I think the three mics of a Decca Tree are supposed to be omnis, so the pointing is not critical. As most omnis are a bit directional at very high frequencies, it may be prudent to point the centre one forward and the two side mics 45 left and right.

I am not a big fan of the Decca Tree, though, find it way too woofly.

As David Spearritt said, all omnis (all polar responses, for that matter) become more directional as the frequency gets higher, so 'aiming' them is part of the art.

I believe many of the early orchestral recordings were made with Neumann M50s, which were omnis but the capsule was mounted in a small sphere which made them considerably more directional at higher frequencies and therefore allowed the engineer to alter the tone of the instruments. This was very useful for adjusting the 'focus' on things like strings, where the centre of the mic could be aimed to favour the front desks (generally the better players) or whereever. Also handy for putting a bit more focus on the woodwinds or other weaker sounds.

Alternatively, they could be faced towards the back of the room or the ceiling to get a clearer room sound and a mellower instrumental sound.

I believe Neumann re-issued the M50 as the M150 some time ago...

DPA offer a similar capability through their APE Acoustic Pressure Equalizers, which are small plastic balls (and other shapes) that can be placed over the end of the microphone (flush with the diaphragm) to alter the polar response at higher frequencies. Interesting things to play with.

You can get a similar effect with DPA's 4015 wide cardioid, which has a high frequency boost on axis.

As for positioning the microphones themselves, I had the good fortune of working as second engineer to a very talented ex-Decca engineer on many orchestral recordings. He used to laugh at all those 'rules' about the positioning of the main three microphones and so on, and said they never really followed them at Decca. He'd just put the microphones where they sounded right, roughly in the triangle position, but he'd ignore symmetry and distance and so on if it didn't sound right. Then he'd rotate them around to get the right tone. His recordings always sounded incredible - huge, open, airy, yet detailed. He is quite famous locally for his ability. I have heard other engineers try to do the Decca Tree following all the 'rules' and so on, and it always ends up sounding like David Spearritt said - woofly!

So, don't be afraid to move those three microphones around. Don't get too stuck in the theory of it all.

By the way, according to this particular ex-Decca engineer the Decca Tree technique (as used by Decca) always had at least six microphones - the 'tree', plus two omni outriggers (one each side of the stage), plus there was always a spot microphone for the tympani. Other spot microphones would then be added as necessary, and all mixed down to stereo. For a large ensemble, those extra microphones made a huge difference. The omni outriggers opened up the space dramatically, allowing the tree microphones to be placed closer for better detail. The tympani spot mic added clarity to the tympani attack and restored its sense of 'drama'. The three tree microphones were never as impressive or interesting without those outriggers and the tympani spot...

Also, we would normally set the mics up so that the left was in front of and above the front desk of violins. The right microphone was positioned similarly on the cellos, and the centre mic would be similarly positioned in front of the violas. Then we'd move them around, up, down, back, forth. Sometimes they were quite a few metres apart, from left to right, but it always sounded good.

I hope that's helpful...

- Greg Simmons

Danielle:

There's some real good information here! Strangely, despite my apparent woody for omni mics, I too am not a big fan of straightforward Decca Tree. I tend to agree with David and think that the closeness to the orchestra creates a bloated and unrealistic sound. Note, this isn't due to proximity effect (omnis...), but it's due to the fact that you are essentially capturing the notes just as they exit the instruments. Much of the magic of orchestral music is what happens AFTER they leave the stage and get mixed with the other instruments.

However, if you decide to try the Decca Tree, it is virtually at your discretion which way you point your mics, but bear in mind, there is some directivity with omni mics, so place carefully. Personally, I do a variation of the Decca Tree with most of my orchestral recordings (I'll say, my "standard" recordings - regular sized orchestra in a performance hall)

What I do is, place outriggers on the left and right sides of the stage (similar to the outside mics in a decca tree, but a little further out). Then, I put a stereo pair on the same stand over the center of the orchestra. The mics are spaced only a little - between 20 and 45 cm and can be panned lightly or folded into a mono center mix (not too many phasing issues creep up at this distance). Then, I can use the outer two mics' levels to create width and the center two to adjust depth. Just as in M/S or Decca, you can't fiddle with the levels too much or it will be plainly obvious that the soundstage is shifting.

Just be careful that in this mode or Decca Tree, that you don't pan too hard left and right. Remember, the orchestra doesn't sit on our far left and far right, we shouldn't pan them that way either. Just a little in from far position should be fine.

J...

It's good to read all this, esp after many years of tweaking my mic's and positions almost identically to what everyon'es describing here, Decca tree config and so on. Sometimes "the rules" just don't work as they're stated.

