Archimago on Greene vs Harley

[opus101]

I'm having trouble parsing your last sentence 'by random trial/error'. So  I take it the answer to my question is 'its random' in which case I agree and I think their process is probably just a strawman. After all, you quoted their marketing materials right?

 

<later> I see you've edited and my conjecture was correct. Thanks.

[pkane2001]

5 minutes ago, opus101 said:

I'm having trouble parsing your last sentence 'by random trial/error'. So  I take it the answer to my question is 'its random' in which case I agree and I think their process is probably just a strawman. After all, you quoted their marketing materials right?

 

I don't get what you don't get. Marketing materials or not, that's the approach I take with listening tests and measurements. If I can't measure something I can identify audibly, I look for a way to measure it. To me, understanding the root cause and being able to find it again is much more valuable than just patching up a problem by guessing.

[opus101]

I'm getting it. Their marketing materials of course want to paint their approach in the best possible light, hence they set up a strawman and demolish that, hence implying they're the truly enlightened ones in the audio business and that others are by implication imbeciles.

[pkane2001]

2 minutes ago, opus101 said:

I'm getting it. Their marketing materials of course want to paint their approach in the best possible light, hence they set up a strawman and demolish that, hence implying they're the truly enlightened ones in the audio business and that others are by implication imbeciles.

 

So do you think that's what I'm doing also?  I'm not using Benchmark to validate my own philosophy. I used it as a shorthand for me not to have to write all that text to describe what my philosophy is. If you want to argue about why my philosophy is wrong, then let's have that discussion. I really couldn't care less about Benchmark or their products or marketing, although I hear they measure well :)

 

[opus101]

Just now, pkane2001 said:

 

So do you think that's what I'm doing also?

 

If your question is 'Do I think you're trying to paint your own products in the best light by marketing them using strawmen?' then the answer's definitely a 'no'. I'm not even clear if you've got stuff to sell.

 

I rather suspect we're talking at crossed-purposes here. I've been focussing on what process Benchmark wish to discredit in their marketing. You've been talking about what your own philosophy is. Two rather different focusses no?

[Rexp]

37 minutes ago, pkane2001 said:

 

What Kal said... 

 

Measurements can show if the reproduction is closer to the recorded signal. If it's different in some specific ways from measurements from another device, this can also tell which one is more true to the original, and what needs to be corrected in the other device. Even if we don't yet know every possible thing to measure, what we do know is still very useful since it allows us to make meaningful, repeatable comparisons and get to the root cause. This article from Benchmark covers my philosophy.

 

Take a real world example, if I record a voice at 24/192 and compare to the original, it sounds close enough to be deemed an accurate reproduction. No doubt the measurements would be similar. Now if I downsample to 16/44 it doesn't sound like the original but the measurements won't reflect this will they? 

[pkane2001]

3 minutes ago, opus101 said:

 

If your question is 'Do I think you're trying to paint your own products in the best light by marketing them using strawmen?' then the answer's definitely a 'no'. I'm not even clear if you've got stuff to sell.

 

I rather suspect we're talking at crossed-purposes here. I've been focussing on what process Benchmark wish to discredit in their marketing. You've been talking about what your own philosophy is. Two rather different focusses no?

 

I'm not selling a thing, and yes, I think we are talking at cross-purposes. Maybe I was a bit lazy by having Benchmark describe my philosophy :)

 

[opus101]

I'm interested in what Benchmark have to say about development using listening tests because in the past I had the pleasure of working with a guy who did (pro) audio design/development using an ABX box he built himself. Here's one of his posts on Gearslutz (the second post on this page) : https://www.gearslutz.com/board/music-computers/542885-paul-frindle-truth-myth-4-print.html

[pkane2001]

30 minutes ago, opus101 said:

I'm interested in what Benchmark have to say about development using listening tests because in the past I had the pleasure of working with a guy who did (pro) audio design/development using an ABX box he built himself. Here's one of his posts on Gearslutz (the second post on this page) : https://www.gearslutz.com/board/music-computers/542885-paul-frindle-truth-myth-4-print.html

 

I like him. Finding it hard to disagree with most of what he said in that thread.

[pkane2001]

52 minutes ago, Rexp said:

Take a real world example, if I record a voice at 24/192 and compare to the original, it sounds close enough to be deemed an accurate reproduction. No doubt the measurements would be similar. Now if I downsample to 16/44 it doesn't sound like the original but the measurements won't reflect this will they? 

 

Of course measurements can reveal if the recording is at 16/44 or 24/192. Do you mean that you can tell the difference between speech recorded at 24/192 and the downsampled version at 16/44? That's not a hard test to perform. In fact, the recent Hi-res test by Mark Waldrep was of a very similar design. Didn't you take part in that test? How did you do?

