Depends on the room. If you listen in a hypo-echoic lava cave, like I used to, room sound isn't a problem.
Not sure how much polar response makes a difference there as I only ever used open baffle (dipole).From my experience in the cave, and other very dry acoustics, I'm kinda on the fence about this one. Maybe it gets down to semantics. In the hypo-echoic cave, the sound was always at, or behind the speakers, never in front or coming into the listening space. But the sound could go back very deep, deeper than I've ever heard, and with layers of depth. The classic "Window on the Sound". The sense of the recording venue space was astounding, even on old mono recordings. But it did completely lack a sense of envelopment, so much so that I was about to install surround channels to juice it up a little.
So I might agree that I never felt completely "in" the recording venue, but more like the venue stretched out in front of me, behind the speakers. That's a trade off I'll gladly take, but I wouldn't mind having my cake and eating it, too.
I've read that a lot - speakers as a sort of giant headphone. But for me it doesn't sound like that - headphones never sound outside of my head and never have any real sense of space like speakers do. As nice as good headphones can sound, the sound is always small and claustrophobic to me.
Headphones will sound good and outside the head if binaural record will be used with your hrtf. But that is not the case with artificial stereo recordings. With speakers as a sort of giant headphones you will not get problem that the sound will be in your head. Crosstalk cancelation filter for loudspeakers must not be perfect in order to make usable sweet spot. And inaudible coloration outside the sweet spot.Plus room reflections added. So the sound will not be inside your head.
From my listening experience i can say that it is something in between, it is closer to the head but not inside. Binaural whispering in one ear over crostalk canceled speakers is a lot better then in not crostalked speaker, but it is not like over headphones. Whispering is closer to the head with crostalk cancelation.
Speakers for a big event need to have a different dispersion patetrn that speakers in a room. And small rooms need different speakers than a big room. Similarly speakers against the wall hardly need baffle step correcrection while speakers out from the wall need more or less depending on the interaction of the room.
Then if you listen near field or some distance away makes a difference too.
In other words there is not a single answer, take your pick of poison.
Then if you listen near field or some distance away makes a difference too.
In other words there is not a single answer, take your pick of poison.
Long post incoming. I've been into HiFi for about 15 years, been a live sound engineer for 10 years, and have been recording various instruments and ensembles for five.
The short version is this: the recording and reproduction of any instrument depends on this - should it be "the musician(s) is/are in my room", or should it be "I am in a venue with the musician(s)"?
To get an idea of how different instruments sound at different angles, here's a good link: http://soundmedia.jp/nuaudk/ IMO, it's worth pausing reading this, follow that link and try a few instruments before proceeding.
If you want "they are in my room", then you could put many microphones on each instrument, and capture a really good impression of the instrument's entire polar response, and then figure out how to play that back in your room. You'll probably need at least 10x amplifier channels and speakers (with correct physical spacing) per instrument.
A drum kit would require a LOT of microphones and playback speakers, because each drum should be considered an instrument with many complex radiating surfaces: think of all the circular harmonics that might be present. You could easily use 20x speakers just to reproduce a snare drum.
Theoretically, though, it could be done. Perhaps a reasonable compromise would be something like 20+ smallish full-range drivers (individually aimable, of course), about eight 8" midbass units and a couple of 15" subwoofers, all attached to CNC-machine-like tracks for moving them to the right position for each piece of music.
All of this assumes, of course, that the musicians are being recorded in an anechoic chamber. Any trace of the original venue would introduce uncorrelated (in terms of recording vs listening room) reflections.
Another option is to aim for "you are at the venue". That method, IMO, is simpler: a relatively straightforward multi-channel capture of the musicians playing at the venue, and then play that back in an anechoic chamber. You might only need 12x microphones, and the same number of wide-range speakers.
NB - I said "surround". Mere 2-channel reproduction cannot convincingly transport you to a choral recital at a cathedral. If we're sat in a cathedral, we will quickly note that while the musicians are in front of us, there is sound all around us. Sure, it's louder in front, but we're losing a lot of spatial information by simply putting all of the rearwards reflections through speakers that are in front of us.
It should be clear that both of these approaches have considerable technical requirements, including the need for an anechoic chamber, which would cause most to pause for a moment before pressing the "Buy It Now" button. Further, neither approach would work for all types of music.
What I'd like to do now is examine whether or not this idea of chasing "High Fidelity" is even worthwhile. "High Fidelity" is the idea that we're aiming for sound reproduction that's as true-to-life as possible.
So, cue up For Those About To Rock by AC/DC. During the intro, the bass drum and hi-hat come in. The hi-hat is being played fairly gently, but a little while later Rudd is really hitting those drums. I have a question: how many of us here have been in a room where a drummer has gone from playing gently, to absolutely smashing the drums?
