Login / Register
 
Tags: , ,

Binaural VS ORTF
New Reply
Subscribe
simonv
Thread Starter
#1
12th October 2007
Old 12th October 2007
  #1
Gear addict
 
simonv's Avatar
 
Joined: Apr 2005
Location: Montreal
Posts: 477

Thread Starter
Send a message via MSN to simonv
simonv is offline
Talking Binaural VS ORTF

Hi all,

I just came upon a funny mic which is a fake head with microphones in the ears.

Just wondering... is that hype?? Or is there any specific reason to use such a mic instead of using the trusty ORTF technique?

Thanks,
Simon
#2
12th October 2007
Old 12th October 2007
  #2
Lives for gear
 
Oroz's Avatar
 
Joined: Oct 2005
Posts: 1,463

Oroz is offline
Well, ORTF are 2 Cardioid microphones. Binaural are two Omni's .

Is this the one that you're talking about?:Georg Neumann GmbH - Products/Current Microphones//

There's a thread about it here somewhere (do a search). I've always had my doubts about the polar patterns of the microphones used in the dummy head (the specs say polar pattern:ear).
#3
26th March 2010
Old 26th March 2010
  #3
Gear addict
 
Mark A. Jay's Avatar
 
Joined: Mar 2010
Location: Ann Arbor, Michigan

Send a message via Skype™ to Mark A. Jay
Mark A. Jay is offline
ORTF vs. Binaural

The differences are quite significant between ORTF and binaural. ORTF was conceived as a form of binaural, but is really stereo (in fact, binaural MUST be based on a mannequin head - any two-channel technique that does not use a mannequin head is stereo, or some form of stereo, but never is it binaural).

Anyway, as I recall, ORTF spaces the mic diaphragms roughly 17 cm apart, and at an angle of 110 degrees. Condensors are most often used.

The idea here is that the directivity of the cardoids purportedly mimics the directivity of our ears, and the spacing (between microphone diaphragms) approximates the ear-to-ear spacing on a 'typical' Human head.

So far, so good.

However, the major distinction between binaural and ORTF is the presence of the head, the artificial ears, and the presence of the microphones in the ear canal(s).

Why is this important and a distinction from ORTF? Let's take a look.

1: The Shadow Effect:

The head itself allows for the possibility of an acoustic shadow being cast between the ears (and thus, the microphoines). In other words, the fact that there is a solid sphere between the microphone elements controls what each microphone 'hears'. By contrast, in the ORTF configuration, what each microphone 'hears' is largely influenced by the polar pattern and angle, but no shadowing effects are possible. This is a very important difference.

Think about the shadowing effect this way: Picture yourself holding a small light bulb in your right hand with your arm extended to your side. As you move your arm towards the center of your body (moving counter-clockwise), more light will illuminate your face, rather than just the right side of your head. Were you to take the small light bulb into your left hand and continue moving it, the left side of your head would now become illuminated, leaving your right ear and most of your face in the dark. Now...imagine holding a (very small) white noise source rather than a light bulb and repeating the experiment.

This is the shadowing effect of which I speak, and is largely a factor at mid and high frequencies. That is, your head effectively blocks and or causes the high(er) frequency to be scattered (diffused). This happens far less with low frequencies which simply 'wrap' around one's head and thus, are really not shadowed nor diffracted - and - they arrive slightly later, and slightly attenuated (depending upon several factors, but frequency playing a major role).

2: Intensity and Arrival Time:

This shadowing effect causes two things to happen - first, form relatively high frequencies (around 300 Hz and up) the ear nearest the sound perceives it as louder (intensity) but the opposite ear senses it a fraction of a second later than the ear nearest the sound (arrival time). For low frequencies, the pressure gradient as sensed by our ears, is close to zero - because our ears are not far anough apart to discern a significant pressure difference in a low frequency wave. Incidentally, this is why humans have a hard time perceiving from which direction a very low frequency comes. If our ears were a meter or so apart, then we could much more readily sense from which direction low frequency sound is coming.

In essence, this is what is happening with sound being incident on the mannequin head (and ears) is a very close facsimilie of what would be incident upon a person whose head was in that particular acoustic space.

The ORTF configuration's directivity is really based on the directivity of the microphones; there is empty space between the microphones.

The directivity of the mannequin head is not based upon the directivity of the microphone element(s) as they are omni-directional transducers. Instead, the directivity of a mannequin head comes from the fact that there is the 'sphere' between the microphones AND they are inside of the artificial ears (save for the AKG D-99, which has no ear canals). Also, there are several binaural heads available for purchase, and in my opinion (having worked with most of them over the past 20 years) the mannequin heads are more alike (in the effect they achieve) than they are dissimilar. Each manufacturer claims theirs to be the best and most technically correct, and while I have seen compelling arguments from all manufacturers, I've no intention of starting a flame war as to whose mannequin head is 'right' and whose is 'wrong' (if that's even a fair word to use).

