an interesting read ...........

[Mikado463]

brought this over from another group and thought it might be appreciated here as well.........

Unreasonable Audio Innovation

Keynote at the AES160th Convention Reception, Cityhall of Copenhagen,May 29th, 2026

purifi-audio.com

 

[m. zillch]

From that link (I've underline the pertinent part):

"The objectivists tribe: standard metrics are sufficient, double-blind testing is the only admissible evidence, and almost nothing is audible. . . .So we have a field that simultaneously argues nothing is audible and yet spends enormous resources trying to make things sound better. At some point we should ask whether those two activities are consistent. Does this sound reasonable? "

Um, "almost nothing is audible"?! Nope, nobody in the objectivist camp I know of has ever said that or anything even remotely close to it, so I stopped reading right there. It's a straw-man argument.

 

[mtrycrafts]

Too bad he didn't mention Toole's work at the National Research Council (NRC), Canada, before Harman.

 

[jinjuku]

Say what?:

What bothers me about both tribes is not that they disagree. It’s that neither is curious. Both have found a way to stop the inquiry and call it a conclusion.

The objective side from my experience is insanely, innately, curious. Instrumentation is used to support hypothesis and ensure you aren't chasing your own tail.

My only standard for subjective claimants is that you need to use your ears only. If you can't trust your ears, well neither can I.

Offer to send subjectivist burned in and unused cables, randomly labeled. No test administrator, fully sighted, subjectivist gets to choose duration and frequency of rolling through the cables and go ahead for good measure, give them 30 days.

These dipshits absolutely will not do it. The other thing they won't do however is challenge the logic. It's one of the few logic bombs I figured out that simply shuts them down.

 

[lovinthehd]

From that link (I've underline the pertinent part):

"The objectivists tribe: standard metrics are sufficient, double-blind testing is the only admissible evidence, and almost nothing is audible. "

Um, "almost nothing is audible"?! Nope, nobody in the objectivist camp I know of has ever said that or anything even remotely close to it, so I stopped reading right there. It's a straw-man argument.

Hope you kept reading....

 

[lovinthehd]

Interesting to at least hear more from Purifi's point of view. Seems their products are well accepted based on measurements at least but I'd like to know more about some of the testing they've done on the audibility points.

 

[mtrycrafts]

Say what?:

What bothers me about both tribes is not that they disagree. It’s that neither is curious. Both have found a way to stop the inquiry and call it a conclusion.

The objective side from my experience is insanely, innately, curious. Instrumentation is used to support hypothesis and ensure you aren't chasing your own tail.

My only standard for subjective claimants is that you need to use your ears only. If you can't trust your ears, well neither can I.

Offer to send subjectivist burned in and unused cables, randomly labeled. No test administrator, fully sighted, subjectivist gets to choose duration and frequency of rolling through the cables and go ahead for good measure, give them 30 days.

These dipshits absolutely will not do it. The other thing they won't do however is challenge the logic. It's one of the few logic bombs I figured out that simply shuts them down.

Or, you could just use identical cables without burn-in, label them randomly and ask them which is burned in which is not.

 

[m. zillch]

Interesting to at least hear more from Purifi's point of view. Seems their products are well accepted based on measurements at least but I'd like to know more about some of the testing they've done on the audibility points.

Keep in mind there's what they say they do vs. what they actually do. They are not necessarily the same thing.

Often these companies use the "great people on both sides" argument (at least in their public statements many will see). Their top concern is not conveying what they actually think but rather they want to offend as few people as possible so they don't alienate any segment of the market. . . .So they adopt a middle of the road or wishy washy argument and pepper it with talking points they hear from both sides.

 

[lovinthehd]

Keep in mind there's what they say they do vs. what they actually do. They are not necessarily the same thing.

Often these companies use the "great people on both sides" argument (at least in their public statements many will see). Their top concern is not conveying what they actually think but rather they want to offend as few people as possible so they don't alienate any segment of the market. . . .So they adopt a middle of the road or wishy washy argument and pepper it with talking points they hear from both sides.

I took more the description of the "sides" to be a bit dramatic and done to make a dramatic point about the outer edges of the situation. Something he might do more in a speech than in a paper. The examples of audibility were vague, tho.

 

[lovinthehd]

Post of Sean Olive's with response by Floyd Toole
<iframe src="https://www.facebook.com/plugins/post.php?href=https://www.facebook.com/sean.olive/posts/pfbid02vQRw9qP7H34JwP7LtNA1U345dcMaRSbFL7VaWPDzc5L1VtfdmHQhcNef2awYWQ7jl&show_text=true&width=500" width="500" height="500" style="border:none;overflow:hidden" scrolling="no" frameborder="0" allowfullscreen="true" allow="autoplay; clipboard-write; encrypted-media; picture-in-picture; web-share"></iframe>

 

[m. zillch]

Now that Olive is recently retired, as I understand it, I wonder if he'll be a bit more forthright about this stuff rather than the somewhat cagey responses I've seen from him in the past?

