professional Sampling: Techniques, Mics...

There seems to be a lot of misunderstanding about the techniques and microphones for professional organ sampling. I am an Audio Engineer and feel the urge to write a statement.... which is, of course, my personal viewpoint and no dogmatic rule!
if you have ANY questions or hate mail :), feel free to contact me!!

1) MICROPHONES

THE ideal Microphone is the omnidirectional pressure Microphone, like the Sennheiser MKH20. The Omni capsule uses a membrane upon a closed chamber, like a barometric sensor. Therefore, all natural frequencies are preserved and the ONLY negative thing about these capsules is their fixed directional pattern.

every other type of Microphone is a modified VELOCITY microphone, which is figure eight in its original form, and can be tuned to cardioid, super/hyper cardioid or shotgun. BUT this costs a lot of bass sensitivity and destroys the ideal linear frequency response. The membranes of these microphones are open to both sides.
For Singing or instrument recording, the colorations from the microphone capsule might be considered great; but for sampling, one must use the most linear and neutral sounding microphone available.

Really good ones (and of course expensive) are: (about 700-1800 USD, each)

Sennheiser MKH 20 (omni)
Microtech Gefell SMS 2000 (modular system, multiple capsules available)
Beyerdynamic MC 723 (with omni capsule, of course!)
Schoeps Colette-Series (VERY high quality, VERY expensive)
Neumann KM 180-Series or KM 100-Series
DPA 4006 (omni)

Of course these microphones are meant to be used with the A-B Stereo technique, which uses two omnis spaced at LEAST 0,5 meters apart. The stereo image is rebuilt completely using the runtime differences of sound, which means a spectacular "depth" but very reduced mono-compatibility. But anyway... MONO? I don´t care about that.
The broadness of the stereo image can be easily controlled by spacing the mics further apart... the more distance, the smaller the stereo image. Small organ, big distance! Huge Organ, 0,5 meters!
This is because the sound has to have about 1,5ms difference from one speaker to the other to be heard totally on one side.
Just try and listen!!

And don´t confuse Mics like AKG C414 or SHURE KSM 44 with real Omnis! These big double membrane mics use a dual membrane to electronically switch between directional patterns. These microphones have VERY severe colorations and unstable (frequency dependent) directional patterns. This might sound GREAT for lead vocals or guitar amps. but NOT for classical music or sampling.

if you consider buying a stereo pair of microphones, be SHURE to buy a factory matched pair!!!!

2) PRE AMPS; RECORDING MEDIA

Using a laptop is a good idea, but you need the right audio interface or Channel Strip. Very good interfaces are the Apogee Rosetta 800 (3400 USD?), the Metric Halo 2882 is very good too, along with MOTU interfaces, RME, NEVE (ultra but VERY expensive) or Solid State Logic (same). Of course DBX or Focusrite pre amps are often used and are very good too. cost effective solutions are missing quality, and nothing is more important than a good mic preAmp and superb A/D converters, otherwise your precious microphone isn´t worth anything!

I recommend using a higher sample rate than 44,1KHz. For organs, 48 - 96 Khz should be fine, and of course you should use 24-bit. This increases the signal-noise ratio to about 143dB instead of 96dB with 16-bit. This can be an important difference!!

Another tip: if you plan to publish the final sample set with 48Khz, 24-bit, for example, you should record the single pipes with 96Khz. When the sample set will be 44,1KHz, record the pipes with 88,2Khz.
This is because down-conversion of sample rates is very tricky and a lossy process, but most software-processors can significantly reduce the negative effects when converting to the exact half of the original sample rate!

3) Last word on RECORDING Technique

I suggested the A-B stereo technique. But be aware that you get a very wet recording with this! Omnidirectional mics sound great but they do not "exclude" diffuse sound (reverberation) like Cardioids do.

If you wish to record DRY, you should use one of the tried´n´true XY or ORTF techniques. If you have any questions about these, contact me!

greetings and awesome recording,

Auriel (samael[aT]inode.at)

Of course you cal buy two AKG C480b microphones. They are really, really good and quite cost effective (700-800 Euro each). Be shure to buy the right capsule for the modular mic system!

And: There is a difference if you want original stereo sampling or mono sampling, panned in stereo afterwards.

Real stereo means you have to record every pipe into a stereo file which leads to a stereo sample file which means huge harddisk space consumption, if I am right about the samples in Hauptwerk.

The easy way to get good results would be a single mic (Omni) for recording all pipes from the same spot (organ console?) and panning them afterwards. Saves a lot of time, money and frustration *G*

Thanks very much for you're interesting facts and opinions. The topics discussed interest me very much.

