• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

The ASR "objectivists tribe" ;) has to read the Keynotes at the AES160th by Lars Risbo

Status
Not open for further replies.
That is a very nice essay and appropriate for the AES, thanks for posting.

JJ here is an expert in psychoacoustics, I am not. ASR has a section on it. Dolby and many university psychology and medicine researchers have ideas. The tuning of mp3 audio compression on music is an example of the limitations of theory.

My background is listening live to acoustic instruments, I would like to see better microphones over time and I'm excited for Schoeps reviving the AES42 microphone. Microphones generally have low mass. Lower noise is always welcome, but we are doing well.

To the instruments themselves, they have been improved to be louder to fill concert halls. There is a lot of ultra close microphone placement on the instrument which is going to get towards nonlinearities at the SPL limit of the microphone. Does the instrumentalist walk to their seat and plug themselves in, or are they on wireless?

In live performance today we have a lot of digital. So we have delay. That delay is going to be heard by the musician in the monitors, and by the audience if there is direct and sound through the recording console and the speakers.

Speculating, we have in our genes a propensity to localize sound to protect ourselves from predators, though as infants our cortex is focused on learning and recognizing voice. So maybe the psychacousticians can tell us more about the auditory cortex.
 
I read the article. Not sure what to say other than our tribe, as represented by me, is not the extreme point of view he mentions. Granted, many of you do behave that way. :) But it is not me. This is why I listen for distortion and don't just run with THD measurements.

Anyway, I asked ChatGPT Pro with thinking model to summarize it. It didn't do a great job but here it is anyway if you want to get an idea. Prompt was: "Forget history... summarize this article: https://purifi-audio.com/blog/tech-notes-1/unreasonable-audio-innovation-21"

The article is a keynote-style essay from PURIFI titled “Unreasonable Audio Innovation,” delivered at the AES 160th Convention Reception in Copenhagen on May 29, 2026. Its central argument is that audio progress is slowed when people turn useful evidence into dogma, whether from the subjectivist or objectivist side. (PURIFI)

Core thesis​

PURIFI argues that audio needs more curious engineering and less tribal certainty. The piece criticizes both camps:

  • Subjectivists: too willing to dismiss measurements and accept unsupported explanations.
  • Objectivists: too willing to treat existing measurements, masking curves, THD, IMD, and conventional blind-test interpretations as complete answers.
The article’s point is not that measurements are bad. It is that standard measurements may not capture every audible failure mode, especially when the ear’s time sensitivity is involved. (PURIFI)

The “missing dimension”: time​

A major theme is that much of audio measurement and audibility thinking is too frequency-domain focused. PURIFI argues that the ear is not just a spectrum analyzer; it is a temporal pattern-recognition system with very fine timing sensitivity. The article points to interaural timing sensitivity and spatial hearing as examples of hearing’s dependence on microsecond-scale timing cues. (PURIFI)

The author argues that conventional masking curves are often overextended. They may show when one simple tone masks another, but that does not automatically prove that distortion is inaudible in complex, time-varying music. The article gives examples such as distortion during exponential sine sweeps and possible crossover group-delay audibility under dynamic conditions. (PURIFI)

Hysteresis distortion example​

PURIFI uses magnetic hysteresis in loudspeaker motors as a case study. The claim is that standard sine-wave tests may make hysteresis look benign, because hysteresis has memory: the distortion depends on prior signal history, not only the instantaneous signal. That means the artifact can be displaced in time from the masking signal, potentially making normal masking assumptions unreliable. (PURIFI)

This is used to justify PURIFI’s engineering philosophy: reduce distortions even when conventional wisdom says they are probably harmless, because the existing tools may not be looking in the right place.

Floyd Toole and “settled science”​

The article is respectful toward Floyd Toole and says his loudspeaker research was rigorous and valuable. But it argues that some later followers overstate what that research proves. In particular, PURIFI says Toole’s work had caveats, including the “circle of confusion” between recordings, rooms, and loudspeakers, and that those caveats are often forgotten. (PURIFI)

The author questions whether preference targets derived from existing speakers, rooms, and listeners should be treated as final truth. The article especially emphasizes that rooms are not only frequency-domain problems; room modes also smear timing and transients. (PURIFI)

THD as a “surrogate marker”​

The article compares audio metrics like THD to surrogate markers in medicine. THD is useful, measurable, and worth optimizing, but it is not the final endpoint. The final endpoint in audio is the listening experience. PURIFI’s warning is that the industry may be “treating the numbers” instead of the actual perceptual result. (PURIFI)

Final message​

The closing call is for engineers to behave more like good scientists: challenge assumptions, run better experiments, update beliefs when evidence changes, and treat disagreement as a path to sharper understanding rather than tribal warfare. The article cites examples involving Wolfgang Klippel, Einstein, Millikan, Bohr, and others to argue for respectful, evidence-driven disagreement. (PURIFI)

My one-sentence summary​

PURIFI’s article argues that audio innovation requires moving beyond both subjective mysticism and objectivist dogma, especially by developing measurements that better account for time-domain hearing, distortion memory effects, room behavior, and the limits of conventional metrics like THD.
 
