Taperssection.com
Gear / Technical Help => Microphones & Setup => Topic started by: Overlay2009 on April 06, 2022, 09:06:29 AM
-
Dear Tapers,
I just subscribed to your wonderful forum
I had some years of experience of live recording in late 90s - early 2000s with my DPA 4060s, Rode NT4, Apogee Mi-Me and various recorders in chronological order like SONY TCD-D5, TASCAM DA-P1, Sony TCD-D100, Microtrack v1 (as bit-bucket), Tascam DR-100 MKII, Sony PCM-M10…
Now, after 15 years of working/travelling abroad with almost all my gear SADLY left in Rome to collect dust, I am back to Italy and I would like to start taping again this summer.
Might I ask you for some wisdom on a setup I intend to test when attending my next loud indoor concert with ugly acoustics
; - )
The key idea is to combine tracks recorded with
1) the Sennheiser Ambeo Smart Headset (binaural)
AND
2) the Rode NT4 pointed to the source - the stage, but maybe also the PAs in really horrible boomy scenarios. Maybe as an alternative to the NT4 I would like to use two Bayer M201TG hypercardiods in X/Y configuration for more flexibility
My intention is to use a high pass filter plugin in DAW to make a sort of crossover between
the low and mid-low part of the binaural recording done with Ambeo Headset, which hopefully will have better bass and ambience
AND
the mids and highs of the XY recording done with the Rode NT4/Bayer M201TG, hopefully more crispy and somehow - if lucky - less subjected to bad reverb
Obviously, compared to a standard M/S configuration, in my setup the S is not given by a figure-eight microphone at 90° COINCIDENT with the mic pointed to the source, but by two omnis spaced by my head, i.e. by binaural recording somehow related to the stereo central mic capture, playing the role of a sort of extended (because stereo) central channel M
I understand reading the Forum since last week that I independently rediscovered what seems to be a quite common approach among tapers, and it can work, as confirmed by Gutbucket:
QUOTE
Re: Relationships between spaced omnis and center mic
https://taperssection.com/index.php?topic=182579.15
« Reply #16 on: June 22, 2017, 05:29:06 PM »
"Well in some cases it can be - low pass the added omni(s) around where the 2-ch setup naturally rolls off and that keeps everything relatively simple.
It won't significantly change the way the 2-ch setup behaves in the range in which it is sensitive."
END OF QUOTE
I understand after a still superficial reading of the treasure of infos in this forum that Gutbucket has since more than a decade developed the strategy I am naively starting to explore at intriguing levels of complexity…
===============================================
MORE DETAIL ON THE SETUP(s) I would like to test, three versions with different SIDEs S
SETUP 1 - SIDE S is given by Sennheiser Ambeo Smart Set
(simple and stealth, I expect less phase issues but less stereo separation)
SETUP 2 - SIDE S is given by two DPA 4060 at the ends of the longest possible telescopic arm
(more cumbersome, it needs two recorders (or something like a Zoom F6 that I do not own) and a longer arm. It could give more stereo separation but maybe it will show comb effect)
SETUP 3 - SIDE S (extended) is given by two hypercardiods (Beyerdynamic M 201 TG) at the ends of the longest possible telescopic arm
(even more bulky, it needs two recorders and heavier mics and arm, but it can be good to catch a PA in desperate reverberant cases)
The FRONT MIC(s) M would be in all three cases
One hypercardioid (Beyerdynamic M 201 TG) or a shotgun (*) pointed to the source
OR
(“extended” M) Two cardioids X/Y (Rode NT4) or two hypercardioids X/Y (Beyerdynamic M 201 TG) pointed to the source
(*) I understand that DSatz – thank you for the beautiful informative encyclopaedic posts! - suggests to avoid shotguns for indoor concerts from far distance. As opposed to good hypercardiods, shotguns' directivity is highly dependent of frequency, so that sounds arriving off axis and reverberant sound are recorded with STRONG colouration.
MY RECONSTRUCTION of DSatz’s reasoning to test if I understand it well
in loud indoors live concerts, for average tickets, with even a good shotgun, whe have
1) THE WEAK DIRECT SOUND of the FAR stage captured as a STRONG uncolored (flat) on axis response
2) A STRONG REFLECTED SOUND FIELD of the same FAR stage captured as a WEAK highly colored (frequency dependent) off axis respose
These two vectors (sound sources) INTERFERE, giving strong frequency-dependent phase cancellations -
with a good hypercardiod, these phase cancellations would be frequency-INdependent.
DSatz’s conclusion: shotguns work only with a ON AXIS CLOSE source surrounded by DIFFERENT off axis sources
OBJECTION by Gutbucket: "the other mics in the array help to hide the off-axis issues of the shotguns"
MY COUNTER-OBJECTION: Are not the other mics in the array subjected by frequency-dependent phase cancellations when combined with the centre mic already "distorted"?
CONCLUSION
what do you guys think of my three configurations, especially the first "stealth" one, namely
Sennheiser Ambeo Smart Headset + Rode NT4 <-----> S + “extended stereo” M (4 tracks in total) ?
All the best
Overlay 2009
-
Hello, welcome to this forum, and (soon) welcome back to live recording.
Just to clarify--the more microphones that you set up close to one another, whose signals you then combine into the same playback channels, the less any theory of stereophonic perception applies. When portable digital recorders began to offer four, six, or eight tracks, many people here started using four, six or eight microphones and combining their signals in various ways. Any two or more microphones that are near each other, (depending on their directional characteristics and how they're aimed and whether you apply crossover filtering as you are thinking of doing) will pick up basically the same sound field. So when you mix their signals together, you will get some really complicated patterns of cancellation and reinforcement. Phase, amplitude and frequency relations become more nearly random.
