Another test finds HomePod frequency response flat, but results potentially meaningless

13567

Comments

  • Reply 41 of 123
    nht said:
    [...] Any speaker design that depends on reflection for part of its performance will get crippled in a chamber.
    One might argue that any speaker that depends on reflections for part of its performance is violating the most basic tenets of high-fidelity anyway, so measurements don't really matter.

    When I've laboured to capture (or create) a carefully crafted sense of sonic "space," the idea that the playback system would deliberately impose the sound of any arbitrary listening space rather than trying to mitigate it is disappointing to say the least. That's one of the reasons Bose was so hated in the audio production community. Amar convinced an entire generation of consumers that adding the sound of their living room to the recording was somehow a good thing.

    The purist will abhor devices that try to artificially "create" (as opposed to simply reproduce) any part of the sound. The pragmatist will recognize that perfect playback is impossible and choose whatever sounds closest or most enjoyable.
    What if most listeners can't or won't labour to capture (or create) a carefully crafted sense of sonic "space," ?
  • Reply 42 of 123
    nht said:
    [...] Any speaker design that depends on reflection for part of its performance will get crippled in a chamber.
    One might argue that any speaker that depends on reflections for part of its performance is violating the most basic tenets of high-fidelity anyway, so measurements don't really matter.

    When I've laboured to capture (or create) a carefully crafted sense of sonic "space," the idea that the playback system would deliberately impose the sound of any arbitrary listening space rather than trying to mitigate it is disappointing to say the least. That's one of the reasons Bose was so hated in the audio production community. Amar convinced an entire generation of consumers that adding the sound of their living room to the recording was somehow a good thing.

    The purist will abhor devices that try to artificially "create" (as opposed to simply reproduce) any part of the sound. The pragmatist will recognize that perfect playback is impossible and choose whatever sounds closest or most enjoyable.
    What if most listeners can't or won't labour to capture (or create) a carefully crafted sense of sonic "space," ?
    The sonic space I refer to is the one I put in the recording. I'm an audio engineer.
  • Reply 43 of 123
    One of the ways to rate a speaker is the relevance of its "positioning".

    Some active speakers will give their best only in a limited number of positionings, only in particular rooms, and only with some kinds of music.
    Some other active speakers would perform well in these variable situations more often than others.

    Now one of the distinctive selling points of the Apple HomePod is its capacity to "hear" the surrounding space and adapt the sound so that the final result for the listener is flat (good).

    That means that if you test HomePods in such an artificial setup as an anechoic chamber, when the HomePod tries to deploy its "listening" function to adapt its sound it will result as if it was positioned in the open air without any reflective surface, near or far, of any kind. But that would not reveal or test its distinctive feature. That doesn't mean that the HomePod is at fault: actually the test is at fault!

    An anechoic chamber would prevent the full operation of such key feature of the HomePod: the mitigation of the "positioning" issue that taints so many traditional speakers. 

    To test such key feature, one should test HomePods against competitors, incl. those much more expensive, in, say, 10 or 20 different rooms (sizes and furniture setups) and 10 or 20 different positioning in each room. 

    I.m.o. the other speakers would perform well only in some rooms/setup and especially in few positionings in these rooms, while fail in most of the others. The very expensive ones might perform well more often than the relatively inexpensive ones.
    Instead, still i.m.o., the HomePod is expected to perform well in most of the rooms, in most of the furniture setups, and in a large variety of positioning in all of the rooms. In this way the HomePod is going to challenge those very expensive speakers that perform well in many situations.

    This is the kind of comparative test that we need in order to give justice to such unique HomePod feature. HomePods are going to sound very good in any circumstance, including any end user's unique room/positioning, more likely than other speakers. 

    Therefore the real-life end-user performance of the HomePod is going to be statistically much better than that of the direct competition.
    edited February 2018
  • Reply 44 of 123
    nht said:
    [...] Any speaker design that depends on reflection for part of its performance will get crippled in a chamber.
    One might argue that any speaker that depends on reflections for part of its performance is violating the most basic tenets of high-fidelity anyway, so measurements don't really matter.

    When I've laboured to capture (or create) a carefully crafted sense of sonic "space," the idea that the playback system would deliberately impose the sound of any arbitrary listening space rather than trying to mitigate it is disappointing to say the least. That's one of the reasons Bose was so hated in the audio production community. Amar convinced an entire generation of consumers that adding the sound of their living room to the recording was somehow a good thing.

    The purist will abhor devices that try to artificially "create" (as opposed to simply reproduce) any part of the sound. The pragmatist will recognize that perfect playback is impossible and choose whatever sounds closest or most enjoyable.
    What if most listeners can't or won't labour to capture (or create) a carefully crafted sense of sonic "space," ?
    The sonic space I refer to is the one I put in the recording. I'm an audio engineer.
    Two questions:
    1. Aren't you setting yourself up for disappointment, as most listeners won't be able to (or care to) truly recreate that sonic "space"?
    2. What if you could create a digital profile of that sonic "space"  (note by note, instrument by instrument, if necessary) -- and package the profile with the space so that the space could be recreated by intelligent speakers available to any listener?