 

[Rexp]

4 minutes ago, pkane2001 said:

 

Of course measurements can reveal if the recording is at 16/44 or 24/192. Do you mean that you can tell the difference between speech recorded at 24/192 and the downsampled version at 16/44? That's not a hard test to perform. In fact, the recent Hi-res test by Mark Waldrep was of a very similar design. Didn't you take part in that test? How did you do?

 

Nevermind

[fas42]

3 hours ago, pkane2001 said:

 

To me, this approach leads to finding real issues and real solutions. Stabbing in the dark at all possible noise sources until everything "sounds just right" simply doesn't fit my temperament, sorry Frank :)

 

 

Interesting the need to use phrases like "stabbing in the dark", and "random trial/error" to describe the process of finding the cause and effect linkage - it's almost as if the actual measuring of some electrical anomaly is more important than the fixing of the issue; that is, if the playback sounds wrong, and you manage to come up with some numbers "that describe it", then you can relax and keep listening to the fault, without being overly concerned with rectifying it 🙂.

 

If one uses a good technique, that of using recordings which highlight the defective behaviour, then it's usually very quick to pinpoint a cause/weakness combination - the hard work is often then to work out a solution which is not expensive, and which delivers a robust - meaning it works under all scenarios - fix. The sessions at the friend up the road of some hours is usually enough to locate where there is a bottleneck in the SQ - it may take weeks to devise a 'smart' resolution, which I leave to him.

 

Most people seem to find it hard to understand the approach, it seems - you listen to a recording you know well, and it's definitely sub-par, probably from noise. Most noise issues come from a lack of physical integrity in some area, or electrical interference - it's quit easy to alter these factors, usually; and the feedback from trying things gives you the knowledge to make the next move.

[pkane2001]

6 hours ago, fas42 said:

 

Interesting the need to use phrases like "stabbing in the dark", and "random trial/error" to describe the process of finding the cause and effect linkage - it's almost as if the actual measuring of some electrical anomaly is more important than the fixing of the issue; that is, if the playback sounds wrong, and you manage to come up with some numbers "that describe it", then you can relax and keep listening to the fault, without being overly concerned with rectifying it 🙂.

 

If one uses a good technique, that of using recordings which highlight the defective behaviour, then it's usually very quick to pinpoint a cause/weakness combination - the hard work is often then to work out a solution which is not expensive, and which delivers a robust - meaning it works under all scenarios - fix. The sessions at the friend up the road of some hours is usually enough to locate where there is a bottleneck in the SQ - it may take weeks to devise a 'smart' resolution, which I leave to him.

 

Most people seem to find it hard to understand the approach, it seems - you listen to a recording you know well, and it's definitely sub-par, probably from noise. Most noise issues come from a lack of physical integrity in some area, or electrical interference - it's quit easy to alter these factors, usually; and the feedback from trying things gives you the knowledge to make the next move.

 

But of course it is "stabbing in the dark", Frank! I'm talking about your method of "fixing" audio faults. You may recall recommending to me to open up my speakers to find out if they have soldered connections AS THE FIRST STEP in troubleshooting my system. Sorry, but if that's not random, then I don't know what is. There's no technique here, it's stabbing in the dark.

 

But yes, I prefer to find the root cause, understand the problem, and do so without resoldering every solder joint on each PCB in my system and then listening for any improvement after each in the hopes that I got it this time!

[Kal Rubinson]

12 hours ago, Rexp said:

Now if I downsample to 16/44 it doesn't sound like the original but the measurements won't reflect this will they? 

That depends on how you measure.  A simple assessment of the digital bitstream will reveal the difference in resolution.  😎

[fas42]

8 hours ago, pkane2001 said:

 

But of course it is "stabbing in the dark", Frank! I'm talking about your method of "fixing" audio faults. You may recall recommending to me to open up my speakers to find out if they have soldered connections AS THE FIRST STEP in troubleshooting my system. Sorry, but if that's not random, then I don't know what is. There's no technique here, it's stabbing in the dark.

 

But yes, I prefer to find the root cause, understand the problem, and do so without resoldering every solder joint on each PCB in my system and then listening for any improvement after each in the hopes that I got it this time!

 

I do indeed ... and what's random about it? Ever since I tweaked my original good rig up a high standard, which made me strongly aware that the integrity of connections, including those in speakers, was critical, the first port of call when I start using a different set of speakers is to dive inside. I'll do a first listen beforehand, and if there is a roughness, an offness to the sound I'll immediately start investigating. On the ToDo list for my cheap actives, but I haven't yet opened them up because the sound was so impressive, from the word go.

 

Remember, it's all about the weakest links; they will dominate the subjective impression of the SQ - how likely is that poor link to be a solder joint on a PCB, versus a cheap quality, push on connector that is a general standard for how the industry does things.

 

Sorry, the Adding Goodness principle has never worked for me - which is why I normally roll my eyes when I hear ambitious, expensive rigs - they have obvious problems, so, yes, a highly likely first step for me would be to open up their speakers ...