Personally I don't like the effect, but no law says that anyone else has to agree. Plus in a given situation there really can be some advantages along with the disadvantages. This is a hobby for most people here; we try to be friendly toward one another and generally hold back from criticizing one another's opinions. But I find that it makes little sense to obsess about the quality of your equipment, then use it in ways that tend to negate that quality. There's a lot of "streetlight effect"--ignoring the weakest link in the chain while obsessively strengthening the already strong links.
One thing that I try to emphasize is that microphones don't only do what you want them to do--they do everything that they do, irrespective of your intentions. When you have a basic two-mike stereo recording and you add signals from additional microphones, you don't just get more of what you particularly wanted the added microphones to pick up; you get a mix that doesn't know and doesn't care which signals came from which microphones for what purpose. Likewise, in a reverberant space and at some distance from the sound sources, the off-axis response of your microphones becomes at least as important as their on-axis response, and some people simply don't want to consider that--they would rather imagine that their microphone "sounds" a certain way, irrespective of the kind of space it's in and how the sound is arriving. Your brain has a huge effect on what you think your ears are hearing, but your brain doesn't exist within preamps and recorders, and can't discount what your nervous system would consider to be irrelevant information; it all goes into the mix together.
Because of their highly irregular off-axis response, certain types of microphones that make good sense in special situations, such as large-diaphragm microphones or shotguns, don't generally make good sense as main microphones for stereo recording in reverberant spaces. Of course they can "work" and even produce very nice-sounding results sometimes--no one should deny that. But the odds of getting such nice-sounding results are quite a bit lower than they would be with well-placed, small-diaphragm microphones that have smooth response both on- and off-axis.
It's one thing to be hampered by the high cost of certain equipment, but to misunderstand how microphones work is another thing entirely. Unfortunately there are some misunderstandings (about microphones, about how digital audio works on a basic level, and others) that have an astonishing "holding power" in audio. It's maybe even comparable to people's opinions about Covid-19 vaccines and mask-wearing--some people strongly prefer to follow their own opinions based on their personal intuition and feelings, even when the evidence of science says something sharply different. Especially we Americans find something admirable in such tenacity, independence of spirit, and willingness to live with the consequences of one's choices. Fortunately, with mistaken beliefs about audio, such character traits generally won't kill you or your loved ones.
--best regards
-
Dear DSatz,
thank you very much for the answer - I agree on the dangers of pseudoscience, from physics to covid treatments
and from several of your posts and others I am finally aware of the shotgun issue
My question above is especially focused on the possibility to use binaural + cemter mic (let us say hypercardioid to avoid the shotgun issue)
The aim is to record in indoor concerts with bad acoustics far from the stage
1) The standard SM configuration seems out of question, because it is effective only close to the stage
2) An alternative "old" solution was
“The once common 3-mic outdoor Grateful Dead taper setup of two PAS shotguns + one center omni” (Gutbucket)
3) A modern (range of possible) alternative(s) - well researched/summarised by Gutbucket and others seems to be
“2meter spaced mini omnis with a center supercard”
“The single forward facing center directional mic plus spaced omnis works especially well at a distance”
I understand that for Gutbucket a shotgun can be an alternative to the HyperCard when used as the center mic to record far from the source in a boomy environment, “limiting the problematic attributes of the pattern, providing clarity from the centrally located, directly forward-facing directional mic without phase interaction problems with another shotgun, or with the omnis since they are spaced far enough to either side of the shotgun-mic.”
My question focused on the possibility of
4) a binaural configuration (Ambeo Smart Headset or DPA 4060s) instead of two 2m-spaced omnis - to add to the center hypercadiod mic
IN ALL THE CASES WHEN an ideal 2m distance between the L & R omnis is not possible for stealth reasons.
I am aware that configuration 4) is NOT ideal because of “phase interaction problems” between the centre mic C and the two quasi coincident omnis (*),
but did somebody in the forum try that configuration?
Best regards
Ovderlay2009
-
I forgot to add the (*) note:
(*) I am not dealing with the issue of the reproduction of the binaural recording on loudspeakers, I had good experiences with it and anyway it seems that the "personalized" HRTF Head-Related Transfer Function is not always a big issue because it can be at least partially inverted, see https://www.dpamicrophones.com/mic-university/binaural-recording-techniques
-
Thank you DSatz for very nice post. I understand what you are explaining us. I appreciate it. I have one question about spaciousness - depth, the feeling like you're in the place of recording. Do professionals compromise on the accuracy and clarity of the recording to achieve this feeling?
I ask this because I think Gutbucket's audience multi-microphone array is also motivated by this effect that even in stereo playback we feel more like we're in the middle of a music hall (I wouldn't like to speak for Gutbucket, I hope he will join). I believe that when the audience microphones are well placed and spaced and then very carefully mixed, it could create this special effect. From my point of view, the disadvantage may be a certain loss of clarity of the recording. But I like this spatial effect a lot. My listening skills are not good, so I can overhear things that are over the edge for someone else.
As you mentioned, it's mostly just a hobby for us. I usually have a predetermined place from which to record. Quite often it is not ideal. I also can't take the test and listen to it while recording. So my recording can be creative only when I record with a multiple microphones. I understand the dangers you mentioned. Perhaps this can be partially reduced by careful placement. Then at home, I can choose one of the pair or try to mix it. When mixing, I try to compare it carefully with a simple pair. It has advantage, I can train my listening skills by this. And I understand, a professional would proceed completely differently.
Overlay 2009, I once recorded with multiple microphones that were hidden. I put two omni on my shoulders, one cardiod between my shoulders towards the front and one between my shoulders towards the back. It was difficult when recording, there were too many wires. But I had fun processing it. In the end, I used an omni, and on the edge of audibility I mixed a cardiod that was pointing forward. I try to setup the main pair that I would use if I had only one option. And try to add a other pair in a totally different configuration - placement and pattern.
