Dude… I cannot thank you enough. This has literally kept me up at night. Like you, it was driving me crazy. One thing- in my case, of course low budget home recording without many mics to choose from, I always found myself wanting- needing a matched stereo pair of mics for recording drums or wide acoustic, etc… so for me mid side has always been my BFF and without it I couldn’t have done many of my earlier recordings. But when tracking acoustic guitar I always ended up with that lean to one side. It made sense to me with drums that the left and right had variance but I could never understand that constant lean with acoustic guitar. It seems so obvious now. Thank you!!!!!!!
Something else to try, take the 414s out of the stereo mount and put them on individual stands (unless you can rotate that rig 90 degrees like the hands on the face of clock). Then aim your cardioid mic capsule the way you used to for mono, capturing the sound you like, except this time, aim the XLR up the neck. Then place your figure 8 capsule perpendicular to even center the sound hole, butted up to the cardioid, with it’s XLR aimed down the body. Now you’re recording the figure8 toward the floor and ceiling though, but with even SPL. The way you’re sitting evenly between the left and right walls, you may wanna sit on a stool so that the guitar is at say 4 feet if you have 8 foot ceilings. Evening the reverb times. Then, when you build the MS Matrix exactly the same as you did in reaper, you’ll effectively be rotating the listeners 90 degrees, but they’ll never know. If your floor and ceiling are totally different materials, carpet/hardwood/tile vs drywall/cloud absorber/diffuser, A) that effect may sound cool, B) you could throw down a rug, blanket, board, change the reverb of the floor to match or to taste. Part of MS Micing is the controlled stereo effect, but part is the natural reverb/room sound. BTW, the way you had it first, you could always lower the level of the Left figure 8. You could also make those left and right tracks into 1 stereo track, then the pan becomes a balance. You’ll also end up EQing the sides, probably rolling off the lows, and that’s easier/cleaner on 1 stereo track. Lastly, as you noted, you don’t need matching microphones for MS, part of the beauty, so maybe you have something else, even a cheapy that fits better in the mix. But you can’t go wrong with a matched pair of 414s! Last thought, when you just pop a LDC into figure 8, pop on some headphones and talk into each side. Most sound very different from one side to the other, the perfect 8 ain’t so perfect. If I recall correctly, the 414’s “backside” is a bit darker, let’s call that the Darkside! Well if that is the case, maybe aim the Darkside at the bright reflective ceiling, and the Frontside at the carpeted floor. Like all things recording, learn your gear and what it can do, experiment, and play. That’s exactly what I saw in your video in the first place!
Great comment, one note. When you talk into opposite sides of a figure 8 while wearing headphones one side will sound different because it is out of phase with your skull. You would need to invert the phase when speaking into the back to avoid that. Your tip on going floor to ceiling blew my mind.
Awesome insight. I imagine now too trying fig 8 in front of sound hole w it’s off axis pointed at the sound so it’s not boomy, then use two other mics cardioid spaced maybe about 12-16 inches equally spaced from central mic like wide ortf. Experiment with levels and panning of these alongside panning the two central +duplication; two tracks at full L/R; two tracks at say 3 oclock. Might have to flip polarity on two occasions but I think I’ll try it. I like idea of louder fuller hard panning and no center track so that other sounds like a snare or vocal can breath separately.
@@chinmeysway There you go. Yeah experiment, have fun, find something new. “If it sounds good, it is good!” One word of caution though: the whole beauty of the Mid-Side Stereo technique is that the sound hits the capsules of the two mics at precisely the same time, thereby mostly eliminating or minimizing phase cancellations. Move the mics apart and different sound gets to different capsules at different times. This isn’t necessarily a problem, just a different phenomenon to factor in when processing.
@@G_handle totally.. yeah i am thinking about that aspect too. im just so interested in creating new mic technics. its interesting that theres a few standards, but im sure they can be thought of as loose guides. i wonder about why the fig8 mic is imperative- if its just to be duplicated as a mono signal, can it just be a cardiod pattern pointed to one of the sides, rather than fig 8, for instance.
@@chinmeysway okay so..... A) Absolutely experiment with every conceivable mic configuration you can come up with. If it sounds good, it is good. And sounding good is subjectively Your opinion, the only opinion that matters. It's Your music. B) The Objective reason why Figure-8 pickup pattern is "imperative" in a "True M.S." technique, is: 1) A figure-8 is a single* diaphragm generating a single mono signal, 2) But it's equally sensitive from both front and back, and has nulls on the sides, 3a) So when you duplicate that single signal (or any single signal), then sum both together, it doubles in loudness...Perfectly. 3b) If you then reverse the polarity of one of the two duplicates, then sum those together, they perfectly null. Because they are identical signals in reverse. So 2+2=4. But 2-2=0. 4a) Now, while the figure-8 "Side" Mic will PERFECTLY phase cancel when summed with its negative self, it will also PARTIALLY phase cancel when summed with a signal from any Other microphone recording the same source. 4b) By placing the "Side" mic's diaphragm as close to the "Mid" mic's diaphragm as possible (co-incident), the 2 Different mics will receive the acoustic energy at about the same TIME, but... 4c) remember that the Figure-8 "Side" mic has its Null facing towards the source, and the "Mid" mic is either and Omni or is directionally facing the source (usually cardioid, but any pattern can be used). So the two mics catch the same source at the same time, but with different sensitivity and directionality. The Side mic is directionally least sensitive where the Mid mic is most sensitive. 4d) Therefore when the acoustic energy is more on one side or the other of the single figure-8 diaphragm, in will physically move in or out from its resting position in the center, and the audio signal will be generated to reflect that push pull with either positive or negative values. 4e) Then that single mono signal's positive or negative waveform value, gets duplicated and just becomes two copies, 4F) When you flip the polarity on one of the copies, they're now still just two signals out of phase with each other, however depending upon which copy you flipped, one will contain useful information from the Right and the other from the Left, but they're not at all stereo yet. 4G) It's only when you Sum either "Side" signal with the "Mid" mic's signal that get a Left or a Right signal. Remember that the "Mid" mic is mono and contains Both the Left and Right information. 5) So now the MS Technique has two tricks up its sleeve: Stereo and Mono. Imagine this: Two Mics, Two Y-Cables, Four Channels on a Mixer. The "Mid" mic is going to channels 1 & 2, the "Side" mic is going to channels 3 & 4. The 'odd' channels are panned Left, the 'even' channels panned Right. If you push up faders 1 & 2 to unity you get a Phantom Center. Pull the faders down. If you push up faders 3 & 4 to unity you get a Phantom Center. But roomy, because the Null was facing the source. Now Invert the Polarity on either 3 or 4. You get a weird Phantom Phase Cancelation acoustically between your speakers. Pull those faders down. Now push 1 & 2 back up, and then blend in 3 & 4 and that Phantom Center will widen out and gain Room around it. Amazing! Thats why you did all this! Push as much Width into it as sounds good to you. Your two Mono Microphones have just created a Stereo Image, that You can control. Every other form of stereo micing, you get what you get. You may prefer the Stereo Image from another technique, but stay tuned... If your mixer has one, press the MONO button. The Left & Right will be summed together, meaning that the two "Side" mic signals will cancel each other out, and you'll be left with a Perfect recreation of the "Mid" mic by itself again. However wide you make your Stereo Mix, if the end user is listening on a Mono system, the wide information just disappears without a trace. So you have a highly controllable Stereo Image, And Perfect Mono Compatibility. 6) Back to the math: Remember: 2 + 2= 4 & 2 - 2= 0 Well instead think: 2 + 2= 4 & 2 + (-2)= 0 2 + (-2)= 0 is what's happening with the "Side" signals when they get summed to Mono. In Stereo mode though, they aren't summed with each other, they're summed with copies of the "Mid" mic's signal that are hard panned Left and Right. So they're not "seeing" each other, they're adding and subtracting with that other diaphragm on two isolated channels, Left and Right. The mono "Mid" mic was capturing all of the Left and Right energy in the air, and now when blended with one side of the figure-8 or the other, its Phase Interaction is either adding when the capsules were going the same way or cancelling when they were going in opposite directions. This works, because you have Both directions from the Same diaphram. So the math for Stereo mode is more like: Left = M + S Right = M + (-S) In Mono mode: Left + Right= Mono So: (M + S) + (M + (-S)) = M + M (In a Mono button it should be (L + R) /2 = Mono. So the level doesn't double) That took Way too much time!!!! Last thought: Back to A! Experiment. If it sounds, good it is good. So if you replace the Figure-8 with 2 SDCs, you won't get the perfect cancelation, but you will get cancelation. And it may sound amazing. And it will almost certainly sound Wider, because the differences will be greater between Left & Right. So understand the "rules", then break em'!!!!!!!!!! (Let me add this disclaimer at the end: This is my understanding. If something ain't quite right, Please tell me. I love to learn, and see being wrong as an opportunity for growth.)
