Mid-Side and Blumlein recording with the Zoom H4n
I was surprised to find that a Mid-Side decoder function is included in the H4n, this is usually a feature of high end field recorders but it’s easy to do in the digital realm so the Zoom engineers could “throw it in” without much added cost. I love fooling with different mic arrangements so I had to set up a couple of figure 8 mics and do some recordings.
Alan Blumlein
In 1931 Alan Blumlein applied for a patent on what we now call stereo sound. He described techniques for creating the auditory illusion of placement and depth using two (or more) channels of recorded and reproduced sound. As part of this patent application he described an array of two bidirectional mics in a coincident X/Y arrangement that we today call a Blumlein array:
Two bidirectional or figure 8 mics in a coincident XY
In the same document he described the Mid-Side mic arrangement. Rather than a crossed pair of mics at 45 degrees to the source, the Mid-Side array has the Mid mic pointing directly at the source and the Side mic at right angles to the source. The Mid mic can be any pattern but the Side mic must be a bidirectional or figure-8 mic.
Two bidirectional or figure 8 mics in Mid-Side
Mid-Side Recording
This article from emusician.com discusses the technical details of Mid-Side recording in some depth. Note that the use of different patterns for the Mid mic gives different results after decoding.
When I record in Mid-Side using my home recording chain it’s startling to see the difference in the levels of the Mid and Side channels. With both channels set to the same gain the Side channel barely shows any signal.
Decoding the Mid and Side channels to Left and Right is easy in a multitrack recording program. Simply place the Mid channel on one track panned to the center. Place the Side channel on a track panned hard left. Then place a copy of the Side channel inverted in polarity on a track panned hard right. When you mix the three tracks down you’ll have two stereo tracks.
Here’s a video I made demonstrating Mid-Side and Blumlein arrays with some demonstrations of the stereo field.
There are a few ? ? ? problems with the video, please accept my apologies. The bubble wrap stereo demo is funny (to me anyway), but the clipping caused by the amazingly loud pops wrecked the stereo field. Listen to the incidental noises for a better idea of how these recording techniques work. And the distortion is not there in the my recording, it was introduced by the YouTube compression process.
Mid-Side and the H4n
I also missed several opportunities to actually address some details of the H4n. For instance, the level metering shows levels after the signal is decoded. I think there’s a good chance of clipping the Mid mic and not seeing it in the decoded signal.
When setting up the Mid-Side connections on the H4n, notice that there’s a menu selection to identify which channel is the Mid and which is the Side.
Let me wrap up by mentioning that I sent my H4n back to the seller. I think it’s a pretty interesting recorder, but for my regular uses the H2 does at least as good a job. When I want to do more serious recording I just turn on my home studio system. I’m looking forward to continued progress in these handheld recorders, though. I just can’t resist the idea of a world class recording chain in my pocket.
This entry was posted on Saturday, March 21st, 2009 at 8:50 am and is filed under Audio, Tutorials. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
Russell Dawkins said in post # 1,
on April 9th, 2009 at 12:07 pm
I noticed the Blumlein recording of the guitar sounded considerably more diffuse than the MS version. I am guessing this is a function of the relatively poor off-axis response of most circular diaphragm condensors, especially large diaphragm condensors.
Although I know you have returned the unit, a repeat of this test with ribbon mics would be enlightening, as ribbons have very good off axis response – at least horizontal, when the ribbon element is vertical, due to the narrow dimension of the ribbon when compared to the diameter of a circular diaphragm.
With ribbons, the MS should sound more like the Blumlein.
I recorded orchestras and choirs for years with a Royer SF12 in MS and so often found that I decoded it 50/50 MS (which is the equivalent of Blumlein) that I eventually went straight to Blumlein to save the degradation of the decoding, which at that time was done in the analog domain.
Connected with all this (off axis response quality considerations) is the reason ribbons do so well with room tone – the response is pretty constant in a torus around the mic, so everything within that is of good fidelity.
I used to wonder why the room that sounded not bad in real life sounded so lousy when captured by a microphone. Then I figured out that when recording at any distance much of what is heard by the mic arrives off axis, and “LDCs” sound variously lousy off axis.
Fran said in post # 2,
on April 9th, 2009 at 2:41 pm
Russell, thanks for stopping by. I appreciate your opinion on the sound of the recording. I’ve only been doing this for a few years and I don’t think I’ll ever catch up on the listening end of things.
A friend of mine has been using an AEA R88 stereo ribbon, and to me his M/S and Blum sound identical. And gorgeous. I was really tempted by a recent Ebay auction but I resisted.
Do you have an opinion on the Schoeps MK8 compared to a pair of ribbons? That’s another “investment” I’ve been resisting.
I noticed on Gearslutz that you use K&H O300s. Any opinion on the O110?
Thanks again for your comments,
Fran
Anton Emery said in post # 3,
on April 15th, 2009 at 7:44 am
Nice post Fran. I have a question, would the following be the correct way to record MS as well?