In regards to an omni-directional microphone vs. a dual diaphragm microphone featuring an omni-directional "mode", you state that the microphones have "very severe" colorations due to the fact that their directional pattern response changes with frequency. Is this really the case for ALL microphones with dual diaphragms?

I'm looking at getting a pair of Rode NT2A microphones so I have a multitude of setups to record with (MS, XY, AB...). Their specifications can be found here. It does show the omni directional response start to warp at around 4kHz but not by much at all? I'm a student so my income isn't that great and I simply can't justify going off and spending AU$1000 on a single microphone and I don't intend to make money from my hobbies (i'm going to fail so much if I ever get into business :-)).

I did a recording of an organ (soon to be released) with a Rode NT4 microphone; contrary to other peoples opinion, the stereo spread was very good considering the width of the organ (it wasn't wide at all) and despite my fears of the response rolling off in the high frequencies, it actually sounded really fantastic in the high and the low ranges. But, the NT4 will simply not do justice on larger organs and other recording methods will need to be used.

Also, as for your comments on panning a single mono recording, wouldn't that kill the natural impulse response of the room?

What is your opinion on MS configurations?

The point is that dual diaphragm microphones do not have the positive features of a real "omni" pressure micrphone because their ccapsule is a velocity capsule; they get the pseudo "omni mode" by electronically "merging" their orignal two figure eight patterns in a specific way. The price for that is a weaker low end, high wind- and shock sensitivity and a strange effect of extreme, sudden bass boost in close proximity to the mic capsule. Be cautious with close-up setups...

of course Doesn´t mean they are bad at all !!! there are very good velocity mics out there.

In most cases the polar diagram tells you how the mic does behave in the 1KHz range. Some polar patterns are extended and tell you more..

The NT2s show pretty much what happens to the cardioid pattern above 4000Hz.. the highter the frequency, the more it becomes a super-cardioid. Would be neat to see the 7KHz polar diagram...
In the end you have to let your ears decide, as always! Test the mic.. set it to omni mode, record, and walk around it while speaking. Same for the other patterns, then listen..

The XY and MS techniques are pure coincidental stereo techniques, there are no differences of sound runtime at all; all stereo imaging is done by loudness differences.
This means good localization of phantom sound sources, but bad, non-spectacular "depth" and "space". ORTF is between coincidental and A/B, so it is good alternative...

MS is interesting because one can set the stereo spread afterwards in the mixing process by making the S+ ans S- signals louder. MS is matrixed and needs to be splitted to 3 seperate Channel strips at the console.
If you can handle MS, give it a try, but compare it to ORTF and AB...

Single Mono panning does indeed crash the stereo reverberance and early reflections, yes. That technique will be really interesting if the new Imulse Response system is implemented in Hauptwerk... Dry, easy mono recordings and an impulse response from the original room, and there you go... nice!

Have i forgotten something?

Thanks for the reply,

A.

Ah yeah: S Microphone would be a figure eight (standard), but try the omni instead of cardioid for the M microphone.

But the MS technique is tricky for Organ Sampling because you have to split signals, phase reverse on of them and so on... in the end you´d have to bounce the files one for one again to get normal stereo interleaved files...
will think about that later, have to go now.

But the MS technique is tricky

Maybe it was a couple of decades ago...

My mic amp has an MS matrix built in; but I record into a computer now, and it is trivial to do the matrixing in the recording software. (In fact I record ambisonic surround, which enables me to generate any coincident mic patterns I wish through simple matrixing.)

Paul

Yes, but not everyone has the Plug-in or the Sequencer software to do this... I guess if you have the money to record with soundfield microphones, you also have the extra money to buy two silly SSL Channel Strips (the ones from the 9000 series) along with ProTools TDM...

I guess if you have the money to record with soundfield microphones

I wish! I use three AKG Blue Line microphones (omni and two figure of 8, which latter I was mistakenly sold at the price of the cheaper capsules) as a "native" horizontal B-format microphone (like the Nimbus setup). The software for any further processing, e.g. from B-format to stereo, is free, except for the VST host I use (AudioMulch).

I am hoping to be able to get this new ambisonic mic soon - whose processing is entirely software-based.

You can get some of my surround recordings (in B-format) from here, as well as many other people's recordings. If you go to my user area (pwhodges) you can also see pictures of my microphone arrangements. To return slightly to topic, several of my recordings there have organs in them (Exeter College, Oxford - a modern Walker in raucous French style; and Oxford University Church - a recent refined-sounding Metzler).

Paul