Thanks for the video, it made me rethink about this problem I had also encountered. I don't think, however, the key issue is the position, but, instead, the similarity of the signals between the M and S mics, after all, the S mic is just a mono signal, so it should always cancel itself no matter the position. Let's suppose that (absurdly) we had the exact same signal in both mics. After the MS processing, in one speaker we would have A+A and in the other speaker we would have A-A, leaving us with 2A on one side and 0 on the other (I am ignoring pan law to simplify the example). So the problem is not the position, but a technique that has a tendency to emphasize one side. When you distance the microphones from the guitar or center them in relation to the source, the S mic gets more reflections and less direct sound, making the signal from both mics more different and less phase aligned, hence, less cancellation between the S flipped side and the mid side occurs resulting in a more centered image. A bigger room will always mean more delay in the S signal, making both signals less similar. Anything that makes both signals more different, like moving mics, using different mics, delaying the S signal to minimize cancellation, or getting rid of the lows in the S side (since most cancellation occurs in the lows because they are longer waves) will help to solve the issue. So, if you like your acoustic sound up close, you can get that in MS by recording as you did (it's preferable if you use a different mic for the sides though, it doesn't even have to be a figure eight), then manually delaying the S signal some milliseconds till it sounds centered and then removing the lows from the sides with a MS EQ. I hope this helps.
While mic placement is always a huge concern, and certainly with the m/s technique, a quick tip I learned (I can't remember where) is to roll off the low end on your sides (the two panned channels) until you see the signal fit in the centre. I think part of this problem has to do with the fact that one side of the figure 8 is favouring the heel or body of the guitar, while the other side is favouring the neck. The bottom side is beefy, the neck side is more airy. So I've found that while it's well and good to find optimal mic position, much of the problem can be solved with rolling off your low end on the two side channels until the signal fits centre. You don't need or want the beefy low end on the sides anyway, that will come from the centre cardioid signal.
@@YHRS Indeed! It's a good solution for me as I do have a tendency to move a bit too much (while trying not to) while playing. It's not so much an issue with other stereo mic techniques, and certainly no issue with a mono record, but it's pretty finicky when recording in M/S. I find the slightest shift of the body can really play hell with the balance.
Hang on hang on- what you need to do is do what this chap did and move the source or the mics so you can eq both side signals together. Bc I guarantee you’re introducing unforeseen phase problems into your recording by eqing individual sides. I’m sure you’ve run into this when trying to eq a single spot mic on a drum kit. It can get weird. So the safest bet is to track it in a way that lessens the need to eq anything at all. That’s my humble opinion. Although, I only just found out WHY my signal on an acoustic was leaning to one side. So what do I know
Of course the less eq the better. Having said that, additive eq is much more problematic than subrtractive. Perhaps I wasn't clear. I'm not suggesting individual eq adjustments to each side, but simply rolling off the low end on both sides together until the signal output focuses in the centre. Usually anywhere up to around 150hz or so off of the sides. Place mics as best you can, and then try it. It works. Never caused my any phase issues. In mid side mixing, the panned left and right channels should be brought up in level a tiny fraction of what the cardioid centre is. So, for example if the cardiod is at -3 the sides would be somewhere around -13. Roughly. It all depends on you record gain, etc. When I mix, I get the centre cardioid to taste, then bring up the sides slowly until I can just hear the width it adds to the mix. Anything beyond that, is too much and unnatural sounding. The sides add a touch of sparkle and width, not body, which is why rolling lows out from them is ok.
oh my, this was HAUNTING me! thanks a lot. I was loosing my mind over this. also, I've seen a fix for this by rotating the array 90 degrees so the stereo field isn't body to neck, but top-down
just a little technical FYI on why this works ... It's not really about timing or reverberation or anything. The side mic gets its figure 8 pattern pattern by sending a signal which is the difference between what hits the front and what hits the back. The dead spot on the sides then happens because sounds coming from the side hit the front and back equally. When turn that mic 90 degrees, the front of the mic hears sounds mainly coming from the left side, and the back hears sounds coming from the right side. The signal is then L - R. When we flip the polarity of the copy that signal become R - L. The mid mic hears sounds coming from both directions equally. So that's L + R. when we leave the mid mic panned down the center, we get L+R going to left speaker, and L+R going to the right speaker. When we bring the original side mic signal (L-R) up panned hard left, the right speaker still just has the mid MIC which is still L+R, but the left speaker now gets a combination of L+R from the mid mic and L-R from the side mic. (L+R) + (L-R) = 2L. The sounds that were coming from the right cancel out and the left speaker only plays sounds that came from the left. Opposite thing happens when we bring up the inverted side signal panned hard right (L+R) + (R-L) = 2R. now it's the sounds that came from the left that cancel out and what we hear is just the sound that came from the right side. Super cool. You get the variable spread because the other side gradually cancels out as you bring up the side mic signals.
M/S is my favorite stereo micing technique but it's not over once you wrap your head around how it works on a singular point-source. The real work begins when you start working it into a mix because it inherently relies upon room reflections at peak fig-8 sensitivity (ie 0/180 degrees). Some direct side information comes in but that info is faster than what comes in at 0/180 of the 8's lobes. This is where the weird phasey stuff crops up from before you raise the mid mic, the direct signal is literally double feeding the lobes of the 8 and that, though eventually smothered by the mid mic, doesn't go away on its own. One solution to tighten that up is move the mid mic back a foot or two and in post use Eventide's Physion to separate the side mic's transients then feed them through a delay 100% wet (0% direct sound) for 0.5-2ms (depending on what frequencies are most causing issues / room size) then re-EQ the mid mic flat and down a few db above 3-4k. Austrian Audio's PolarDesigner can help if you have too much low in the sides vs mid, acting like a phase perfect crossover so that can be sorted. Most people won't notice the difference between a recording done this way and one done without it on full spectrum monitors -> then they play it back on, say, a cellphone with speakers that just CAN'T reproduce 8k+ as well as 3-8k and they'll scratch their heads and wonder why all the sheen now sounds like crusty diapers.
Great work! In depth as it should be! 10:47 If someone wonder why 86x plugins pop up like this, and how to ''solve'' it: Click on Fx>Right click on desired 86x plugin>Run as>Embed bridged UI...
I think any stereo mic that’s placed that close to the guitar will give you a similar left/right balance because that is the actual stereo image at that location. If you want the stereo image to resemble what you’d hear in a house concert, the mics need to be further away. When moved further back the mid mic will be picking up more of the room ambiance relative to the instrument, losing much of the presence you get from close mic’ing. That effect can be mitigated by switching to a mic with higher directivity-e.g., a hypercardiod. I haven’t experimented with that myself-I’m just thinking about the physics.
Thanks, this was helpful. After watching this I tried with the Cardioid mic at around the 12th fret sweetspot and the figure of 8 mic past the bridge of the guitar. I was concerned there migh be a phase issue, but no, I got a very nice stereo recording, centred nicely and with the nice tone from good mic positioning.
Now ad a second cardioid above the fig-8 to record your (singing) voice in m/s stereo too. You can give it a slight upward pointing angle and the guitar one a slight downward pointing one.
Hey man great Video I definately learned something. I almost clicked of in the beginning of the video because you said XY. I was just binging a lot of videos about MS and yours did not disappoint. I just thought i clicked the wrong video in the beginning.