Binaural vs ORTF Demo - Real Time

As an aside, I have recorded several live concerts using an ORTF configuration in as space very close to the space of the mannequin head (bit taking care to be above it or below it so as to not be influenced by the shadow of the head) and the differences range from enormous to significant, but this depoends upon many factors including the volumke (in cubic meters) of the hall, its acoustics, source material, etc) but never is the ORTF version, whien listened to over headphones, never is it anywhere near as realistic as the binaural version.

I have even normalized the ORTF microphones so they have the same frequency response as the microphones in the mannequin head and normalized the sensitivity (so that the only difference was based on the physics of the microphone spacing and placing, and not microphone frequency response or perceived loudness - a fair comparison and control measure) and constructed A/B B/A A/B sequences of live performances.

What I have found is that even when you take away the differences due to the frequency response between the microphones used in ORTF and those in the mannequin head, in these A/B B/A demos (where the comparisons are immediate) when it's binaural, you hear the fullness - as though you were 'there' anbd when it switches to ORTF or to XY or anotehr technique, there is still separation, but it just seems 'artificial', and then when you hear the binaural come back, you realize just how stark the contrast between the two approaches can be.

Again, this also depends on many other factors, such as the style of music, the mount of low frequency (versus high frequency) is present and so on, but the differences are real nonetheless.

Does this mean that binaural is 'better' than ORTF? No. Like everything in life, some may prefer ORTF, or XY, or M-S to binaural. However, it is fair to say that the binaural recording will be the most technicallly accurate and realistic rendition - but again - some people may prefer the other configurations on an aesthestic level.

Closing Remarks:

Anyway, I have rambled far too long. If you want to hear a variety of binaural examples, check out my website under 'audio samples'. Just look up my profile and you'll find the url for my site (I'm trying not to be too self-promoting here, but I also happen to think that if you audition the binaural samples, the realism will be self-evident).

Mark
PS: Apart from recording, I have worked - for many years - in signal processing, noise and vibration, sound quality, and Human perception. Much of what I have written in this post is based on the things I have observed in that role.

Last edited by Mark A. Jay; 1st June 2010 at 12:22 AM.. Reason: Changed a typoe "Cardoids" to "condensors" as referenced in another post). Sorry I didn't go back sooner to correct.
#4
26th March 2010
Old 26th March 2010
  #4
Lives for gear
 
didier.brest's Avatar
 
Joined: Dec 2008
Posts: 2,407

didier.brest is offline
Thank you much Mark for you very interesting post. I've two questions:

1) It has been often said that binaural works well for headphone listening, but not for loudspeaker listening. What do you think about this reserve?

2) Is binaural recording of any interest for recording at close distance a single instrument, a piano for instance ?
#5
26th March 2010
Old 26th March 2010
  #5
Lives for gear
 
rumleymusic's Avatar
 
Joined: Jul 2008
Location: San Diego, CA
Posts: 2,129

rumleymusic is offline
Quote:
1) It has been often said that binaural works well for headphone listening, but not for loudspeaker listening. What do you think about this reserve?
His site does say "Music for the headphone generation"
#6
26th March 2010
Old 26th March 2010
  #6
Gear addict
 
Mark A. Jay's Avatar
 
Joined: Mar 2010
Location: Ann Arbor, Michigan

Send a message via Skype™ to Mark A. Jay
Mark A. Jay is offline
Of Binaural, Headphones, and Environment

My 'official' Engineering / signal-processing answer to your first question is this:

It depends.

I know that may sound amusing or like I am trying to avoid the question, but as a degreed Engineer, I believe that many answers to questions really have this at their core. However, to more directly address your question, here are my thoughts (and why I say 'it depends', as you will see):

In principle, I agree with the statement that binaural recordings are best suited for headphones. The reason for this is very simple - using headphones does two things:

a) It puts the sound that was incident at each ear of the mannequin at the appropriate ear of the listener.

b) It eliminates any possibility of cross-talk between left and right channels.

In other words, because the headphones are resting upon (or in some cases, in) the ear, the left ear driver of the headphones reproduces ONLY what the left mannequin ear recorded, and the right ear driver of the headphones does the same.

When loudspeakers are used, the listener no longer has the "guarantee" that his / her left ear will only hear sound from the left speaker, and that his / her ear will only hear sound from the right speaker. Invariably, there will be some degree of cross-talk; some of what is heard in the left ear will be information from the right channel, and vice versa.

So, here is where 'it depends' comes into play. Let's assume two sets of 'boundary conditions':

Scenario "A" - A Very "Dead" Room:

Suppose that you had two relatively compact speakers in front of you. Now, suppose that these were set up in a room whereby most of the room was heavily damped (due to absorptive material such as drapes, absorptive panels, etc).

If you sat very close to the speakers (i.e. near-field) then chances are, you would experience (primarily) the direct sound from the speakers and little of the reflected sound. In this way, most of the binaural effect would be preserved. However, this effect will also be a function of the speaker's dispersion characteristics (versus frequency) and of course, the listener-to-loudspeaker distance.