    I ask the 2nd question this way, because Apple has spent a lot of time, talent and dollars on the homePod.  I suspect that we're only seeing the tip of the iceberg of the technology and A/V products.  I've been observing Apple since 1978, and that's the way Apple rolls!

    Note:  I realize that others are working with similar tech, e.g. the Beolab 90 speakers at $80,000 a pair -- but I believe that Apple is better positioned to bring this to the prosumer -- even the hoi polloi.

    IDK if you are compensated for the number of songs/albums albums sold -- or just do it for the pride and satisfaction.  I suspect both motives would be more rewarding if you could reach a significantly larger number of listeners.
    edited February 2018
  • Reply 45 of 123
    HomePod Mono or Stereo?

    Just play this song on your HomePod, Journey Of The Sorcerer by Eagles.

    At the very beginning the banjo reverb is growing from right to center, then from left to center.
    It's easy to tell if you listen to it on a pair of earplugs or headphones.
    Now if the same effect (at the very beginning the banjo reverb is growing from right to center, then from left to center) is reproduced also by the HomePod in your room, then it is stereo indeed, otherwise, as I expect, it is mono.
    I'm sure there are many more famous songs that could help to achieve the same distinction.

    https://open.spotify.com/track/3OcBH9Vzd1UwJkQd3r1dVG?si=CwRmT4lwTK2uline5z03HQ
  • Reply 46 of 123

    nht said:

    foggyhill said: 
    It's not mono either
    This is one of several claims Apple makes about the HomePod that make me scratch my head a little...

    Apple says the HomePod beams "direct" sound to the middle of the room and "ambient" sound to the left and right. But then Apple ALSO claims that the HomePod sounds the same no matter where you are in the room. Those are conflicting statements.

    Since people are saying the sound of the HomePod does remain very consistent as one moves around the room, the logical conclusion is that there can't be much in the way of directional cues coming out of it. If there were, listeners would perceive a change in the sound as they moved around.

    If that assessment is correct, it means that despite what Apple's marketing claims, it's really essentially a mono device.
    In the same way that a violin is essentially a mono device...
    I'm pretty sure you meant that as a funny, but just in case you didn't, my response would be "Yeah, kinda, except not."

    The sense of "ambience" we hear from the violin is the result of reflections off the surfaces in the room. Those reflections are the sound of the room though, not the instrument per se. In the case of music reproduction, the goal is to hear the reflections from the room in which the recording was made, not the listening space.

    Apple's claim of beam forming is actually a reasonable alternative to a spaced pair of speakers. What I'm having trouble with is reconciling that with Apple's other claim that the HomePod sounds the same no matter where you are in the room. Both cannot be true. Either the HomePod creates directional signals to separate difference and sum components, which would result in it sounding different in the middle of the room than it does to one side, *OR* the sound doesn't change as you move around, meaning it's essentially omnidirectional and thus does NOT provide any kind of meaningful directional cues. It can't be both.

    Since those who have shared their experience with the HomePod have told us that the sound remains very consistent as they move around the listening space, the conclusion is that there isn't much of a directional component to its output, meaning it's essentially mono. One may still experience a sense of space and ambience, but it's the ambience of the room in which the HomePod is placed, not the one captured in the recording. That's contrary to the primary objective of a reproduction system, which is to as accurately as possible reproduce the sound of the recording without adding or removing anything.

    That doesn't mean it can't be fun to listen to -- just look at how successful Bose was with its Direct/Reflecting designs -- it just isn't consistent with an "audiophile" approach or experience, that's all.

    Consider a HomePod placed against a wall (the most likely positioning).

    What if the front 3 tweeters are used to create the direct sound that covers the majority of the listening area in front of the speaker (hence why moving around the room doesn't change the sound much) and 2 tweeters on each side at the rear are using beamforming to bounce the ambient sound off the walls to the left & right for a wider soundstage?

    I don't think it has to be one or the other.
    I've gone over it again and again in my head (which I acknowledge is seriously flawed!) and I keep coming back to the same conclusion...

    Put aside *how* the sound is directed. What do you HEAR? If the answer is "It sounds the same everywhere," it must mean directional cues aren't very pronounced. If there was any appreciable distinction between left and right, it would not be possible to hear the same thing on one side of the room as the other. It doesn't matter how the HomePod is delivering the signal if the result is not apparent to the listener.

    Maybe bouncing the difference signals off the walls the way you describe results in a wash of non-directional room reflections. This could create a sense of "space" but it's the sound of the listening room, not the recording. In addition to being an affront to the concept of accurate reproduction ( :D ), it means the listener still isn't getting any directional cues, just some scattered mush. It might sound more interesting than just a direct-radiating single speaker, but it's not creating any kind of soundstage. It isn't even pseudo-stereo -- it's mono with reverb.