Many thanks for explaining the unbalanced stereo image think. I was playing around with M/S recordings this evening. And ran into exactly the same problem/question. Which left me a bit confused. And by change i found you episode. Solved ;-)
It does matter which panned side copy gets phase inverted. A figure eight mic has a front and back side. In the front side, positive pressure creates a positive voltage. In the back side, a positive pressure creates a negative voltage. This is why the center of the figure eight is null. So always pan the uninverted side track to the side the front of the side mic was facing. Always pan the inverted side track to the side the back of the side mic was facing. This will keep the stereo image correct and phase with the center correct. A good check is to play the two side tracks, one inverted, the other not, unpanned, with the center track muted and listen if they perfectly cancel each other out. It is important to note that when done correctly with side tracks panned equally in opposite directions and played back in mono, you will hear only the center mic because the two side tracks will cancel out. It is therefore highly advisable to ensure the center mic, by itself, is getting a good sound since, in mono, that's all you will hear. Another suggestion is to aim the whole array at an angle to the guitar using the center mic to set the aim. I mic acoustic guitars, banjos, mandolins, and octave mandolins a couple of times a week for live rehearsals and performance of an acoustic band. I use a single small diameter condenser - a KM 84. I place it perpendicularly at about the neck/body joint but aimed at an angle towards the body. The capsule is closer to the fretboard but the main lobe is aimed down the body but further away from it. It gets the body without aiming directly into the sound hole. I had to experiment with distances and angles to get this right. Note also the fretboard is further into the near field of the mic than the body, so the fretboard proximity effect is greater than the body's proximity effect. Too much proximity effect on the body can make it boxy and boomy and bassy sounding.
Sorry to be a smart ass, but in the first nanosecond that I saw you were recording a guitar with essentially an XY configuration at a central point, it was pretty clear why it would sound lopsided. Glad you figured it out.
To fix this even more you need the microphones to be different… even though you use two of the c414s in different polar patterns, they both are responding to the source nearly identically still as to capsule design, which means you will have phase cancellation issues between one of the side channels and the mid channel… bruce swedien says you need something like a ribbon microphone for the sides and a tube mic for the centre… they just need to be totally different for it to work so you get the least amount of phase cancellation as possible…. But your experiment was fascinating the way it altered the stereo fields balance by moving the microphones directly at the sound source… this makes sense because the capsule on the side signal c414 is positioned perpendicular and so the capsule is affected by the speed of sound hitting that round capsule at nanosecond differences from the front to the back of the capsule and this time delay of the speed of sound is actually creating a phase imbalance even moreso as the capsule is collecting info at the front and back of the capsule… which makes complete sense that a non-direct position will gradient the capsule surfaces absorption of the direct sound … and lead to those phase inversions, as the angle gradient of capture smears the phase.. i suppose a smaller capsule will also help mitigate this …so im gonna try that also to enhance my m/s technique.. i use a beyer m130 for the side mic and c414 xl2 for the mid. Thanks for the excellent video!
Hi, Thank you for the vid, awesome tutorial. I tried it and it works great. Quick note. when you take 2 of the same signal and you invert the phase, in mono they cancel each other out, in phase they become additive. In this case you have 3 signals, 2 in phase and one out. The out of phase side is now the weaker side and everything will lean to that side.
Oh this seems accurate perhaps (there’s several ideas on why this is happening it’s kinda wacky!). Yet if an out of phase side /signal is weaker don’t you mean things would lean to the other side rather? I’ll have to try it out.
I also went down this same road.... could not figure out why I had this L/R imbalance issue, figured it out, and adjusted the mic positions accordingly. Of course, you can also keep the two L/R tracks on separate faders and simply adjust individually for better balance that way. Nice video!
the mid qne rthe in phase signal are adding db, roughly 3 db on the left plus its left sub mix from the center panning. when you pan full left, you basically add 3 db more to that signal, so if your center in -12 db, then your full left-in phase will be roughly -9, and your full right will be about -15 db relative to the left channel due to the phase cancellation. These numbers are not exact, just illustrative of a general principle. The midside mixing technique works better with a more complex signal like a band or group of instruments that actually propogate differently into space, and so have different content coming into both sides of the figure-8 pickup pattern. (If the guitar were just a simple sign wave, it would totally be cancelled out in the phase-flipped channel. But, since it is actually a complex group of frequencies that propogate differently into space, they only pattially cancel out. )
I would try keeping the mics in the original position at a little greater distance but turning the side mics at a rectangle of the sound hole this should fix the loudness issues without disturbing to much of the stereo image. (Mid side mics not perfectly at right angles)
I'm glad you figured out why it was leaning left. (no political jokes here, please! ;) ) If you put your head where you put your mics, then listened critically, the sound would come from the left! BTW, if you wish to keep left and right correct in your stereo sound field, it does matter which way you pan the inverted channel. If you always point the front face of the Fig-8 to the left while recording, the non-inverted channel should be panned left and the inverted channel panned right. (IIRC!) Of course, if that level of accuracy doesn't matter to your recording, do whatever!
also.... you could turn both microphones a bit, so that the mid mic is looking toward the the sound hole but is placed around the 12th fret... this way the back of the mic will pick up more sound and the front side will pick-up less direct sound but you get less boomy sound.
shyeeeeet... i knew that changing fase is the golden key to the right chest! few years ago i was trying to mix friends vocals to the bands mix. just for entertaining purposes, because engineer was working on this simultaniously. the thing is i accidentally found that option - 'change of phase'. ito sounds beatufil in front of her clear vocals. but... still with making some natural reaserch i've made three tracks and dayum... couldn't blend them to the instrumental mix at any effort. too bad that this vid wasn't on YT back then...
Another thing to help focus the image to center is to apply highpass filter to the side information. (Our ears key in on high frequencies anyways with respects to stereo information). So 0-120hz is not doing much in the side except being boomy.
@@NkechiR The mid-side technique will not produce a 'true' stereo formation. Mid-side is mostly used to create a pseudo stereo environment for 'monophonic' sources. It also has music Production applications (the process) however the best way is to record in mid-side, first. Could you do it? Sure, you can do anything you want. Will it achieve an accurate representation of the choir and the people, in their respective locations in Stereo? No. The benefit of a mid/side recording is more for say a 'stereo/mono guitar rig' or solo singer, which would totally work for a 'lead' vocalist in the choir. Or lead instrument, because you are taking the 'side' Mic (polar pattern), which is actually turned Sideways to the sound source, doubling it that input, and panning one left and the other right (while phase inverting) one of them. This 'pseudo' stereo information gets then blended with a directional microphone input (direct at the sound source, cardioid pattern). So, the effect is that the Left and Right information that cancels/doubles with the Mid, produces a phase difference left to right (which is what 'stereo' sound is). The difference in mid-side is how this effect is being produced. For instance, I like sending the 'sides' to reverb differently than to the 'Mid' and often you can control these volumes independently (Adding energy in a chorus to rhythm guitars) or if the vocal seems lost in the mix, you can mid-side to get a bigger sound. In the end, I think I would mic a full Choir in true 'stereo'.
@@ryangraffius1474 ahh! 😁 Thanks a lot for taking time to reply. Being very new to the recording world, i will have to read through your explanation several times to understand it well. Given what you said, can u please tell me, in your best opinion, how to go from this th-cam.com/video/xmnDNt2EAOs/w-d-xo.html (us) to this th-cam.com/video/KR_wYJgAgmQ/w-d-xo.html Thanks a lot.
very interesting - it demonstrates this MidSide technique isn't just an artificious gimmik but a proper way of capturing the ambience of sound - I'd say traditional stereo and MS do capture the 3d of sound (ambience and more) in different ways...but the MS method offers much more control (in post production). Then, didn't you noticed the same bias to the left when recording with a typical stereo mic? I mean did you normally record with the mic to the end of the neck? (not direct to the soundhole?).
Feel like I need to help you out a bit here, it's called the 12/12 technique mate. 12th fret up, 12 inches from that point, simple and works every time.