So, if you think about it, the closer you move to the speakers, the more you are in the direct field. The more you are in the direct field, the more this arrangement approximates the boundary conditions provided by headphones. Again, the key word here is "approximates"...

Scenario "B" - A Comparatively "Live" Room:

For a fair comparison, imagine that you now have the same speakers in front of you, only now, the room, having the same volume / geometry as the room in scenario "A", has had most to all of its absorptive material removed.

Under such conditions, it will be very difficult to listen primarily to the direct field. Instead, you will predominantly hear the reflected sound - which constitutes a lot of cross-talk. In this case, the binaural image will almost certainly erode into something far from binaural.

So in essence, the binaural listening experience, like all listening experiences, really relies upon the boundary conditions.

In other words, using headphones makes the listening experience independent on the acoustic boundary conditions in which the listener finds himself / herself - and this is true whether the source material is conventional stereo or binaural. The difference however is that in the binaural case, the spatial cues (left-to-right, front-to-back) will all be preserved when headphones are used.

Tchad Blake (who has done a fair amount of binaural recording) was once asked a similar question, and had this to say (excerpt taken from the article found at this link: BINAURAL EXCITEMENT ):

"I know people say that the problem with binaural is that there's no actual centre, that it's very difficult to get things to sound right in front, but I've never found that to be the case. Or, at least, only when you record something that's very close to your face -- then you can get the feeling that it's coming from behind. Binaural isn't perfect, you don't get 100% accurate localisation, but I simply love what it does. It sounds fantastic and really natural and spatial on headphones. I also love the way it translates to the speakers. It's no longer truly binaural when you listen to speakers, but it has its own sound, which is really cool, almost like a fishbowl effect. You get this depth that you don't get from any other kind of stereo miking technique. And finally, it's great for field recording, because it's cheap, really portable and really simple, and I'm not carrying anything in my hands, so I can walk around, take pictures, do all sorts of things."

Now...on to your next question, with regard to proximity...

Again, the answer here is "it depends". I have recorded solo piano or other instrument recitals, and ultimately it comes down to (once again...drumroll please... the boundary conditions).

What do I mean by this? OK, revisit the two scenarios that I discussed for playback of binaural over speakers, only now, imagine that you are going to record in the two spaces.

Clearly, if you were recording in a space (leike scenario "A", above) then it would be very easy to get primarily the direct sound - regardless of whether you used binaural microphones or conventional ones. In other words, the relatively 'dead' charcteristics of the room would mean that you would get very little room effect in the recording.

So, a lot would depend on the source (instrument, or vocal etc) and its timbre. If the source were one that produced frequencies that were easily absorbed by the room, then yes, you would get primarily the direct field, and placement (mic to source) would become very important.

Now consider the opposite ' scenario "B". In this case, the reverberation of the room is dominant, and starts to look like a true diffuse field (where measured pressure is independent of location). In such a case, whether using a binaural microphone or a traditonal approach, the room will play a major role in what is recorded.

So, in essence, binaural can't 'fix' bad acoustics nor more than a conventional technique. However, I will say this - where the acoustics are truly awful and highly reverberant, a (dimensionally) small microphone placed very close to the source or instrument will be dominated by the source and not by the room. However, this isn't necessarily realistic.

So, it seems like all things, there's a compromise. Does the project require accuracy and realism, or does the project require being part of a multi-track process etc. With binaural, it's a throwback to the days of direct to disc (for those who remember vinyl) - fi you get it wrong, there's little chance of an overdub.

Now, if you want to see and hear something really cool (Flamenco Guitar, recorded binaurally), check out this link: Humble Voice // Video // Snakecharmer-Binaural by Ottmar Liebert

All that was used here was the Neumann type KU-100 (my personal favorite) and a digital recorder. If you visit the site, please be sure to use headphones (but hey, they even remind you of that).

Anyway, it's a cool piece of music, and very convincing in its realism.

Phew...I hope that answered your questions. - Mark
#7
26th March 2010
Old 26th March 2010
  #7
Lives for gear
 
Joined: Oct 2008
Posts: 776

Ozpeter is offline
Quote:
Music for the headphone generation
Given that 80% of the world's youth spends 90% of its leisure time wearing earphones, I'm surprised that binaural sound hasn't aroused more interest in recent years. I suppose it's more relevant to classical/acoustic music than rock/pop.

Mike Skeet is another name that comes to mind when binaural sound is mentioned - Binaural Music and British Music label Recording Mike Skeet
#8
26th March 2010
Old 26th March 2010
  #8
Lives for gear
 
doug hazelrigg's Avatar
 
Joined: Jun 2009
Posts: 1,999

doug hazelrigg is offline
I am NOT trying to dismiss Mark's knowledge of the subject, but to my ears ORTF and binaural pretty much always sounded close if not the same
#9
27th March 2010
Old 27th March 2010
  #9
Lives for gear
 
Joined: Oct 2008
Posts: 776

Ozpeter is offline
I'd love to see some studies done (or the results of existing ones) on individual stereo perception. For instance, I once worked with a producer who appeared unable to tell the difference between stereo and mono on a classical recording (after I apologised for accidentally sending him a mono headphone feed for about an hour).

