Can Loudspeakers Accurately Reproduce The Sound Of Real Instruments...and Do You Care?

[Hifi]

I think that the quality of recording is more important than the quality of the replay equipment. A bad recording will be a bad recording regardless of what you use to listen to it. A top quality recording will still shine on a modest but decent set up.

A bad recording should sound exactly like a bad recording on a highly resolving two channel audio system

 

[MRC01]

I'm not suggesting that speakers are capable of lower distortion than microphones (they aren't) ...

With Room EQ Wizard, I measure lower total distortion using Rode NT1A mics, than I do with the UMIK-1. All else the same. I just assumed (since I don't really know, haven't seen mic distortion measurements) the UMIK-1 had flatter FR but higher distortion, and the speakers are clean enough to tell the difference.

 

[andreasmaaan]

With Room EQ Wizard, I measure lower total distortion using Rode NT1A mics, than I do with the UMIK-1. All else the same. I just assumed (since I don't really know, haven't seen mic distortion measurements) the UMIK-1 had flatter FR but higher distortion, and the speakers are clean enough to tell the difference.

That seems plausible to me. MiniDSP doesn't publish distortion specs for the UMIK-1, but I wouldn't be at all surprised if a good Rode mic has lower distortion.

 

[JJB70]

A bad recording should sound exactly like a bad recording on a highly resolving two channel audio system

Very true, but recorded music is about capturing the performance. Therefore the recording is the most critical part of the audio chain (well, excepting the music and performance itself). Audio playback equipment is a tool, it exists to serve music. I would much rather listen to great music well recorded on a competent system than to exquisitely replayed junk.

 

[MRC01]

I've seen (more accurately, heard!) all combinations:
A) Recordings that sound fine on bad equipment (car stereo, clock radio) but sound bad on good equipment
B) Recordings that sound bad on bad equipment but sound good on good equipment
C) Recordings that sound bad on all equipment
D) Recordings that sound good on all equipment

 

[JJB70]

I've seen all combinations:
A) Recordings that sound fine on bad equipment (car stereo, clock radio) but sound bad on good equipment
B) Recordings that sound bad on bad equipment but sound good on good equipment
C) Recordings that sound bad on all equipment
D) Recordings that sound good on all equipment

I think that the main issue in recorded music these days is that music is being mastered to play well on small BT speakers and car audio systems. Hence the compression, I think we all hoped that the loudness war was finished and standards would improve, but now it is more about optimising recordings for the equipment which will be used by the vast majority of listeners. And much as I hate it I really can't blame the record labels for that, however it'd be nice if they offered versions for hi-fi playback.

 

[MRC01]

I've heard this view before but it's belied by evidence: "special release" high bit rate recordings are often more dynamically compressed than the originals. If any version of a recording was going to be played on good equipment, this would be the one. Yet even here they squash the life out of the music. Then speciously claim it is "high definition" simply because it uses more bits, even though it's so compressed it could be encoded in 8-bit!
Dynamic compression is only 1 of several ways to ruin a recording. They also EQ to the point it sounds painfully artificial on a good system having linear FR. The end result is just garbage.
Plenty of blame to spread around: the musicians, the studios, and reviewers who are too ignorant or timid to call this spade a spade.
Fortunately, classical and some other genres are largely free of this insanity. It seems focused on popular genres.
Like you, I'd love to see certified "audiophile" versions with minimal post-processing. Physical discs and high bit rate downloads would be ideal for this, since the general public doesn't buy them.

 

[andreasmaaan]

I've heard this view before but it's belied by evidence: "special release" high bit rate recordings are often more dynamically compressed than the originals. If any version of a recording was going to be played on good equipment, this would be the one. Yet even here they squash the life out of the music. Then speciously claim it is "high definition" simply because it uses more bits, even though it's so compressed it could be encoded in 8-bit!
Dynamic compression is only 1 of several ways to ruin a recording. They also EQ to the point it sounds painfully artificial on a good system having linear FR. The end result is just garbage.
Plenty of blame to spread around: the musicians, the studios, and reviewers who are too ignorant or timid to call this spade a spade.
Fortunately, classical and some other genres are largely free of this insanity. It seems focused on popular genres.
Like you, I'd love to see certified "audiophile" versions with minimal post-processing. Physical discs and high bit rate downloads would be ideal for this, since the general public doesn't buy them.