Thanks for the channel, good content, presented in a friendly communication style. Wishing you all the very best for 2020! I've been scratching my head over this one! You basically recorded 2 guitar tracks simultaneously, one directional mic as per normal. The second mic (at same distance), you made a copy of this, reversed the phase and panned them hard L + R. I'd be interested to know if you went back and just changed the phase on channels L+R of the omni mic, if the track would lean o the right? Cos if it's just I signal doubled etc...... Also I know why you wanted the directional and figure 8 mics at the same distance and perpendicular in axis to each other (maybe like XY micing for phase). Might it be worth experimenting with the figure 8 mic in terms of distance/angle, so that on it's own it produces a balanced tone in term of treble, bass and room ambience? After all the phase may not be a problem due to it not detecting in the guitar a straight line (as the directional mic does), but by mostly sound reflections. Probably not clearly expressed! I fancy having a go at this sometime, maybe I'd try swivelling the figure 8 mic so bass and treble of that signal are in ok balance on it's own then copy the rest of the process you did (directional mic, double track phase and panning etc.) PS I've recently been setting up some old recording gear that hasn't been used for too long, and have watched a quite few youtubers on DAW recording, you are one of the few I have kept subscribed to. Many Thanks and keep up the good work.
Thanks a lot! I did find that changing which of the L/R tracks is phase reversed did change which way the stereo image leaned. The reason for having both mics' capsules as close to each other as possible is for phase coherency between the two mics. When you change the distance of one mic or the other, you'll start to get different frequencies that'll start to cancel out or reinforce each other, and it'll start to sound hollow and kinda weird unless they're at just the right distances and loudness in the mix. X/Y, Mid/Side, and Blumlein are all stereo techniques knows as "coincident pairs", where the idea is to get the capsules close enough together that they don't start to cause phase issues if they overlap in the mix. Best of luck in your own experiments and trials. It's fun to play with stuff like this :)
@@YHRS Yeah after reading up a bit I agree with your points and can appreciate the mic setup. Thanks for including the info in your reply about changing the phase on both the figure 8 tracks and the stereo going the other way. Thought came to me - could you zoom in close on the waveforms of the cardiod single channel and only the quieter of the stereo split recordings. Just nudge the quiet side a few miliseconds on the timeline, there might be a sweet spot. I think this might solve the phasing problem, as long as the offset length doesn't mess with the overall impression of timing. (or do the reverse of phase reinforcement!). Since sound travels about 300 feet per second, 3 feet = about 10ms offset, which I guess is much further than a typical band rehearsal/live distance on stage.
I used this technique once and did get a very skewed image, I did manage to get a good L+R balance by backing the mics away a bit though, I wonder though, how does the guitar sound if you start applying EQ to the mid channel? Phase problems? would it be a requirement to EQ all 3 together on a buss using a Linear phase EQ? I can't wrap my mind around the potential problems I could create. Maybe I should spend a few hours and experiment some on this technique. Nice video, good to hear a little shout out for Warren also.
Have you tried micing the guitar on a different axis? i.e instead of micing the guitar in terms of left and right (body on the left and neck on the right) you rotate the mics so that its parallel instead of perpendicular to the guitar? I find that it eliminates the "piano effect" of stereo micing where its more bass heavy on one side than the other.
That's an interesting idea! I'll have to give it a try. My main concern would be that the distance from the mic array to the floor/ceiling would be much shorter, causing earlier reflections and not as much "space" around the middle mic. I'll give it a try, though.
Except bass to treble is running horizontal if the strings are to be considered but unsure if mics pick that nuance up if back a bit so hmm never mind. Your idea is to assume treble is coming out of the neck, bass out of the body tho yeah? Yah u might be onto something. It’s confusing as to how subtle this affect is tho I guess, for me.
Hi, could you please tell me what your input gain was set at for both tracks? I’ve made one mid-side recording (Martin D28) and it turned out pretty good. When you moved the mics further away didn’t you have to increase gain to compensate? I’ve watched more than a few YT videos on this technique and none have mentioned anything about setting gain.
My gain settings are always just "not clipping". I usually turn up the gain until it clips when I do something loud, then back it off from there. The relative gain between the two mics doesn't matter a whole lot either, since you'll be manipulating their relative gain in post when you blend in the side mics. As long as they're not clipping and not so quiet that they exaggerate the noise floor, then I'd call that the ideal gain :)
What if you turn the blumlein pair sideways so the side mic isnt pointing left and right but up and down? Wouldn't the volume be balanced and you can still have the mid mic where you want it
Anyone have any problem with floor noise with a ribbon side mic??? It’s hard Getting my ribbon to around -18db without hiss when not facing the sound source. Would appreciate some input thanks
I use Reaper for recording my audio. You can try it free at www.reaper.fm/. After the trial period expires, it remains fully functional but nags you to purchase the $60 license. I think that's a great price for such a useful application.
The whole point of mid/side is to first have a mono sound that you love. Your initial microphone position was correct, it is where you love the mono sound. What you should be doing is to route your whole mid/side setup into a regular stereo channel, and from there, balance the audio through panning, without affecting the mid/side configuration. That way you'll keep the perfect mono sound and you will have a balanced stereo image. Also, if you start by routing your microphone signals through auxiliar channels, you don't need to duplicate the recording and you can monitor your stereo image live.
For how infrequently I've used mid-side so far, it just hasn't been worth the investment to get a mid-side decoder so I can fit all 3 signals into a single stereo feed. For now, I'll be stuck doing it in software. But once it's set up in software, the monitoring and playback works great.
Yeah so none of this worked for me. Mo matter where I position the guitar in relation to centre microphone. As soon as I bring in the centre mic it just leans to the left.
It's possible to pan a mid-side recording that is off center without re-recording the part. I made a video that shows how to do this here: th-cam.com/video/3MK02soeEJA/w-d-xo.html&ab_channel=MalcolmPaterson
Sorry man but you don’t get it. There is NO reason that either side of the fig 8 could be louder unless you have some problem in your DAW. You should have done more and better research regarding MS technique before you record and post a video. You are just adding to the YT confusion.
Figure 8 is not stereo. You need two sets of leads and two capsules to achieve true stereo. Your "stereo" is from the cardioid being combined with the figure 8, or visa versa. And only one side of the figure 8 is necessary for a stereo image when combined with another mic. Your recordings in the living room are "chirpy."
I don't recall any point in the video where I said that a figure-8 mic is stereo. The stereo part of M/S is created (as demonstrated in the video) by duplicating, panning, and inverting the polarity of the duplicate of the side mic.
![BOWKYLION - วิงวอน (ex-change) [Official MV]](http://i.ytimg.com/vi/cLdnZWDaO-w/mqdefault.jpg)
Thank you for this, I love when people document their mistakes to guide newbies to the craft
Dude… I cannot thank you enough. This has literally kept me up at night. Like you, it was driving me crazy.
One thing- in my case, of course low budget home recording without many mics to choose from, I always found myself wanting- needing a matched stereo pair of mics for recording drums or wide acoustic, etc… so for me mid side has always been my BFF and without it I couldn’t have done many of my earlier recordings. But when tracking acoustic guitar I always ended up with that lean to one side. It made sense to me with drums that the left and right had variance but I could never understand that constant lean with acoustic guitar. It seems so obvious now. Thank you!!!!!!!
Something else to try, take the 414s out of the stereo mount and put them on individual stands (unless you can rotate that rig 90 degrees like the hands on the face of clock).
Then aim your cardioid mic capsule the way you used to for mono, capturing the sound you like, except this time, aim the XLR up the neck.
Then place your figure 8 capsule perpendicular to even center the sound hole, butted up to the cardioid, with it’s XLR aimed down the body.
Now you’re recording the figure8 toward the floor and ceiling though, but with even SPL.
The way you’re sitting evenly between the left and right walls, you may wanna sit on a stool so that the guitar is at say 4 feet if you have 8 foot ceilings.
Evening the reverb times.
Then, when you build the MS Matrix exactly the same as you did in reaper, you’ll effectively be rotating the listeners 90 degrees, but they’ll never know.
If your floor and ceiling are totally different materials, carpet/hardwood/tile vs drywall/cloud absorber/diffuser, A) that effect may sound cool, B) you could throw down a rug, blanket, board, change the reverb of the floor to match or to taste.
Part of MS Micing is the controlled stereo effect, but part is the natural reverb/room sound.
BTW, the way you had it first, you could always lower the level of the Left figure 8.
You could also make those left and right tracks into 1 stereo track, then the pan becomes a balance.
You’ll also end up EQing the sides, probably rolling off the lows, and that’s easier/cleaner on 1 stereo track.