I have. The volume difference, in a 100-capacity venue (bigger than most living rooms), is this: at the start, you'd easily be able to talk over the hi-hat. No problem at all. When things get going, you'd be shouting at the person next to you, and there would still be communication difficulties.
Next question: does anyone seriously want that dynamic range in their living room? Do you really want to start a piece of music and think "that's a sensible volume", or things to then get so loud you're blinking on every snare drum hit, and the crash cymbal sets your ears ringing? I'm assuming, of course, that your HiFi system can cope with such demands.
What do we do about the singer, who can't possibly keep up with the drums in terms of outright volume? Do we sacrifice vocals entirely in the name of realistic performance?
It's the job of the recording studio to take all of that raw energy (in terms of dynamic range etc) and present it to us, the listeners, in a format that will work in a domestic situation. That means a compressed dynamic range. That means that some microphones will need to be turned up louder than others, so that an artistically pleasing version of a musical event may be presented.
Other artistic tweaks may also be applied during the recording process: things as simple as the position of a microphone on a guitar amp can make a large difference to the sound. Here's a video:
which demonstrates that quite nicely.
So, HiFi people, which mic position is correct? The answer is all of them. The amp is making all of those sounds, and it's just how we capture it that's changing.
There are many other changes that may be made, including (but not limited to) EQ/compression on some/all channels, and some mix busses (ie, you might take the drum mix and put just that through a compressor/limiter - some engineers find that adds some "glue" to the drum sound, keeping it all together), and then there's all the potential effects. Reverbs, delays, chorus, etc etc etc. The list is almost endless, and we're still well within the realm of a low/mid-tier digital mixing desk. Modern recording studios have way more processing power available. We haven't considered for a moment the difference that a different mic preamp/compressor/EQ might make to the sound. They're not all created equal, after all.
Side-note: it's also worth noting that there's no requirement for the original "take" of each performance is the good one. If you record one instrument at a time, then you might be hearing the guitarist's 2nd attempt at that solo, and the singer's 3rd run through. You might also be hearing a jigsaw of takes, put together to achieve one good finished performance.
At the end of all that, though, the band/management/whoever will be called upon to listen to the finished mix, to make sure everything is balanced the way it should be, that the performances are good.
It's my opinion that we should be aiming for the reproduction of that sound. The one that the band hears through the monitors at the studio, and says "yep, that's the one". It's their music, and they're happy with how it's represented. In theory, at least.
After drawing that conclusion, I put together a fairly simple AV system, with four speakers* and a decent Denon AVR as the centre of it, and have been enjoying music and movies ever since. Not particularly for the fidelity, but for the enjoyment of musical performance.
* Simple surround setup: main speakers have 2x8"+compression driver, and the two satellites are SB65s in small sealed boxes. The AVR runs them down to 150Hz, and bass gets re-routed to the front speakers. When the music is recorded in a multi-channel format, the AVR does a good job of figuring out which content to put where, and the immersion is much improved over stereo. It did take a considerable amount of time to get the surround levels and delays dialled in correctly, but now (when called upon) effects can pan along the side of my room, fairly seemlessly.
I chose a horn for >1kHz because my speakers are fairly close to the side walls, soI wanted to avoid the walls being illuminated too much. I also like the usual benefits of compression drivers: they can go loud without complaining, and also lower than most dome tweeters.
Chris
The short version is this: the recording and reproduction of any instrument depends on this - should it be "the musician(s) is/are in my room", or should it be "I am in a venue with the musician(s)"?
To get an idea of how different instruments sound at different angles, here's a good link: http://soundmedia.jp/nuaudk/ IMO, it's worth pausing reading this, follow that link and try a few instruments before proceeding.
If you want "they are in my room", then you could put many microphones on each instrument, and capture a really good impression of the instrument's entire polar response, and then figure out how to play that back in your room. You'll probably need at least 10x amplifier channels and speakers (with correct physical spacing) per instrument.
A drum kit would require a LOT of microphones and playback speakers, because each drum should be considered an instrument with many complex radiating surfaces: think of all the circular harmonics that might be present. You could easily use 20x speakers just to reproduce a snare drum.
Theoretically, though, it could be done. Perhaps a reasonable compromise would be something like 20+ smallish full-range drivers (individually aimable, of course), about eight 8" midbass units and a couple of 15" subwoofers, all attached to CNC-machine-like tracks for moving them to the right position for each piece of music.
All of this assumes, of course, that the musicians are being recorded in an anechoic chamber. Any trace of the original venue would introduce uncorrelated (in terms of recording vs listening room) reflections.