-
Welcome to the forum Overlay2009,
It's somewhat of an odd thing we do here, each person involved having somewhat different goals, motivations, ways of thinking and going about it.. and I value and appreciate the differences in taste, approach, and musical focus of all contributors. Thankfully, or regrettably depending on your perspective, there is no one single correct answer (thankfully I say). Much of how to best go about this is driven primarily by practical constraints and secondarily by one's personal thresholds in dealing with those constraints.
I try to determine important foundational relationships between the experience of being at a live performance, acoustics, recording technique, and the playback listening experience as a way of working toward understanding and attempting to describe those relationships in useful ways to myself and others. I tend to post about the more complex and unusual approaches around here in part because they interest me, yet also because many have proven useful in pursuit of my own goals. That said, the way I approach things will not necessarily be the best approach for others. Beware of the perils. Most often simple is best, all things considered.. just not overly-simple for the sake of simplicity.
Here are a few things I consider to be the most important because they precede and set the stage for everything else, including which approaches will be most appropriate for you: Know yourself; Know the situations in which you will be involved; Learn to correctly discern what you are hearing (both at a live event and upon playback); Trust your ears and how you hear things (corollary is to question what you think you are hearing); Keep things simple enough to remain manageable and fun (recognizing that simple means different things to different folks); Perfection is attainable, yet often an unreasonable goal and the enemy of the good; Be pleasant and respectful with everyone regarding the way in which you go about pursuing whatever it is you do (to the musicians, audiences, venues, folks here, folks listening at home..)
Some more specific, if less fundamentally important insights:
A perfect stereo recording can be made with just two microphones (corollary- To do so consistently requires near-perfect situations); Having options is good, especially when you are working with otherwise limited ones; The more channels and microphones used, the more complicated things get (which presents increasingly complicated problems in addition to additional opportunities); The complications of additional microphones and channels intended to be mixed together tend to increase in a geometric or even exponential way; Seek a balance of simplicity and complication which works appropriately for you, your situations and goals.
More on your specific questions to follow..
-
Somewhat more specific answers-
The recording position in the venue tends to be the most important thing we as tapers (sometimes) have control over. Effort put into recording from a more favorable position tends to pay greater dividends in the quality of the resulting recording than most everything else.
The simplest and most appropriate ways to use multiple channels tends to be:
1) Record two (or more) stereo pair configurations that work well on their own, then compare then afterward and choose the best sounding one. This is not only the best way I know of to figure out the differences between various stereo microphone configurations, it serves to usefully "hedge the bet" given the taper constraints under which we are making recordings.
2) Record a direct feed from the soundboard (mixing desk) if possible, in addition to the audience perspective pair of microphones. Unless you've really nailed the multichannel approach, this works as well or better than combining multiple microphones out in the audience. This also serves to greatly "hedge the bet", particularly in not so great reverberant rooms.
More complications begin arise when mixing multiple microphones that all share the same general position placed out in the audience. However, it sounds like this is what you want to explore (and it is what I like to do) so enough disclaimers, let's get on with that-
My recommendations for tapers pursuing an audience-position multichannel microphone approach:
1) Four channels is enough to start (ok, six if 2 are SBD feed)
2) Forgo your usually preferred near-spaced stereo pair. It is optimized in terms of pickup-pattern, spacing, and angle for a single pair of microphones alone. You are now re-configuring to optimize for four microphones instead of two. Different game.
3) Instead of a 2-channel near-spaced microphone configuration optimizing for both spacing and angle, use a wider than normal spaced configuration plus a coincident configuration in the center. Even though there are now 4 microphone channels, this represents just one additional microphone position in space (3 rather than 2) which in combination with the increased spacing reduces the potential for phase interaction complexities. If you have it, a soundboard feed can take the place of the coincident center pair.
4) Outdoors or in a good room start with 3' wide omnis and a coincident pair in the center (Mid/Side or X/Y). Go up to 5' wide with the omnis. If you can't arrange get the omnis wide enough, put a baffle between them (such as your head using the Senn Ambeo headset), or use somewhat more directional microphones and angle them apart to compensate.
5) Indoors in a less than stellar reverberant room, get as close to the sweet spot as you can. Get a SBD feed if you can. Replace the omnis with directional microphones spaced somewhat less far apart, but still twice as far or more than a typical near-spaced stereo pair. If the room is decent, angle them +/-45 degrees. If the room is challenging, point all microphones directly at the sound sources of interest - that is, at the band itself, back-line cabinets and PA speakers. Point the spaced directionals directly at the PA speakers (PAS) If this results in a particularly narrow angle between the microphones, compensate by increased the spacing between them if you can.
The above is something of a condensed conclusion of my OMT multichannel microphone array approach developed over the years - based in theory, verified empirically.
Almost time for me to head out. I'll address your specific questions about mixing and unconventional Side channel options tomorrow.
-
To get some of this out of the way..
MY RECONSTRUCTION of DSatz’s reasoning to test if I understand it well
in loud indoors live concerts, for average tickets, with even a good shotgun, whe have
1) THE WEAK DIRECT SOUND of the FAR stage captured as a STRONG uncolored (flat) on axis response
2) A STRONG REFLECTED SOUND FIELD of the same FAR stage captured as a WEAK highly colored (frequency dependent) off axis respose
These two vectors (sound sources) INTERFERE, giving strong frequency-dependent phase cancellations -
with a good hypercardiod, these phase cancellations would be frequency-INdependent.
DSatz’s conclusion: shotguns work only with a ON AXIS CLOSE source surrounded by DIFFERENT off axis sources
OBJECTION by Gutbucket: "the other mics in the array help to hide the off-axis issues of the shotguns"
MY COUNTER-OBJECTION: Are not the other mics in the array subjected by frequency-dependent phase cancellations when combined with the centre mic already "distorted"?
For the purpose of our discussion, the on-axis and off-axis response of a single microphone isn't interfering with itself. Technically it is, but that's all happening "within the microphone itself" and this is something beyond our control, other than choosing a different microphone all together. One of the primary concerns when mixing multiple microphones together is the interaction between them (the sum of all left panned mics in the Left channel, the sum of all right panned mics in the Right channel, and the sum of any center panned mics with both Left and Right channels).