 

[Eppie]

I don't think his one phrase about the objectivist tribe negates the rest of what he was trying to say in the article. I found it an interesting read and recommend you read the whole article.

After the article was a link to another article relating to baffle step. For anyone curious about baffle step in regards to speaker design, it's a good introduction.

What is the Baffle Step?

Why do we round the corners?

purifi-audio.com

 

[m. zillch]

When the guy makes generalizations regarding the "Objectivist camp" who exactly speaks for them? Sometimes I even see major outlets which from my perspective are pretty much indeed "objectivists" say almost completely different things. Take this I just read, with the pertinent part I've underlined, for example

"At its rated 200 watts per channel, the Amplitude16 [~$13K] delivered 80dB SINAD (0.01% THD+N), which remains very good and well below the threshold of audibility. "

source: Recent Audioholics review based on objective measurements

vs. this other objectivist outlet comparing SINAD values it has measured:

source: https://www.audiosciencereview.com/forum/index.php?threads/sinad-measurements.4071/page-13


enlarged section in question:

Can I safely assume "red" means it should be avoided, ASR? If so notice how slightly more than half of the "red" units on ASR's color chart fall between 80 to 90 SINAD? . . . So which is it guys? Units falling between 80-90 "are very good and fall well below the limit of audibility"? Or should they instead be avoided?

For any sane listening environment I personally lean more towards Gene's assessment, by the way. Think about this: even state-of-the-art phono playback costing hundreds of thousands of dollars would have SINAD values around maybe 60 to 70, on their best day, and that's only if they are perfect condition records without any pops and ticks. Also consider this: all music recorded prior to the 80/90s used analog studio tape recorders where the "red line" peaks were routinely placed at even up to 3% THD (to record as high above the noise floor as humanely possible), and in terms of SINAD that would be (pretending for the moment the format is noise free and only has THD contributing to its SINAD value) around 30. [Granted this value, 30, is just for the brief musical peaks, not a sustained value.]

 

[m. zillch]

Here's another way to think about it. I can't find THD specs for the legendary J37 tape recorder which made the Beatles late 60s records such as Sgt Pepper, as well as Floyd's Dark Side of the Moon,

but I have found a claimed noise figure:

"The characteristics highlight a bandwidth that goes from 30 to 15,000 kHz (+- 2 dB) and a signal-to-noise ratio that is truly exceptional for its period, at around 76 decibels in RMS weighted measure. "
source

So even if we pretend for the sake of argument that somehow magically it has no distortion whatsoever, so its SINAD figure [THD+N, expressed in dB] is entirely from noise alone, (and we'll also pretend its unweighted value—what's used for SINAD— is just as good as its weighted value, whereas usually they differ) it means whenever we hear that music we are being exposed to, at best, 76 dB SINAD sound. . . Yet despite being down low in the "red" section of ASR's master SINAD chart, when released on a good format that music is quite listenable, isn't it?

 

[jinjuku]

When the guy makes generalizations regarding the "Objectivist camp" who exactly speaks for them? Sometimes I even see major outlets which from my perspective are pretty much indeed "objectivists" say almost completely different things. Take this I just read, with the pertinent part I've underlined, for example

"At its rated 200 watts per channel, the Amplitude16 [~$13K] delivered 80dB SINAD (0.01% THD+N), which remains very good and well below the threshold of audibility. "

source: Recent Audioholics review based on objective measurements

vs. this other objectivist outlet comparing SINAD values it has measured:

source: https://www.audiosciencereview.com/forum/index.php?threads/sinad-measurements.4071/page-13


enlarged section in question:

View attachment 79719

Can I safely assume "red" means it should be avoided, ASR? If so notice how slightly more than half of the "red" units on ASR's color chart fall between 80 to 90 SINAD? . . . So which is it guys? Units falling between 80-90 "are very good and fall well below the limit of audibility"? Or should they instead be avoided?

For any sane listening environment I personally lean more towards Gene's assessment, by the way. Think about this: even state-of-the-art phono playback costing hundreds of thousands of dollars would have SINAD values around maybe 60 to 70, on their best day, and that's only if they are perfect condition records without any pops and ticks. Also consider this: all music recorded prior to the 80/90s used analog studio tape recorders where the "red line" peaks were routinely placed at even up to 3% THD (to record as high above the noise floor as humanely possible), and in terms of SINAD that would be (pretending for the moment the format is noise free and only has THD contributing to its SINAD value) around 30. [Granted this value, 30, is just for the brief musical peaks, not a sustained value.]

The objective part is true. The measurements and data reproducible by others. That's the point.

