Audio Science: Does it explain everything about how something sounds?

[jkeny]

Oh, I don't think those are the two groups at all. Show me a post, any post on WBF, where anyone here has taken that position. I think the two groups you're talking about are actually people who believe there's a lot to be gained from existing science and expect that manufacturers are using it if it's real, and those who obsess over, or profit from, that which cannot be heard, has not been tested, and will not be verified because scientists and the big manufacturers with the resources won't chase those ghosts.

But at the fringes of the high end, John, baseless speculation, even about the limits of human hearing, is an opportunity. Go for it. There are more than enough audiophiles who already believe they hear what mortals and the best testing equipment available cannot, and they'll hear everything you imagine.

Tim

Tim, it's ironic that you deny what I wrote & then proceed to confirm it by the rest of the contents of your post

 

[jkeny]

Oh, I don't think those are the two groups at all. Show me a post, any post on WBF, where anyone here has taken that position. I think the two groups you're talking about are actually people who believe there's a lot to be gained from existing science and expect that manufacturers are using it if it's real, and those who obsess over, or profit from, that which cannot be heard, has not been tested, and will not be verified because scientists and the big manufacturers with the resources won't chase those ghosts.

Apart from speakers & room treatment where else can you point to a lot to be gained from using audio science?

But at the fringes of the high end, John, baseless speculation, even about the limits of human hearing, is an opportunity. Go for it. There are more than enough audiophiles who already believe they hear what mortals and the best testing equipment available cannot, and they'll hear everything you imagine.

Tim

Tim, I find these comments insulting

 

[Fitzcaraldo215]

Thank you Fitzcaraldo215. I started the thread and asked the questions hoping to learn more about what audio science actually is and what it can tell us. I had a "largely incomplete and superficial understanding of what audio science really is." Posts like yours above and those from Atmasphere, 853guy and others, have shed some light on my query and I have enjoyed, and learned from, the discussion.

I think I understand a little more about audio science now, and I am sorry that you found the title to the thread to be naive and a set up. I did not mean it to be either, and I should add, that I was certainly not in fear of learning anything new.

Sorry, Peter. I meant no personal offense. Best wishes.

 

[Atmasphere]

Tony,
have you given any thought to what mechanism might account for this?

The ear can detect phase shift in spectrums of frequency even though it cannot with individual tones.

I worked on a phono state in an MFA Magus preamp that was sounding bright, so much so it sounded almost digital (this was about 25 years ago; digital is a lot better now...). Turns out that there was a network so that the EQ went to 'flat' at 50KHz. I took the network out (restoring the RIAA slope at 50KHz; yes I know that the RIAA does not spec that high but look at the electronics of any cutter system and you will see that the curve is there at that frequency anyway and nearly all phono circuits have no troubles going that high either) and the brightness went away. What was happening is that there was phase shift in the electronics as a result of the change at 50KHz. Even though there was no hope of hearing anything at that frequency, the ear translates phase and distortion issues into tonality. In this case the phase artifacts were present to 1/10th the frequency in question, IOW down to 5KHz and increasing effect with increased frequency. It was easily heard- which is why I was asked to see if there was anything that could be done.

This is why bandwidth is important- not because we can hear that high but so that the electronics does not introduce phase shift artifacts within the audio passband. To this end 2Hkz to 200KHz insures that no phase shift will be present in the audio passband. Those that think 20-20KHz is sufficient are dealing with what was known about the human ear/brain system from 60 years ago and are a bit out of date.

However, one should not be in the least surprised if simple models prove inadequate to explain the world in detail.

+1 on that

 

[Steve williams]

Tim, I find these comments insulting

I do too as well as a few others who have reported Tim's post

Tim

I have to tell you that having read your most recent posts to this thread that I agree with Jon Keny as well as with he who reported the post

I appreciate debate Tim but quite honestly you do nothing to your position with posts like these as they serve only to alienate members.Kindly address the post and leave your trolling out of it

 

[FrantzM]

The ear can detect phase shift in spectrums of frequency even though it cannot with individual tones.

I worked on a phono state in an MFA Magus preamp that was sounding bright, so much so it sounded almost digital (this was about 25 years ago; digital is a lot better now...). Turns out that there was a network so that the EQ went to 'flat' at 50KHz. I took the network out (restoring the RIAA slope at 50KHz; yes I know that the RIAA does not spec that high but look at the electronics of any cutter system and you will see that the curve is there at that frequency anyway and nearly all phono circuits have no troubles going that high either) and the brightness went away. What was happening is that there was phase shift in the electronics as a result of the change at 50KHz. Even though there was no hope of hearing anything at that frequency, the ear translates phase and distortion issues into tonality. In this case the phase artifacts were present to 1/10th the frequency in question, IOW down to 5KHz and increasing effect with increased frequency. It was easily heard- which is why I was asked to see if there was anything that could be done.