[Jud]

On 11/24/2020 at 11:54 AM, pkane2001 said:

I often can't predict what my wife would prefer, and I've spent most of my life with her, observing her preferences and talking to her about her choices thousands of times.

 

Perhaps you are evaluating her preferences subjectively rather than objectively. 😉

[pkane2001]

15 minutes ago, Jud said:

 

Perhaps you are evaluating her preferences subjectively rather than objectively. 😉

 

That's the problem! 

[fas42]

18 minutes ago, pkane2001 said:

 

That's the problem! 

 

So, does she give the thumbs up on how your system sounds, at all times ... ? 😈

[pkane2001]

9 minutes ago, fas42 said:

 

So, does she give the thumbs up on how your system sounds, at all times ... ? 😈

 

Of course! But for some reason, only when she hears it from the kitchen...? I hear that this is proof-positive that the system sounds excellent 😜

[Rexp]

7 hours ago, Kal Rubinson said:

That depends on how you measure.  A simple assessment of the digital bitstream will reveal the difference in resolution.  😎

So you agree that measurements can determine the accuracy of audio reproduction?

[Kal Rubinson]

8 minutes ago, Rexp said:

So you agree that measurements can determine the accuracy of audio reproduction?

Determinacy, overall. is not entirely possible today.   I do believe that measurements assess accuracy for the parameters measured.  

[Rexp]

1 minute ago, Kal Rubinson said:

Determinacy, overall. is not entirely possible today.   I do believe that measurements assess accuracy for the parameters measured.  

I'll take that as a no

[Jud]

With regard to the poll question: Apologizing in advance for what might be seen as a cop-out, my answer would be "None of the above."

 

I certainly agree we're at a point where measurements of everything thought to be significant in electronics and acoustics are quite excellent. So then what's my problem with Greene and Archimago? (Preliminarily, I'll mention that I have no problem at all with Greene's recommendation that we get the listening room acoustics and speaker response reasonably close to right.)

 

I don't think we're at a sufficient level of maturity with regard to measurement of human perception. A couple of points along these lines:

 

- When scientists do testing that relies on verbal responses to consciously perceived phenomena, the fact that this may not reflect all that is going on with the test subjects must be taken into account. Obviously relying on a patient to say "I feel feverish" or to say whether they have a low or high fever not only isn't as accurate as a simple thermometer, it may not be the most accurate description of how sick the patient feels. When we rely on verbal responses to consciously perceived phenomena to tell us whether subjects like a piece of music better or worse or have perceived a change, this may not in fact be the most accurate description of whether any reaction has taken place on the subconscious or even molecular level (dopamine release for pleasure, the equivalent for a subconscious reaction to unpleasant stimulus).

 

- While equipment does better than humans with respect to measurement, it hasn't yet caught up to the human brain in terms of pattern matching for many functions, including when a piece of music sounds as if "you are there" (for either the real venue if there is one, or the virtual venue for a studio recording). Certainly the electromechanical and acoustic parts of the chain (speakers, mics, the acoustic characteristics of recording and listening venues) introduce larger measurable distortions than the purely electronic stages. But I know of no equipment that measures to what extent each of these sorts of distortion prevents us from having that "you are there" feeling, and to what extent they don't bother us or cause us to lose that feeling of verisimilitude. Optical and audible illusions teach us that gross distortions of visual and audio input can look or sound to us exactly like the "real thing." (Remember, stereo itself is built on a quite reliable audible illusion.) Note also that in many cases we respond to such illusions - distortions - differently from each other. A distortion that passes by you not even consciously heard may set my teeth on edge, or at least destroy that feeling of verisimilitude.

 

So - equipment measurements, I think we're mostly doing fine. Measurements of what's going on inside our brains when we listen, we're not there yet, and I'm not willing to rely on conscious verbal responses as a substitute.

[fas42]

50 minutes ago, Jud said:

- While equipment does better than humans with respect to measurement, it hasn't yet caught up to the human brain in terms of pattern matching for many functions, including when a piece of music sounds as if "you are there" (for either the real venue if there is one, or the virtual venue for a studio recording). Certainly the electromechanical and acoustic parts of the chain (speakers, mics, the acoustic characteristics of recording and listening venues) introduce larger measurable distortions than the purely electronic stages. But I know of no equipment that measures to what extent each of these sorts of distortion prevents us from having that "you are there" feeling, and to what extent they don't bother us or cause us to lose that feeling of verisimilitude. Optical and audible illusions teach us that gross distortions of visual and audio input can look or sound to us exactly like the "real thing." (Remember, stereo itself is built on a quite reliable audible illusion.) Note also that in many cases we respond to such illusions - distortions - differently from each other. A distortion that passes by you not even consciously heard may set my teeth on edge, or at least destroy that feeling of verisimilitude.

 

 

Precisely so ...

[Kal Rubinson]

2 hours ago, Rexp said:

I'll take that as a no

OTOH, I would not make a subjective choice without it.