    Another possibility is that the people who are reporting "it sounds the same everywhere in the room" are not particularly careful or critical listeners. Maybe when they say "it sounds the same" what they really mean is "I don't hear any glaringly obvious changes when I move around."

    If it seems like I'm criticizing the HomePod for not being a studio-grade device, I'm not. It's a really positive step in the market segment it occupies. I just think it's important to be accurate in our descriptions and realistic in our expectations. Inaccurate and/or exaggerated claims cause subsequent buyers to be disappointed, and diminish the perceived value of products that really DO deliver high-end results.
    muthuk_vanalingam
  • Reply 47 of 123
    tmaytmay Posts: 3,832member

    nht said:

    foggyhill said: 
    It's not mono either
    This is one of several claims Apple makes about the HomePod that make me scratch my head a little...

    Apple says the HomePod beams "direct" sound to the middle of the room and "ambient" sound to the left and right. But then Apple ALSO claims that the HomePod sounds the same no matter where you are in the room. Those are conflicting statements.

    Since people are saying the sound of the HomePod does remain very consistent as one moves around the room, the logical conclusion is that there can't be much in the way of directional cues coming out of it. If there were, listeners would perceive a change in the sound as they moved around.

    If that assessment is correct, it means that despite what Apple's marketing claims, it's really essentially a mono device.
    In the same way that a violin is essentially a mono device...
    I'm pretty sure you meant that as a funny, but just in case you didn't, my response would be "Yeah, kinda, except not."

    The sense of "ambience" we hear from the violin is the result of reflections off the surfaces in the room. Those reflections are the sound of the room though, not the instrument per se. In the case of music reproduction, the goal is to hear the reflections from the room in which the recording was made, not the listening space.

    Apple's claim of beam forming is actually a reasonable alternative to a spaced pair of speakers. What I'm having trouble with is reconciling that with Apple's other claim that the HomePod sounds the same no matter where you are in the room. Both cannot be true. Either the HomePod creates directional signals to separate difference and sum components, which would result in it sounding different in the middle of the room than it does to one side, *OR* the sound doesn't change as you move around, meaning it's essentially omnidirectional and thus does NOT provide any kind of meaningful directional cues. It can't be both.

    Since those who have shared their experience with the HomePod have told us that the sound remains very consistent as they move around the listening space, the conclusion is that there isn't much of a directional component to its output, meaning it's essentially mono. One may still experience a sense of space and ambience, but it's the ambience of the room in which the HomePod is placed, not the one captured in the recording. That's contrary to the primary objective of a reproduction system, which is to as accurately as possible reproduce the sound of the recording without adding or removing anything.

    That doesn't mean it can't be fun to listen to -- just look at how successful Bose was with its Direct/Reflecting designs -- it just isn't consistent with an "audiophile" approach or experience, that's all.

    Consider a HomePod placed against a wall (the most likely positioning).

    What if the front 3 tweeters are used to create the direct sound that covers the majority of the listening area in front of the speaker (hence why moving around the room doesn't change the sound much) and 2 tweeters on each side at the rear are using beamforming to bounce the ambient sound off the walls to the left & right for a wider soundstage?

    I don't think it has to be one or the other.
    I would assume that there is some beam forming with the two tweeters that are 51 degree off axis on either side in the front. Indeed, it's possible to bias the set up with up to a 26 degree rotation; not sure what will happen that would make any difference. Beamforming seems to be necessary for the two tweeters at the sides and the two at the back that are at plus and minus 26 degree to the wall.

    I'm assuming a symmetric rectangular room with the HomePod on a centerline orthogonal to a wall. Odd configurations will require more asymmetry in the beam forming. 

    I don't have an expertise in speakers; I'm just thinking of the most common case for setup.
  • Reply 48 of 123
    dinoone said:
    HomePod Mono or Stereo?

    Just play this song on your HomePod, Journey Of The Sorcerer by Eagles.

    At the very beginning the banjo reverb is growing from right to center, then from left to center.
    It's easy to tell if you listen to it on a pair of earplugs or headphones.
    Now if the same effect (at the very beginning the banjo reverb is growing from right to center, then from left to center) is reproduced also by the HomePod in your room, then it is stereo indeed, otherwise, as I expect, it is mono.
    I'm sure there are many more famous songs that could help to achieve the same distinction.

    https://open.spotify.com/track/3OcBH9Vzd1UwJkQd3r1dVG?si=CwRmT4lwTK2uline5z03HQ
    I am playing this (from iTunes Store) on 2 homePods:

    1. first, as  a roll-your-own stereo as described in a prior post.  You get separation and there is a sweet spot
    2. second, as just 2 homePods playing the song.  You get separation everywhere -- no sweet spot.
    tmay
  • Reply 49 of 123
    dinoone said:
    One of the ways to rate a speaker is the relevance of its "positioning".

    Some active speakers will give their best only in a limited number of positionings, only in particular rooms, and only with some kinds of music.
    Some other active speakers would perform well in these variable situations more often than others.

    Now one of the distinctive selling points of the Apple HomePod is its capacity to "hear" the surrounding space and adapt the sound so that the final result for the listener is flat (good).