Well, I am sorry to say, but there is not much I can take away from that text.

First, a straw man:
I recently read a post on a prominent audio forum that said, paraphrasing only slightly: “the science is now settled, so why aren’t all speakers designed to perfection?”
Click to expand...

And after much talking he arrived at the following, and it is certainly not even a little bit new:
That sustained, respectful, evidence-driven disagreement is how the physics got science gets done.
Click to expand...
Emphasis by me.
If you do not bring any evidence, well, then why bother? This has nothing to do with tribalism (although that exists, too).
 
To me, it's intellectually dishonest to criticize one tribe for being 'tribal' and put the onus on them to rescue the industry, while completely ignoring the other tribe's negative contribution to the development of better audio reproduction.

Also, he says it's 'not science, it's engineering' then uses examples of great scientists to describe 'the ideal method'. Whoops. That wouldn't have passed my desk.
 
Sokel said:
I'll say it one more time, mixing "audibility" * with engineering belongs only to the bin thread to win debates interchanging the two in any way that suit us.

Right or wrong, @Lars Risbo is my exact way of thinking about audio, a wonderful, evolving, exciting field for the curious.

*Edit: of course I'm talking about the nihilistic point of view about audibility, hence the quotes.
Click to expand...

Based on the fact that, even when spending a ton of money, the results are very diminishing... I think we're at the peak of what we can do with audio reproduction given our own limitations. Many audio discussions turn into the assumption that our hearing capability exceeds equipment capabilities. Which may be the case when things aren't optimized, but overall I don't buy it anymore.

The next wave of improvement will inevitably be bio-engineering our sensory abilities... and I want none of that, personally. There's a reson why evolution chose our hearing should just be like it is. When it comes to bio-hacking, no one seems to think about potential and very real potental negatives. May some of the new things you hear detrimental to your enjoyment or even mental health as your brain processes stuff it wasn't suposed to?

I believe Purifi and Ncore and what not measures sublimely and sounds great. But there is a BIG leap of faith into credibly claiming the differences are hearable.
 
pablolie said:
Based on the fact that, even when spending a ton of money, the results are very diminishing... I think we're at the peak of what we can do with audio reproduction given our own limitations. Many audio discussions turn into the assumption that our hearing capability exceeds equipment capabilities. Which may be the case when things aren't optimized, but overall I don't buy it anymore.

The next wave of improvement will inevitably be bio-engineering our sensory abilities... and I want none of that, personally. There's a reson why evolution chose our hearing should just be like it is. When it comes to bio-hacking, no one seems to think about potential and very real potental negatives. May some of the new things you hear detrimental to your enjoyment or even mental health as your brain processes stuff it wasn't suposed to?

I believe Purifi and Ncore and what not measures sublimely and sounds great. But there is a BIG leap of faith into credibly claiming the differences are hearable.
Click to expand...
I agree with this when it comes to electronics, but I read his speech as talking about speakers / drivers primarily. The hysteresis thing is a big selling point for the "ushindi" drivers.

I don't think audio technology is anywhere near endgame when it comes to production OR reproduction. To @EERecordist 's point, mics are nowhere near totally solved. The raw sound from a mic doesn't usually sound that convincing compared to what you just heard in person.

An ideal would be to stick a mic in a seat in a concert hall, then stick some kind of speaker in your room at home, and the output is totally convincing and realistic. From that POV there is a long way to go. Achieving that requires very specific conditions and a lot of money in 2026.
 
kemmler3D said:
...

An ideal would be to stick a mic in a seat in a concert hall, then stick some kind of speaker in your room at home, and the output is totally convincing and realistic. From that POV there is a long way to go. Achieving that requires very specific conditions and a lot of money in 2026.
Click to expand...

Great point, and I wish we could discuss it over a good coffee or glass of wine, since we're local and I have always enjoyed your contributions.

I find myself arguing in circles at times, it is a complicated topic.