Binaural sound perhaps is perhaps the most individually-perceived form of stereo. People will listen to the same recording, and some will even claim to hear front/rear/height clues, and others will hear unfocussed normal stereo. A good one for a doctorial thesis, that.
#10
27th March 2010
Old 27th March 2010
  #10
Lives for gear
 
didier.brest's Avatar
 
Joined: Dec 2008
Posts: 2,407

didier.brest is offline
Quote:
Originally Posted by Binaural_Mark View Post
Phew...I hope that answered your questions. - Mark
Yes! Thank you a lot. The flamenco guitar experience is great.
#11
27th March 2010
Old 27th March 2010
  #11
Lives for gear
 
Joined: Oct 2008
Location: Espoo Finland
Posts: 1,070

Petrus is offline
Quote:
Originally Posted by Ozpeter View Post
I'd love to see some studies done (or the results of existing ones) on individual stereo perception. For instance, I once worked with a producer who appeared unable to tell the difference between stereo and mono on a classical recording (after I apologised for accidentally sending him a mono headphone feed for about an hour).

Binaural sound perhaps is perhaps the most individually-perceived form of stereo. People will listen to the same recording, and some will even claim to hear front/rear/height clues, and others will hear unfocussed normal stereo. A good one for a doctorial thesis, that.
There are a loads of studies done on these subjects, also doctoral level, and basically there is really not much to add to those. To find them you have to access academic databases, many of them are not available on the net (50 year old stuff) and many also are not in English (mostly German for obvious reasons). Stereo preception has been studed since thirties, nothing new in it.
#12
27th March 2010
Old 27th March 2010
  #12
Lives for gear
 
Teddy Ray's Avatar
 
Joined: Mar 2009
Location: Carolina is where they'll bury me.
Posts: 7,095

Teddy Ray is offline
read the life and times of alan dower blumlein. --robert charles alexander

and Handbook of recording engineering

the microphone book

and sound recording (those 3 by eargle)

youll be set.

no need for foreign languages or diving deep into university archives. right there in those books.
__________________
"I would shoot a man if he put me through autotune" - Charlie Louvin
#13
27th March 2010
Old 27th March 2010
  #13
Gear addict
 
Mark A. Jay's Avatar
 
Joined: Mar 2010
Location: Ann Arbor, Michigan

Send a message via Skype™ to Mark A. Jay
Mark A. Jay is offline
Binaural vs. ORTF Demo

Hey all, Doug said "I am NOT trying to dismiss Mark's knowledge of the subject, but to my ears ORTF and binaural pretty much always sounded close if not the same".

All I can say is this - as I am new to this forum and don't fully know my way around just yet, I have to become acclimated. However, once I do, I will dig through some of my tracks that I recorded simultaneously (binaural and ORTF syncjronously) and construct an A/B B/A A/B (switching back and forth between ORTF and binaural) type mp3 file, and post it here (if that sort of thing is allowed, which I assume it is).

I know - mp3s are compressed, but if I make the demo a 320 kbps file it'll preserve most of the fidelity. If I am truly industrious, I will write two versions of this comparison (but I may only write one). Those versions would be:

a) ORTF and binaural tracks normalized in terms of loudness

NOTE: this is a bit tricky because the frequency response of the microphones used to record the performances differs (i.e. the ORTF pair and mannequin head do not share similar frequency response). Therefore, I can normalize based on overall level or a frequency-specific region. In this instance, the comparison is a bit 'faulty' because there are teo differences - frequency response and sensitivity. In other words, if the mics I use in the ORTF had the same frequency response and sensitivity as the mics in the mannequin head, then the only difference in the demo would be due to the differnce in boundary conditions (that is, the physical properties of each configuration).

b) ORTF and binaurl tracks equalized and normalized. In this version of the demo, the frequency response of the ORTF microphones is made to be that of the mincrophones in the binaural mannequin. That's step 1. Step 2 is then to normalize the loudness levels. Once that's done, you now have a very fair and unbiased comparison between the two techniques, because the eq and normalization (of one type to the other) will make it so. That is, the loudness and timbre would see the same, with the only difference between the ORTF and binaural then being the spatial attributes.

c) I know I said two demo possibilities, but as I was typing what I did about 'demo b' the following came to mind. If I have time, I could take the inverse approach of demo 'b'. That is, make the binaural mics have the same frequency response as the cardiods used in the mannequin, and then normalize for loudness. In this variant, the timbre and loudness would both be normalized (as in demo 'b'), only in this case to the frequency response of the cardiods.

In any event, I will:

1) Look into where I can post mp3 demo materials on this site.

2) Dig up / construct some ORTF demo files as outlined.

3) Post the files (assuming it is possible)

4) Wait for feedback / comments / so forth.

In closing, let me say how cool I find this place (and the people here) to be.