Two mics, a cardoid for the mid mic, and a figure eight mic for the side mic. The mid mic is assigned to a mono track, the figure 8 mic to a stereo track. Record my guitar, then invert the polarity on one of the channels of the stereo track using a Logic plugin. Then do you adjust the Mid to Side balance by adjusting the volume of the respective tracks? Or does it involve panning as well.
Anton
Fran Guidry said in post # 4,
on April 15th, 2009 at 8:06 am
Hey, Anton,
The setup you describe is exactly what you need. The inverted track is the one on the “back” side of the mid mic. So if the mid mic is arranged with its front side facing right, you invert the left track.
That should do it.
Are you cutting more tracks at home? I need to book some time with Doug myself, so I can try out that R88!
Thanks for stopping by,
Fran
Anton Emery said in post # 5,
on April 16th, 2009 at 8:17 am
Not cutting any tracks at home at the moment, just lusting over some recording gear. A friend might be leaving his MOTU 8pre and some mics in my care for a bit, so perhaps i will have some gear to experiment with. I would like to get an interface and some mics for myself eventually, but with doing the recording for album at Doug’s, it just doesnt make sense at the moment.
I hope to head down to Doug’s in the next few months to finish up the tracks for my cd. Got 4 more to go. Its really nice working at his place, such a relaxing environment, and nice guitars sitting around to play.
Anton
Miguel Ramses said in post # 6,
on October 12th, 2010 at 6:31 am
I’m totally a beginner with mics and all this stuff and i’m just starting to learn a lot from all this=) Thank God there are people like you=) I have a problem with my zoomh4n. Although this does not have anything to do with mid side or blumlein, I just wonder how you guys are able to achieve complete silence when recording without the sound of..Actually I turned on all the noise reductions and lo cut filters and I could still here the background and it pretty much is annoying. Can anyone help me please=(
Fran Guidry said in post # 7,
on October 12th, 2010 at 7:20 am
There’s an old saying: “Silence is golden” – in recording we can say “Silence costs a lot of gold.” Getting rid of background noise takes a lot of money and time and effort, you basically have to build a second room inside your room, seal every crack, build double thick walls, and on and on. This is one big reason why traditional studios were so expensive.
Those of us recording at home simply have to put up with background noise. I wait until midnight to do my recording, close all the doors and windows, and if a car drives by during a quiet passage I do it over.
Sorry,
Fran
david said in post # 8,
on March 6th, 2011 at 3:57 pm
hi
i have a zoom h4n and is a pretty good device as i have an old tdm station too..
but i like the on the go thing that it does ..
i would ask if somebody tried how is this device with ribbon mics.
ribbon capsule usually always need a bit of more gain .. but i’m not sure if the zoom recorder can handle the gap..and then give a very low recording resulting cutting resolution e make a useless or to noisy recording
thanks for helping me
Fran Guidry said in post # 9,
on March 6th, 2011 at 5:48 pm
Hi, David,
If you compare the H4n to the other recorders in this post: http://www.homebrewedmusic.com/2009/03/07/first-look-at-the-zoom-h4n/ I think you’ll agree that the H4n is a bit noisy compared to a high quality recording chain. The lower output of a ribbon mic will result in a less satisfactory signal to noise ratio. Of course, if you have a loud source there will be no problem – something like a horn or drum set or electric guitar should work very well.
Fran
david said in post # 10,
on March 8th, 2011 at 4:00 pm
thanks Fran
i’m reading the comparsions .. yes is more noisy that what i supposed..
at this point i really have to leave the ribbon mic idea…
the only way to do a good job is to use any kind of mic but with an external micpre,
the one i saw is the Sound Devices MixPre wich is very good with 66 Db of gain
not so expensive ..but them is not in budget purposes anymore … i think
at this point i might need a condenser
and focus on something that have a good sensitivity, find a good fig 8 or multipattern mic
also .. good for the buck
any suggestion beetween NT2A OKTAVA or whatever…
i already have couple of mic for covering vocals cabs drums and acustic guitar
so this time i should focus on a MS ambient recording .. probably a bright ,or better , flat response mic
thanks again for helping me
Fran Guidry said in post # 11,
on March 8th, 2011 at 5:05 pm
David, I’ve used the NT2a, the NT2000, the Shure KSM44, the AT4050, and the CAD M179 as figure 8 mics. The CAD is the least expensive. To me, the difference between mics is generally pretty subtle. You can hear some of these mics, but only in cardioid, in some of the other posts here.
Fran
Andrew Kraus said in post # 12,
on June 23rd, 2011 at 2:08 pm
Thanks for the very interesting write up on Blumlein mid-side and plain ole Blumlein. To my ears, the plain ole Blumlein sounded more natural, and the spatial locations were more distinct. It also sounded less mid-range boomy to use a technical term.
Regarding your “set-up”: I’m interested in the details of your Blumlein bar. It looks as if you’re using a “On-Stage” 19in boom screwed on top of a mic stand, and you’ve mounted your microphone (couldn’t make out the brand or type from the video) shock mounts to clamps (were these MPC 10 clamps?) screwed onto the bar.