How much authenticy do I want? So often, when attending live concerts (which I do regularly, I have a subscription to Symphony Hall and attend several other concert venues...) the question is... do I want that exact sound at home. My answer would be "no". Of course I enjoy live concerts, otherwise I would not bother, but in 90% of cases I feel I trade "pure" sound quality (at home) for the thrill of being there live albeit sacrificing sound quality (venue, setup, loudness level, annoying crowds). Then again, I often use the word "artificial" when talking about recorded music, and listening through our beloved systems *is* a definition of artificial, isn't it? :-) A lot in audio discussions seems to turn a bit circular...
:)
 
Alexander Gödde said:
willing to change their views based on new data
Click to expand...
Unfortunately, there is little new data being produced. We can all claim we are open to new ideas and research but there is so little of it that it's an empty gesture.
 
amirm said:
The article is respectful toward Floyd Toole and says his loudspeaker research was rigorous and valuable. But it argues that some later followers overstate what that research proves. In particular, PURIFI says Toole’s work had caveats, including the “circle of confusion” between recordings, rooms, and loudspeakers, and that those caveats are often forgotten. (PURIFI)
Click to expand...

I wouldn't call those "caveats" that in any way impact the findings - they are the reality of taking something from a controlled environment to the real, uncontrolled world. Watch out for other factors that may infulence the solid and proven fundamental findings. This happens anywhere. Newton didn't measure the falling apple in a 60mph gale, but gravity still stands. :-)

If Purifi claims they have defeated the influence of "recordings, rooms and loudspeakers", I want to vape some of the stuff they use in their engineering teams! :-)
 
"THD as a “surrogate marker”
The article compares audio metrics like THD to surrogate markers in medicine. THD is useful, measurable, and worth optimizing, but it is not the final endpoint."

I have worked in many medicine areas where surrogate markers are used extensively. In many cases, these surrogate markers had demonstrated "HARD ENDPOINTS" as well, so even the regulatory authorities allow them.

To me a very striking example is vaccination for HPV. Theoretically, eliminating these viruses in the young would result in a lower incidence of uterine and related genital cancers. But you can't do a 40 or 50 year old study to show that giving the vaccine to kids would results in these adults having less cancers. So they chose "warts" as a "surrogate endpoint". The existence of genital warts was a good predictor of future genital cancers, so the trials had prevention of warts as an endpoint. When the vaccines were approved, they only had warts prevention as "data". Only decades later it was demonstrated that children who were vaccinated against HPV did not get genital cancers, and that warts did behave as an accurate surrogate marker.

Blood pressure control, LDL levels are good examples of useful surrogate markers.* But you will find these days that many people believe that GLP-1 weight loss is a good surrogate marker for preventing death or other complications. The science HERE is not completely settled. Maybe soon, but there are confounding variables of weight loss itself as a predictor of survival. The statistical analysis will be very tough and controversial. For example, some claim that GLP-1 drugs help with osteoarthritis but there is no biological plausibility for this. It is lower body weight that reduces the impact of osteoarthritis [pain], but people make claims...

When antivirals for hepatitis C were approved, the FDA wanted to know if viral eradication did prevent liver cancers and death. That was a very controversial subject, but now it is clear that viral eradication can be considered a cure.

If Lars wants to say that THD is not fully established, and he says that "listening experience" IS the endpoint, then he has to show the methods and demonstration for this argument. Otherwise, that statement is pure subjectivism. For now!

EDIT: * despite blood pressure control and LDL levels being widely accepted as good surrogate markers of survival (death), a drug manufacturer still has to conduct studies to demonstrate the hard point if they wish to make the claim. Otherwise, they can only claim LDL or BP control WITHOUT the hard endpoint. It costs a lot of money and takes a lot of time to do these studies. Many pharmacies companies try to take some shortcuts but in the end, a good study is what is needed.
 
Last edited:
amirm said:
It is that standard measurements may not capture every audible failure mode . . .
Click to expand...

AudioInstruments.png
 
Last edited:
amirm said:
...audio innovation requires ... developing measurements that better account for time-domain hearing, distortion memory effects, room behavior, and the limits of conventional metrics like THD.
Click to expand...
Sounds reasonable. So, let's develop those better measurements and do some human subject trials to see if they are actually audible.

I fail to understand the "both-sides-ism" that a cursory reading of the article would imply.
 
pablolie said:
But isn't that the case with most hobbies?
Click to expand...
To put a finer point on your comment: I used to work as a recording engineer so I have a pretty good idea how the sausage is made. And perusing the various threads on this site re what participants are listening to, a fairly large number of us are listening to 40 to 70 year old recordings (e.g. Dark Side of the Moon is 55, Kind of Blue is 68).

