![]() |
Recording for 5.1
Hi
I was told my someone (who for some reason couldn't explain to me why) that it takes two microphones to accurately reproduce the surround sound effect, and any production that requires this must record with two microphones. I don't see many productions actually using two booms or lavs, especially for dialog. So basically my questions are: 1. Do we need two stereo microphones? 2. If yes, then does that mean using two mics on every dialog, sound effect, etc? 3. What is the basic post production workflow - which mic is used for L/R, and how does one decide that on set? I'd appreciate any help, and if anyone can point to any other literature on the net, that'll be great, too. Thanks! |
Re: Recording for 5.1
Quote:
Quote:
Quote:
Waiting until you are on the set is rather too late to decide these kinds of things. Study some books and videos about production sound and learn how it is done before just assuming that you must have a stereo microphone. |
Re: Recording for 5.1
Thanks for replying. You're right, I actually meant two mono mics.
I've always used mono mics (own an ME66 still) and lavs for my feature and other production work. My feature was mixed in stereo, but the audio was just split 50:50; didn't have the budget for anything fancy. But I did intend to mix in 5.1 and someone told me you can't mix in 5.1 with just a mono channel. That's what I hoped to find out - whether there is any truth to that statement. If, hypothetically speaking, one wanted a feature film mix for theatrical release, that is. |
Re: Recording for 5.1
Quote:
|
Re: Recording for 5.1
You've got lots of info, but most of it is a little 'displaced' - correct for some things, but very wrong for another.
For conventional media products, dialogue is mono. Stereo is reserved for the music and effects. Practically all 5.1 sound is produced in the studio in post. It's to do with context. Let's say shot one is of an evil looking priest saying a prayer in a cathedral, shot from a long way away to set the scene, you can see the aisles, the pews, the walls and probably the organ - it looks BIG. Next shot is a close up of him removing something from his pocket while saying the prayer out loud. In shot 1, we hear the sound of his words from a distance, so quiet and reverberation quite evident - probably actually recorded close in, (let's ignore dialogue replacement for this example). So in post, his voice is thinned out, reverb added and then it's blended with the acoustic sound of a cathedral in stereo - plus the rear channels adding to the hugeness of the space. In shot 2 - his voice is much closer - because the camera is right in there - yet the same huge space continues in the side and rear channels - the centre mono channel has a closer perspective to match the camera angle - but the other atmos tracks give the viewer the feeling of big space. Maybe the stereo front channels also have small effects - footsteps, maybe a little bit of hubbub from other people in the space, same at the back. Sort of immerses the viewer. So if we assume the actual space was recorded, then we'd probably have a radio and/or a shotgun for the priests voice pickup and either spaced microphones or an X/Y pair for recording the ambience - perhaps with another set in the rear. This wild track would be edited to remove the odd shout or bang, and then the mixer would blend this all in to produce the surround sound. If you had a camera with so-called 5.1 sound it would NOT sound the same because real sound doesn't sound like what we 'think' it does. The only time this doesn't apply is when the product is a live recording of an orchestral or choral work, and perhaps other sources that have a natural acoustic balance - folk bands with no electric instruments. In this case a stereo recording is the main requirement. You have a choice of deciding to change audio perspective with the camera shots, or to leave it as it is. This is a common issue. When recording an interview with three people, do you pan the middle one to the centre and the outer people left and right, to match their real position, or do you keep them mono and perhaps record the traffic passing in stereo? See the problem? The Physics To record stereo needs two mics. To record stereo with good imaging is much more difficult. Are you looking for effect, or realism? The test for realism is simple. listen with your eyes closed and point to the featured sound source. Open your eyes and see if it is where you are pointing. Most stereo recordings with pictures fail this test miserably. We are not really recording stereo, we're recording two channels - not at all the same. You do see mobile stereo recording. It's quite common. A short shotgun like a 416 or similar, with a sideways facing fig-8 mic inside a zeppelin type windshield. You record in M/S - so mono for the centre and then the fig-8 mic provides the stereo side-to-side information. In post you can accurately control the width of the recording. Turning all these sources into 5.1, complete with mono centre, stereo front, effects and sub bass is a complex process! Planning for stereo is not a simple process. M/S 2 channel is probably the most useful technique for a single mic type shoot (even though there are really two!) |
Re: Recording for 5.1
The final film/video sound track is PRODUCED, it is not RECORDED. Even monaural finished sound tracks are mixed to produce the final product. It is never just the raw sound that was recorded. Even for fast-moving TV news they overlay voice-overs, and add wild-sound clips, etc.
