Poll

Want to see Dave have a look at Bybee Technology's audio Quantum Purifiers?

Yes
18 (23.7%)
No
58 (76.3%)

Total Members Voted: 75

Voting closed: October 21, 2017, 11:05:41 am

Author Topic: Bybee's Lament  (Read 28669 times)

0 Members and 1 Guest are viewing this topic.

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: Bybee's Lament
« Reply #125 on: September 28, 2017, 07:32:00 pm »
Right, so you don't design it to sound 'right', you design it to work right and then fiddle. As you don't know what, in a technical sense, makes it sound right, that fiddling is just random. If that fiddling isn't just random, then you do know what makes it right, in a technical sense, and thus ought to be able to measure that. What I'm chipping away at here is the claim to 'designing' the sound. If it is design, then it is quantifiable and, by corollary, if you can't quantify it then it isn't design it's something else which I don't have a name for.

What differentiates this something else from audiophoolery? I'm not trying to be insulting by asking that, I'm inviting you to offer a reasoned defence of your methodology that differentiates it from all the other unquantifiable arsing around much beloved of the audiophool community.
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 

Offline Alex Nikitin

  • Super Contributor
  • ***
  • Posts: 1149
  • Country: gb
  • Femtoampnut and Tapehead.
    • A.N.T. Audio
Re: Bybee's Lament
« Reply #126 on: September 28, 2017, 08:17:19 pm »
Right, so you don't design it to sound 'right', you design it to work right and then fiddle. As you don't know what, in a technical sense, makes it sound right, that fiddling is just random. If that fiddling isn't just random, then you do know what makes it right, in a technical sense, and thus ought to be able to measure that. What I'm chipping away at here is the claim to 'designing' the sound. If it is design, then it is quantifiable and, by corollary, if you can't quantify it then it isn't design it's something else which I don't have a name for.

What differentiates this something else from audiophoolery? I'm not trying to be insulting by asking that, I'm inviting you to offer a reasoned defence of your methodology that differentiates it from all the other unquantifiable arsing around much beloved of the audiophool community.

It is not as simple as this. The base design gives you a platform to work on. The tuning is not random, it is quite targeted, as listening usually gives you a pretty good idea where to look and what to try and change - it comes from experience. It may involve changing the time constants in various places, DC operating points, sometimes the layout, power supply parameters, thermal constants and arrangements. Usually these changes do not involve a substantial change in overall measured performance, if the base design is correct. Sometimes this kind of search can be guided by measurements (for example, when I did design a CD-player with CS4396 DAC chip in it, just released at that time, I was not happy with the sound the chip produced and discover that lowering the DAC reference voltage by about a volt, from 5V to 4V, and substantially increasing values of some capacitors around the chip gives considerably better measured performance and somewhat unsurprisingly, a better sound to boot). In some of my designs I did include a user accessible tuning knob, as I've found that a certain parameter can be changed in a particular system for the best performance.

My approach is quite straightforward, actually. Even the very best technically (and expensive) designs are not perfect, what matters is how the electronics reacts to a musical signal going through it. The footprint of the system can be benign from the music perception point of view, and can be unpleasant even if all usual technical parameters are very good. The idea of tuning is to make the system's signature as benign as possible so it does not interfere with our perception. For that reason I do not believe in deliberate lowering the performance, "colouring" of the sound and all that crap. The system should be as transparent as possible, and as technically perfect as possible (for a given budget), and these two sides are not mutually exclusive.

Cheers

Alex
 

Offline kalel

  • Frequent Contributor
  • **
  • Posts: 880
  • Country: 00
Re: Bybee's Lament
« Reply #127 on: September 28, 2017, 08:20:51 pm »
I have no experience whatsoever in audio design, nor do I own any high end audio equipment.

But the problem with listening based adjustments could be that every person hears things differently (or one problem at least).
Not only the "built-in lowpass" that changes over time (this is probably accounted for by simply not using the high end when recording/mixing songs), but other factors that may be present in how someone perceives sound.

I usually listen to sound on my computer, mp3 player or phone.
With different devices, I can usually hear a difference, mostly because they are (most likely) measurably different.
With different earphones (usually cheap ones) I can also hear a difference. E.g. on one set there is more bass, on another there is more treble and the bass is quite difficult to hear at all. I'm almost sure these differences could be measured and displayed with FFT using even quite low quality measurement gear.