Now how you personally interpret those results and report them, is well, subjective.

 

[m. zillch]

Say what?:

What bothers me about both tribes is not that they disagree. It’s that neither is curious. Both have found a way to stop the inquiry and call it a conclusion.

The objective side from my experience is insanely, innately, curious. Instrumentation is used to support hypothesis and ensure you aren't chasing your own tail.

Exactly! The guy is building a false narrative, a straw man, so as to tear it down. In truth only one of the two sides have "found a way to stop inquiry and call it a conclusion". [For example, by their refusal to actually participate in blind testing to back their claims.]

I can't tell you how many times people accuse me and others of (paraphrased):
"Not listening with your ears and instead relying on instruments"

but in truth I jump at the opportunity to participate in properly conducted listening comparisons. Problem is they are rare events/opportunities which often require quite a bit of prep work. [Not necessarily expensive to undertake, mind you, just time consuming to prepare.]

 

[mtrycrafts]

My thought would be that if a certain SINAD and above is consistently not audible, what is there more to research.

 

[Auditor55]

Say what?:

What bothers me about both tribes is not that they disagree. It’s that neither is curious. Both have found a way to stop the inquiry and call it a conclusion.

The objective side from my experience is insanely, innately, curious. Instrumentation is used to support hypothesis and ensure you aren't chasing your own tail.

My only standard for subjective claimants is that you need to use your ears only. If you can't trust your ears, well neither can I.

Offer to send subjectivist burned in and unused cables, randomly labeled. No test administrator, fully sighted, subjectivist gets to choose duration and frequency of rolling through the cables and go ahead for good measure, give them 30 days.

These dipshits absolutely will not do it. The other thing they won't do however is challenge the logic. It's one of the few logic bombs I figured out that simply shuts them down.

If you were allowed to file a class action suit against some of those companies for fraud, I guess a great many of those snake oil products would go away.

 

[Auditor55]

When the guy makes generalizations regarding the "Objectivist camp" who exactly speaks for them? Sometimes I even see major outlets which from my perspective are pretty much indeed "objectivists" say almost completely different things. Take this I just read, with the pertinent part I've underlined, for example

"At its rated 200 watts per channel, the Amplitude16 [~$13K] delivered 80dB SINAD (0.01% THD+N), which remains very good and well below the threshold of audibility. "

source: Recent Audioholics review based on objective measurements

vs. this other objectivist outlet comparing SINAD values it has measured:

source: https://www.audiosciencereview.com/forum/index.php?threads/sinad-measurements.4071/page-13


enlarged section in question:

View attachment 79719

Can I safely assume "red" means it should be avoided, ASR? If so notice how slightly more than half of the "red" units on ASR's color chart fall between 80 to 90 SINAD? . . . So which is it guys? Units falling between 80-90 "are very good and fall well below the limit of audibility"? Or should they instead be avoided?

For any sane listening environment I personally lean more towards Gene's assessment, by the way. Think about this: even state-of-the-art phono playback costing hundreds of thousands of dollars would have SINAD values around maybe 60 to 70, on their best day, and that's only if they are perfect condition records without any pops and ticks. Also consider this: all music recorded prior to the 80/90s used analog studio tape recorders where the "red line" peaks were routinely placed at even up to 3% THD (to record as high above the noise floor as humanely possible), and in terms of SINAD that would be (pretending for the moment the format is noise free and only has THD contributing to its SINAD value) around 30. [Granted this value, 30, is just for the brief musical peaks, not a sustained value.]

Many of those golden-eared ones are in their late 60's and 70's, worried about SINAD.

 

[m. zillch]

My thought would be that if a certain SINAD and above is consistently not audible, what is there more to research.

Problem is there is no one magic transition number for all people in all circumstances. As an analogy when is a room "hot"? For some people if the room rises to even 74 F (23.3C) they run to immediately adjust the thermostat, whereas for people like me I don't usually notice anything is off until the room hits more like ~80 F (26.7C). I suppose the humidity of the room also would vary how humans perceive the heat just like how different music changes our tolerance levels of poor SINAD performance.

My main problem is SINAD [the current, trendy way to describe THD + N but in dB rather than as a percentage] is it smashes together two completely unrelated performance metrics into one number. Me? I want to know an amplifier/DAC's noise (almost always hiss, but technically in rare circumstances these days, also hum) and its THD (total harmonic distortion as well as other forms of distortion, including intermodulation distortion, IM) independently, thank you very much.

Smashing these two values together to concoct a singular number, SINAD, would be like inventing a new spec to quote for cars' performance that smashes together both its 0-60 mph acceleration with its MPG figure (miles/gallon, say under EPA Highway conditions). Um, hello, sure both things are important but in many/most circumstances I'd want to know what they are on their own.