And especially for stereo or higher output track counts, it is RARE that you go out into the location(s) and record 4-channel (never "5.1") The mix of sounds (including which channels they sound from) are produced artificially during the post-production mix-down. And very frequently using source clips that are monaural. And the background sounds "ambience" are typically NOT recorded on the set, but somewhere that sounds like what is needed. That MAY be in the same place the dialog was recorded, but it is not unusual for the ambience sounds to come from somewhere else. Or even artificially created from various pieces like wind, traffic, birds, etc. etc. The object of dialog recording is to EXCLUDE all extraneous sounds so that the editor has the maximum flexibility to combine foreground and background sounds to produce the desired final effect. So even if you want the ambient sounds from the set location, it is better to collect them before or after the actual video setup. |
Re: Recording for 5.1
Quote:
|
Re: Recording for 5.1
Recording effects in stereo can be a time-saving short-cut assuming the editor is skilled enough to use stereo clips properly. There are some things that are naturally stereo, like a passing train, etc. But it is also possible to create a convincing stereo passing train sound from a monaural source clip (by panning). And you could make the argument that it is easier to do it with a monaural source because then the location, speed, etc are fully under the control of the editor.
|
Re: Recording for 5.1
Quote:
Also, that means the person who told me it was not professional to record mono sound for a 5.1 mix was incorrect, am I right? |
Re: Recording for 5.1
Quote:
Quote:
Quote:
|
Re: Recording for 5.1
"I can get a 5.1 mix done in post production without any worries."
Even with pro mixer, there's worries. And you still have no control, once it's in the hands of the consumer... equipment, quality, placement, but that goes for mono or stereo too. 5.1 throws more factors into the equation. "Would I be missing out on anything by going fully mono?" If it's all dialog, mono is fine. Music tracks and some atmos' are nice in stereo, the dialog is still mixed mono... Panned to center, which comes out of both left and right speakers equally, which created a 'phantom center.' "Also, that means the person who told me it was not professional to record mono sound for a 5.1 mix was incorrect, am I right"? Right, this person does not know much about sound acquisition for picture. |
Re: Recording for 5.1
Thanks Rick! Could you tell me what the advantage is in recording effects in stereo? Does it help enhance the reality in post or is it just because it gives the mixer more options to play with?
The reason I ask is, it seems illogical to expect a stereo mic on set to capture the 'final space' of the edited film, and the effort seems like a compromise - for example, how different is it from capturing the same effect with two mono mics side by side? |
Re: Recording for 5.1
There are several Surround Sound mics produced or techniques...
I have used a borrowed Holophone on live to air broadcasting with stunning results. SoundField: Benefits of a SoundField System H3-D Mitra 3D Mic Pro : 3D Mic Pro Blumlein Pair - Wikipedia, the free encyclopedia I have looked at the H2n | ZOOM and wonder what results it may deliver in the 360 deg mode for film production. |
Re: Recording for 5.1
Quote:
|
Re: Recording for 5.1
Quote:
Two mono mics placed side by side would likely have phase abnormalities (or worse) when summed to mono.. |
Re: Recording for 5.1
To all...I didn't participate in this discussion because I've always believed in the old saying, 'Tis better to keep your mouth shut and let others think you a fool, then open your mouth and remove all doubt." Heh, heh!