    That means that if you test HomePods in such an artificial setup as an anechoic chamber, when the HomePod tries to deploy its "listening" function to adapt its sound it will result as if it was positioned in the open air without any reflective surface, near or far, of any kind. But that would not reveal or test its distinctive feature. That doesn't mean that the HomePod is at fault: actually the test is at fault!

    An anechoic chamber would prevent the full operation of such key feature of the HomePod: the mitigation of the "positioning" issue that taints so many traditional speakers. 

    To test such key feature, one should test HomePods against competitors, incl. those much more expensive, in, say, 10 or 20 different rooms (sizes and furniture setups) and 10 or 20 different positioning in each room. 

    I.m.o. the other speakers would perform well only in some rooms/setup and especially in few positionings in these rooms, while fail in most of the others. The very expensive ones might perform well more often than the relatively inexpensive ones.
    Instead, still i.m.o., the HomePod is expected to perform well in most of the rooms, in most of the furniture setups, and in a large variety of positioning in all of the rooms. In this way the HomePod is going to challenge those very expensive speakers that perform well in many situations.

    This is the kind of comparative test that we need in order to give justice to such unique HomePod feature. HomePods are going to sound very good in any circumstance, including any end user's unique room/positioning, more likely than other speakers. 

    Therefore the real-life end-user performance of the HomePod is going to be statistically much better than that of the direct competition.
    You're absolutely right about everything except the part of the anechoic chamber somehow "hurting" the measurements of the HomePod.

    The reason the HomePod "listens" is to correct room anomalies. The anechoic chamber doesn't HAVE any anomalies so there's nothing for the HomePod to "fix." It is, like you say, the equivalent of testing the HomePod 100 feet in the air over an open field.

    I suppose it's possible the HomePod NEEDS room reflections to sound good, but it would be disappointing if it does. The concept of superimposing the sound of a random room over the sound of the recording is philosophically icky. It's not a problem if listeners like that effect, but it's definitely contrary to any notion of "audiophile" performance.
    edited February 2018 muthuk_vanalingam
  • Reply 50 of 123

    nht said:

    foggyhill said: 
    It's not mono either
    This is one of several claims Apple makes about the HomePod that make me scratch my head a little...

    Apple says the HomePod beams "direct" sound to the middle of the room and "ambient" sound to the left and right. But then Apple ALSO claims that the HomePod sounds the same no matter where you are in the room. Those are conflicting statements.

    Since people are saying the sound of the HomePod does remain very consistent as one moves around the room, the logical conclusion is that there can't be much in the way of directional cues coming out of it. If there were, listeners would perceive a change in the sound as they moved around.

    If that assessment is correct, it means that despite what Apple's marketing claims, it's really essentially a mono device.
    In the same way that a violin is essentially a mono device...
    I'm pretty sure you meant that as a funny, but just in case you didn't, my response would be "Yeah, kinda, except not."

    The sense of "ambience" we hear from the violin is the result of reflections off the surfaces in the room. Those reflections are the sound of the room though, not the instrument per se. In the case of music reproduction, the goal is to hear the reflections from the room in which the recording was made, not the listening space.

    Apple's claim of beam forming is actually a reasonable alternative to a spaced pair of speakers. What I'm having trouble with is reconciling that with Apple's other claim that the HomePod sounds the same no matter where you are in the room. Both cannot be true. Either the HomePod creates directional signals to separate difference and sum components, which would result in it sounding different in the middle of the room than it does to one side, *OR* the sound doesn't change as you move around, meaning it's essentially omnidirectional and thus does NOT provide any kind of meaningful directional cues. It can't be both.

    Since those who have shared their experience with the HomePod have told us that the sound remains very consistent as they move around the listening space, the conclusion is that there isn't much of a directional component to its output, meaning it's essentially mono. One may still experience a sense of space and ambience, but it's the ambience of the room in which the HomePod is placed, not the one captured in the recording. That's contrary to the primary objective of a reproduction system, which is to as accurately as possible reproduce the sound of the recording without adding or removing anything.

    That doesn't mean it can't be fun to listen to -- just look at how successful Bose was with its Direct/Reflecting designs -- it just isn't consistent with an "audiophile" approach or experience, that's all.

    Consider a HomePod placed against a wall (the most likely positioning).

    What if the front 3 tweeters are used to create the direct sound that covers the majority of the listening area in front of the speaker (hence why moving around the room doesn't change the sound much) and 2 tweeters on each side at the rear are using beamforming to bounce the ambient sound off the walls to the left & right for a wider soundstage?

    I don't think it has to be one or the other.
    I've gone over it again and again in my head (which I acknowledge is seriously flawed!) and I keep coming back to the same conclusion...

    Put aside *how* the sound is directed. What do you HEAR? If the answer is "It sounds the same everywhere," it must mean directional cues aren't very pronounced. If there was any appreciable distinction between left and right, it would not be possible to hear the same thing on one side of the room as the other. It doesn't matter how the HomePod is delivering the signal if the result is not apparent to the listener.