Lastly, as you noted, you don’t need matching microphones for MS, part of the beauty, so maybe you have something else, even a cheapy that fits better in the mix. But you can’t go wrong with a matched pair of 414s!
Last thought, when you just pop a LDC into figure 8, pop on some headphones and talk into each side.
Most sound very different from one side to the other, the perfect 8 ain’t so perfect.
If I recall correctly, the 414’s “backside” is a bit darker, let’s call that the Darkside!
Well if that is the case, maybe aim the Darkside at the bright reflective ceiling, and the Frontside at the carpeted floor.
Like all things recording, learn your gear and what it can do, experiment, and play.
That’s exactly what I saw in your video in the first place!
Great comment, one note. When you talk into opposite sides of a figure 8 while wearing headphones one side will sound different because it is out of phase with your skull. You would need to invert the phase when speaking into the back to avoid that.
Your tip on going floor to ceiling blew my mind.
Awesome insight. I imagine now too trying fig 8 in front of sound hole w it’s off axis pointed at the sound so it’s not boomy, then use two other mics cardioid spaced maybe about 12-16 inches equally spaced from central mic like wide ortf. Experiment with levels and panning of these alongside panning the two central +duplication; two tracks at full L/R; two tracks at say 3 oclock. Might have to flip polarity on two occasions but I think I’ll try it. I like idea of louder fuller hard panning and no center track so that other sounds like a snare or vocal can breath separately.
@@chinmeysway There you go. Yeah experiment, have fun, find something new. “If it sounds good, it is good!”
One word of caution though: the whole beauty of the Mid-Side Stereo technique is that the sound hits the capsules of the two mics at precisely the same time, thereby mostly eliminating or minimizing phase cancellations. Move the mics apart and different sound gets to different capsules at different times. This isn’t necessarily a problem, just a different phenomenon to factor in when processing.
@@G_handle totally.. yeah i am thinking about that aspect too. im just so interested in creating new mic technics. its interesting that theres a few standards, but im sure they can be thought of as loose guides. i wonder about why the fig8 mic is imperative- if its just to be duplicated as a mono signal, can it just be a cardiod pattern pointed to one of the sides, rather than fig 8, for instance.
@@chinmeysway okay so.....
A) Absolutely experiment with every conceivable mic configuration you can come up with. If it sounds good, it is good. And sounding good is subjectively Your opinion, the only opinion that matters. It's Your music.
B) The Objective reason why Figure-8 pickup pattern is "imperative" in a "True M.S." technique, is:
1) A figure-8 is a single* diaphragm generating a single mono signal,
2) But it's equally sensitive from both front and back, and has nulls on the sides,
3a) So when you duplicate that single signal (or any single signal), then sum both together, it doubles in loudness...Perfectly.
3b) If you then reverse the polarity of one of the two duplicates, then sum those together, they perfectly null. Because they are identical signals in reverse. So 2+2=4. But 2-2=0.
4a) Now, while the figure-8 "Side" Mic will PERFECTLY phase cancel when summed with its negative self, it will also PARTIALLY phase cancel when summed with a signal from any Other microphone recording the same source.
4b) By placing the "Side" mic's diaphragm as close to the "Mid" mic's diaphragm as possible (co-incident), the 2 Different mics will receive the acoustic energy at about the same TIME, but...
4c) remember that the Figure-8 "Side" mic has its Null facing towards the source, and the "Mid" mic is either and Omni or is directionally facing the source (usually cardioid, but any pattern can be used). So the two mics catch the same source at the same time, but with different sensitivity and directionality. The Side mic is directionally least sensitive where the Mid mic is most sensitive.
4d) Therefore when the acoustic energy is more on one side or the other of the single figure-8 diaphragm, in will physically move in or out from its resting position in the center, and the audio signal will be generated to reflect that push pull with either positive or negative values.
4e) Then that single mono signal's positive or negative waveform value, gets duplicated and just becomes two copies,
4F) When you flip the polarity on one of the copies, they're now still just two signals out of phase with each other, however depending upon which copy you flipped, one will contain useful information from the Right and the other from the Left, but they're not at all stereo yet.
4G) It's only when you Sum either "Side" signal with the "Mid" mic's signal that get a Left or a Right signal. Remember that the "Mid" mic is mono and contains Both the Left and Right information.
5) So now the MS Technique has two tricks up its sleeve: Stereo and Mono.
Imagine this:
Two Mics, Two Y-Cables, Four Channels on a Mixer.
The "Mid" mic is going to channels 1 & 2, the "Side" mic is going to channels 3 & 4.
The 'odd' channels are panned Left, the 'even' channels panned Right.
If you push up faders 1 & 2 to unity you get a Phantom Center. Pull the faders down.
If you push up faders 3 & 4 to unity you get a Phantom Center. But roomy, because the Null was facing the source.
Now Invert the Polarity on either 3 or 4. You get a weird Phantom Phase Cancelation acoustically between your speakers. Pull those faders down.
Now push 1 & 2 back up, and then blend in 3 & 4 and that Phantom Center will widen out and gain Room around it.
Amazing! Thats why you did all this!
Push as much Width into it as sounds good to you.
Your two Mono Microphones have just created a Stereo Image, that You can control.
Every other form of stereo micing, you get what you get.
You may prefer the Stereo Image from another technique, but stay tuned...
If your mixer has one, press the MONO button.
The Left & Right will be summed together, meaning that the two "Side" mic signals will cancel each other out, and you'll be left with a Perfect recreation of the "Mid" mic by itself again.
However wide you make your Stereo Mix, if the end user is listening on a Mono system, the wide information just disappears without a trace.
So you have a highly controllable Stereo Image,
And Perfect Mono Compatibility.
6) Back to the math:
Remember: 2 + 2= 4 & 2 - 2= 0
Well instead think: 2 + 2= 4 & 2 + (-2)= 0
2 + (-2)= 0 is what's happening with the "Side" signals when they get summed to Mono.
In Stereo mode though, they aren't summed with each other, they're summed with copies of the "Mid" mic's signal that are hard panned Left and Right. So they're not "seeing" each other, they're adding and subtracting with that other diaphragm on two isolated channels, Left and Right.
The mono "Mid" mic was capturing all of the Left and Right energy in the air, and now when blended with one side of the figure-8 or the other, its Phase Interaction is either adding when the capsules were going the same way or cancelling when they were going in opposite directions. This works, because you have Both directions from the Same diaphram.
So the math for Stereo mode is more like:
Left = M + S
Right = M + (-S)
In Mono mode:
Left + Right= Mono
So:
(M + S) + (M + (-S)) = M + M
(In a Mono button it should be (L + R) /2 = Mono. So the level doesn't double)
That took Way too much time!!!!
Last thought: Back to A!
Experiment.
If it sounds, good it is good.
So if you replace the Figure-8 with 2 SDCs, you won't get the perfect cancelation, but you will get cancelation.
And it may sound amazing.
And it will almost certainly sound Wider, because the differences will be greater between Left & Right.
So understand the "rules", then break em'!!!!!!!!!!
(Let me add this disclaimer at the end: This is my understanding. If something ain't quite right, Please tell me. I love to learn, and see being wrong as an opportunity for growth.)
Thanks for the video, it made me rethink about this problem I had also encountered. I don't think, however, the key issue is the position, but, instead, the similarity of the signals between the M and S mics, after all, the S mic is just a mono signal, so it should always cancel itself no matter the position. Let's suppose that (absurdly) we had the exact same signal in both mics. After the MS processing, in one speaker we would have A+A and in the other speaker we would have A-A, leaving us with 2A on one side and 0 on the other (I am ignoring pan law to simplify the example). So the problem is not the position, but a technique that has a tendency to emphasize one side. When you distance the microphones from the guitar or center them in relation to the source, the S mic gets more reflections and less direct sound, making the signal from both mics more different and less phase aligned, hence, less cancellation between the S flipped side and the mid side occurs resulting in a more centered image. A bigger room will always mean more delay in the S signal, making both signals less similar. Anything that makes both signals more different, like moving mics, using different mics, delaying the S signal to minimize cancellation, or getting rid of the lows in the S side (since most cancellation occurs in the lows because they are longer waves) will help to solve the issue. So, if you like your acoustic sound up close, you can get that in MS by recording as you did (it's preferable if you use a different mic for the sides though, it doesn't even have to be a figure eight), then manually delaying the S signal some milliseconds till it sounds centered and then removing the lows from the sides with a MS EQ. I hope this helps.