I smiled at the description of the typmani mics; I often do that, in spite of the horrified looks of some players - "What are you mic'ing the TYMPANI for!??!!?" is the usual question. Usually it takes too long to explain that I'm actually looking for a little detail (like the mallet hitting the head, rather than just the boom of the lower end). The tymp players "Get it", though, and not too long ago I got into a whole discussion with one player who was switching between various Mallets for the piece at hand; a Mozart overture was one hardness/mallet, and a Mahler symphony was entirely different sticks. This fellow was really into it, and begged me to let him hear the difference in the headphones afterwards. It really mattered to him, and I appreciated that. I always chat with him about it when we're working on the same gig.

Ditto for the harp and other touch-up mics. There's often a section or instrument that could use a little help in the mix with an extra mic or two, depending on the hall. It might be small hand percussion or bells, but sometimes it's nice just to have that option.

The "Decca tree" is a good place to start, but there are always more options if you're able to experiement a little.

I have never used this technique but read an article on Decca Tree by Ron Streicher [Mix online, 2003] the other day and thought I'd post the link.

http://mixonline.com/recording/applications/audio_decca_tree/

First of all- Greg, it is great to see you here. For those that don't know him, he's the guy that made my job in Sydney recording at the opera house in July happen. A great engineer and a great guy...

Anyways, as he mentioned, the original decca tree was made actually with M49's and soon after replaced with M50's. The polar response of that microphone is such that the upper mid to high frequencies are *very* directional. The lower ones being quite omni directional. Because most of our sense of imaging comes from upper frequency content, you actually get a pretty good image from this.

The original tree was in a "T" shape with arms of approximately 1 meter and the position was over the conductor's head. The three mics panned Left, Right and Center respectively. For a successful tree to be done, it requires a bit of tweaking for the hall... while all three mics are supposed to be at the same gain setting, it doesn't always work that way. Sometimes I find that a tweak of the middle mic can make a huge difference. Also the relative positioning/spacing can also make or break the sound. The outer mics are generally mostly facing out from the "T" but angled inwards a touch (so the high-frequency pickup is of actual instruments rather than ambience- it also can help narrow the image a bit). I usually aim the middle mic towards the woodwind section (which lowers the need for a woodwind touchup mic).

Today, there are precious few mics out there that really excell for the Decca tree without some sort of modification though the use of the Balls that Greg mentioned. DPA and Shoeps make the spheres and I'm pretty sure that Neumann does, too (I can't remember). For a straight-out, unmodified mic, I like the Shoeps MK21 capsule in a decca tree. The Sennheiser MKH800 can also work well, but doesn't have quite the image (the sound, though is devine).

I don't think the tree is great for all cases, but when it works it is fantastic. The wide sound of film scoring is from the use of a decca tree. I also like using it for chorus with orchestra recording as it gives a sense of size that you otherwise wouldn't normally get. I love the image that coincident micing gives (and when on a tight time schedule for setup, I won't do anything else), but a decca tree can sound really good.

If the conductor lets me, my next big chorus and orchestra recording will use a tree- probably using UM 57 microphones. Should be interesting to see what that sounds like. (It is a performance of the Britten War Requiem)

--Ben

Some excellent thoughts on the venerable tree. I do not think there is another approach that yields such a vivid result, which is why I am not a fan of this approach with anything less than a major orchestra string section in a great room. It is simply too revealing. Also, the first desk violins and celli end up unaturally too far to the left and right for me. I like upper and lower strings to have more width than separation.

Another option is the Mercury/ex-Telarc 3 omnis equally spaced across-- it literally mixes itself. A variation of that is a close A-B omni pair in the center rather than a single mic.

Finally, a Tony Faulkner array consisting of a wide carbon-fibre stereo bar with an MK2H on each end, and a wide ORTF pair of MK21s in the middle. The MK2H's usually end up about -6dB or -8dB,depending on how wet and boomy the hall is.

Rich

Thanks for the intro and kind words, Ben. That was a great concert!

I’m not sure how much more help this will be, but the ex-Decca engineer I used to work with is an Irish guy called Brendan Frost. He won an Emmy for his recording of La Boheme some years ago, which was also a Decca tree recording.

For the recordings I worked on with him, we used Neumann M149s for the tree. We also used a matched pair of B&K (now DPA) high voltage omnis as the outriggers, and an AKG 414 or similar to spot the tymp(s).

I say ‘we’ because the microphone choice was usually up to me, and I’d get the best stuff I could based on what the client wanted to achieve. Brendan was one of those true professionals who could probably get a killer sound from two tin cans and a length of string, so he was never too fussed about the gear. If he saw Neumanns, AKGs, DPAs, or similar mics that had an omni option, he was happy.

He would invariably use omnis for everything unless he *really* had to get something a bit more directional. He was, and still is, a real omni expert, and probably the greatest recording engineer I’ve ever worked with (er, except for Ben, of course!). Totally intuitive - he'd take a quick listen to the mix, hobble out to the stage (he was in bad need of a hip replacement back then), bump one of the microphones an inch or so, and it would make a major difference. I used to joke to others about how he always knew exactly which microphone of the six or more to adjust, and he usually got it right, first time every time. Amazing...

- Greg Simmons