    Maybe bouncing the difference signals off the walls the way you describe results in a wash of non-directional room reflections. This could create a sense of "space" but it's the sound of the listening room, not the recording. In addition to being an affront to the concept of accurate reproduction ( :D ), it means the listener still isn't getting any directional cues, just some scattered mush. It might sound more interesting than just a direct-radiating single speaker, but it's not creating any kind of soundstage. It isn't even pseudo-stereo -- it's mono with reverb.

    Another possibility is that the people who are reporting "it sounds the same everywhere in the room" are not particularly careful or critical listeners. Maybe when they say "it sounds the same" what they really mean is "I don't hear any glaringly obvious changes when I move around."

    If it seems like I'm criticizing the HomePod for not being a studio-grade device, I'm not. It's a really positive step in the market segment it occupies. I just think it's important to be accurate in our descriptions and realistic in our expectations. Inaccurate and/or exaggerated claims cause subsequent buyers to be disappointed, and diminish the perceived value of products that really DO deliver high-end results.

    Do you have access to a homePod on which you can test your conclusion?
  • Reply 51 of 123
    zoetmb said:
    That's not a flat frequency response.  Back in the hi-fi days, the measurement (with pink noise input) had to be from 20 to 20,000 Hz +/- 3db with no more than X% distortion (preferably 1% or under).

    But even if it was flat, a flat frequency response won't sound very good to most people.  That's why hi-fi equipment had tone controls and loudness compensation (which increased the bass and treble at low volume levels according to the Fletcher-Munson curve and which gradually had less effect as the volume was raised).

    I happened to be at an Apple store yesterday to listen to the HP.   IMO, it has an over-emphasized bass.   On new, digitally recorded material, it sounded pretty good.  But on tracks I tried from the 1960's-1970's (which I was pleasantly surprised were in the Apple Music library), it sounded kind of dull.   And while the Apple person was a bit confused and had to look things up, a single unit is NOT stereo.   At first, he claimed it was because "it contains lots of speakers", but when I told him that the number of speakers was not relevant and that it sounded mono to me, he did some more research and admitted it would take two for stereo and that stereo is not currently supported even with two units, but it would be with some future update.   

    Most of the reviews said the sound was great but that Siri sucked.  But in my tests and even in the noisy store environment, Siri did really well - it never played the wrong track or artist that I called out and it properly heard me say "louder" and "softer" and "pause".  It was far more accurate than the Apple TV achieved in the same environment. 
    Not being two separate stereo speakers isn’t the same as being mono. The HP does channel separation including left and right, bouncing them off the walls in different directions. 

    Direct sound is beamed to the middle of the room, while ambient sound is diffused into left and right channels and bounced off the wall.”

    ...that is functionally different than mono.  
    tmay
  • Reply 52 of 123
    trydtryd Posts: 135member
    lorin schultz said:

    That's one of the reasons Bose was so hated in the audio production community. Amar convinced an entire generation of consumers that adding the sound of their living room to the recording was somehow a good thing.

    The purist will abhor devices that try to artificially "create" (as opposed to simply reproduce) any part of the sound. The pragmatist will recognize that perfect playback is impossible and choose whatever sounds closest or most enjoyable.
    Youare not adding the sound of your living room to the recording. You are realizing that the room plays an important role in the sound you hear and you try to use that to your advantage. I used to have a set of Sonab OA-116 loudspeakers (OA means OrthoAcoustic and means that the speaker uses the rooms acoustics as part of the experience), and the sense of being there was fantastic. You did not get pin-pointing of the different instruments, but you don't get that in a concert hall anyway. I was in the opera on Tuesday and directional information is very limited. You get front-back and right-left, but that is all. Artificial soundstaging, like panning of instruments, I can do without.
  • Reply 53 of 123

    foggyhill said: 
    It's not mono either
    This is one of several claims Apple makes about the HomePod that make me scratch my head a little...

    Apple says the HomePod beams "direct" sound to the middle of the room and "ambient" sound to the left and right. But then Apple ALSO claims that the HomePod sounds the same no matter where you are in the room. Those are conflicting statements.

    Since people are saying the sound of the HomePod does remain very consistent as one moves around the room, the logical conclusion is that there can't be much in the way of directional cues coming out of it. If there were, listeners would perceive a change in the sound as they moved around.

    If that assessment is correct, it means that despite what Apple's marketing claims, it's really essentially a mono device.
    If you have a small shelf system with speakers two or three feet away, and you walk 15-20 feet away from it, does that system become mono? the speakers aren’t on your left and right anymore so you’re not getting a stereophonic effect to your ears, but it’s still not mono. 

    you cant just redefine words because you feel like it. 
  • Reply 54 of 123

    The reason the HomePod "listens" is to correct room anomalies. The anechoic chamber doesn't HAVE any anomalies so there's nothing for the HomePod to "fix." It is, like you say, the equivalent of testing the HomePod 100 feet in the air over an open field.