I totally agree with this, except I think the purpose of these hi-res releases is not to offer a higher fidelity alternative, but rather to cater to an audiophile market that is convinced that more bits and more samples = better sound. If of these labels were really interested in releasing genuinely higher fidelity versions of the recordings (putting aside of course those few labels that actually are), they would be doing it in 16/44.1, or at modestly high-res rates (24/48 or 24/88.2, for example).

 

[Blumlein 88]

I totally agree with this, except I think the purpose of these hi-res releases is not to offer a higher fidelity alternative, but rather to cater to an audiophile market that is convinced that more bits and more samples = better sound. If of these labels were really interested in releasing genuinely higher fidelity versions of the recordings (putting aside of course those few labels that actually are), they would be doing it in 16/44.1, or at modestly high-res rates (24/48 or 24/88.2, for example).

I agree. It is a shame they go out of their way to make them worse at the same time. I even believed at one point they did this to make sure it sounded different believing customers would be satisfied if they heard a difference. And they would assume it was a positive difference. No longer even think that. I also no longer think it is about targeting users who listen with earbuds or in automobiles.

It is very simple. People who do this work now genuinely believe this slammed version is superior. Given a chance to work on such re-masterings they believe they are slamming it in ways not available way back when on the original issue and they've made it better. It is insane and I don't know what you can do about it. I've had the conversation with a couple such people, and shown examples and nothing registers. Louder is better period. If you are a paying customer who insists otherwise, they'll reluctantly give you what you want. They think you've made a foolish decision.

 

[Blumlein 88]

I've seen (more accurately, heard!) all combinations:
A) Recordings that sound fine on bad equipment (car stereo, clock radio) but sound bad on good equipment
B) Recordings that sound bad on bad equipment but sound good on good equipment
C) Recordings that sound bad on all equipment
D) Recordings that sound good on all equipment

Category A is because bad equipment doesn't portray everything on the recording some of which is just bad. Though there are a few sub-categories of how this can work.

Category B is usually your genuine top notch recordings.

I think category C is the definition of a definitely BAD recording.

I also think category D is usually something that isn't a real highly resolved recording. It can be pleasant and enjoyable, and might be the work of a good recording craftsman who makes use of the limited material they have to work with.

 

[MattHooper]

I recently put my little tiny Spendor S 3/5 monitors in to my main system. (Normally they are used as "tv speakers" for our flatscreen set up).

Because I like mixing up my speakers, I do this every once in a while with the Spendors and, like every time before, I'm just knocked out by the sound.

Boy does the timbral quality of instruments every sound "right" (to me) through these things. My pal has yet another pair of expensive (over 20K) full range speakers in to review, which I'd been listening to. I'd take the tiny, cheap Spendors in a heartbeat over those big speakers, just because their tone sounds "right" to me and the big speakers sound impressive but "a bit off." (Too dark).

The most striking thing about the spendors is their way with vocals. These things are marvels with the human voice. Voices sound so human, so rounded, soft, timbrally bang on and "project" in space with so much gestalt of real voices. I've played a wonderful acapella recording of the Carpenters (Without A Song) doing harmonies. With eyes closed, it sounds more like real people singing to me than anything I've experienced, save maybe the MBL speakers. In track after track from different artists, vocals, either distant harmonies or up front vocals, all sound more human, less artificial, and so now I'm digging out all my vocal stuff.

I wonder what specifically in this old BBC design makes them sound so particularly "human" with voices, whether it's some combination of the somewhat lively sealed-cabinet design, the particular driver materials, or the particular voicing, or what. But they are convincing on voice in a way the otherwise highly-engineered speakers I've heard are not.

 

[TG1]

I basically agree with you, but just not for the reason that I think you're suggesting In terms of causing air pressure fluctuations that are the same as those that would be produced by a bow being drawn across a cello, a loudspeaker can actually get extremely close. By that I mean that it can produce something very close to the same resonances and harmonics that a cello can, so close that any differences between what the mic recorded and what the speaker reproduces may well be inaudible, or extremely close to inaudible.

The bigger issue is that the signal the speaker receives is either:

1. a recording of a cello which contains both direct and reflected sounds from the space in which the cello was played, or
2. a close-mic'd or anechoic recording of a cello that contains (more or less) only the direct sound from the cello

If it is (1), upon playback the loudspeaker is forced to be the source of all direct sound and all reflections, but instead of these coming from different locations around the listening room as they would if a real cello were there, all these sounds originate in the listening room from the location of the speaker.