Different music genres (or specific songs) can have different levels as well. For example, some will be more bassy or mostly bass, others might have a focus on mid-treble.
So, the first type might sound better on one set of earphones/headphones/speakers and the second type might sound better on another set  (or even device that outputs it).

I think it's quite complicated to find "the best" gear for either all types of music or all people. While more than one person might agree on a cheap/awful set of earphones (that has a significantly limited range) compared to decent more expensive ones. I wish that all earphones could/would be comparatively measured, so that one could look at a "frequency response chart" to find the differences. That would surely make some cheap hidden gems easier to find.
« Last Edit: September 28, 2017, 08:24:26 pm by kalel »
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Bybee's Lament
« Reply #128 on: September 28, 2017, 08:59:46 pm »
Well there have been new developments (couple years back)  backed by experiments and published by the AES that (sample)time is crucial for pinpointing the source of a sound signal and that humans are able to go as low as 10us. That would mean a samplerate of 200kHz not for frequency but for correct timealignment.

What happens if you move your head 3mm to the left when listening to stereo speakers?
« Last Edit: September 28, 2017, 09:03:12 pm by Fungus »
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: Bybee's Lament
« Reply #129 on: September 28, 2017, 09:03:13 pm »
Sounds pretty normal to begin with but then we get to:

Even the very best technically (and expensive) designs are not perfect, what matters is how the electronics reacts to a musical signal going through it.

Are we to take it from that, that you believe there is some difference between a "musical signal" and a sine wave? That one can somehow design objectively for good performance with a sine wave but that a "musical signal" is somehow ineffably different? I'll telegraph the punchline here, it's 'superposition'.

Quote
The footprint of the system can be benign from the music perception point of view, and can be unpleasant even if all usual technical parameters are very good. The idea of tuning is to make the system's signature as benign as possible so it does not interfere with our perception. For that reason I do not believe in deliberate lowering the performance, "colouring" of the sound and all that crap. The system should be as transparent as possible, and as technically perfect as possible (for a given budget), and these two sides are not mutually exclusive.

And at that point, I've got to say, what you've said doesn't sound any qualitatively different from the kind of language I've heard artists use to describe their painting, drawing, installations or whatever. What it doesn't sound like is engineering. To quote Monkey from the TV ads, "I say keep it TEA". (This is an obscure joke, that only watchers of UK TV ads who are also member of TEA will get.)
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6459
  • Country: nl
Re: Bybee's Lament
« Reply #130 on: September 28, 2017, 09:07:09 pm »
What happens if you move your head 3mm to the left when listening to stereo speakers?
That is a very little deviation, for me personally nothing really changes.
 

Offline Don Hills

  • Regular Contributor
  • *
  • Posts: 159
Re: Bybee's Lament
« Reply #131 on: September 28, 2017, 09:14:04 pm »
About time to repeat this:
"The time resolution of a 16 bit, 44.1khz PCM channel is not limited to the 22.7µs time difference between samples. The actual minimum time resolution is equivalent to 1/(2pi * quantization levels * sample rate). For 16/44.1, that is 1/(2pi * 65536 * 44100), which is about 55 picoseconds. To put that in perspective, light travels less than an inch in that time.

Shannon and Nyquist showed that as long as you keep all components of the input signal below half the sampling frequency, you can reconstruct the original signal perfectly - not just in terms of amplitude, but in terms of temporal relationships too. They only addressed sampling, and assumed infinite resolution in amplitude. With a digital signal the precision is limited by the number of amplitude steps, leading to the above formula. "
 
The following users thanked this post: schmitt trigger

Offline Alex Nikitin

  • Super Contributor
  • ***
  • Posts: 1149
  • Country: gb
  • Femtoampnut and Tapehead.
    • A.N.T. Audio
Re: Bybee's Lament
« Reply #132 on: September 28, 2017, 09:21:26 pm »
Sounds pretty normal to begin with but then we get to:

Even the very best technically (and expensive) designs are not perfect, what matters is how the electronics reacts to a musical signal going through it.

Are we to take it from that, that you believe there is some difference between a "musical signal" and a sine wave? That one can somehow design objectively for good performance with a sine wave but that a "musical signal" is somehow ineffably different? I'll telegraph the punchline here, it's 'superposition'.