While mic placement is always a huge concern, and certainly with the m/s technique, a quick tip I learned (I can't remember where) is to roll off the low end on your sides (the two panned channels) until you see the signal fit in the centre. I think part of this problem has to do with the fact that one side of the figure 8 is favouring the heel or body of the guitar, while the other side is favouring the neck. The bottom side is beefy, the neck side is more airy. So I've found that while it's well and good to find optimal mic position, much of the problem can be solved with rolling off your low end on the two side channels until the signal fits centre. You don't need or want the beefy low end on the sides anyway, that will come from the centre cardioid signal.
Thanks for the tips, I think that's a good idea to do some hi and lo passing on the side signals to de-clutter things a bit!
@@YHRS Indeed! It's a good solution for me as I do have a tendency to move a bit too much (while trying not to) while playing. It's not so much an issue with other stereo mic techniques, and certainly no issue with a mono record, but it's pretty finicky when recording in M/S. I find the slightest shift of the body can really play hell with the balance.
Hang on hang on- what you need to do is do what this chap did and move the source or the mics so you can eq both side signals together. Bc I guarantee you’re introducing unforeseen phase problems into your recording by eqing individual sides. I’m sure you’ve run into this when trying to eq a single spot mic on a drum kit. It can get weird. So the safest bet is to track it in a way that lessens the need to eq anything at all. That’s my humble opinion.
Although, I only just found out WHY my signal on an acoustic was leaning to one side. So what do I know
I guess you are suggesting hi passing both sides together. Not individually, correct?
Of course the less eq the better. Having said that, additive eq is much more problematic than subrtractive. Perhaps I wasn't clear. I'm not suggesting individual eq adjustments to each side, but simply rolling off the low end on both sides together until the signal output focuses in the centre. Usually anywhere up to around 150hz or so off of the sides. Place mics as best you can, and then try it. It works. Never caused my any phase issues. In mid side mixing, the panned left and right channels should be brought up in level a tiny fraction of what the cardioid centre is. So, for example if the cardiod is at -3 the sides would be somewhere around -13. Roughly. It all depends on you record gain, etc. When I mix, I get the centre cardioid to taste, then bring up the sides slowly until I can just hear the width it adds to the mix. Anything beyond that, is too much and unnatural sounding. The sides add a touch of sparkle and width, not body, which is why rolling lows out from them is ok.
Great content. This was my suspicion after the last time using this technique. Thanks for confirming and demonstrating!
Thank you for sharing... A very useful technique indeed... I will be experimenting later today. Kind Regards. Andy
Best of luck! Hope it turns out well for you!
oh my, this was HAUNTING me! thanks a lot. I was loosing my mind over this.
also, I've seen a fix for this by rotating the array 90 degrees so the stereo field isn't body to neck, but top-down
just a little technical FYI on why this works ... It's not really about timing or reverberation or anything. The side mic gets its figure 8 pattern pattern by sending a signal which is the difference between what hits the front and what hits the back. The dead spot on the sides then happens because sounds coming from the side hit the front and back equally. When turn that mic 90 degrees, the front of the mic hears sounds mainly coming from the left side, and the back hears sounds coming from the right side. The signal is then L - R. When we flip the polarity of the copy that signal become R - L. The mid mic hears sounds coming from both directions equally. So that's L + R. when we leave the mid mic panned down the center, we get L+R going to left speaker, and L+R going to the right speaker. When we bring the original side mic signal (L-R) up panned hard left, the right speaker still just has the mid MIC which is still L+R, but the left speaker now gets a combination of L+R from the mid mic and L-R from the side mic. (L+R) + (L-R) = 2L. The sounds that were coming from the right cancel out and the left speaker only plays sounds that came from the left. Opposite thing happens when we bring up the inverted side signal panned hard right (L+R) + (R-L) = 2R. now it's the sounds that came from the left that cancel out and what we hear is just the sound that came from the right side. Super cool. You get the variable spread because the other side gradually cancels out as you bring up the side mic signals.
M/S is my favorite stereo micing technique but it's not over once you wrap your head around how it works on a singular point-source. The real work begins when you start working it into a mix because it inherently relies upon room reflections at peak fig-8 sensitivity (ie 0/180 degrees). Some direct side information comes in but that info is faster than what comes in at 0/180 of the 8's lobes. This is where the weird phasey stuff crops up from before you raise the mid mic, the direct signal is literally double feeding the lobes of the 8 and that, though eventually smothered by the mid mic, doesn't go away on its own. One solution to tighten that up is move the mid mic back a foot or two and in post use Eventide's Physion to separate the side mic's transients then feed them through a delay 100% wet (0% direct sound) for 0.5-2ms (depending on what frequencies are most causing issues / room size) then re-EQ the mid mic flat and down a few db above 3-4k. Austrian Audio's PolarDesigner can help if you have too much low in the sides vs mid, acting like a phase perfect crossover so that can be sorted. Most people won't notice the difference between a recording done this way and one done without it on full spectrum monitors -> then they play it back on, say, a cellphone with speakers that just CAN'T reproduce 8k+ as well as 3-8k and they'll scratch their heads and wonder why all the sheen now sounds like crusty diapers.
Thank you! I was having the same issue and it was driving me nuts! Makes sense.
Great work! In depth as it should be!
10:47 If someone wonder why 86x plugins pop up like this, and how to ''solve'' it: Click on Fx>Right click on desired 86x plugin>Run as>Embed bridged UI...
I think any stereo mic that’s placed that close to the guitar will give you a similar left/right balance because that is the actual stereo image at that location. If you want the stereo image to resemble what you’d hear in a house concert, the mics need to be further away. When moved further back the mid mic will be picking up more of the room ambiance relative to the instrument, losing much of the presence you get from close mic’ing. That effect can be mitigated by switching to a mic with higher directivity-e.g., a hypercardiod.
I haven’t experimented with that myself-I’m just thinking about the physics.
Thanks, this was helpful. After watching this I tried with the Cardioid mic at around the 12th fret sweetspot and the figure of 8 mic past the bridge of the guitar. I was concerned there migh be a phase issue, but no, I got a very nice stereo recording, centred nicely and with the nice tone from good mic positioning.
Now ad a second cardioid above the fig-8 to record your (singing) voice in m/s stereo too. You can give it a slight upward pointing angle and the guitar one a slight downward pointing one.
sounds like the perfect songwriter set up! gotta try it out soon. thanks!
Wouldn’t you get a lot of spill into the guitar mics?
Terrible advice
You should try the ORTF stereo technique and do a comparison. Thanks for the heads up on Paz. It looks like a useful tool.
Hey man great Video I definately learned something. I almost clicked of in the beginning of the video because you said XY. I was just binging a lot of videos about MS and yours did not disappoint. I just thought i clicked the wrong video in the beginning.
Many thanks for explaining the unbalanced stereo image think. I was playing around with M/S recordings this evening. And ran into exactly the same problem/question. Which left me a bit confused. And by change i found you episode. Solved ;-)
Haha - solved a problem which I was really struggling to understand!!! Thanks a lot - so simple!
It does matter which panned side copy gets phase inverted. A figure eight mic has a front and back side. In the front side, positive pressure creates a positive voltage. In the back side, a positive pressure creates a negative voltage. This is why the center of the figure eight is null. So always pan the uninverted side track to the side the front of the side mic was facing. Always pan the inverted side track to the side the back of the side mic was facing. This will keep the stereo image correct and phase with the center correct. A good check is to play the two side tracks, one inverted, the other not, unpanned, with the center track muted and listen if they perfectly cancel each other out.
It is important to note that when done correctly with side tracks panned equally in opposite directions and played back in mono, you will hear only the center mic because the two side tracks will cancel out. It is therefore highly advisable to ensure the center mic, by itself, is getting a good sound since, in mono, that's all you will hear.