This is why bandwidth is important- not because we can hear that high but so that the electronics does not introduce phase shift artifacts within the audio passband. To this end 2Hkz to 200KHz insures that no phase shift will be present in the audio passband. Those that think 20-20KHz is sufficient are dealing with what was known about the human ear/brain system from 60 years ago and are a bit out of date.




+1 on that

Well that is an explanation brought in by Science ... I am fine with that ...

 

[Phelonious Ponk]

Apart from speakers & room treatment where else can you point to a lot to be gained from using audio science?


Tim, I find these comments insulting

Sorry. I find the comment above alarming from a component designer, so perhaps we're even.

Tim

 

[jkeny]

Sorry. I find the comment above alarming from a component designer, so perhaps we're even.

Tim

Huh?
Edit: I see that I left a word off my post to which you commented on above - it should have read "Apart from speakers & room treatment where else can you point to a lot MORE to be gained from using audio science?"

 

[Don Hills]

Common frequency domain approaches are based on linear systems theory. The ear and the brain's acoustic processing are not linear. There is really nothing to explain. The linear approximation is good for first order accuracy, hence the common tests and measurements. However, one should not be in the least surprised if simple models prove inadequate to explain the world in detail.

Ah, OK, thanks. I was just wondering if you'd given it a little more thought than "because non-linear". A deeper understanding of the mechanisms could lead to improvements in audio processing.

 

[Don Hills]

The ear can detect phase shift in spectrums of frequency even though it cannot with individual tones.

... In this case the phase artifacts were present to 1/10th the frequency in question, IOW down to 5KHz and increasing effect with increased frequency
.... Those that think 20-20KHz is sufficient are dealing with what was known about the human ear/brain system from 60 years ago and are a bit out of date. ...

Understood. But it doesn't prove that 20 Hz - 20 KHz is not enough. The problem was not the phase shift at 50 KHz. The problem was that there was phase shift within the audible range.

 

[Tony Lauck]

Understood. But it doesn't prove that 20 Hz - 20 KHz is not enough. The problem was not the phase shift at 50 KHz. The problem was that there was phase shift within the audible range.

The problem is use of terms like "audible range", making a fair and intelligent discussion difficult and possibly unpleasant.

 

[stehno]

I installed a system in one of my cars years ago using all Focal speakers and an amp with a tube input section. I took the interior apart and added a ton of noise suppression to all the interior panels, floor, firewall, etc and the difference was unbelievable. It was a 1992 AWD Talon/Eclipse with the Mitsu 2L turbo motor, and after the install it was as quiet as a big Mercedes and the doors sounded like a bank vault when they shut. These days I couldn't care less about the system in my car but back then it was fun...

That reminds me.

In 1978 I owned a rather large 1969 Oldsmodbile Delta 88. I installed a Pioneer SuperTuner in the glovebox and had a pair of AudioVox coaxial speakers. I removed the tweeters from the coaxial speakers and placed the tweeters inside the stock dashboard speaker with the 6 x 9 woofers in the rear package shelf.

So the high frequencies would spray across the entire windshield up front out of the stock speaker grill with the mids and lows coming from the rear. For that era, it really sounded quite nice. Chuck Mangione never "Felt So Good."

I'm not sure but this may have been one of the very first instances of separate drivers in an automobile.

 

[Phelonious Ponk]

Huh?
Edit: I see that I left a word off my post to which you commented on above - it should have read "Apart from speakers & room treatment where else can you point to a lot MORE to be gained from using audio science?"

The MORE doesn't help. Sorry, I still find that to be a rather shocking statement from someone who designs and builds audio components. Maybe I'm not understanding you.

Tim

 

[esldude]

The problem is use of terms like "audible range", making a fair and intelligent discussion difficult and possibly unpleasant.

I find a refusal to at least nominally agree on an audible range makes for an even more difficult situation in which to carry on discussion.

In the most recent example it wasn't the 50 khz phase shift that was directly audible as Don Hills said. It was the phase changes it also caused in the range below 20 khz. Saying a 50 khz filter alters below 20 khz sound isn't controversial, odd, or difficult to understand. Saying results at 50 khz change the sound is more opaque, and likely to lead to miscommunication.

The old guideline of getting response that was -3db down out to 200 khz recognizes you need that for flat response without phase issues to 20 khz. Paired with knowledge that some collections of high frequency sounds get timed together like lower frequency tones makes it all seem reasonably explained and understandable. Nothing mysterious about it really.

 

[esldude]

That reminds me.

In 1978 I owned a rather large 1969 Oldsmodbile Delta 88. I installed a Pioneer SuperTuner in the glovebox and had a pair of AudioVox coaxial speakers. I removed the tweeters from the coaxial speakers and placed the tweeters inside the stock dashboard speaker with the 6 x 9 woofers in the rear package shelf.

