Skip to main content

I've had the Cowboy Junkies' Trinity Sessions in pretty heavy rotation. Very cool sound. I read that it was recorded using ambisonic microphone techniques, so I started to read a bit about it. Very interesting. Invented by mathematicians! Has anyone around here used it? Any info on the Trinity Sessions? (Some of the info doesn't add up - it looks like you need at least four tracks for ambisonic, but they reportedly recorded direct to a Betamax.?)

Comments

BobRogers Fri, 09/25/2009 - 07:22

I heard it when it first came out, but didn't get too deeply into it. The singer in our folk group wanted to do a similar version of Sweet Jane, so I bought it and started listening closely. Now that I'm more interested in acoustic recording it really caught my interest. I just ordered the CD/DVD of the recent sessions. They seem to be recorded with very different techniques. Still - a great sounding church.

DIY ambisonic is definitely a possibility. Might be a good undergraduate research project.

anonymous Sat, 09/26/2009 - 11:41

BobRogers wrote: I've had the Cowboy Junkies' Trinity Sessions in pretty heavy rotation. Very cool sound. I read that it was recorded using ambisonic microphone techniques, so I started to read a bit about it. Very interesting. Invented by mathematicians!!! Has anyone around here used it? Any info on the Trinity Sessions? (Some of the info doesn't add up - it looks like you need at least four tracks for ambisonic, but they reportedly recorded direct to a Betamax.?)

I created and maintain the FAQ on Ambisonic Surround Sound. Happy to help, but it gets complicated.

First, some general stuff. An Ambisonic mic can produce B-Format. This captures the full-sphere soundfield encountered at the mic, and consists of four channels, W, X, Y, Z. For horizontal-only replay, the Z (height) channel is dropped leaving three channels. These three channels can then be matrixed down to two, the result being called UHJ. UHJ is mono and stereo compatible, and can also be decoded back to surround sound. Unfortunately, this last part can not be done without compromise; no compromise surround sound requires three- or four-channel B-Format.

With the Trinity Sessions -- the original CD, not the recent sessions -- they recorded two-channel UHJ. However, when this is decoded to surround sound, the musicians come out in strange places. This is because they were in strange places, suggesting that the UHJ was used as a stereo effect and was not intended to be decoded to surround sound. Fortunately, domestic Ambisonic decoders have a "Super Stereo" mode in addition to a UHJ mode. This includes a width control which allows the stereo image to be compressed to mono-like or expanded into a horseshoe around the listener. Decoding the Trinity Sessions with Super Stereo allows the musicians to be pushed back to the front, where they "belong", while still being surrounded by the wonderful acoustic of the Trinity Church. This allows you to have your sonic cake and eat it too.

See, I told you it would get complicated. For more info, the Ambisonic FAQ is available on my website at http://members.tripod.com/martin_leese/Ambisonic/.

Today, on Ambisonia.com (http://www.ambisonia.com) there are over 200 pieces in B-Format available for free download. Almost all of these are full-sphere. Ad hoc software decoders are also available (but only for B-Format, not UHJ or Super Stereo). Note that Ambisonia.com will soon be moving to a new website, SoundOfSpace.com.

Share and enjoy,
Martin

BobRogers Sat, 09/26/2009 - 12:33

Martin-

Welcome to recording.org!

I've been all over the tripod and Ambisonia web sites recently. I'm a math prof at Virginia Tech. I'm currently teaching partial differential equations and beginning work on an undergraduate text on the subject. I'm planning on including some examples from acoustics and this looks like a great application of spherical harmonics. Since I want to work with students on this I'm far more interested in DIY projects. I have enough ribbon figure eights and a condenser omni to set up a rudimentary native B-format mic. The capsules are bigger than ideal. I also want to set up the decoding in the DAW myself rather than use a plugin. I'm going to get hold of Gerson's papers for more info on that. I think I understand the position decoding, but I have to read more on the EQ adjustments.

I'm sure I'll be after you with more questions as this project progresses.

anonymous Sat, 09/26/2009 - 14:22

BobRogers wrote: Martin-

Welcome to recording.org!

Thank you, but I am not planning on staying. Just passing through. (However, I will be notified of any posts to this particular thread.)

BobRogers wrote: I've been all over the tripod and Ambisonia web sites recently. I'm a math prof at Virginia Tech. I'm currently teaching partial differential equations and beginning work on an undergraduate text on the subject. I'm planning on including some examples from acoustics and this looks like a great application of spherical harmonics. Since I want to work with students on this I'm far more interested in DIY projects. I have enough ribbon figure eights and a condenser omni to set up a rudimentary native B-format mic. The capsules are bigger than ideal. I also want to set up the decoding in the DAW myself rather than use a plugin.

Making your own recording with a "native" B-Format mic sounds like an excellent way to start. The other way to start is to buy a TetraMic from Core for about $1000, http://www.core-sound.com/. (I am just a domestic listener, and have never made a recording in my life.)

One of the problems with Ambisonics is that it is very easy to create a very poor decoder. I would strongly reccommend you start with an existing decoder. Later, if you choose, develop your own. A good decoder incorporates things like Speaker Distance Compensation (also called Near-Field Correction) and Shelf Filters.

Existing decoders are listed at (Dead Link Removed).

BobRogers wrote: I'm going to get hold of Gerson's papers for more info on that. I think I understand the position decoding, but I have to read more on the EQ adjustments.

I'm sure I'll be after you with more questions as this project progresses.

Rather than start with the Gerzon papers -- there are a dump truck full -- try these three, particularly the third: http://www.ai.sri.com/ajh/ambisonics/BLaH.html.

Have fun,
Martin

BobRogers Sat, 09/26/2009 - 17:13

MartinLeese wrote: Rather than start with the Gerzon papers -- there are a dump truck full -- try these three, particularly the third: http://www.ai.sri.com/ajh/ambisonics/BLaH.html.

Those are great! Just what I was looking for.

Despite your advice, I will be working on my own decoder first. The primary goal at this point is to teach students about spherical harmonics, etc. Any musical benefit will be gravy.