Another suggestion is to aim the whole array at an angle to the guitar using the center mic to set the aim. I mic acoustic guitars, banjos, mandolins, and octave mandolins a couple of times a week for live rehearsals and performance of an acoustic band. I use a single small diameter condenser - a KM 84. I place it perpendicularly at about the neck/body joint but aimed at an angle towards the body. The capsule is closer to the fretboard but the main lobe is aimed down the body but further away from it. It gets the body without aiming directly into the sound hole. I had to experiment with distances and angles to get this right. Note also the fretboard is further into the near field of the mic than the body, so the fretboard proximity effect is greater than the body's proximity effect. Too much proximity effect on the body can make it boxy and boomy and bassy sounding.
18” away i. The larger room sounded much better to my ears. Thanks for sharing.
Agreed, I liked the ambiance that the larger room added. If I were doing a solo acoustic guitar track, that'd be my first place to try to record!
This was super helpful. Thanks for posting. I was having the exact same issue and this solved the problem.
That's great to hear, I'm glad it helped!
Sorry to be a smart ass, but in the first nanosecond that I saw you were recording a guitar with essentially an XY configuration at a central point, it was pretty clear why it would sound lopsided. Glad you figured it out.
To fix this even more you need the microphones to be different… even though you use two of the c414s in different polar patterns, they both are responding to the source nearly identically still as to capsule design, which means you will have phase cancellation issues between one of the side channels and the mid channel… bruce swedien says you need something like a ribbon microphone for the sides and a tube mic for the centre… they just need to be totally different for it to work so you get the least amount of phase cancellation as possible…. But your experiment was fascinating the way it altered the stereo fields balance by moving the microphones directly at the sound source… this makes sense because the capsule on the side signal c414 is positioned perpendicular and so the capsule is affected by the speed of sound hitting that round capsule at nanosecond differences from the front to the back of the capsule and this time delay of the speed of sound is actually creating a phase imbalance even moreso as the capsule is collecting info at the front and back of the capsule… which makes complete sense that a non-direct position will gradient the capsule surfaces absorption of the direct sound … and lead to those phase inversions, as the angle gradient of capture smears the phase.. i suppose a smaller capsule will also help mitigate this …so im gonna try that also to enhance my m/s technique.. i use a beyer m130 for the side mic and c414 xl2 for the mid. Thanks for the excellent video!
Hi, Thank you for the vid, awesome tutorial. I tried it and it works great. Quick note. when you take 2 of the same signal and you invert the phase, in mono they cancel each other out, in phase they become additive. In this case you have 3 signals, 2 in phase and one out. The out of phase side is now the weaker side and everything will lean to that side.
Oh this seems accurate perhaps (there’s several ideas on why this is happening it’s kinda wacky!). Yet if an out of phase side /signal is weaker don’t you mean things would lean to the other side rather?
I’ll have to try it out.
I like your style, dude.
Thank you, sir! This really helped. I was wondering the same thing!
I also went down this same road.... could not figure out why I had this L/R imbalance issue, figured it out, and adjusted the mic positions accordingly.
Of course, you can also keep the two L/R tracks on separate faders and simply adjust individually for better balance that way.
Nice video!
the mid qne rthe in phase signal are adding db, roughly 3 db on the left plus its left sub mix from the center panning. when you pan full left, you basically add 3 db more to that signal, so if your center in -12 db, then your full left-in phase will be roughly -9, and your full right will be about -15 db relative to the left channel due to the phase cancellation. These numbers are not exact, just illustrative of a general principle. The midside mixing technique works better with a more complex signal like a band or group of instruments that actually propogate differently into space, and so have different content coming into both sides of the figure-8 pickup pattern. (If the guitar were just a simple sign wave, it would totally be cancelled out in the phase-flipped channel. But, since it is actually a complex group of frequencies that propogate differently into space, they only pattially cancel out. )
I would try keeping the mics in the original position at a little greater distance but turning the side mics at a rectangle of the sound hole this should fix the loudness issues without disturbing to much of the stereo image. (Mid side mics not perfectly at right angles)
Thanks, got to try this now, would like to see more of this type content, keep up the good work, all the best for 2020
Makes we wanna try it , great work ,thanks
I'm glad you figured out why it was leaning left. (no political jokes here, please! ;) ) If you put your head where you put your mics, then listened critically, the sound would come from the left! BTW, if you wish to keep left and right correct in your stereo sound field, it does matter which way you pan the inverted channel. If you always point the front face of the Fig-8 to the left while recording, the non-inverted channel should be panned left and the inverted channel panned right. (IIRC!) Of course, if that level of accuracy doesn't matter to your recording, do whatever!
Great video, I have to say that guitar sounds great even with out a mic. Awesome!
thank a lot！solve my problem！
Thanks so much! Yeah, I love that guitar a whole lot. It's just a nice, rich acoustic sound.
also.... you could turn both microphones a bit, so that the mid mic is looking toward the the sound hole but is placed around the 12th fret... this way the back of the mic will pick up more sound and the front side will pick-up less direct sound but you get less boomy sound.
shyeeeeet... i knew that changing fase is the golden key to the right chest! few years ago i was trying to mix friends vocals to the bands mix. just for entertaining purposes, because engineer was working on this simultaniously. the thing is i accidentally found that option - 'change of phase'. ito sounds beatufil in front of her clear vocals. but... still with making some natural reaserch i've made three tracks and dayum... couldn't blend them to the instrumental mix at any effort. too bad that this vid wasn't on YT back then...
very helpful. thanks mate
Thank you for the video! definitely, a problem I had... I fase align the side and the mid signals on the DAW and that seems to work :)
A small correction: Ribbon mics don't "tend to have figure 8 capabilities". They're all figure 8, naturally, no need to switch them to that pattern.
Another thing to help focus the image to center is to apply highpass filter to the side information. (Our ears key in on high frequencies anyways with respects to stereo information). So 0-120hz is not doing much in the side except being boomy.
I use a Neumann U87 mid and a Royer 121 side.
I bet that's a terrific sounding setup! Those are some drool-worthy mics!
@@ryangraffius1474 please, can i record a small choir (about 20 singers) with the mid-side technique?
If yes, how do i go about it?
@@NkechiR
The mid-side technique will not produce a 'true' stereo formation. Mid-side is mostly used to create a pseudo stereo environment for 'monophonic' sources. It also has music Production applications (the process) however the best way is to record in mid-side, first.
Could you do it? Sure, you can do anything you want. Will it achieve an accurate representation of the choir and the people, in their respective locations in Stereo? No.
The benefit of a mid/side recording is more for say a 'stereo/mono guitar rig' or solo singer, which would totally work for a 'lead' vocalist in the choir. Or lead instrument, because you are taking the 'side' Mic (polar pattern), which is actually turned Sideways to the sound source, doubling it that input, and panning one left and the other right (while phase inverting) one of them.
This 'pseudo' stereo information gets then blended with a directional microphone input (direct at the sound source, cardioid pattern). So, the effect is that the Left and Right information that cancels/doubles with the Mid, produces a phase difference left to right (which is what 'stereo' sound is). The difference in mid-side is how this effect is being produced.
For instance, I like sending the 'sides' to reverb differently than to the 'Mid' and often you can control these volumes independently (Adding energy in a chorus to rhythm guitars) or if the vocal seems lost in the mix, you can mid-side to get a bigger sound.
In the end, I think I would mic a full Choir in true 'stereo'.
@@ryangraffius1474 ahh!
😁
Thanks a lot for taking time to reply.
Being very new to the recording world, i will have to read through your explanation several times to understand it well.
Given what you said, can u please tell me, in your best opinion, how to go from this th-cam.com/video/xmnDNt2EAOs/w-d-xo.html (us) to this th-cam.com/video/KR_wYJgAgmQ/w-d-xo.html
Thanks a lot.
very interesting - it demonstrates this MidSide technique isn't just an artificious gimmik but a proper way of capturing the ambience of sound - I'd say traditional stereo and MS do capture the 3d of sound (ambience and more) in different ways...but the MS method offers much more control (in post production).
Then, didn't you noticed the same bias to the left when recording with a typical stereo mic? I mean did you normally record with the mic to the end of the neck? (not direct to the soundhole?).
Thanks for the video! Very interesting.
You are very welcome! Thanks for watching!
Very helpful. Thank you!