If it is (2), there's a theoretical problem, because there is no single correct location in an anechoic chamber or with a close mic to record the cello (you can't put a mic at the same location from which the cello produces sound). So you've captured only the sound of the cello as it was at one specific location.

Moreover, the loudspeaker has a polar response that is different from a cello. So when it replays the recording taken by method (2), it will send it out into the listening room with a different polar pattern than the cello had IRL.

This is why, generally, the best a loudspeaker can hope to do is to accurately reproduce what's on the recording - whatever that is - not the cello that was recorded. Notwithstanding this limitation, good loudspeakers can actually do this very well (i.e. produce near identical resonances, harmonics, etc. as contained in the recording of the cello or whatever instrument(s) happen to be present).

Well, I haven't been able to listen to set ups that much better than pretty good, and I happen to think that all that extra dosh is paying for something, especially after listening to what a very high end turntable could do with a record that was a bit shit on my player.
You're right, anyway, my reason was based far more on a subjective impression of hearing live and recorded music. It would be interesting to do an experiment to see whether a trained ear could always hear the difference.

 

[TG1]

I’m going to, at least, try to answer.....

The very same way I have heard live voices and those same voices played back on a system. I hear, in total, enough information that I know it’s the very same person. I have the same experience with acoustic guitar listening and then playback on a sound system. As a teenager my roommate played his drums as we listened to The Clash and Elvis Costello. The live drums sound just like playback drums. I noticed the biggest sound difference was his live Kik verse the recorded Kik drum...but both were real After my ears and brain processed. I think different artist personally tune their Kik to wide variations and recording of Kik is often handled differently from various recording studios. The quick and hard hit of the snare was spot on. In total the recording of drums sounds like drums in a decent Two Channel audio system.

So yes the source is different.... but with enough information your brain will say it’s real. . . And there is some science that supports.

You stated your LP of solo cello would sound different than a solo cello playing live in your room. It absolutely should sound different even if speakers were “approximating” to a extremely high degree. That is in a very large part to do with you are listening to a recorded cello in a likely brilliant space (church?) and playing back in , guessing, a room with four walls and average ceiling. Two very different sounds. Your brain should still process enough information to enjoy some authenticity. I think things get spooky real , in a Two Channel audio system, when the playback “Room” is fixed. For now I will use “fixed” loosely. If one can get playback room issues tamed it doesn’t require “big league “ equipment to make “magically real sounding” sound.

Again, you have the advantage over me there, because I haven't really fixed my room, just done what I can with isolation. It's encouraging that even a modest system can sound very real, because I am not going to chase high end equipment. I've read a bit about fixing the room and now I will read some more.

 

[andreasmaaan]

Well, I haven't been able to listen to set ups that much better than pretty good, and I happen to think that all that extra dosh is paying for something, especially after listening to what a very high end turntable could do with a record that was a bit shit on my player.

It's really a question of when diminishing returns kick in IMO, also keeping in mind that at all price levels there are excellent and poor examples, and I'm not convinced that in audio the general quality:rubbish ratio changes much as one moves up in price. But yeh, there are speakers in the mid-high price bracket in particular ($10-20K-ish) that are clearly better than anything under $5k - or at least, they are as good but also play lower and louder. I'm not convinced spending more than $10-20k adds any value whatsoever, and I actually think spending less but setting a system up well is likely to give far better results than buying expensive gear alone. In fact, I think huge amounts are wasted in this hobby on gear - especially electronics - that would be much better spent on room treatment and calibration.

Re: turntables, they are already at a major disadvantage compared to digital since the mechanics of vinyl playback mean it's impossible to produce a turntable that isn't outperformed by e.g. the DAC in a smartphone, but of course they can still sound very nice.

You're right, anyway, my reason was based far more on a subjective impression of hearing live and recorded music. It would be interesting to do an experiment to see whether a trained ear could always hear the difference.

Something like this?

 

[TG1]

I think that the main issue in recorded music these days is that music is being mastered to play well on small BT speakers and car audio systems. Hence the compression, I think we all hoped that the loudness war was finished and standards would improve, but now it is more about optimising recordings for the equipment which will be used by the vast majority of listeners. And much as I hate it I really can't blame the record labels for that, however it'd be nice if they offered versions for hi-fi playback.