So the high frequencies would spray across the entire windshield up front out of the stock speaker grill with the mids and lows coming from the rear. For that era, it really sounded quite nice. Chuck Mangione never "Felt So Good."

I'm not sure but this may have been one of the very first instances of separate drivers in an automobile.

Talk about separate drivers. I purchased a 1967 Camaro once. It had something like 11 or 13 speakers in it (yes an odd number though stereo). Various sizes, locations and types. The previous owner's theory was all speakers have uneven response. It only gets more uneven in different parts of a car. Averaging of many types of response was more likely to result in an average response close to flat. So your money was better spent on many cheap speakers than a couple of expensive speakers. He even was quite convinced of how well it all worked. I still remember the strange look on his face, the life in his eyes as he proudly explained it.

Now of course, it was loud and just about the mush you would expect. I removed all of them, and put two pretty okay 6x9's in the hat rack. Much improvement. Of course that was all driven by a Craig 8 track head unit with maybe 10 real wpc.

 

[Don Hills]

The problem is use of terms like "audible range", making a fair and intelligent discussion difficult and possibly unpleasant.

Tony,
I'm sorry, I was imprecise. For "audible range", please read "20 Hz - 20 KHz".

 

[jkeny]

I'm asking you what advances you see happening in the audio area (other than room & speakers) using current audio science in other words current stock audio measurements?

The MORE doesn't help. Sorry, I still find that to be a rather shocking statement from someone who designs and builds audio components. Maybe I'm not understanding you.

Tim

 

[Groucho]

I'm asking you what advances you see happening in the audio area (other than room & speakers) using current audio science in other words current stock audio measurements?

How do we know what is an advance and what isn't if we don't know what it is measured against? If the aim is to reproduce the signal that the microphones picked up and was committed to the recording, 3m from the speaker in an anechoic chamber then the system that comes closest in absolute terms is a digital recording, DSP driver correction, solid state amps and active speakers. Marginal future improvements may be possible in terms of absolute accuracy. I would also claim that this will sound the best in a real room.

If the aim is something less precise (but not necessarily different) like "reproducing the sound of the live event" then science has nothing to say on the matter - listening tests don't count because true (as opposed to pseudo-) science has no validity in the field of aesthetic judgement. What I am certain of, though, is that if strict accuracy of the waveform at the speaker is not 'the answer' (I think it is), then the 'answer' will lie in DSP algorithms, not re-hashing old technology or messing about with wires.

(Sorry if that's insulting )

 

[jkeny]

How do we know what is an advance and what isn't if we don't know what it is measured against?

Don't really know what you're taking about? Are Dacs, amplifiers preamps not already perfected according to current audio measurements? So what do you mean that you don't know what to measure against?

If the aim is to reproduce the signal that the microphones picked up and was committed to the recording, 3m from the speaker in an anechoic chamber then the system that comes closest in absolute terms is a digital recording, DSP driver correction, solid state amps and active speakers. Marginal future improvements may be possible in terms of absolute accuracy. I would also claim that this will sound the best in a real room.

If the aim is something less precise (but not necessarily different) like "reproducing the sound of the live event" then science has nothing to say on the matter - listening tests don't count because true (as opposed to pseudo-) science has no validity in the field of aesthetic judgement. What I am certain of, though, is that if strict accuracy of the waveform at the speaker is not 'the answer' (I think it is), then the 'answer' will lie in DSP algorithms, not re-hashing old technology or messing about with wires.

(Sorry if that's insulting )

 

[Atmasphere]

Understood. But it doesn't prove that 20 Hz - 20 KHz is not enough. The problem was not the phase shift at 50 KHz. The problem was that there was phase shift within the audible range.

Don, here's an engineering principle for you: Phase shift components are generally considered to exist to 1/10th the cutoff frequency in the case of a HF rolloff (assuming a 6db slope, even lower if the cutoff is more profound); to 10X the cutoff frequency in the case of LF rolloff. So 20-20KHz is insufficient for electronics if they are going to be considered to be 'hifi' because they won't.

I find a refusal to at least nominally agree on an audible range makes for an even more difficult situation in which to carry on discussion.

In the most recent example it wasn't the 50 khz phase shift that was directly audible as Don Hills said. It was the phase changes it also caused in the range below 20 khz. Saying a 50 khz filter alters below 20 khz sound isn't controversial, odd, or difficult to understand. Saying results at 50 khz change the sound is more opaque, and likely to lead to miscommunication.

The old guideline of getting response that was -3db down out to 200 khz recognizes you need that for flat response without phase issues to 20 khz. Paired with knowledge that some collections of high frequency sounds get timed together like lower frequency tones makes it all seem reasonably explained and understandable. Nothing mysterious about it really.

Exactly +1