Mark
#14
27th March 2010
Old 27th March 2010
  #14
Gear addict
 
Mark A. Jay's Avatar
 
Joined: Mar 2010
Location: Ann Arbor, Michigan

Send a message via Skype™ to Mark A. Jay
Mark A. Jay is offline
Ozpeter had this to say: "Given that 80% of the world's youth spends 90% of its leisure time wearing earphones, I'm surprised that binaural sound hasn't aroused more interest in recent years. I suppose it's more relevant to classical/acoustic music than rock/pop."

Indeed. It seems that the headphone generation are a natural audience for binaural. As far as why there is not more popular music out there that is recorded binaurally (and thus, being 'missed') has (in my opinion) a lot to do with the fact that it tends to be more relevant to classical and acoustic (as Oz mentioned) becuase the modus operandi of such performances is 'there is but one take'. In other words, symphonic, choral, acoustic performances are generally done with the idea that (unless stated expressedly to the contrary) no overdubs or sweetening will be done.

On the other hand, I have recorded some great live rock and pop shows (Sam Roberts Band, Sloan, The Reefermen) that turned out wonderfully. However, in those instances, the FOH engineer(s) know / knew their bands and their crews quite well, and thus, delivered a great FOH mix. Therein lies the down-side for binaural - it will be a very accurate 'document' of what was performed, but it will also possibly show the warts and all of a performance, because unlike multi-track, close-miked, and D.I. driven 'live' recordings that will be overdubbed later, the binaural (or ORTF, or XY, or M-S or whatever) will only be as good as the FOH mix, and what will be recorded will be an amalgm of the FOH mix, and the acoustics of the sapce. Lastly, of course, how the band / ensemble performs is of course the deciding factor.

Question: If I could get clearance from one of the bands (pop / rock / folk / classical) or symphonies that I have recorded, would the readers be interested in hearing samples? I'm just curious...

I had heard (for classical fans) that the Milwaukee Symphony Orchestra purchased a Neumann KU-100 (what I typically use) and has been recording their concerts binaurally and purportedly, these are available for download.

It's my belief (and I know this is going to make me seem like a 'hater' to some) that many musicians rely upon multiple takes, punch-ins, overdubs and the like, and again, with binaural (or any 'simple', live two-channel approach that does not rely upon a borad send from the FOH mix), everybody has to be 'on' (i.e. no mistakes) because you can't re-do a flubbed guitar lick or bass part (for example, taken from a D.I. into a front end).

Mark
#15
27th March 2010
Old 27th March 2010
  #15
Lives for gear
 
John Willett's Avatar
 
Joined: Feb 2008
Location: Oxfordshire, UK
Posts: 7,902

John Willett is offline
Smile

Quote:
Originally Posted by Binaural_Mark View Post
Anyway, ss I recall, ORTF spaces the mic diaphragms roughly 17 cm apart, and at an angle of 110 degrees. Cardiods are most often used.
ORTF *has* to be cardioids and is specified as 17cm and 110 degrees - any other polar-pattern, distance or angle is not ORTF.


Quote:
Originally Posted by didier.brest View Post
1) It has been often said that binaural works well for headphone listening, but not for loudspeaker listening. What do you think about this reserve?
Yes, this is true - but using a Jecklin or Schneider disk does convert much better on loudspeakers while still retaining a lot of what true binaural does.


NB: I am looking in from an extremely slow and very expensive GPRS link abroad, so have not had the chance to fully read this excellent thread - I will read it properly when I get back to Blighty.
__________________
John Willett
Sound-Link ProAudio Ltd.
Circle Sound Services

President - Fédération Internationale des Chasseurs de Sons

(and lots more - please look at my Profile)
#16
27th March 2010
Old 27th March 2010
  #16
Lives for gear
 
DanDaMan's Avatar
 
Joined: Feb 2009
Location: Bristol, England
Posts: 854

DanDaMan is offline
Quote:
Originally Posted by simonv View Post
Hi all,

I just came upon a funny mic which is a fake head with microphones in the ears.

Just wondering... is that hype?? Or is there any specific reason to use such a mic instead of using the trusty ORTF technique?

Thanks,
Simon
It's not hype, Binaural is designed to pick up sound exactly as we humans detect sound where as ORTF is a stereo mic technique which is very good at providing accurate localization of sound.
#17
27th March 2010
Old 27th March 2010
  #17
Lives for gear
 
DanDaMan's Avatar
 
Joined: Feb 2009
Location: Bristol, England
Posts: 854

DanDaMan is offline
Quote:
Originally Posted by Oroz View Post
Well, ORTF are 2 Cardioid microphones. Binaural are two Omni's .


I did not know that, but now you say it, It makes perfect sense
#18
27th March 2010
Old 27th March 2010
  #18
Lives for gear
 
Plush's Avatar
 
Joined: Jul 2003
Location: EARS/Chicago
Posts: 5,458

Plush is offline
MSO Download Store: Home


At the above address select binaural recordings under "browse by category." There you can audition and buy several excellent binaural recordings that I did for the Milwaukee Symphony. We used the Neumann 100 dummy head and hung that head for all concerts.