Feel like I need to help you out a bit here, it's called the 12/12 technique mate. 12th fret up, 12 inches from that point, simple and works every time.
Those mic stands look pretty solid, which brand is it?
Thanks for the channel, good content, presented in a friendly communication style.
Wishing you all the very best for 2020!
I've been scratching my head over this one!
You basically recorded 2 guitar tracks simultaneously, one directional mic as per normal.
The second mic (at same distance), you made a copy of this, reversed the phase and panned them hard L + R.
I'd be interested to know if you went back and just changed the phase on channels L+R of the omni mic, if the track would lean o the right? Cos if it's just I signal doubled etc......
Also I know why you wanted the directional and figure 8 mics at the same distance and perpendicular in axis to each other (maybe like XY micing for phase).
Might it be worth experimenting with the figure 8 mic in terms of distance/angle, so that on it's own it produces a balanced tone in term of treble, bass and room ambience? After all the phase may not be a problem due to it not detecting in the guitar a straight line (as the directional mic does), but by mostly sound reflections.
Probably not clearly expressed!
I fancy having a go at this sometime, maybe I'd try swivelling the figure 8 mic so bass and treble of that signal are in ok balance on it's own then copy the rest of the process you did (directional mic, double track phase and panning etc.)
PS I've recently been setting up some old recording gear that hasn't been used for too long, and have watched a quite few youtubers on DAW recording, you are one of the few I have kept subscribed to.
Many Thanks and keep up the good work.
Thanks a lot!
I did find that changing which of the L/R tracks is phase reversed did change which way the stereo image leaned.
The reason for having both mics' capsules as close to each other as possible is for phase coherency between the two mics. When you change the distance of one mic or the other, you'll start to get different frequencies that'll start to cancel out or reinforce each other, and it'll start to sound hollow and kinda weird unless they're at just the right distances and loudness in the mix. X/Y, Mid/Side, and Blumlein are all stereo techniques knows as "coincident pairs", where the idea is to get the capsules close enough together that they don't start to cause phase issues if they overlap in the mix.
Best of luck in your own experiments and trials. It's fun to play with stuff like this :)
@@YHRS Yeah after reading up a bit I agree with your points and can appreciate the mic setup. Thanks for including the info in your reply about changing the phase on both the figure 8 tracks and the stereo going the other way.
Thought came to me - could you zoom in close on the waveforms of the cardiod single channel and only the quieter of the stereo split recordings. Just nudge the quiet side a few miliseconds on the timeline, there might be a sweet spot. I think this might solve the phasing problem, as long as the offset length doesn't mess with the overall impression of timing. (or do the reverse of phase reinforcement!).
Since sound travels about 300 feet per second, 3 feet = about 10ms offset, which I guess is much further than a typical band rehearsal/live distance on stage.
I love your videos. A+
Position figure 8 mic so front of it is facing headstock. Also, use gain plugin to balance the L and R sides
I used this technique once and did get a very skewed image, I did manage to get a good L+R balance by backing the mics away a bit though, I wonder though, how does the guitar sound if you start applying EQ to the mid channel? Phase problems? would it be a requirement to EQ all 3 together on a buss using a Linear phase EQ? I can't wrap my mind around the potential problems I could create. Maybe I should spend a few hours and experiment some on this technique. Nice video, good to hear a little shout out for Warren also.
Have you tried micing the guitar on a different axis? i.e instead of micing the guitar in terms of left and right (body on the left and neck on the right) you rotate the mics so that its parallel instead of perpendicular to the guitar? I find that it eliminates the "piano effect" of stereo micing where its more bass heavy on one side than the other.
That's an interesting idea! I'll have to give it a try. My main concern would be that the distance from the mic array to the floor/ceiling would be much shorter, causing earlier reflections and not as much "space" around the middle mic. I'll give it a try, though.
Except bass to treble is running horizontal if the strings are to be considered but unsure if mics pick that nuance up if back a bit so hmm never mind. Your idea is to assume treble is coming out of the neck, bass out of the body tho yeah? Yah u might be onto something. It’s confusing as to how subtle this affect is tho I guess, for me.
Hi, could you please tell me what your input gain was set at for both tracks? I’ve made one mid-side recording (Martin D28) and it turned out pretty good. When you moved the mics further away didn’t you have to increase gain to compensate? I’ve watched more than a few YT videos on this technique and none have mentioned anything about setting gain.
My gain settings are always just "not clipping". I usually turn up the gain until it clips when I do something loud, then back it off from there. The relative gain between the two mics doesn't matter a whole lot either, since you'll be manipulating their relative gain in post when you blend in the side mics. As long as they're not clipping and not so quiet that they exaggerate the noise floor, then I'd call that the ideal gain :)
@@YHRS Perfect and excellent answer thas earned you a new subscriber! Thank you and you’ll probably hear from me again.
🎶 BOWIE’S IN SPACE 🎶
I've had success staying at the neck joint and pointing the mid mic slightly more neckward.
Hi nice Sir…
Can we use one condenser mic 🎙 I have rode Nt1-1 and
Other dynamic mic 🎤 SM-57 I have?
Pls reply…🙏💓
What if you turn the blumlein pair sideways so the side mic isnt pointing left and right but up and down? Wouldn't the volume be balanced and you can still have the mid mic where you want it
Cool demonstration. Try it on your drum kit. Very nice on drums sometimes.
th-cam.com/video/upG1KaKExMY/w-d-xo.html seemed to cover the kick drum interestingly.
Will do!
The 18 inch distance in the large room is best.
Anyone have any problem with floor noise with a ribbon side mic??? It’s hard Getting my ribbon to around -18db without hiss when not facing the sound source. Would appreciate some input thanks
What mic is it/ what preamp?
hey Bro. what is the software you used in recording? hoping you can see my comments, I want to change my software that I am using, . I am beginner.
I use Reaper for recording my audio. You can try it free at www.reaper.fm/. After the trial period expires, it remains fully functional but nags you to purchase the $60 license. I think that's a great price for such a useful application.
@@YHRS Thank you so much for that information. If I purchase 60 dollars is it lifetime or yearly?
Maybe you should start with getting your monitors at the correct height.
The whole point of mid/side is to first have a mono sound that you love. Your initial microphone position was correct, it is where you love the mono sound. What you should be doing is to route your whole mid/side setup into a regular stereo channel, and from there, balance the audio through panning, without affecting the mid/side configuration. That way you'll keep the perfect mono sound and you will have a balanced stereo image. Also, if you start by routing your microphone signals through auxiliar channels, you don't need to duplicate the recording and you can monitor your stereo image live.
For how infrequently I've used mid-side so far, it just hasn't been worth the investment to get a mid-side decoder so I can fit all 3 signals into a single stereo feed. For now, I'll be stuck doing it in software. But once it's set up in software, the monitoring and playback works great.
That sounds like a wise method
Yeah so none of this worked for me. Mo matter where I position the guitar in relation to centre microphone. As soon as I bring in the centre mic it just leans to the left.
Then maybe make a follow up video!
It's possible to pan a mid-side recording that is off center without re-recording the part. I made a video that shows how to do this here:
th-cam.com/video/3MK02soeEJA/w-d-xo.html&ab_channel=MalcolmPaterson
Just use single mic or double record the performance.
I
Using an omnis as center mic is better. Les boom
Sorry man but you don’t get it.
There is NO reason that either side of the fig 8 could be louder unless you have some problem in your DAW.
You should have done more and better research regarding MS technique before you record and post a video. You are just adding to the YT confusion.
u talk to much and show to little
Figure 8 is not stereo. You need two sets of leads and two capsules to achieve true stereo. Your "stereo" is from the cardioid being combined with the figure 8, or visa versa. And only one side of the figure 8 is necessary for a stereo image when combined with another mic. Your recordings in the living room are "chirpy."
I don't recall any point in the video where I said that a figure-8 mic is stereo. The stereo part of M/S is created (as demonstrated in the video) by duplicating, panning, and inverting the polarity of the duplicate of the side mic.
@@YHRS ~12:37
I’m new but finding that ms is kinda overrated
Hat being said I think the reason it’s uneven is because the guitar is asymmetric per se. it’s louder on the body than the neck.
I honestly don't use it much. But it's a great tool for the toolkit, and can come in really handy sometimes.