    Somewhere, in all the info flow about the homePod, I've read that with Airplay 2 -- multiple homePods can be setup (in a room) to listen to each other and adjust to each other -- as well as each adjusting to room anomalies.
  • Reply 55 of 123

    larrya said:
    sully54 said:
    Can we all just give this a rest and enjoy the HomePod? Or don’t. You do you. But these reviews are getting quite ridiculous. It’s like people are tripping over each other trying to get the farthest decimal place. 

    when I’m listening to music, graphs and numbers are honestly the last things on my mind. 
    We are in rationalization mode now to explain the less than unanimous reviews. For many, admiration of Apple and its products is now a religion.  If you like it, buy and enjoy it. But learn to accept that others, who are just as intelligent, may not come to the same conclusions. 
    And the chastizing of Apple and its products has long been an equal religion full of irrational zealots. 
  • Reply 56 of 123

    nht said:

    foggyhill said: 
    It's not mono either
    This is one of several claims Apple makes about the HomePod that make me scratch my head a little...

    Apple says the HomePod beams "direct" sound to the middle of the room and "ambient" sound to the left and right. But then Apple ALSO claims that the HomePod sounds the same no matter where you are in the room. Those are conflicting statements.

    Since people are saying the sound of the HomePod does remain very consistent as one moves around the room, the logical conclusion is that there can't be much in the way of directional cues coming out of it. If there were, listeners would perceive a change in the sound as they moved around.

    If that assessment is correct, it means that despite what Apple's marketing claims, it's really essentially a mono device.
    In the same way that a violin is essentially a mono device...
    I'm pretty sure you meant that as a funny, but just in case you didn't, my response would be "Yeah, kinda, except not."

    The sense of "ambience" we hear from the violin is the result of reflections off the surfaces in the room. Those reflections are the sound of the room though, not the instrument per se. In the case of music reproduction, the goal is to hear the reflections from the room in which the recording was made, not the listening space.

    Apple's claim of beam forming is actually a reasonable alternative to a spaced pair of speakers. What I'm having trouble with is reconciling that with Apple's other claim that the HomePod sounds the same no matter where you are in the room. Both cannot be true. Either the HomePod creates directional signals to separate difference and sum components, which would result in it sounding different in the middle of the room than it does to one side, *OR* the sound doesn't change as you move around, meaning it's essentially omnidirectional and thus does NOT provide any kind of meaningful directional cues. It can't be both.

    Since those who have shared their experience with the HomePod have told us that the sound remains very consistent as they move around the listening space, the conclusion is that there isn't much of a directional component to its output, meaning it's essentially mono. One may still experience a sense of space and ambience, but it's the ambience of the room in which the HomePod is placed, not the one captured in the recording. That's contrary to the primary objective of a reproduction system, which is to as accurately as possible reproduce the sound of the recording without adding or removing anything.

    That doesn't mean it can't be fun to listen to -- just look at how successful Bose was with its Direct/Reflecting designs -- it just isn't consistent with an "audiophile" approach or experience, that's all.

    Consider a HomePod placed against a wall (the most likely positioning).

    What if the front 3 tweeters are used to create the direct sound that covers the majority of the listening area in front of the speaker (hence why moving around the room doesn't change the sound much) and 2 tweeters on each side at the rear are using beamforming to bounce the ambient sound off the walls to the left & right for a wider soundstage?

    I don't think it has to be one or the other.
    I've gone over it again and again in my head (which I acknowledge is seriously flawed!) and I keep coming back to the same conclusion...

    Put aside *how* the sound is directed. What do you HEAR? If the answer is "It sounds the same everywhere," it must mean directional cues aren't very pronounced. If there was any appreciable distinction between left and right, it would not be possible to hear the same thing on one side of the room as the other. It doesn't matter how the HomePod is delivering the signal if the result is not apparent to the listener.

    Maybe bouncing the difference signals off the walls the way you describe results in a wash of non-directional room reflections. This could create a sense of "space" but it's the sound of the listening room, not the recording. In addition to being an affront to the concept of accurate reproduction ( :D ), it means the listener still isn't getting any directional cues, just some scattered mush. It might sound more interesting than just a direct-radiating single speaker, but it's not creating any kind of soundstage. It isn't even pseudo-stereo -- it's mono with reverb.

    Another possibility is that the people who are reporting "it sounds the same everywhere in the room" are not particularly careful or critical listeners. Maybe when they say "it sounds the same" what they really mean is "I don't hear any glaringly obvious changes when I move around."

    If it seems like I'm criticizing the HomePod for not being a studio-grade device, I'm not. It's a really positive step in the market segment it occupies. I just think it's important to be accurate in our descriptions and realistic in our expectations. Inaccurate and/or exaggerated claims cause subsequent buyers to be disappointed, and diminish the perceived value of products that really DO deliver high-end results.

    Do you have access to a homePod on which you can test your conclusion?
    Now what fun would THAT be? The enjoyment in this derives from the mental exercise. Actually finding an answer would bring the fun to an end!