Of particular interest is the Greig Suite No. 1 from Peer Gynt. These are for headphone listening only.

There has been much research done on binaural recording and it is one of the earliest types of stereo. The current model Neumann 100 head is also equalized for listening on loudspeakers.
__________________
Atelier HudSonic, Chicago

EARS-Chicago, Engineering And Recording Society

http://www.ears-chicago.org/
Deaf before Dishonor

http://soundcloud.com/hudson-fair
#19
27th March 2010
Old 27th March 2010
  #19
Lives for gear
 
Piedpiper's Avatar
 
Joined: Jan 2009
Location: Between the Notes, Iowa
Posts: 3,085

Piedpiper is offline
Quote:
Originally Posted by Plush View Post
MSO Download Store: Home


At the above address select binaural recordings under "browse by category." There you can audition and buy several excellent binaural recordings that I did for the Milwaukee Symphony. We used the Neumann 100 dummy head and hung that head for all concerts.

Of particular interest is the Greig Suite No. 1 from Peer Gynt. These are for headphone listening only.

There has been much research done on binaural recording and it is one of the earliest types of stereo. The current model Neumann 100 head is also equalized for listening on loudspeakers.
How is it equalized differently?

I have used a Crown SASS-P stereo mic that is a variant on binaural. It uses PZM capsules in a configuration that mimics the dummy head without literally using a head. I position my speakers in a right triangle with the listening position, relatively close, 6 feet from me and 9 1/2 feet from each other. This works well for recordings using stereo micing. On non stereo miced recordings I sometimes prefer to move back a foot or so depending.
__________________
Tim Britton
producer, engineer, musician, audio sales
http://www.piedpiperprod.com
http://uilleanpipes.com

row, row, row your boat...
#20
27th March 2010
Old 27th March 2010
  #20
Gear addict
 
Mark A. Jay's Avatar
 
Joined: Mar 2010
Location: Ann Arbor, Michigan

Send a message via Skype™ to Mark A. Jay
Mark A. Jay is offline
SASS-P vs. Mannequin Heads

Quote:
Originally Posted by Piedpiper View Post
How is it equalized differently?

I have used a Crown SASS-P stereo mic that is a variant on binaural. It uses PZM capsules in a configuration that mimics the dummy head without literally using a head. I position my speakers in a right triangle with the listening position, relatively close, 6 feet from me and 9 1/2 feet from each other. This works well for recordings using stereo micing. On non stereo miced recordings I sometimes prefer to move back a foot or so depending.
I have used both of these types (various mannequin heads - Neumann, Head Acoustics, Bruel & Kjaer, Cortex, et al and the SASS-P stereo microphone), and while the SASS-P does a nice job of producing a good stereo signal, it's not binaural.

It literally comes down to the physics - the dimensions of the mannequin head (and also, its shape) play a big role in the inter-ear delay times and shadowing effects. The shape and dimensions of the (artificial) ears is also important, and (particularly) in the measurement community, a subject of often heated debate. That is, what ear dimensions and shapes are the 'right' ones and which are not. You wouldn't think this would be such a big deal, but when you really start thinking about all of the variables that comprise such a device...

Anyway, in measurement and research, artificial test subjects (i.e. crash dummies and the like) are modeled after specific percentile male / female height and weight (i.e. the 50th percentile male, 5th percentile female, etc) and I believe that there has been work done on classification of ear shapes and geometries (I know...but different things interest different people). As far as I know, ISO have not yet homolugated a 'standard' ear shape, and I think much of this is probably political as each manufacturer of heads has a vested interest, certainly from a noteriety if not financial standpoint (for having 'their' ear design chosen as a standard). Frankly, what the ISO should do is set engineering politics aside, and define a standard ear shape (and probably head shape as well). In this way, as a minimum, the ear (and hopefully) head dimensions, damping, mass, and so on could all be standardized - but I don't know how close to reality that is, because like any industry, product differentiation among competitive products always affects sway if not hold it.

Anyway, if you look at the head (proper) dimensions of the various binaural microphones out there, they are all very similar, save for the shape of the ears and the types of equalization that can be used (free-field (FF), diffuse field (DF), and some seemingly proprietary ones like Independent-of-Direction (ID).

Some binaural microphones however, include a torso as part of mannequin head assembly (and some manufacturers even sell absorptive vests that fit upon the torso so that the diffraction that would be produced by the torso can be minimized). I can see how, in some settings, with some ensembles / groups, the presence / absence of a torso could have an influence on what ultimately reaches the ears of the mannequin and thus, what ultimately reaches the ears of someone listening to the binaural recording. I can also see, however, how in some instances, the presence or absence of a torso is a moot point.

Anyway, the dimensional differences between the SASS-P and any of the mannequin heads out there is pretty big. Unless the SASS-P were equipped with a DSP to attempt to mimic the HRTF of a mannequin head, then it's still stereo.