Another option is to aim for "you are at the venue". That method, IMO, is simpler: a relatively straightforward multi-channel capture of the musicians playing at the venue, and then play that back in an anechoic chamber. You might only need 12x microphones, and the same number of wide-range speakers.
NB - I said "surround". Mere 2-channel reproduction cannot convincingly transport you to a choral recital at a cathedral. If we're sat in a cathedral, we will quickly note that while the musicians are in front of us, there is sound all around us. Sure, it's louder in front, but we're losing a lot of spatial information by simply putting all of the rearwards reflections through speakers that are in front of us.
It should be clear that both of these approaches have considerable technical requirements, including the need for an anechoic chamber, which would cause most to pause for a moment before pressing the "Buy It Now" button. Further, neither approach would work for all types of music.
What I'd like to do now is examine whether or not this idea of chasing "High Fidelity" is even worthwhile. "High Fidelity" is the idea that we're aiming for sound reproduction that's as true-to-life as possible.
So, cue up For Those About To Rock by AC/DC. During the intro, the bass drum and hi-hat come in. The hi-hat is being played fairly gently, but a little while later Rudd is really hitting those drums. I have a question: how many of us here have been in a room where a drummer has gone from playing gently, to absolutely smashing the drums?
I have. The volume difference, in a 100-capacity venue (bigger than most living rooms), is this: at the start, you'd easily be able to talk over the hi-hat. No problem at all. When things get going, you'd be shouting at the person next to you, and there would still be communication difficulties.
Next question: does anyone seriously want that dynamic range in their living room? Do you really want to start a piece of music and think "that's a sensible volume", or things to then get so loud you're blinking on every snare drum hit, and the crash cymbal sets your ears ringing? I'm assuming, of course, that your HiFi system can cope with such demands.
What do we do about the singer, who can't possibly keep up with the drums in terms of outright volume? Do we sacrifice vocals entirely in the name of realistic performance?
It's the job of the recording studio to take all of that raw energy (in terms of dynamic range etc) and present it to us, the listeners, in a format that will work in a domestic situation. That means a compressed dynamic range. That means that some microphones will need to be turned up louder than others, so that an artistically pleasing version of a musical event may be presented.
Other artistic tweaks may also be applied during the recording process: things as simple as the position of a microphone on a guitar amp can make a large difference to the sound. Here's a video:
So, HiFi people, which mic position is correct? The answer is all of them. The amp is making all of those sounds, and it's just how we capture it that's changing.
There are many other changes that may be made, including (but not limited to) EQ/compression on some/all channels, and some mix busses (ie, you might take the drum mix and put just that through a compressor/limiter - some engineers find that adds some "glue" to the drum sound, keeping it all together), and then there's all the potential effects. Reverbs, delays, chorus, etc etc etc. The list is almost endless, and we're still well within the realm of a low/mid-tier digital mixing desk. Modern recording studios have way more processing power available. We haven't considered for a moment the difference that a different mic preamp/compressor/EQ might make to the sound. They're not all created equal, after all.
Side-note: it's also worth noting that there's no requirement for the original "take" of each performance is the good one. If you record one instrument at a time, then you might be hearing the guitarist's 2nd attempt at that solo, and the singer's 3rd run through. You might also be hearing a jigsaw of takes, put together to achieve one good finished performance.
At the end of all that, though, the band/management/whoever will be called upon to listen to the finished mix, to make sure everything is balanced the way it should be, that the performances are good.
It's my opinion that we should be aiming for the reproduction of that sound. The one that the band hears through the monitors at the studio, and says "yep, that's the one". It's their music, and they're happy with how it's represented. In theory, at least.
After drawing that conclusion, I put together a fairly simple AV system, with four speakers* and a decent Denon AVR as the centre of it, and have been enjoying music and movies ever since. Not particularly for the fidelity, but for the enjoyment of musical performance.
* Simple surround setup: main speakers have 2x8"+compression driver, and the two satellites are SB65s in small sealed boxes. The AVR runs them down to 150Hz, and bass gets re-routed to the front speakers. When the music is recorded in a multi-channel format, the AVR does a good job of figuring out which content to put where, and the immersion is much improved over stereo. It did take a considerable amount of time to get the surround levels and delays dialled in correctly, but now (when called upon) effects can pan along the side of my room, fairly seemlessly.
I chose a horn for >1kHz because my speakers are fairly close to the side walls, soI wanted to avoid the walls being illuminated too much. I also like the usual benefits of compression drivers: they can go loud without complaining, and also lower than most dome tweeters.
Chris