    But seriously, Apple hasn't released it here yet. Even if it was available now, our present financial priorities preclude even a relatively modest purchase of something we don't need just to satisfy my curiosity.

    Besides, that's what we have YOU for! :)

    Speaking of you, you haven't responded to my post asking you to please share your source(s) of information about HomePod-specific metadata for sound shaping/equalization. Is there a reason you don't want to?
    muthuk_vanalingam
  • Reply 57 of 123
    nht said:
    zoetmb said:
    That's not a flat frequency response.  Back in the hi-fi days, the measurement (with pink noise input) had to be from 20 to 20,000 Hz +/- 3db with no more than X% distortion (preferably 1% or under).

    But even if it was flat, a flat frequency response won't sound very good to most people.  That's why hi-fi equipment had tone controls and loudness compensation (which increased the bass and treble at low volume levels according to the Fletcher-Munson curve and which gradually had less effect as the volume was raised).

    I happened to be at an Apple store yesterday to listen to the HP.   IMO, it has an over-emphasized bass.   On new, digitally recorded material, it sounded pretty good.  But on tracks I tried from the 1960's-1970's (which I was pleasantly surprised were in the Apple Music library), it sounded kind of dull.   And while the Apple person was a bit confused and had to look things up, a single unit is NOT stereo.   At first, he claimed it was because "it contains lots of speakers", but when I told him that the number of speakers was not relevant and that it sounded mono to me, he did some more research and admitted it would take two for stereo and that stereo is not currently supported even with two units, but it would be with some future update.   

    Most of the reviews said the sound was great but that Siri sucked.  But in my tests and even in the noisy store environment, Siri did really well - it never played the wrong track or artist that I called out and it properly heard me say "louder" and "softer" and "pause".  It was far more accurate than the Apple TV achieved in the same environment. 

    Mmm... I certainly hope that Apple persons don't become typical Sound Store Lizards!
    Your get stereo from the tweeters (i.e. treble only) but not the mids or bass.  So it will sound mono for a lot of material.  But if you own a home pod try going into GarageBand and create a track with separate left and right and then listen.

    The kid was right and zoetmb wrong.  The number of speakers do matter, home pod supports stereo but its a tiny little thing so not a whole lot of separation possible especially if it doesn't have a surface to bounce the left quadrant and right quadrant tweeters off of.  

    This also explains why Apple says it sounds the same around the room (the one "woofer" handling mids and bass) and but has ambient directional cues left and right.  It can only widen it's treble sound stage via the tweeters.  So there will be some directional cues but not enough for most material to matter beyond "ambient".

    Note that "true" stereo will not be supported even when Apple releases support for two units.  Apple says "stereo-like" in it's product literature because you won't designate one home pod as left and the other right.  Instead it will use beam forming from both to try to create the "immersive" sound stage.  
    Nope, apple doesn’t use that term “stereo-like” on their product page at all. It says:

    ”Create stereo sound with a second HomePod.

    Put another HomePod in the same room and they automatically detect and balance each other. With advanced beamforming capabilities, a HomePod pair is able to create a wider, more immersive soundstage than a traditional stereo pair“

    https://www.apple.com/homepod/


    ...balance here refers to left and right.
    tmay
  • Reply 58 of 123
    nhtnht Posts: 4,494member

    nht said:

    foggyhill said: 
    It's not mono either
    This is one of several claims Apple makes about the HomePod that make me scratch my head a little...

    Apple says the HomePod beams "direct" sound to the middle of the room and "ambient" sound to the left and right. But then Apple ALSO claims that the HomePod sounds the same no matter where you are in the room. Those are conflicting statements.

    Since people are saying the sound of the HomePod does remain very consistent as one moves around the room, the logical conclusion is that there can't be much in the way of directional cues coming out of it. If there were, listeners would perceive a change in the sound as they moved around.

    If that assessment is correct, it means that despite what Apple's marketing claims, it's really essentially a mono device.
    In the same way that a violin is essentially a mono device...
    I'm pretty sure you meant that as a funny, but just in case you didn't, my response would be "Yeah, kinda, except not."

    The sense of "ambience" we hear from the violin is the result of reflections off the surfaces in the room. Those reflections are the sound of the room though, not the instrument per se. In the case of music reproduction, the goal is to hear the reflections from the room in which the recording was made, not the listening space.
    You will not accurately recreate the reflections from the room the recording was made and nothing from the listening space unless you use headphones.

