal to noise ratio that can handle being amplified a LOT.

And in that game, only top quality gear, used by people who understand the techniques of good recording will generally fill the bill.

I applaud your approach to learning all you can about sound.

And I hope this helps a little with one of the trickier parts of understanding how microphones ACTUALLY work.

Good luck

[Eric Gulbransen]

Thank you Bill. You are right, I had yet another concept completely backwards.

Considering how long it takes me to decrypt posts like some of the ones above, I'm guessing my recordings will never be ready for the big screen. But for sure they can be many times better than what I capture today, so I am here - asking and reading and fumbling my way toward being far better.

I listened to some of the frogrecordist's tracks once and was instantly captivated. I could have sworn they were not single recordings I was listening to, but instead at least 5 individual sources all working independent of each other - yet also together to paint a picture I could not only hear, but almost see. That was a year ago, and all I had at the time was the on-board mic that came with my HD200. Today I use that mic as a door stop and don't even plug xlr cables in to my camera anymore. Although I have to admit, carrying the tripod, backpack, field recorder, mics, lenses, and popcorn to mark my trail gets a little trying at times. I need to leave earlier, learn a LOT more, and I need to be more patient.

I'm a carpenter. Ironically, I built a sound studio for a client last spring. Major production. Two layers of walls completely independent of each other, two layers of "sheetrock" on the OUTSIDE of the building, sandwiched between two layers of plywood and siding/roofing. Separate trenches to run wires for mics vs electric. No part of one wall could touch any part of the other. List goes on and on. Guy was a mad scientist. Crazy hair, no color to his complexion, walked around hunched over from wearing such thick glasses all day long. And his people skills were completely horrid (I can get along with almost anyone..). He did some big screen work, but I couldn't get a word out of him. It killed me.

So here's the question; is there an interim stage between mad scientist and stock on-board mics? Like, if I have half a room's worth of space and decent hardware/software/knowledge/creativity, can I do this? Capture respectable captivating sound? Or do I need cut all ties to my friends, build an egg crate foam-lined Hee Man Woman Haters Club, and run out to Lens Crafters for some half inch glass?

Thanks again for all the help guys,

Eric

[Eric Gulbransen]

Just wanted to say thanks guys. I struggled in post quite a bit. Not because I don't understand your help, but more because I couldn't get my Soundtrack Pro to accept the project from Final Cut Pro. In the end I had to do the work to the audio separately, blindly, then bring it into the FCP timeline. Wasn't an idea situation for me, but still I wanted to say thanks because at least now I know what I'm after - whereas before I also stunk, but had no idea which way to turn to improve.

Here's what we ended up with.

California Sea Lions

Thank you again. Very generous of you to take the time.

[Seth Bloombaum]

Eric, that's a nice little piece. Congratulations.

No 1/2" thick glasses or hee-man club neccessary. In my not-so-humble opinion, good enough is good enough. I'd rather be going on to the next project than finessing stuff only I will see and hear on the last project. Of course it is very difficult to retain objectivity about that...

Not getting a proper workflow between FCP and Soundtrack Pro is a pisser, though. Just happens that as of today I have the loan of a sexy new iMac loaded up with software. Being more of a PC-editor on Vegas, where very sophisticated sound capabilities are right in the NLE, I'm looking forward to learning more about FCP workflows that include Soundtrack Pro.

Not that I'm abandoning Vegas, but I teach in an FCP-centric program... and it's time for us to start showing more sophisticated audio capabilities. We'll see.