In most indoor rooms, the reverberant sound heard in most audience locations is far stronger than most folks realize (in comparison to the level of direct-arriving sound). However, the presence of a PA system can change this balance significantly, and is one of the things that makes taping amplified performances which feature PA reinforcement considerably different than recording a fully acoustic performances.
Specifically on the bit quoted above. I've not actually used traditional shotgun mics in my arrays and am not specifically recommending them. However I have heard numerous recordings that have used them successfully. I also have experience with how a multiple microphone technique can serve to "hide" the less than good off-axis behavior of other microphones in the array when arranged and mixed appropriately. Based on that I have speculated about their potential use, yet use supercardioid as the most-directional pattern in my own setups.
Specifically on your counter-objection: This is where "arranged and mixed appropriately" comes in to play. To avoid the problem, it is advantageous for the channels which will be summed together to feature content which differs sufficiently from each other (even if they all sound similar in isolation). To achieve that essentially means spacing them far enough apart from each other, angling them in sufficiently different directions from each other, or both.
-
> For the purpose of our discussion, the on-axis and off-axis response of a single microphone isn't interfering with itself. Technically it is, but that's all happening "within the microphone itself" and this is something beyond our control, other than choosing a different microphone all together.
Please check your logic here. It makes zero difference whether two signals mix within a microphone or within a mixer or recorder; the outcome will be the same, including any and all destructive interference.
The old adage about the hammer and the nails applies here quite directly--things don't obligingly turn themselves into nails just because you have a hammer in your hand--even a very good one. If an option exists to choose "a different microphone [al]together", why rule that out in advance? Microphones with special characteristics (e.g. a polar pattern that varies greatly at different frequencies) can be problem-solvers for particular types of situations, but outside of those situations, such microphones are far less likely to deliver optimal results.
(Please excuse my picking on just this one part of an otherwise highly informative post ...)
-
^Yes, yes. All true and agreed. Thanks for calling this out.
I sort of held my nose when typing that.. in an attempt at directing attention to the interaction between multiple microphones based upon how they are arranged, without confusing things by getting into what is going on within each microphone itself (even though in both cases the underlying basis of constructive/destructive interference is the same).
My "For the purpose of our discussion" disclaimer doesn't actually stand up in terms of the underlying physical interactions. It was only meant to avoid a complicating side-discussion on shotgun/interference-tube microphones.
Overlay2009, if you are going to try using these multichannel microphone array techniques, use supercardioids/hypercardioids rather than of shotguns in them!
-
I'm not much of an open taping guy with stands and the like, but do have solid recording experience from 1970 to today.
Take as many microphones and put them in any array you would like, just put them on separate channels on a multi-channel recorder.
On the other hand if you're trying to load more than one microphone per channel, I think you will not get as good results as you would with one microphone per channel. It is similar to on the fly soundboard mixes to two channel, with lots of room for error.
I may not be reading the question very well, so excuse me if I've answered incorrectly.
-
That's actually an excellent point, one that has been presumed in the discussion so far, but should be clarified.
In this day and age of relatively inexpensive multichannel recorders I just assume tapers are recording each input channel separately. That includes all microphones and soundboard feeds. For folks wanting to combine mics, or even mics + SBD, save all mixing decisions until afterward when you can do it right, monitor clearly and play around with various level combinations and other things until you find the most appropriate combination prior to committing to any sub-mix or final stereo mix. These days I can't imagine ever suggesting to sum anything prior to recording, and likely wouldn't have suggested trying to do so it "back then" either.
To do so otherwise (mixing channels ahead of the recorder) is just asking for trouble and very difficult to get right IMO. Yet in the past, before multichannel recorders became common, inexpensive and of good quality, doing it that way was not unknown in the taper community. One particular instance that comes to mind, which older tapers will recall and younger tapers may have heard about is the Nakamichi 3-channel microphone mixer that was packaged with the Nakamichi microphone set (similarly a number of 2-ch Nakamichi tape recorders featured 3-inputs) which some Grateful Dead tapers employed in the 1980's for recording 3-microphone setups directly to 2-channel cassette (this is veering dangerously close to the shotgun discussion). Another is a few Oade GD tapes from the mid 80's where Doug combined spaced omnis and a center hypercardoid pair ahead of the recorder, and in interviews talks about about putting a lot of thought, time and effort into determining the exact spacing and mix levels beforehand. That they didn't continue to use that technique tells you something.
Record all channels separately, mix them later by ear.
-
I have one question about spaciousness - depth, the feeling like you're in the place of recording. Do professionals compromise on the accuracy and clarity of the recording to achieve this feeling?
I suppose it depends on how one defines accuracy. Or rather, accurate in what sense? Clear, accurate diction? Accurate sense of space and place? Timbre? Imaging? All of these? If so in what order of importance?
I find the most appropriate amount of depth-dimension and spaciousness - the "you are there-ness" qualities - to vary with the way in which the recording is intended to be experienced. Playback over quality headphones or a really good stereo playback allows for more of it. True surround playback allows for more still. More compromised listening situations do not, in which case the listening experience is better with greater emphasis on direct sound clarity. On better systems, recordings optimized for those compromised situations will sound flatter and less involving than they otherwise would be. In the compromised playback situations. recordings optimized for really good systems can sound less clear and more distant. It's difficult to determine a compromise that is "just-right" and works decently well most everywhere - which is an important part of the art of mastering.
This gets back to recording everything on separate tracks to defer mix decisions until later when you can scratch your head over it all you like. I find that trying to dial in a decent mix live (made to a stereo mix track on the same multichannel recorder, or to create a live-mix to patch out to another taper) just doesn't ever seem to work out just right. Even if I'm returning for something like a second night run, and have dialed in a good mix of the previous night's performance (same band, same venue, same recording location), often that won't translate to the second night without some careful listening and adjustment.