I am disappointed with this reference to a "superposition" . It is substantially different. A continuous sine wave or even a number of sine waves do not represent it accurately, neither does a noise type signal.  Music has a number of qualities which distinguish it from other sounds. Music contains transients and these transients contain the information - which is perceived up to a large degree on a subconscious level, producing emotions. Large part of musical perception is down to its rhythmical structure, which is positioned well below the audible frequency range and thus can be affected by a low frequency time constants, including thermal behaviour of a circuit - in seconds and tens of seconds time frame. And so on. On a simple level it looks simple. If we talk about the best achievable quality of reproduction - not so simple.

Cheers

Alex
« Last Edit: September 28, 2017, 09:38:40 pm by Alex Nikitin »
 
The following users thanked this post: Kjelt

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6459
  • Country: nl
Re: Bybee's Lament
« Reply #133 on: September 28, 2017, 09:40:37 pm »
About time to repeat this:
"The time resolution of a 16 bit, 44.1khz PCM channel is not limited to the 22.7µs time difference between samples. The actual minimum time resolution is equivalent to 1/(2pi * quantization levels * sample rate). For 16/44.1, that is 1/(2pi * 65536 * 44100), which is about 55 picoseconds. To put that in perspective, light travels less than an inch in that time.

Shannon and Nyquist showed that as long as you keep all components of the input signal below half the sampling frequency, you can reconstruct the original signal perfectly - not just in terms of amplitude, but in terms of temporal relationships too. They only addressed sampling, and assumed infinite resolution in amplitude. With a digital signal the precision is limited by the number of amplitude steps, leading to the above formula. "
I believe you but do not understand unless this is valid for audio only. So how can I reproduce two dirac pulses one on the left channel and one on the right channel from 0 to max (65535)with a raise time of 10ns each and a time difference between the two of 5us with a sample frequency which is timealigned between the two channels (same clock) of 44,1 kHz  :-//
 

Offline xrunner

  • Super Contributor
  • ***
  • Posts: 7496
  • Country: us
  • hp>Agilent>Keysight>???
Re: Bybee's Lament
« Reply #134 on: September 28, 2017, 09:42:50 pm »
But the problem with listening based adjustments could be that every person hears things differently (or one problem at least).

LOL really. There is no doubt that the mechanical parts of each individual person's ear - hammer, anvil, stirrup and cochlea are not exactly identical. So when the audiophiles rant on about how a power strip encased in granite and filled with a rare gas (hey neat idea!) makes the music sound better, how can they claim that it will be better when they don't know who's receiving the signal?

So when you think about the minute flaws each individual's ear - who can really say what minutia each of us hear. Perhaps there should even be a hearing test to determine if a person can even be an audiophile.  :-//

It might save people a lot of money not having to buy the best gear.  :-DD
I told my friends I could teach them to be funny, but they all just laughed at me.
 

Offline Don Hills

  • Regular Contributor
  • *
  • Posts: 159
Re: Bybee's Lament
« Reply #135 on: September 28, 2017, 11:41:33 pm »
...
Shannon and Nyquist showed that as long as you keep all components of the input signal below half the sampling frequency, you can reconstruct the original signal perfectly - not just in terms of amplitude, but in terms of temporal relationships too.  ...  "
I believe you but do not understand unless this is valid for audio only. So how can I reproduce two dirac pulses one on the left channel and one on the right channel from 0 to max (65535)with a raise time of 10ns each and a time difference between the two of 5us with a sample frequency which is timealigned between the two channels (same clock) of 44,1 kHz  :-//

See the part I bolded above. Your example violates that requirement. If you were to start with the pulses as analogue signals and digitise them with an ideal anti-aliasing filter (as required by the Theorem), then convert back to analogue with an ideal anti-imaging (reconstruction) filter, you would see a time delay between the channels of 5us.
« Last Edit: September 28, 2017, 11:46:00 pm by Don Hills »
 
The following users thanked this post: Kjelt

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: Bybee's Lament
« Reply #136 on: September 29, 2017, 12:57:27 am »
I am disappointed with this reference to a "superposition" .

I'm sure you are, because it falls into the category of inconvenient facts. Everyone here who didn't sleep through "signal and systems" will know all music is, is a rather odd collection of sine waves in superposition.  Although music in itself might be viewed as special, mystical or magical, as a signal there is nothing special, mystical or magical about music, it is at the end of the day, just another signal. If you can reproduce a sine wave accurately you can reproduce music accurately. If a reproduction system has properties that genuinely affect the music being played though it that effect will be measurable with test equipment just as it will with non-music signals.