Have I got any of this right?
Thanks much for the article.
PS – the mics I’ve decided on for classical piano are CM-414 from Dave Thomas at Advanced Audio in Canada. To my ears they are darn close to the sound of the old AKG C-414s, a very nice microphone for what I do.
Fran Guidry said in post # 13,
on June 23rd, 2011 at 3:12 pm
Hi, Andrew. The mics are a pair of Rodes, NT2000 and NT2a. They differ in their electrical switching options but use the same capsule and spec the same.
The stand is a short DR boom, designed for drum miking, I think. The On-Stage device you see is an elbow. It then holds a part made by Sabra Som, something like this: http://www.oktava.com/inc/sdetail/284
I bought a hex shaped aluminum bar to replace the very short steel bar I got with the Sabra Som parts. Additional bits from Sabra Som connected the shock mounts from Rode. It wasn’t a really satisfactory setup, the aluminum bar isn’t really strong enough to support two fairly heavy LD mics.
Have fun with your new mics!
Fran
David Powsner said in post # 14,
on July 9th, 2012 at 6:33 am
Fran,
I came across your site while trying to understand M/S mode on my H4N, which I just purchased (yes, rather late “in the game,” as this product goes). I understand your discussion of the use of M/S mode on the H4N when the inputs are coming from auxiliary mics. However, I’m curious why the H4N also permits use of M/S mode with the on-board mic’s only. My guess is that, in such a configuration, the internal circuitry converts the X-Y signal from those mics into M/S form. I’m not sure the utility of this–other than for movie-making or others who demand their audio signals in that “format.” Any thoughts?
Thanks,
Dave
Fran Guidry said in post # 15,
on July 9th, 2012 at 8:03 am
Wow, I certainly missed that detail. I just examined the documentation for the H4n and found the setting you refer to. Since I don’t have an H4n to try I can only conjecture.
The options in this area allow you to set the side input to MUTE, which would then give you the Mid mic only, resulting in a mono output.
As for your idea that the setting will deliver an “encoded” mid-side pair of signals, I can only suggest that you test to see if that’s the result. As a rule, mid-side in encoded form will have a much larger Mid signal and smaller Side, and when they’re processed by an external matrixing tool you’ll have a stereo result. I’d be interested to hear what the result will be.
Fran
David Powsner said in post # 16,
on July 9th, 2012 at 12:23 pm
Fran,
Thanks.
I’ve begun experimenting and got a result that seemed partially consistent with guess but not fully: the “mid” audio (basically, my speaking directly in front of the on-board mics) got encoded into the left channel as I would have guessed, given the settings I provided. However, the side audio (i.e., my speaking to the left and, then, to the rightof the on-board mics) seemed to get encoded into the left and right channels. My testing was fairly lame, though, so I’ll try more and report back.
Dave
Fran Guidry said in post # 17,
on July 9th, 2012 at 2:04 pm
I’m sorry, David, but I forgot one crucial fact – the built-in mics are replaced by mics connected via the 1/8″ stereo mic input. So if you attached a mid-side array through that connector you could decode via this menu selection.
It honestly makes basically no sense to apply mid-side decoding to an XY pair of cardioids.
Fran
Matej Kovacic said in post # 18,
on March 14th, 2017 at 4:09 pm
Hi,
I wanted to ask you how good decoding system is in the Zoom H4n PRO? I do the field recordings also and I was thinking to try once what advantages could I have to try this way of recordings outdoors. I don´t know if it would have any advantage.. Just thoughts..
And I wanted to ask you something about the MS Matrix. I don´t have the Figure Of 8 Mic. If I place 2 Condenser Cardioid Mics in 180° would make the same effect of Figure Of 8 like if I would have the FO8 mic?
Thanks for your answers!
Matej
Fran Guidry said in post # 19,
on March 14th, 2017 at 5:15 pm
I haven’t used an H4nPro so I can’t offer hands-on experience. But decoding Mid/Side is really very simple so I would expect that the performance is limited by other issues.
Back to back cardioids can be used but you’ll need to invert one of the mics then sum the two in order to synthesize your side mic, adding a bit of complexity. And the distance between the two mics reduces the quality of the Mid/Side capture. However, this the approach used by Zoom in their mid-side systems, as do others. In fact, a dual diaphragm switchable mic in Fig 8 mode is just two cardioids back to back but quite close together.
Fran
Richard Straight said in post # 20,
on January 24th, 2018 at 4:54 pm
Recording guitar and vocal on H4n with no need to overdub anything, so only need two tracks. But I want to have some of both on both left and right channel. Using a very nice condenser for the vocal and internal guitar pick inside the Martin D-41 recording both at the same time. Do I need to input into a mixer and then out into an aux in to accomplish this?
Fran Guidry said in post # 21,
on January 26th, 2018 at 7:04 pm
I haven’t had my H4n for a long time, and I didn’t explore it in great detail, but I believe you can configure the H4n to record two mono tracks then mix them to taste. You might try the Zoom forum with your question: http://zoomforum.us/viewforum.php?f=15
Fran