However...

The SASS-P does have a boundary of sorts between the microphone elements, so some shadowing is taking place but it's not the same as found in a mannequin microphone.
#21
27th March 2010
Old 27th March 2010
  #21
Lives for gear
 
Piedpiper's Avatar
 
Joined: Jan 2009
Location: Between the Notes, Iowa
Posts: 3,085

Piedpiper is offline
There will always be those who wind out on details of questionable value and those who summarily dismiss even a cursory thought in that direction. Where would you say, in your educated opinion, lies an appropriate jumping off point with say, the concern about ear shape and head size in this discussion? And how does the difference between a well implemented dummy head and the SASS-P show up?
#22
27th March 2010
Old 27th March 2010
  #22
Lives for gear
 
doug hazelrigg's Avatar
 
Joined: Jun 2009
Posts: 1,999

doug hazelrigg is offline
Back in the 80's I was on an ORTF kick where I overdubbed EVERYTHING (except maybe bass guitar) using the same 2 mic setup. I'd even output a keyboard pad to a single speaker and track it in ORTF. You basically had to mix everything in your head beforehand -- you had to plan on each instrument's position before recording. With practice it worked pretty good. It had this kind of pleasant phasey effect. I would do wierd things like shake a tamborine and then walk around in the soundfield in front of an ORTF setup -- sounded unusual but incredibly real. This was mostly Pop music along with some Hard Rock. You'd run out of tracks real fast so you had to plan carefully ahead.

Today, when I do record something in stereo, I use M-S. It's much easier to mix if you're compiling overdubs
#23
28th March 2010
Old 28th March 2010
  #23
Gear addict
 
Mark A. Jay's Avatar
 
Joined: Mar 2010
Location: Ann Arbor, Michigan

Send a message via Skype™ to Mark A. Jay
Mark A. Jay is offline
SASS-P and Binaural

Quote:
Originally Posted by Piedpiper View Post
There will always be those who wind out on details of questionable value and those who summarily dismiss even a cursory thought in that direction. Where would you say, in your educated opinion, lies an appropriate jumping off point with say, the concern about ear shape and head size in this discussion? And how does the difference between a well implemented dummy head and the SASS-P show up?
I'm sorry if I came across as dismissive in that last post. I may have been too focused on describing differences technically and issues rather than focusing on the bigger issue - and your point is valid. It may have sounded like I was sweeping awau the aestheticvalue of the SASS-P, and that wasn't my intention.

In any good recording (I think), the result is an amalgam of the artistic as well as the mathematical and theoretical - so you are right to focus on the end result (which is ultimately appreciated as an art / craft, rather than a mathematical exercise), because after all, that's truly all that matters.

As far as the ear shape goes, frankly - I just don't know what the 'technically correct' answer to this one is. That's kind of why I had groused a bit about a standard for such a shape being non-existant. There are some very clever people out there (mostly working in reasearch at various universities) who have given this issues a great deal more thought (and have mountains of data upon which to make an informed decision) than I have and as such, are far better equipped to chime in about the subtleties that the pinnae, its sahpe, density andf so on have upon one's hearing (timbre / directivity-wise). All I know is that this is something with which we are all acquainted (for instance, when we effectively extend our pannae when we cup our ears).

Again, if the SASS-P sounds good and fits the aesthetic as envisioned by the concerned parties that's all that matters. Right?

As far as the ear / lack of ear on a mannequin vs. the SASS-P, I see that as significant. Mathematically ... and percepturally.

Thus, a well-implemented mannequin head and a SASS-P placed close to it will sound quite different, most likely due to their differences in frequency response, but even if that were corrected (one to the other's response, SASS-P to mannequin or vice versa), the ears and ear canal play a big role in what is perceived in the recording (what is effectively incident upon the microphone elements within each ear canal).

Mark
#24
28th March 2010
Old 28th March 2010
  #24
Lives for gear
 
Plush's Avatar
 
Joined: Jul 2003
Location: EARS/Chicago
Posts: 5,458

Plush is offline
Hello Piedpiper,

You can read some AES papers and Neumann literature to learn about the various iterations of the Neumann head. One of the trickiest subjects is getting binaural head recordings to also play back properly on loudspeakers.

Even though binaural recordings are made to be listened to on headphones, much research has gone in to getting the good recorded stereo from the head to be reproduce-able on loudspeakers. Currently the Neumann 100 head is also a very good stereo mic.

Crown SASS mics are OK but they are not binaural for the reasons already stated. The rubber ear and the ear canal must be present with a real barrier recording method to be real binaural. Combining pressure zone mics with a semi-barrier is falsely advertised as binaural. I am not a fan of their tonal quality either.