I ask this because I think Gutbucket's audience multi-microphone array is also motivated by this effect that even in stereo playback we feel more like we're in the middle of a music hall (I wouldn't like to speak for Gutbucket, I hope he will join). I believe that when the audience microphones are well placed and spaced and then very carefully mixed, it could create this special effect. From my point of view, the disadvantage may be a certain loss of clarity of the recording. But I like this spatial effect a lot. My listening skills are not good, so I can overhear things that are over the edge for someone else.
That's it in a nutshell, and what I seek in my own recordings. The spatial qualities are no substitute for clarity, which I consider to be more fundamental, and can at times be at odds with it with lesser playback systems as mentioned above. This is one of the blessing/curse things about these techniques. The extra "realness element" they can provide I find delicious and worth the effort, but in turn demands attention and a careful touch to dial it in just right. Without making that effort you can easily end up with a mess. Along with the additional freedom comes a need to responsibly manage it well. Using these methods requires additional commitment that cannot be avoided, but I feel that is a good problem to have as long as one is okay with dealing with it! Know yourself.
[..]Then at home, I can choose one of the pair or try to mix it. When mixing, I try to compare it carefully with a simple pair. It has advantage, I can train my listening skills by this. And I understand, a professional would proceed completely differently.
This is what I do as well. If I find my multi-microphone array recording effort doesn't sound significantly better than a straight two-microphone recording made at the same performance from a similar location, I know I'm doing something is wrong.
Thanks for interest and post.
-
Dear Gutbucket,
thank you very much for the detailed multi-post answer
I am eager to test a “fine tuned” setup based on your wide-range suggestions in the present post– plus other posts of yours I read last week, including the fascinating “Oddball Microphone Technique” pdf summary …
After reading your present post, maybe because of a my background (a BA in physics almost completed before trashing everything for the misguided choice of working in advertising :-( ) the only part that was a bit unclear for me is precisely the same passage commented by DSatz, namely that “For the purpose of our discussion, the on-axis and off-axis response of a single microphone isn't interfering with itself”… but I now understand that you want to consider the single microphone interactions as an independent variable for practical purposes
More on this later I hope
Thanks also to ‘kuba e’ for the comment and regarding Scooter123’s reminder about independent channels, I would be forced to do it anyway because the Sennheiser Ambeo records independently using my ipod as bitbucket…
All the best
Overlay 2009
-
Just a short comment on this,
Gutbucket: “The spatial qualities are no substitute for clarity, which I consider to be more fundamental”
I entirely agree. Moreover, the attempt to be faithful/to reconstruct the soundstage is strongly genre-dependent
Only for classical music and partly Jazz/acoustics a realistic reconstruction of the live spatial image is meaningful, for rock concerts or other sources the spatial dimension is imagined/artificially created from the start, and the live version is not meant to carefully reproduce an “original” source /”studio” version. The live performance is always artificial compared with a studio recording that is itself “artificial”, I mean constructed. this I think is well known - the point I want to make
with my maybe trivial comment is that our recordings are themselves, in a sense, creative reconstructions where we can try to place instruments/events where we prefer/hope/can strating from an already highly artificial live dimension...
-
After reading your present post, maybe because of a my background (a BA in physics almost completed before trashing everything for the misguided choice of working in advertising :-( ) the only part that was a bit unclear for me is precisely the same passage commented by DSatz, namely that “For the purpose of our discussion, the on-axis and off-axis response of a single microphone isn't interfering with itself”… but I now understand that you want to consider the single microphone interactions as an independent variable for practical purposes
More on this later I hope
Simplify that to- "best results tend to come from quality microphones that have polar patterns which are smooth and well-behaved off-axis". This is true for any stereo microphone recording, but especially so for taper style recording where the microphones are generally placed farther from the primary source of interest than most other types of recording, which tends to increase the proportion of sound arriving to the microphones off-axis, and it's perceptual importance.
Moreover, the attempt to be faithful/to reconstruct the soundstage is strongly genre-dependent
Only for classical music and partly Jazz/acoustics a realistic reconstruction of the live spatial image is meaningful, for rock concerts or other sources the spatial dimension is imagined/artificially created from the start, and the live version is not meant to carefully reproduce an “original” source /”studio” version. The live performance is always artificial compared with a studio recording that is itself “artificial”, I mean constructed. this I think is well known - the point I want to make
with my maybe trivial comment is that our recordings are themselves, in a sense, creative reconstructions where we can try to place instruments/events where we prefer/hope/can strating from an already highly artificial live dimension...
In an acoustic sense its all equally real as performed. The documentarian way of looking at it (which I find to be the most common taper mindset) is attempted capture and reproduction of the soundfield present at the event "as it was". Doesn't matter what that event was, the reality being documented is the sound in the room at the recording position.
That's seems a simple, clear goal. But the reality is that every recording will be significantly different than the sound at the actual event, and even considerably different from each other. Fortunately the musical content is the primary thing, and that generally translates well regardless of how closely the experience of listening to the recording emulates the experience of being at the performance.
We are most certainly crafting auditory illusions, and the listener is a willing participant in the illusion. I think the most important part of the entire recording and reproduction chain is the listener's willing suspension of disbelief. Even if they don't recognize it consciously, listeners want to convince themselves the recording they are listening to sounds real. Our job is to provide sufficient, good-enough cues to allow that suspension of disbelief to be engaged easier and more completely.
I try to find an appropriate balance between "what it sounded like there" and "what sounds most enjoyable and convincing on playback", which represents a creative reconstruction, as you term it. The use of multi-microphone arrays are an attempt to extend the sense of immersion and convincing imaging beyond the front soundstage, by getting the sound arriving from all directions as well as the front sounding as natural and convincing as possible. It's not especially important for translating or documenting the music, yet helps translate a more convincing sense of being present at the live performance.