Quote
It is substantially different. A continuous sine wave or even a number of sine waves do not represent it accurately, neither does a noise type signal.  Music has a number of qualities which distinguish it from other sounds. Music contains transients and these transients contain the information - which is perceived up to a large degree on a subconscious level, producing emotions. Large part of musical perception is down to its rhythmical structure, which is positioned well below the audible frequency range and thus can be affected by a low frequency time constants, including thermal behaviour of a circuit - in seconds and tens of seconds time frame. And so on. On a simple level it looks simple. If we talk about the best achievable quality of reproduction - not so simple.

Oh c'mon. Perceiving rhythmical structure is not going to be affected by a ten second thermal time constant. I can perceive a tarantella rhythm out of my speakers, respond to it emotionally, dance down the stairs (in time) and be ten feet up the road well out of ear shot before 10 seconds have passed. If a long time constant was going to have an effect on my emotional perception it'd have to chase me down the street, something that does not come as second nature to electronics unless it also happens to contain the additional elements propulsion, guidance and warhead. Engineering has definitely left the building and wo-wo, or at best conjecture, is being offered up in its place.
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 
The following users thanked this post: newbrain

Offline MK14

  • Super Contributor
  • ***
  • Posts: 4527
  • Country: gb
Re: Bybee's Lament
« Reply #137 on: September 29, 2017, 02:42:01 am »
I am disappointed with this reference to a "superposition" .

I'm sure you are, because it falls into the category of inconvenient facts. Everyone here who didn't sleep through "signal and systems" will know all music is, is a rather odd collection of sine waves in superposition.  Although music in itself might be viewed as special, mystical or magical, as a signal there is nothing special, mystical or magical about music, it is at the end of the day, just another signal. If you can reproduce a sine wave accurately you can reproduce music accurately. If a reproduction system has properties that genuinely affect the music being played though it that effect will be measurable with test equipment just as it will with non-music signals.

Quote
It is substantially different. A continuous sine wave or even a number of sine waves do not represent it accurately, neither does a noise type signal.  Music has a number of qualities which distinguish it from other sounds. Music contains transients and these transients contain the information - which is perceived up to a large degree on a subconscious level, producing emotions. Large part of musical perception is down to its rhythmical structure, which is positioned well below the audible frequency range and thus can be affected by a low frequency time constants, including thermal behaviour of a circuit - in seconds and tens of seconds time frame. And so on. On a simple level it looks simple. If we talk about the best achievable quality of reproduction - not so simple.

Oh c'mon. Perceiving rhythmical structure is not going to be affected by a ten second thermal time constant. I can perceive a tarantella rhythm out of my speakers, respond to it emotionally, dance down the stairs (in time) and be ten feet up the road well out of ear shot before 10 seconds have passed. If a long time constant was going to have an effect on my emotional perception it'd have to chase me down the street, something that does not come as second nature to electronics unless it also happens to contain the additional elements propulsion, guidance and warhead. Engineering has definitely left the building and wo-wo, or at best conjecture, is being offered up in its place.

So in summary, you are basically saying that THEORY and PRACTICE and REAL-LIFE are all EXACTLY the same.

Let's throw away all our real life musical instruments, such as Pianos, and give all the musicians all sine wave function generators, with knobs to twiddle.

tl;dr
Replace all Orchestras with some sine wave oscillators. They will sound EXACTLY the same.
 

Offline MK14

  • Super Contributor
  • ***
  • Posts: 4527
  • Country: gb
Re: Bybee's Lament
« Reply #138 on: September 29, 2017, 02:59:55 am »
Example
In theory, an old 2 watt resistor, selling for $200, with a highly suspicious technical explanation on how it works. Would NOT work.

But in practice (because of the Placebo effect), with some people, if you are persuasive enough. It CAN work and improve the sound quality (in their perception/opinion).

There are limitations, in real audio systems, such as distortion and other (usually) undesirable effects. Which we as humans can usually perceive. So using various electronics techniques, (such as filtering, etc etc), are legitimate ways of improving the perceived sound quality, of audio systems.
Even if in theory it should not be necessary, or would even make the sound worse.
« Last Edit: September 29, 2017, 03:01:58 am by MK14 »
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Bybee's Lament
« Reply #139 on: September 29, 2017, 03:05:46 am »
Let's throw away all our real life musical instruments, such as Pianos, and give all the musicians all sine wave function generators, with knobs to twiddle.