I am not an expert concerning the equalization applied in the Neumann 100 head to get it to sound good on speakers. I am a practitioner who has used the head for a lot of recordings. It is a very worthwhile purchase in my opinion.
#25
28th March 2010
Old 28th March 2010
  #25
Lives for gear
 
didier.brest's Avatar
 
Joined: Dec 2008
Posts: 2,407

didier.brest is offline
Hi Plush,

for which applications (orchestral, chamber, solo ?) the KU 100 is especially suitable ?
#26
28th March 2010
Old 28th March 2010
  #26
Gear addict
 
MBishopSFX's Avatar
 
Joined: Feb 2005
Location: NE Ohio
Posts: 397

MBishopSFX is offline
I've used the KU-100 on a couple of hundred major orchestral and small group (including jazz and blues) recording sessions. I love this mic and am amazed at all the uses I've found for ol' Fritz over the years. On orchestral sessions, the KU-100 is the center "stereo pair" with omni outriggers and other mics added as needed. I've used the KU-100 in that position simply because the imaging it achieves cannot be matched by other stereo mic techniques. The KU-100 has also made a terrific drum kit overhead "pair."

The vast majority of my sessions are not considered to be binaural because of the additional mics employed. Here are a few projects in which the KU-100 was the sole pickup. These are binaural recordings that play best in headphones, but are also successful on speakers.

Delta Crossroads : Robert Lockwood, Jr. : Concord Music Group

From The Isles To The Courts : Ensemble Galilei : Concord Music Group

Angeli Music Of Angels : Ensemble P.A.N. With Tapestry : Concord Music Group

.
__________________
With Best Regards,

Michael Bishop

Learn why Everything's Better in 5/4!
http://Recording.Pro
#27
28th March 2010
Old 28th March 2010
  #27
Gear addict
 
Larry Elliott's Avatar
 
Joined: Aug 2006
Location: Auckland , New Zealand
Posts: 476

Send a message via AIM to Larry Elliott Send a message via Skype™ to Larry Elliott
Larry Elliott is offline
Quote:
Originally Posted by MBishopSFX View Post
I've used the KU-100 on a couple of hundred major orchestral and small group (including jazz and blues) recording sessions. .... On orchestral sessions, the KU-100 is the center "stereo pair" with omni outriggers and other mics added as needed.

Angeli Music Of Angels : Ensemble P.A.N. With Tapestry : Concord Music Group

.
Thanks for this Michael.

The Music of Angels is spectacular!

Larry
#28
28th March 2010
Old 28th March 2010
  #28
Lives for gear
 
didier.brest's Avatar
 
Joined: Dec 2008
Posts: 2,407

didier.brest is offline
Thanks! I prefer the ensemble recordings where the stereo image is so much natural.
#29
28th March 2010
Old 28th March 2010
  #29
Lives for gear
 
Joined: Oct 2008
Posts: 776

Ozpeter is offline
Hmmm. I listened on headphones to several samples of the 'Angels' recording and found the exaggerated width to be very distracting. Not so bad on speakers. I opened some samples in Audition together with a Kings College Cambridge choral recording for comparison, and checking the phase analysis the amount of out-of-phase components in the 'Angels' recordings was very obvious - the 'ball' in the display was often bouncing around in the rear half rather than the front. Using Audition's stereo width effect, and separately the Voxengo MSED VST MS processing effect, it was possible by eye and ear to obtain a similar sound on the Kings College recording by adding about 6dB to the side channel. Or one could add about 6dB to the mid channel of the 'Angels' recording and the perceived width became more 'normal' and the phase display likewise - though the sound was significantly less 'lush' of course.

In other words, perhaps one can emulate the general sound of a binaural recording simply by adding to the side channel information of a normal recording. However, as I've mentioned before, stereo perception from one individual to another can vary widely and what I hear may well differ from what others hear (or from what I'm supposed to be able to hear).

Was the width of this recording manipulated at all, or is what we're hearing more or less the natural image the dummy head produced?
#30
29th March 2010
Old 29th March 2010
  #30
Gear addict
 
Broken's Avatar
 
Joined: Jun 2004
Posts: 452

Broken is offline
Quote:
Originally Posted by didier.brest View Post
Thank you much Mark for you very interesting post. I've two questions:

1) It has been often said that binaural works well for headphone listening, but not for loudspeaker listening. What do you think about this reserve?

2) Is binaural recording of any interest for recording at close distance a single instrument, a piano for instance ?
1. Binaural only works well for loud speaker if you secure the listeners head in the right place. It is totally useless for anything other than lab listening tests.

It's ideal for headphones.

2. Binaual is great if you want to capture the elevation and azimuth. It isn't perfect and localisation errors happen around specific angles. This is due to the fact that we rely a lot on head movements to pin-point a sound source.

Head tracking helps but this technology isn't widely used.....yet.
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+ 
 
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Nu-tra / Remote Possibilities in Acoustic Music & Location Recording
13
MeatPye / Music Computers
0
sourceslut / Remote Possibilities in Acoustic Music & Location Recording
17
octatonic / Q & A with Tchad Blake
25
jbuntz / Remote Possibilities in Acoustic Music & Location Recording
11

Forum Jump

SEO by vBSEO ©2011, Crawlability, Inc.