However, I have made a copy of the discussion and added it to my 'Recordist' info file. Thank you all very much. Best regards, J. |
Re: Recording for 5.1
Quote:
Which brings me back to my question: why is it that effects have to be recorded in stereo, but dialog needn't? What are the pitfalls of recording mono effects? |
Re: Recording for 5.1
Quote:
In fact, when I am recording production sound, my preference would be that the space where the scene is shot (and the dialog recorded) is so QUIET that there IS NO other sound to record for background or atmosphere, etc. REMEMBER that the objective during shooting is to capture the dialog and ONLY the dialog as cleanly as possible. The fewer sounds there in that space the better. Quote:
Quote:
Quote:
Note that none of these work very well as generic questions. There are often exceptions on both sides. And very frequently you will need to make trade-off decisions based on your resources (equipment, personnel, schedule, experience, budget) vs. what the producer wants for the production. |
Re: Recording for 5.1
I guess the question is that when you are watching a movie, what you hear is rarely real. Certain things are required before you can edit sound. A clean track of dialogue - with as little background and as much artiste as possible, with no reverberation. You then have the choice of it sounding close - or, if you perhaps back off the low end, and add appropriate reverberation, you can make the person sound as if they are in any space, to suit the visuals. Reverberation is a subject in itself. Our brain uses reflections of the main sound to determine how far away it is, and what kind of building the subject is in.
So in my example, the sound of the priest at a distance would be the product of the editor working on the 'close' sound to make it sound distance. Then when we switch to closeup, the sound image changes too to reflect what a voice would sound like to match the new image. If you go to a cathedral and record the ambients sounds there, then a stereo microphone technique such as X/Y or M/S would produce the nicest stereo image if we want realism. However, the differences between one mic, in mono, or these two mics is subtle, and rarely overt. If you use a spaced microphone technique - recording sounds in two or more separate tracks, then you get much better separation. You also get time delays. Let's say your two spaced mics are a long way apart - maybe 25m each way. Then there will be real sync problems if somebody near the left mic claps their hands. The other mic may pick this up as a separate event, when the sound arrives there. So to our ears this sounds wrong. You could sync them up, but then some sounds will be even further apart time wise. An X/Y or M/S recording would not have this problem. For work in post, what you need are enough sources to let you create what you need. So maybe you want to add bells, where there weren't any - so you use sound effects. Mono bells, with some matching reverb to simulate what they really would have sounded will work fine. It's not real, you are creating a sound stage - you are creating a 'space' where one doesn't exist. It's magic. The person who told you that recording in mono is standard is probably quite correct. However, if you wish to record more than one thing, you need more channels. Note not specifically stereo. This is why pro cameras never label their audio channels as left and right, just A and B or 1 and 2. I can't say I ever record stereo with my bigger cameras, I have the on camera mic and maybe a radio receiver - but it isn't stereo. You mentioned dialogue in stereo - this again should be two channels if you have two people and two microphones. The editors nightmare is distant sound. It's almost impossible to make a distant perspective sound close. No magic plug-in yet to remove reverberation and coloration. |
Re: Recording for 5.1
Its an interesting subject and upon thinking on it further of why FX were recorded in mono years ago, the answer might be that mono recorders were the standard for location sound guys, then came the stereo Nagras and DAT machines but were there great stereo mics and easily available for location at that time?
These days multi track FX recordings are easily possible (and easily handled in a work flow) so maybe its time we all think at possibly a better way of doing FX recording and not be locked in with preconceived paradigms of the past. Almost of these hand held / portable digital recorders used today (no names mentioned) are way superior to any analogue tape machine used for location sound recording in the past. |
Re: Recording for 5.1
It's all about contrast and whilst location buzz tracks can be useful for filling in dialogue edits or adding filler to mask edits I tend to prefer creating all my stereo and surround backgrounds from scratch as it means I can control them and layer up ready for mixing or panning in the 5.1 sound field.
Dialogue is best recorded in mono as that is what it will end up as in the final mix but multi track recording to allow clean tracks is also OK, yes the dialogue may be panned or even placed thru processor to give space or depth but it all originates in mono. Sound effects can be recorded in mono or stereo depending on what they are but I have never encountered any 5.1 recording in all my years as a dubbing mixer or editor. Yes it may have its uses for sport and live event and the calrec soundfield mic is still used on such events but it is not a 5.1 mic just a multi channel mic that can produce several channels of sound in the same way that taking several shotgun or other mics and panning them around the 5.1 soundfield can produce. The .1 or LFE is not designed to be active all the time anyway and is there to add "low frequency effects" as a when required, if you leave it on or feed sound to it all the time you are doing it wrong and just adding low frequencies when they may not be required. All the 5.1 I have ever done has been created in post using specialist mixing hardware such as the AMS Neve DFC and the source material was pretty much all mono or stereo. For the past 15 years my main mic for FX recording has been this sony prosumer mic into a mini disc or a marantz recorder: |
Re: Recording for 5.1
Quote:
|
Re: Recording for 5.1
If its bagpipes I would personally recommend that the director makes it a oscar winning silent movie like The Artist!