Although I obviously value all these things, they are all "nice to have" rather than "need to have" aspects of a good live music recording.
-
Apologies for turning your thread into a philosophical dissertation!
-
We are most certainly crafting auditory illusions, and the listener is a willing participant in the illusion. I think the most important part of the entire recording and reproduction chain is the listener's willing suspension of disbelief. Even if they don't recognize it consciously, listeners want to convince themselves the recording they are listening to sounds real. Our job is to provide sufficient, good-enough cues to allow that suspension of disbelief to be engaged easier and more completely.
I try to find an appropriate balance between "what it sounded like there" and "what sounds most enjoyable and convincing on playback", which represents a creative reconstruction, as you term it. The use of multi-microphone arrays are an attempt to extend the sense of immersion and convincing imaging beyond the front soundstage, by getting the sound arriving from all directions as well as the front sounding as natural and convincing as possible. It's not especially important for translating or documenting the music, yet helps translate a more convincing sense of being present at the live performance.
Although I obviously value all these things, they are all "nice to have" rather than "need to have" aspects of a good live music recording.
This is such a great discussion. I feel the exact same way. I am documenting the performance at the show, but I am definitely taking creative license to create what I feel is the most pleasing aural experience as part of the mastering process. That could be EQ, saturation, adding stereo width, etc. Limitations are obviously your own ear and monitoring, but knowing your system and listening through different systems can mitigate some of those concerns.
-
Just a short comment on this,
Gutbucket: “The spatial qualities are no substitute for clarity, which I consider to be more fundamental”
I entirely agree. Moreover, the attempt to be faithful/to reconstruct the soundstage is strongly genre-dependent
Only for classical music and partly Jazz/acoustics a realistic reconstruction of the live spatial image is meaningful, for rock concerts or other sources the spatial dimension is imagined/artificially created from the start, and the live version is not meant to carefully reproduce an “original” source /”studio” version. The live performance is always less artificial compared with a studio recording that is itself “artificial”, I mean constructed. this I think is well known - the point I want to make
with my maybe trivial comment is that our recordings are themselves, in a sense, creative reconstructions where we can try to place instruments/events where we prefer/hope/can strating from an already highly artificial live dimension...
Fixed?
-
To wrap up my more philosophical contributions to this discussion, before returning to the specific questions in the original post, let me say that I very much agree with DSatz's comment that there is significant "streetlight effect" in the taper community--ignoring the weakest link in the chain while obsessively strengthening the already strong links, and that best results will generally be found with well-placed, small-diaphragm microphones that have smooth response both on- and off-axis.
Where we may differ a bit is identification of that weakest link, or more likely, how best to deal with it. I think the weakest link in the chain for live concert tapers, and the most basic thing that differentiates our type of recording from most others, is the "well-placed" thing (recording position in the venue), and secondarily the lack of opportunity for tapers to listen through the recording chain when setting up the recording gear, allowing for important decisions and corrections to the setup prior to pressing the record button. If I could put the microphones anywhere I wanted and adjust the microphone arrangement while listening during an extended soundcheck, I'd be much more content to rely on a simple two-channel stereo pair recording arrangement. Unfortunately this is rarely the situation for concert tapers, and I feel these two things in combination represent our weakest link.
With regards to this forum, I find the primary value of using additional microphones and recording channels to be greater flexibility in achieving a good recording when working under those constraints - and I feel this is often not sufficiently recognized as being the fundamental difference in circumstance with this type of recording. Also, for myself at least, the ability to consistently achieve a strong sense of "sounds like I'm right there" by way of using more complex techniques is just as strong.
-
The key idea is to combine tracks recorded with
1) the Sennheiser Ambeo Smart Headset (binaural)
AND
2) the Rode NT4 pointed to the source - the stage, but maybe also the PAs in really horrible boomy scenarios. Maybe as an alternative to the NT4 I would like to use two Bayer M201TG hypercardiods in X/Y configuration for more flexibility
MORE DETAIL ON THE SETUP(s) I would like to test, three versions with different SIDEs S
SETUP 1 - SIDE S is given by Sennheiser Ambeo Smart Set
(simple and stealth, I expect less phase issues but less stereo separation)
SETUP 2 - SIDE S is given by two DPA 4060 at the ends of the longest possible telescopic arm
(more cumbersome, it needs two recorders (or something like a Zoom F6 that I do not own) and a longer arm. It could give more stereo separation but maybe it will show comb effect)
SETUP 3 - SIDE S (extended) is given by two hypercardiods (Beyerdynamic M 201 TG) at the ends of the longest possible telescopic arm
(even more bulky, it needs two recorders and heavier mics and arm, but it can be good to catch a PA in desperate reverberant cases)
The FRONT MIC(s) M would be in all three cases
One hypercardioid (Beyerdynamic M 201 TG) or a shotgun (*) pointed to the source
OR
(“extended” M) Two cardioids X/Y (Rode NT4) or two hypercardioids X/Y (Beyerdynamic M 201 TG) pointed to the source
If using two microphones (keeping things simple) and wanting to improve your odds of making a good recording in a difficult acoustic, I suggest first determining the location of the best recording position in the venue and setting up there if possible. If not possible, choose the best location that is practical. Then arrange your Beyerdynamic microphones in a stereo configuration determined by the width of the PA and stage as viewed from that position and the suggested microphone angle and spacing combination indicated on the improved PAS table linked in my signature. You might want to run the Sennheiser Ambeo Headset separately to provide a point of comparison.
If recording more than two channels with the intent of combining them later, I strongly suggest recording all channels to the same recorder. This requires a recorder capable of recording more than two channels of course.