Regardless of what we do with reproduction, we're not going to improve on the source material, and if the lead vocals were recorded on an SM-58, you can't tell me that a $50k speaker is going to make it any better.

Now, if the technology was progressing to where current recordings using the latest technology were that much better sounding than the "old favorite" recordings I'd be interested. But while I'm not up on the latest trends in the record industry, my limited sample set indicates that it is not. So, I agree we should spend our time more productively - by listening to music.
 
test1223 said:
Hello,


I think everyone here in the town hall of "the objectivists tribe" ;) has to read the great keynote by Lars Risbos.
purifi-audio.com

Unreasonable Audio Innovation

Keynote at the AES160th Convention Reception, Cityhall of Copenhagen,May 29th, 2026
purifi-audio.com purifi-audio.com

Thank you @Lars Risbo for your great Keynotes.

Best
Thomas
Click to expand...

It is interesting. Thanks for providing the link.
 
About objectivists being "too willing to treat existing measurements, masking curves, THD, IMD, and conventional blind-test interpretations as complete answers."

I think that is a mischaracterization, or at least usage of a way too big brush. I consider myself an objectivist, but never ever have I thought that the available measurements and tests constitute a complete answer. I just like the scientific approach, so I'd say that they constitute the best rational answer we have at the moment, and anybody challenging that answer should provide objectively valid arguments to support their challenge.
 
Lars is right about many things, but I'll make a few contrary comments:
  • He is right that psychoacoustics is an incomplete field. Not only very complex, but very underfunded.
  • LF localization has very unclear boundaries. Most psychoacoustical experiments stop at 500Hz, and the available data below that is scant. Localization at 30Hz is not clearly supported by evidence, at least in terms of the kind of precise direction sensing normal at high frequencies. At VLF, as far as I've read and experienced, the most that happens is imaging effects rather than any kind of precision. The fact that externalization is occurring (when you hear sound outside of your head) is proof that there is some spatial information, but it's a stretch to call that localization. These are separate effects, a simple fact being that you can localize sound even without externalization, with headphones. Either way, the research isn't very clear cut here, not the way it is at higher frequencies. Not enough work has been done.
  • He has made the claims about hysteresis distortion many times. As far as I know there is no experimental proof for this kind of haze in speaker drivers. Hysteresis isn't only magnetic (Lars has talked about "grains" in the magnet structure affecting sound in a long interview a few years ago, which I can't take seriously); slot-based cardioid speakers have acoustic hysteresis distortion which is clearly measurable (in the D&D 8c prominently). I think he usually leans on his experience with the clean sound of his drivers when asked what he means. I think the correct psychoacoustic mechanism at play is called roughness, which occurs when tones are close together. As with other things, research is scant.
  • In terms of masking, it has always had spectral, temporal and spatial components: sounds are harder to hear when coming from a similar direction, with a similar bandwidth and frequency distribution and when occurring closely in time. His example doesn't make sense: the fact that you can hear a quiet tone means that it has been unmasked because some of the occlusive factors have moved out of the way, in time, space or spectrum. THD plots as we know them only show the spectral component. There's a lot more to study if we want to meaningfully disambiguate effects.
  • The work to find distortion mechanisms and push the numbers lower is always meaningful. It doesn't need to be supported by audiophile consumerism (although it is, by lower-number chasers) and doesn't need to lead to subjective effects to be worth the time.
Overall he took the opportunity to speak loosely and expansively. I think he would have done better to focus on his work whenever he had the impulse to reach for something bigger. Would have been more interesting.
 
Mr. Swordfish said:
To put a finer point on your comment: I used to work as a recording engineer so I have a pretty good idea how the sausage is made. And perusing the various threads on this site re what participants are listening to, a fairly large number of us are listening to 40 to 70 year old recordings (e.g. Dark Side of the Moon is 55, Kind of Blue is 68).

Regardless of what we do with reproduction, we're not going to improve on the source material, and if the lead vocals were recorded on an SM-58, you can't tell me that a $50k speaker is going to make it any better.

Now, if the technology was progressing to where current recordings using the latest technology were that much better sounding than the "old favorite" recordings I'd be interested. But while I'm not up on the latest trends in the record industry, my limited sample set indicates that it is not. So, I agree we should spend our time more productively - by listening to music.
Click to expand...
Thank you.
I think a lot of audio dialectic these days is confused in the ways we think about science vs philosophy vs metaphysics, and they are all thrown into the mix without enough discipline into what belongs into which.
 
Status
Not open for further replies.
Back
Top Bottom