More seriously most surround desks also have a divergence control for surround so you can adjust the sound stage and in the case of the AMS Neve consoles I helped to design when I worked for them we have an A/B wide control that you can use to change the width of any stereo signal, you can also automate it so it is possible to adjust the soundstage dynamically to suit the shot. As for sound stage you make it suit the flow of the pictures and it may be that locking a stereo sound stage suits the overall sequence more than trying to pan things around to create soundstage movement. |
Re: Recording for 5.1
Quote:
|
Re: Recording for 5.1
Quote:
|
Re: Recording for 5.1
Quote:
|
Re: Recording for 5.1
Quote:
it outputs in A/B format but my consoles also had the A/B wide controls so I could remove the S signal and just have the M or mono content if required. If you are not familiar with M/S mics they are basically a mono cardioid mic forward facing with a figure of eight mounted in the same sound plane, you then matrix the signals together to get a stereo output adn if you adjust the level of the S signal you can change the width of recording. The stereo matrix to get an AB output is as follows: Mono= cardioid capsule Left= M+S cardioid and fig of eight out of phase Right= M-S cardioid and fig of eight in phase So if I record everything FX wise in stereo and then remove the S content using my A/B wide control I am left with what the mono capsule picked up. As the mic if very phase coherant it is also possible to sum the A/B stereo output to get a good mono signal. |
Re: Recording for 5.1
Thank you for the explanation. I'll do a bit more research into all this, now that I know what to look for. Appreciate it.
|
Re: Recording for 5.1
I've always understood that:
L = (M + S) R = (M - S) In other words, the "front" or "positive pressure" side of the figure-eight is pointing to the left. And, incidentally, the same technique is used for FM stereo broadcasting. You sum (L+R) / 2 and use that to modulate the main carrier (which is the only thing picked up on a mono receiver). Then you subtract (L-R) / 2 (which is equivalent to the "side" signal) and use that to modulate the 38kHz stereo subcarrier. The receiver converts what is essentially M/S back to L/R stereo. Anyway, the formulas I have always seen appear to be the opposite sign (for the side mic) compared to the formula you give. Not a huge issue, except the wrong formulas would effectively exchange the left and right channels (and the violins really should be on the left). I'm curious where you found the info that you posted. |
Re: Recording for 5.1
You are probably right Greg as it's 25 years since I did my stereo sound training course so I think I got them the wrong way round and I have corrected my original post.
M/S mic's are very useful and having the ability to control M and S content of an A/B signal on the AMS Neve DFC consoles I have used has been very useful. You can also get this free plug in which allows you to adjust the width of any stereo signal: http://www.brainworx-music.de/en/plugins/bx_solo |
Re: Recording for 5.1
OK Gary, thanks. I'm quite confident in those formulas because I use them at least weekly if not more often when working on old phono recordings or cleaning noise in other sources.
I sometimes need to cut out a very tiny bit of audio (I'm talking of a few msec) to get rid of some noise which otherwise can't be removed. I want to cut on zero crossings, but the info in the two channels does not cross at the same point in time. So I convert to M/S, cut on the zero crossing of the M channel, ignore the S channel and do a very brief gain dip to zero at the edit point if necessary, and finally convert from M/S back to L/R. (Actually it's more complex than that, but that's a reasonable summary.) Of course, come to think of it, that editing technique would work with your formulas too, except that the M channel would appear in the usual R channel space, rather than the usual L channel space. Hmmm. But I have checked a few references, and the formulas are correct as given. Carry on! |
All times are GMT -6. The time now is 11:35 PM. |
DV Info Net -- Real Names, Real People, Real Info!
1998-2025 The Digital Video Information Network