It can also be done using separate recorders that are digitally clock-linked so as to remain synchronized, but that's probably not in the cards. Technically if can also be done using two entirely unconnected recorders but honestly that's a real hassle and likely to be more trouble than its worth. Recording to two separate recorders that are clock-linked requires more effort and complication before, during, and afterward. You'll need to wire to recorders together, make sure they are clock linked and recording together, and afterward get the starting point of the files perfectly synchronized. To do the same using two entirely separate un-linked recorders requires somewhat less complication before and during recording, but is a lot more complicated afterward. You'd then need to not only synchronize the start of the files, but also alter them so they will play at exactly the same speed and remain synchronized throughout until the end. Honestly, its probably not worth it in comparison to the cost of a four or six channel recorder these days, where all channels will be perfectly synchronized with each other without any effort.
Alright, lets assume you are using your Ambio binaural pair as the foundation of the recording and want to add a single Beyer hypercardioid to that to increase clarity. First, you'll want to keep the geometric relationship between the binaural mics and the hypercardioid the same throughout the entire recording. Don't allow the microphones to move relative to each other. If they do you are likely to hear phasing once combined. Binaural on your head and the hypercard in a hat maybe? Non-moving dummy head fro the binaurals? Either way, point the hypercard directly at the loudest, clearest source, likely the closer PA speaker, and orient the binaural pair so that the the sound as heard from that position is Left/Right balanced in an auditory sense, regardless of how things appear to the eye. Don't face everything stage-center based on what you see, face toward the apparent "acoustic-center" even if that means turning more toward the left or right. This applies to any microphone configuration.
If you use the Rode X/Y mic in place of the single Beyer hypercard, or both Beyers in X/Y, do the same, except recording 4 channels instead of 3. If you use the DPA omnis instead of the Senn binaural headset (on your head or spaced farther apart somehow) approach it the same way. I think you get the picture.
Once home, transfer all the files to the computer and play around. This is the fun and educational part. You can try whatever you want, whatever you think of. Listen to each part alone. Get each part sounding as good as it can in isolation. Play with various level combinations. Try everything full range, try implementing crossovers if you want, try different equalization between each parts (which in effect acts as sort of a less domineering cross-over), try equalizing the resulting combination and how that might affect your level combinations or EQ of the parts. And so on and so forth. It's your baby to raise as you see fit. Go back regularly and compare with your straight 2-channel stereo pair, or someone else's recording of the same event. Question what you are hearing and what you decide sounds best. Take listening breaks and return with fresh ears. Listen elsewhere. Ask what a friend thinks. We'll be here with advice if you want.
-
Dear Gutbucket,
I am very grateful for your detailed analysis of my suggested setups. I cannot wait to try them armed with the PAS table... and upload some sample to the group, maybe with a before and after post-production comparison...
After your pointing out so clearly the potential delay issues I could have trying to synchronize independent recordings made with unlocked stereo recorders without a common clock, I now understand that the Ambeo could be a challenge not only because of phase differences between its binaural setting and the NT4 or the two Beyer M 201, but also because unfortunately Ambeo's Apogee AD converter section cannot connect to a multitrack recorder like the Zoom F6, because its digital output can connect only to ipad/ipod/iphone, using them as a bitbuckets through a Lightning cable...
What do you think about using a software like “PluralEyes” to quickly solve the syncho issue almost in an automated way? It is used for video with multi source audio. Have you ever used it? https://www.maxon.net/en/red-giant-complete/pluraleyes
Maybe is there a plugin for WaveLab or IZotope?
-
I've not used PluralEyes myself, but am aware of how it works. Cool stuff for aligning separate video sources.
Others here will have experience with it. However, I doubt it will work well enough for aligning audio recorded on two separate recorders for two reasons:
1) The alignment needs to be way more accurate for audio intended to be mixed together. Consider that video frame rate is generally around 30 frames per second, while audio samples rates are 41,100 samples per second or greater. Audio recorded on the same recorder will retain that level of accuracy across all channels. Granted, absolute sample accurate alignment isn't necessary, but you will need to get the two within a few milliseconds of each other..
2) ..and retain that relationship throughout the recording without one source running faster than the other causing the two to "drift" apart over time. That requires stretching the length of the faster running source or shrinking the length of the slower running source to match the other. I don't think PluralEyes is capable of this. It typically requires using a resampling routine in a DAW along with some math to figure out the right amount of stretch, or a specific time-stretching function that does the math for you after identifying common beginning and end points of the two files.
I use Samplitude to do it which has an Elastic Audio function that eliminates the need to do the math. I'm sure it can be done in WaveLab, and other TS'ers who use that program may describe how to do so or point to you discussion threads here about it.
-
:banging head: My experience with PluralEyes 4 has not been a good one.
Thanksgiving 2021 was when I bought and started trying to install :o PluralEyes 4. Vaporware to me so far.
I have been trying to install it through the "MAXON" portal since November 2021. The program launches, and after five seconds "POOF" it is gone.
I have spent many hours in the troubleshooting rabbit-hole with Doug Y and some other individuals (Tim, Rich, Micah) at ToolFarm, and have tried the four(4) new MAXON versions that they have come up with, since we started. Today I give up.
I have decided to attempt to install it on my wife's brand new win10 laptop. It pains me that I may not ever be able to use PluralEyes on the workstation it was intended for, but I will make an attempt to install it and OBS on the new machine, to eliminate the question of this problem being something wrong with my fios network connection (their latest hair-brained idea).
This is quite off-topic, pardon me and pm me for details on this saga if interested..
-
Thank you for the very nice posts. It's a very interesting read. I'm sorry to join in the delay. I've been away from the computer for a longer time.