    Apple's claim of beam forming is actually a reasonable alternative to a spaced pair of speakers. What I'm having trouble with is reconciling that with Apple's other claim that the HomePod sounds the same no matter where you are in the room. Both cannot be true. Either the HomePod creates directional signals to separate difference and sum components, which would result in it sounding different in the middle of the room than it does to one side, *OR* the sound doesn't change as you move around, meaning it's essentially omnidirectional and thus does NOT provide any kind of meaningful directional cues. It can't be both.
    I explained why they can make that claim.  The directional cues are limited to what the tweeters can provide.  Most material will be omni with some directional cues.  Whether it is "meaningful" depends on the source material.
    Since those who have shared their experience with the HomePod have told us that the sound remains very consistent as they move around the listening space, the conclusion is that there isn't much of a directional component to its output, meaning it's essentially mono. One may still experience a sense of space and ambience, but it's the ambience of the room in which the HomePod is placed, not the one captured in the recording. That's contrary to the primary objective of a reproduction system, which is to as accurately as possible reproduce the sound of the recording without adding or removing anything.
    That doesn't mean it can't be fun to listen to -- just look at how successful Bose was with its Direct/Reflecting designs -- it just isn't consistent with an "audiophile" approach or experience, that's all.
    You mean the typical "audiophile" approach of buying very expensive gear and not building a listening room?  This is like buying expensive  home theater gear without a dedicated light and audio controlled HT room.

    The audiophile approach has been consistently stupid over the years.  I knew folks with $30,000+ speakers in dens/living rooms with craptastic acoustics.  You can only just shrug your shoulders.
    edited February 2018 tmay
  • Reply 59 of 123
    nhtnht Posts: 4,494member
    nht said:
    zoetmb said:
    That's not a flat frequency response.  Back in the hi-fi days, the measurement (with pink noise input) had to be from 20 to 20,000 Hz +/- 3db with no more than X% distortion (preferably 1% or under).

    But even if it was flat, a flat frequency response won't sound very good to most people.  That's why hi-fi equipment had tone controls and loudness compensation (which increased the bass and treble at low volume levels according to the Fletcher-Munson curve and which gradually had less effect as the volume was raised).

    I happened to be at an Apple store yesterday to listen to the HP.   IMO, it has an over-emphasized bass.   On new, digitally recorded material, it sounded pretty good.  But on tracks I tried from the 1960's-1970's (which I was pleasantly surprised were in the Apple Music library), it sounded kind of dull.   And while the Apple person was a bit confused and had to look things up, a single unit is NOT stereo.   At first, he claimed it was because "it contains lots of speakers", but when I told him that the number of speakers was not relevant and that it sounded mono to me, he did some more research and admitted it would take two for stereo and that stereo is not currently supported even with two units, but it would be with some future update.   

    Most of the reviews said the sound was great but that Siri sucked.  But in my tests and even in the noisy store environment, Siri did really well - it never played the wrong track or artist that I called out and it properly heard me say "louder" and "softer" and "pause".  It was far more accurate than the Apple TV achieved in the same environment. 

    Mmm... I certainly hope that Apple persons don't become typical Sound Store Lizards!
    Your get stereo from the tweeters (i.e. treble only) but not the mids or bass.  So it will sound mono for a lot of material.  But if you own a home pod try going into GarageBand and create a track with separate left and right and then listen.

    The kid was right and zoetmb wrong.  The number of speakers do matter, home pod supports stereo but its a tiny little thing so not a whole lot of separation possible especially if it doesn't have a surface to bounce the left quadrant and right quadrant tweeters off of.  

    This also explains why Apple says it sounds the same around the room (the one "woofer" handling mids and bass) and but has ambient directional cues left and right.  It can only widen it's treble sound stage via the tweeters.  So there will be some directional cues but not enough for most material to matter beyond "ambient".

    Note that "true" stereo will not be supported even when Apple releases support for two units.  Apple says "stereo-like" in it's product literature because you won't designate one home pod as left and the other right.  Instead it will use beam forming from both to try to create the "immersive" sound stage.  
    Nope, apple doesn’t use that term “stereo-like” on their product page at all. It says:

    ”Create stereo sound with a second HomePod.

    Put another HomePod in the same room and they automatically detect and balance each other. With advanced beamforming capabilities, a HomePod pair is able to create a wider, more immersive soundstage than a traditional stereo pair“

    https://www.apple.com/homepod/


    ...balance here refers to left and right.
    Ooops.  Now I gotta go find where I read that.  But my recollection was reading that Apple stated that you won't be able to designate left and right home pods.
  • Reply 60 of 123
    Exactly what is a typical living room? A huge  LR in a McMansion with vaulted ceilings and in a smallish Manhattan apartment with a flat ceiling are going to be very different. Blinds vs curtains, leather vs cloth upholstery, carpet vs hard surface floor, acoustic ceiling vs smooth, open vs door exits, the presence of bookshelves, the shape of the room, textured wallpaper vs painted and many other factors influence how sound behaves. Nobody I can think of has a living room that similar to anyone else’s.

    I know Apple has invested a lot in DSP, but it can only do so much.

    In the Apple Store (Saddle Creek) that speaker sounds like a cheap boom box. Largish room, hard floors, high ceilings, lots of glass. no soft furniture or curtains. That should normally make a speaker set flat to sound bright. It didn’t sound bright. It sounded like the bass was turned up all the way up trying to compensate for the room and failed unless you like the sound of a boom box.

    Also, didn’t see the HomePod getting too much attention.
Sign In or Register to comment.