Pianos are sine wave function generators.
« Last Edit: September 29, 2017, 03:19:42 am by Fungus »
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 12288
  • Country: au
Re: Bybee's Lament
« Reply #140 on: September 29, 2017, 03:09:58 am »
Let's throw away all our real life musical instruments, such as Pianos, and give all the musicians all sine wave function generators, with knobs to twiddle.

tl;dr
Replace all Orchestras with some sine wave oscillators. They will sound EXACTLY the same.

Given enough sine wave oscillators and the ability to adjust them as quickly as necessary - then yes they will.

What fails here is the practicality of doing so.
 

Offline MK14

  • Super Contributor
  • ***
  • Posts: 4527
  • Country: gb
Re: Bybee's Lament
« Reply #141 on: September 29, 2017, 03:13:51 am »
Let's throw away all our real life musical instruments, such as Pianos, and give all the musicians all sine wave function generators, with knobs to twiddle.

Pianos are sine wave function generators.

Yes, it is. But it also adds all sorts of resonances and other effects to the sound. Making it (for some people), an enjoyable experience to listen to someone playing music on it.

I have one or more function generators, capable of producing, nice sine waves. For some mysterious reason, nobody has asked me to play music on it for them, by twisting the frequency dial, up and down.

Maybe I should contact my nearest Orchestra and offer to play my function generator, as part of their Orchestra ?
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 12288
  • Country: au
Re: Bybee's Lament
« Reply #142 on: September 29, 2017, 03:16:24 am »
Call it a Theremin and you might get a gig.
 

Offline MK14

  • Super Contributor
  • ***
  • Posts: 4527
  • Country: gb
Re: Bybee's Lament
« Reply #143 on: September 29, 2017, 03:18:12 am »
Call it a Theremin and you might get a gig.

True.

If we were back in the 1960's, we could probably make lots of money, with electronically generated music.
Or play lazy, and buy shares in Microsoft, when 1975 arrives.
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: Bybee's Lament
« Reply #144 on: September 29, 2017, 03:19:21 am »
Replace all Orchestras with some sine wave oscillators. They will sound EXACTLY the same.

Let me guess, you did sleep through "signals and systems", that or you've no formal electronics training.  It is possible to create any arbitrary waveform by simply summing the right set of sine waves at the right amplitudes - ask Monsieur Fourier. Further, it is a fundamental property of linear systems that you get the same result if you process all those summed sine waves through them at once as you do if you process all those sine waves individually and then add them together afterwards. That's the superposition theorem and that's what means a bode plot produced from a swept sine wave is equivalent to a bode plot produced by feeding broadband noise through a system, or even feeding music through. All of which is a long winded way of saying that there is nothing special about music as a signal, it's just another signal.
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 
The following users thanked this post: fonograph

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Bybee's Lament
« Reply #145 on: September 29, 2017, 03:21:10 am »
In theory, an old 2 watt resistor, selling for $200, with a highly suspicious technical explanation on how it works. Would NOT work.

But in practice (because of the Placebo effect), with some people, if you are persuasive enough. It CAN work and improve the sound quality (in their perception/opinion).

Yes.

There are limitations, in real audio systems, such as distortion and other (usually) undesirable effects. Which we as humans can usually perceive. So using various electronics techniques, (such as filtering, etc etc), are legitimate ways of improving the perceived sound quality, of audio systems.
Even if in theory it should not be necessary, or would even make the sound worse.

Yes, but the limitations aren't where the audiophools tell you they are. The limitations aren't in the source so 24bit sampling or 192kHz sample rate won't make any difference.  They aren't in the cables or interconnects, they're not really in the amplifiers (not once you get past a few hundred $), they're mostly in the speakers and listening environment.

nb. An ideal speaker doesn't (and cannot) exist. An ideal speaker is a tiny point source but a point source would have very little low frequency response. Speakers with low frequency response have large transducers so they lose stereo image. It's a tradeoff between the two.

You can either:
a) Use large, powerful speakers in a larger, anechoic room and stand well back (so the speaker cones look tiny to you).
b) Use smaller speakers and sit very close to them (near field listening).
c) Use headphones. This is really just 'b' taken to the extreme.

All methods have their pros and cons.
« Last Edit: September 29, 2017, 03:24:30 am by Fungus »
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 12288
  • Country: au
Re: Bybee's Lament
« Reply #146 on: September 29, 2017, 03:24:45 am »
I have one or more function generators, capable of producing, nice sine waves. For some mysterious reason, nobody has asked me to play music on it for them, by twisting the frequency dial, up and down.