To that philosophical part, I would add that what we do not know, we cannot hear. Someone has talent and is sensitive to music. But we who do not have talent do not hear a lot of things. These things do not exist from our point of view. Someone has to point it out before we can focus on it. Only when we focuses on it, it begins to exist. And even those who have a huge talent, they will make a different recordings. (that's great, there are more ways to create the same quality recordings)
The posts about the depth-dimension and spaciousness are very interesting for me. I have only few of my recordings made by microphones array and I'm just playing with it. But I would have one example that caught my ear. It is a stereo sbd matrix by Dusborne with two audience recordings. I don't know how he could sync it. But the result is great for me. The depth of recording reminds me of what Gutbucket does with his microphone arrays. It sounds very similar to me. I believe that the depth was created by mixing the two different audience recordings. The loss of clarity I had in mind is when I compare it to the pure sbd or aud. I am able to tolerate this loss of clarity. And I can imagine that a similar style of depth could be created also in an acoustic music. The question is how other people hear it.
matrix: https://archive.org/details/gd1991-06-09.128299.mtx.dusborne.flac16
sbd: https://archive.org/details/gd1991-06-09.138333.sbd.miller.flac16
aud1: https://archive.org/details/gd1991-06-09.fob.brennecke-young.GEMS.95641.flac16
aud2: https://archive.org/details/gd1991-06-09.AKG451.Darby.118811.Flac1644
-
Hi kuba e,
I just gave a listen to Playing in the Band in two recordings you linked above. My thoughts-
The soundboard recording is a complete rendition of the stage-mic'd "direct sound", without any on-stage source missing or under-represented. Although elements in it seem overly hard-panned, they are all in good balance with each other. This is an example of what I'd call a complete soundboard feed, in contrast to some soundboard feeds from smaller venues where things like the guitars or other sources that were particularly loud on stage (and thus in need of less PA reinforcement in the room) and therefor often have less representation in the soundboard feed, making for something I think of as that an incomplete soundboard feed.
Although this soundboard recording is complete in the above sense, it doesn't convey any ambience or sense of the place in which the performance occurred, and contains no audience involvement at all. It is clean and clear, yet unnatural and "sterile" in comparison to what someone at the concert would have experienced. In a way it represents only a portion of the listening experience and is lacking much of the sound and excitement of actually being at the event.
In the version with the audience microphones mixed in, I hear audience involvement and space around the instruments. That sense of space is subtle, but is enough to sound more natural, less unnaturally flat and studio-like - still not exactly like what someone at the concert would hear, or even an idealized version of such, yet a significant improvement to my way of thinking.
I don't mean to be too critical, but the mixed version is not without problems. I hear some alignment/timing issues smearing things a few minutes in, and there is not enough audience or ambience to sound really convincing to me. However, I found the loss of clarity you mention mostly just a frequency balance issue, correctable with some EQ. In both cases I used the built-in Winamp EQ of the embedded archive player to correct the response of some crappy in-ear headphones that I have plugged directly into this computer in addition to adjusting the recording itself as needed to sound right overall. Comparing the two resulting EQ curves, I found I had dialed in a bit more emphasis between 5k-14k in the mixed version to achieve the same apparent frequency balance, and with that the clarity between the two was similar.
In my opinion, although the version which includes the audience microphones is a big improvement, I still find myself wanting more sense of actually being at Buckeye Lake Music Center in 1991, listening to the band play there and interact with the audience from a realistic perspective within it. As it is, I can sort of imagine myself seemingly suspended well above the audience, heard more as a somewhat disembodied distant roar of applause in the quieter parts.
To me, achieving that sense of "as if there realism" is the challenge, the icing on the cake. It's hard to nail. It's hard to get a realistic audience reaction without highlighting specific audience contributions that are unwanted. It's hard to get sufficient direct sound clarity along with a goods sense of actually being in the performance hall surrounded by other like-minded folks enjoying the experience, without drawing attention away from the music performance. Each aspect of the recording working together to help convey a believably real experience of being present at that particular live performance, in all its "in the present moment" glory.
What is fitting by way of this example is that from the start, more so than most, The Grateful Dead were very much interested in minimizing the isolation inherent in the metaphoric 4th wall between band and audience. What I'm advocating for in live performance music recordings feels to me like something of an extension of that same desire - reproducing the truly "live" essence of a live performance which tends to involve a lot of information flowing both ways across that invisible line of demarcation.
-
A few other thoughts on soundboard recordings and the one above in particular-
With a complete soundboard recording such as that one, I'd be happy running just a pair of wide omnis, and/or the rear-facing microphones out in the audience. I don't really need the microphones in the array that are trying to pickup as much direct-sound clarity as possible from the PA. In this case the soundboard is the primary source and the audience microphones contribute what the soundboard is lacking, and can potentially provide that without as much overlap and potential conflict in that way.
However, I almost always lean on the audience microphones more strongly than the soundboard feed most of the time, and I'm not going to forgo the microphone channels in my audience-position microphone array which focus on direct sound from the stage and PA, in part because I won't know how good the SBD feed really is until later. I generally prefer to only use as much soundboard as necessary - most often to help clear up and reinforce the vocals, along with anything else that needs help. One way I know I'm on the right track is when I find don't really need the soundboard feed even though I might have recorded it.
-
^ and with that in mind, I now see that you've also linked the two AUD sources in addition to the straight SBD and the mix (matrix) of that with the two AUDs. If I get some time over the next few days I'll give all of these a more proper listen. One thing I want to listen for is how well the AUD sources stand on their own, how much I feel the addition of SBD to them might improve things, or if that is even needed. In my opinion, the straight SBD definitely benefits from the addition of AUD. It will be interesting to determine if the AUDs sound like they would benefit from some SBD or not. If they are really good, they won't need it - and kudos to the tapers that made them.. along with everyone there, all involved in making that moment what it was.
Let me know your thoughts on these recordings.
-
Thank you for the review and explanation. Yes, the matrix is not aligned in many places. I guess it was a lot of work to match the three independent stereo tracks. For some reason, where it sounds good, I like this sound a lot. Thanks for the tip, I'll try to tune it with eq. Both audience recordings are good. One has a lot of direct sound, the other is a little further away. When I have more time, I'll try to choose one track to align it and try to make my own matrix to see how each resource affects. I also like less soundboard and more audience.