I wonder if this is happening with new release vinyl. I had to chuck the re-release of Inner Visions in the trash, it was so bad. I've never heard the original on vinyl so could not compare, but I find it hard to believe it would have been that bad. Of course, as someone pointed out to me a while back, being older means my hearing will inevitably not be what it was, so you never know whether it is you or the recording.

 

[TG1]

It's really a question of when diminishing returns kick in IMO, also keeping in mind that at all price levels there are excellent and poor examples, and I'm not convinced that in audio the general quality:rubbish ratio changes much as one moves up in price. But yeh, there are speakers in the mid-high price bracket in particular ($10-20K-ish) that are clearly better than anything under $5k - or at least, they are as good but also play lower and louder. I'm not convinced spending more than $10-20k adds any value whatsoever, and I actually think spending less but setting a system up well is likely to give far better results than buying expensive gear alone. In fact, I think huge amounts are wasted in this hobby on gear - especially electronics - that would be much better spent on room treatment and calibration.

Re: turntables, they are already at a major disadvantage compared to digital since the mechanics of vinyl playback mean it's impossible to produce a turntable that isn't outperformed by e.g. the DAC in a smartphone, but of course they can still sound very nice.



Something like this?

Had no idea it was so problematic.

 

[TG1]

It's really a question of when diminishing returns kick in IMO, also keeping in mind that at all price levels there are excellent and poor examples, and I'm not convinced that in audio the general quality:rubbish ratio changes much as one moves up in price. But yeh, there are speakers in the mid-high price bracket in particular ($10-20K-ish) that are clearly better than anything under $5k - or at least, they are as good but also play lower and louder. I'm not convinced spending more than $10-20k adds any value whatsoever, and I actually think spending less but setting a system up well is likely to give far better results than buying expensive gear alone. In fact, I think huge amounts are wasted in this hobby on gear - especially electronics - that would be much better spent on room treatment and calibration.

Re: turntables, they are already at a major disadvantage compared to digital since the mechanics of vinyl playback mean it's impossible to produce a turntable that isn't outperformed by e.g. the DAC in a smartphone, but of course they can still sound very nice.



Something like this?

I know a record shop where you can get shot for saying stuff like that. I don't know about DAC, never used it. I have an £800 cd player (original rrp - I bought second hand) and a turntable that would probably retail for around the same now if it were made, and while the CD player is great, when I get the turntable set up properly the sound is better. However, so many things can screw with the signal it can be a real bind. I'm lucky in that I am not near any roads, so traffic vibration isn't an issue, and it's in a different room, so speaker vibration is minimised, but some hum has crept back in (no idea why, just has). Odd things make a difference. putting sandbags under the shelving unit the other components are in helped a bit.

 

[andreasmaaan]

Had no idea it was so problematic.

Vinyl? Given what's going on mechanically on the surface of a record at the microscopic level, I find it amazing it works as well as it does.

Imagine a tiny needle having to hover above a spinning target, moving back and forth through a groove less than a millimetre thick at up to 20,000 times per second - in two perpendicular planes simultaneously. The fact that it works at all I find mind-boggling

 

[andreasmaaan]

I know a record shop where you can get shot for saying stuff like that. I don't know about DAC, never used it. I have an £800 cd player and a turntable that would probably retail for around the same now if it were made, and while the CD player is great, when I get the turntable set up properly the sound is better. However, so many things can screw with the signal it can be a real bind. I'm lucky in that I am not near any roads, so traffic vibration isn't an issue, and it's in a different room, so speaker vibration is minimised, but some hum has crept back in (no idea why, just has). Odd things make a difference. putting sandbags under the shelving unit the other components are in helped a bit.

Yeh and actually I wasn't intending to imply that vinyl sounds bad, just that whereas arguably very few DACs these days produce any audible noise or distortion, there isn't a turntable on the planet that doesn't produce some. But that doesn't mean every DAC is audibly perfect, or that higher fidelity necessarily sounds better to all ears and with all recordings.

 

[Hifi]

Sometimes the subtleties revealed in vinyl playback are more noticeable than on digital playback of same music ...to my ears in my listening room. There is no way anyone can make a “blanket” statement that All CDs and digital music sound better than vinyl because “distortion “. I have some very quiet vinyl.