This is the practicality issue.

You will need quite a few oscillators to mimic a piano - and that will be for just one note.  You will also have to be able to make hundreds of adjustments per second to get that note sounding right.  Add some chords and you're really going to be working hard.  Put that into a score and a hundred people would not be able to create enough for one instrument.

It is not impossible ... just completely impractical.
 

Offline MK14

  • Super Contributor
  • ***
  • Posts: 4527
  • Country: gb
Re: Bybee's Lament
« Reply #147 on: September 29, 2017, 03:32:22 am »
Replace all Orchestras with some sine wave oscillators. They will sound EXACTLY the same.

Let me guess, you did sleep through "signals and systems", that or you've no formal electronics training.  It is possible to create any arbitrary waveform by simply summing the right set of sine waves at the right amplitudes - ask Monsieur Fourier. Further, it is a fundamental property of linear systems that you get the same result if you process all those summed sine waves through them at once as you do if you process all those sine waves individually and then add them together afterwards. That's the superposition theorem and that's what means a bode plot produced from a swept sine wave is equivalent to a bode plot produced by feeding broadband noise through a system, or even feeding music through. All of which is a long winded way of saying that there is nothing special about music as a signal, it's just another signal.

I agree with that part of your argument, in that a signal is a signal. Even if one of the signals, has been created/made/transferred in a different or complicated way. Such as by using multiple sine waves.
Assuming that all the necessary features of the original sound signal, have been reproduced, correctly. As necessary.

But humans perception of sound (music), is probably significantly influenced by all sorts of things. Some of which, are probably nothing to do with sound at all. Such as lighting, smells, ambiance etc.
 

Offline MK14

  • Super Contributor
  • ***
  • Posts: 4527
  • Country: gb
Re: Bybee's Lament
« Reply #148 on: September 29, 2017, 03:41:21 am »
Yes, but the limitations aren't where the audiophools tell you they are. The limitations aren't in the source so 24bit sampling or 192kHz sample rate won't make any difference.  They aren't in the cables or interconnects, they're not really in the amplifiers (not once you get past a few hundred $), they're mostly in the speakers and listening environment.

nb. An ideal speaker doesn't (and cannot) exist. An ideal speaker is a tiny point source but a point source would have very little low frequency response. Speakers with low frequency response have large transducers so they lose stereo image. It's a tradeoff between the two.

You can either:
a) Use large, powerful speakers in a larger, anechoic room and stand well back (so the speaker cones look tiny to you).
b) Use smaller speakers and sit very close to them (near field listening).
c) Use headphones. This is really just 'b' taken to the extreme.

All methods have their pros and cons.

I agree.

I think, to an extent, improving the sound quality of things (i.e. quality Hi-Fi systems with quality speakers), is good and useful. But audiophools can take things too far, and as you have just been saying, they are worrying about things which scientifically will (in theory and practice), make absolutely no difference/improvement, whatsoever. (Ignoring the Placebo effect, which is another matter).
 

Offline MK14

  • Super Contributor
  • ***
  • Posts: 4527
  • Country: gb
Re: Bybee's Lament
« Reply #149 on: September 29, 2017, 03:46:50 am »
I have one or more function generators, capable of producing, nice sine waves. For some mysterious reason, nobody has asked me to play music on it for them, by twisting the frequency dial, up and down.

This is the practicality issue.

You will need quite a few oscillators to mimic a piano - and that will be for just one note.  You will also have to be able to make hundreds of adjustments per second to get that note sounding right.  Add some chords and you're really going to be working hard.  Put that into a score and a hundred people would not be able to create enough for one instrument.

It is not impossible ... just completely impractical.

That makes a lot of sense.
In the same way that ray tracing can make great 3d graphics. I think there are similar systems, for creating sounds, just using pure computation. They are called Physical modeling synthesizers. So nowadays, we can use such methods to create sounds/music, if you want to.
I.e. Because it is being created inside a computer, it does not matter if you need hundreds or thousands of sine wave generators, as they are essentially free, will take up no physical space, and stay perfectly in sync and in tune.
Unlike real life analogue function generators (without PLL or other tricks, up their sleeve).

https://en.wikipedia.org/wiki/Physical_modelling_synthesis
« Last Edit: September 29, 2017, 03:51:39 am by MK14 »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf