Audio Science Review Discussion

Off Topic
570

Detection threshold depends on spectrum of distortion and not just its total. This is why I show the FFT spectrum in every review and don’t just quote this number. Sadly no manufacturer I know shows the spectrum of any distortion so one must measure to have the data to analyze.

Note also that SINAD includes both distortion and noise. Invariably products that have high distortion also have less than ideal noise figures. And noise can easily be audible in sensitive tweeters and when music is not playing (i.e. masking doesn’t apply).

As to 1% THD being enough, that relies on masking and listener abilities/training. There is no straightforward research to tell you what to aim for. Fortunately we can get DACs that have vanishingly low distortion that is below threshold of hearing in the absolute. That is no analysis of masking is needed. Since these DACs have very reasonable cost anyway, there is not reason to buy high distortion DACs.

I don’t know why anyone would be in the market for a $5,000 DAC on basis of performance and hence “R&D.” It is almost a rule that the more you, the worse the distortion and noise. Check out this $6,000 PS Audio DSD DAC for example:

We have distortion spikes that are well above noise floor of CD (96 dB) for heaven’s sake. Its noise floor is above that as well:

This is how its combined noise and distortion compare to a $250 DAC&Amp from Topping:

It gets even worse if you sweep frequencies:

Here are my subjective impressions for what it is worth:

"I started the testing with my audiophile, audio-show, test tracks. You know, the very well recorded track with lucious detail and “black backgrounds.” I immediately noticed lack of detail in PerfectWave DS DAC. It was as if someone just put a barrier between you and the source. Mind you, it was subtle but it was there. I repeated this a few times and while it was not always there with all music, I could spot it on some tracks.

Next I played some of my bass heaving tracks i use for headphone testing. Here, it was easy to notice that bass impact was softened. But also, highs were exaggerated due to higher distortion. Despite loss of high frequency hearing, I found that accentuation unpleasant. With tracks that had lisping issues with female vocals for example, the DS DAC made that a lot worse."

You honestly think this kind of “R&D” needs to be paid for? They reinvented the DAC but got it wrong.

Sorry what? If they have controlled listening tests that shows their DAC to a) sound different and b) to sound better than another DAC, they would not want to disclose that? For what possible reason? There is nothing but goodness in putting such information in their marketing material. There is proof of this in how Harman through its Revel speaker line documents such:

“The performance of Revel Concerta™ on-wall loudspeakers has been extensively evaluated in a variety of on-and near-wall installations through blind listening tests in the Revel Multichannel Listening Lab. … None of the competing systems tested under identical conditions performed as well. Remarkably, Revel engineers could not identify a single competing on-wall system they considered suitable for on-wall use, based on either laboratory measurements or double-blind listening tests.”

Are you saying Harman is giving away some secret here by disclosing that they perform double blind tests like this?

To be clear, I am perfectly good if a DAC cost thousands of dollars if a) it had state of the art measurements and b) had superb build quality and looked great. A couple of years ago I tested a $20,000 DAC that had amazing build but testing showed clear bug in implementation of ESS DAC that even $100 DACs have fixed.

So definitely disagree. Showing measurements used to be the standard in 1970s and 1980s. Not only were companies not afraid of showing such specs, but they would go through extensive documentation in their marketing material to explain how they achieved it. Patents can be used to protect the design.

We need transparency both in performance and specifications. There is no defense for doing otherwise. We can’t being asked to spend literally thousands of dollars on a story and fantastical marketing.

12 Likes
571

For an audiophile this sounds absurd. Like knowing everything about a country after reading books about it even though you’ve never been there.
For a technophile it’s nice if life could be that simple. But one day you run into something that wasn’t predicted in the model and you might get confused.
It’s also the technological truth until a new version of the model is released.

2 Likes
572

Knowledge is power. Even reading about countries before you ever stepped foot inside.

Take that from someone that has had to work around the world… it is better to know things than not , in many cases.

11 Likes
573

The absurdity is that many audiophiles don’t seem to be aware, that state of the art science regarding electrical engineering, acoustics and psychoacoustics has advanced to a point, where any new discoveries will refine, rather than topple over, currently accepted knowledge.

So the best approach today is to get electronics with inaudible imperfections for a budget and splurge on best in class most neutral measuring speakers with most even off axis behavior and step response, then treat the room as sensibly as possible while applying state of the art DSP to optimize room response.

Any tilting of the spectral balance may easily be effected ad libidum via DSP thereafter.

Put some extra cash on the counter for esthetics, if that’s important to you.

This way you could get to the real fun in the hobby as swiftly as possible, namely listening to and discovering new music.

17 Likes
574

It would be nice if masking diagrams somehow can be integrated in the numbers. Distortion in loudspeakers is usually much higher, a lot have theoretically audible 3rd harmonic distortion in the mid frequencies. A really very very bad amp could add more distortion in that band, I guess then that amp-distortion might become audible.

No. Your subjective listening (!) confirms that this dac has a (groundplane?) noise issue and emphasizes on the highs. Has a causal relationship between this noise and a ‘clearer’ sound ever been properly investigated? I noticed more HF energy from my DAC when I had a groundloop through the usb power ground.

In the 90-s I worked for a large manufacturer of electro-acoustic products that wrote an AES paper and found a competitor implementing it exactly like it was in the paper in their product. So they decided not to publish any research outcome or technical details anymore. I agree it is frustrating that we now have to read al that marketing nonsense targeted at our emotions in stead of performance but doesn’t microsoft or any company do that too? Unfortunately most audio journalists aren’t of any help either.
A nice site with detailed information about what happened to the audio magazines including an investigation which add followed what positive review is this (in German):
http://www.hifimuseum.de/hifi-magazine.html
There is a lot of information here on classic audio recorders and all that is wrong with it, entertaining and informative.

But I meant someone who believes to know everything about a country by only reading books and not going there. Doing both is best, as you said.

Thanks to Roon we don’t have to spend hours in recordshops not finding what could have been nice new music to us. I do like this discovery of music a lot and don’t want to be disturbed by thinking of the frequency response. That could also happen with the use of only dsp: thinking of changing a filter. I guess most of us found a way to keep the audiophile hobby away from the music-listening from time to time.

2 Likes
575

Totally absurd, esp the speakers and to a slightly lesser degree the amp (and I’ll stay far out of the way out of this whole DAC thing). Last time I went to audition speakers, one of them was Larsen 3’s. I was grooving along nicely to some jazz with them, and I forget what else, thinking these speakers just might be the one, and then I put on one of my favorite Cat Power tracks (The Greatest) and it was as if the song had been totally remixed - and not for the better. I quickly passed on those speakers, and I have no idea how they measured and don’t care.

The statement above proves to me (and I’m sure others) that the measurements above all types have no clue what they are listening to. No baseline, just the ol’ ‘sounds good to me’ trope. I’m sure it does, and that’s great, but that has nothing to do with how it measures. In my opinion. And I also don’t think one needs to spend gobs of money to get great hi-fi, but one does need to LISTEN!

1 Like
576

That sums it up quite nicely.

See quote of my post farther up and take some time reading up on available work by scientists in the field.

:person_facepalming:

5 Likes
577

@Arjan_H Thanks for sharing that link. Hilarious!

Luckily, this is all in the hobby area and not in politics or business. And so there are no dead people in the verbal virtual Internet wars and global (word) battles in the various forums, only slightly mentally injured people and of course masses of “winners”.

hifimuseum.de - Sie sind im Bereich : Hifi Psyche. But there are more categorizations worth defining including some folks here! And for some this is a business.

I know it’s difficult to respond to all the points being made and the 4 hour wait makes it even harder. If only humans could behave well.

One item that has been mentioned at least at a couple of times by @steven44 that @Amir_Majidimehr sells Revel speakers. Can this claim of conflict of interest be cleared up?

578

I founded a company called Madrona Digital about 15 years ago. Our core business was custom integration (full house automation, security, lighting, etc.) but we thought we could also do some retail business. This was a mistake and after a few years, we closed that part, moved out of the retail location and are 100% focused on CI business and work out of office space. Bulk of our business are very wealthy individuals who want the services I mentioned. They have no interest, need or desire for hi-fi gear. To serve that business, we are dealers for Harman as we use a lot of their crown amplifiers for whole house sound (and for some commercial jobs), and some of their in-wall speakers. We have also done a few high-end home theaters.

As part of above, we can sell Revel speakers but since we have no showroom, inventory or customers, that is just an ability we have. Occasionally an ASR member reaches out to me to buy some Revel speakers. I give them a price and sometimes they buy it, sometimes they go and buy it elsewhere.

Unless I forget, I am very up front with the potential conflict of interest and stated it right in the review:

" NOTE : our company, Madrona Digital is a dealer for Harman products (and hence JBL) for custom integration business. So while there is no benefit to us one way or the other how this review goes, feel free to read whatever bias you want into my subjective remarks."

Keep in mind that since bulk of my review involves measurements that are standardized, I have little ability to skew the results in favor of any product. This kind of concern would be a huge issue for a subjectivist reviewer who can say anything and you have no ability to fact check. Not so with me. Measurements speak especially since there are other published measurements of Revel speakers.

To that end, I have to speak the truth when a Harman product doesn’t do well such as above:

*JBL AC25 Listening Tests and Equalization
As I mentioned, I went into this testing without realizing the need for filters. The result was boomy sound with strange emphasis of some frequencies in mid-range. At first I did not want to attempt to develop a filter but took a shot at it anyway:"

I have also posted negative reviews of other Harman products, much to anger of friends I have in the company. Again, measurements speak loudly and my ethics are not for sale in this manner.

Also note how in every review the second sentence is always where a speaker came from. Good luck finding that in other reviews even though it is part of US laws to provide such disclosure.

Keep in mind that Revel speakers tend to do well in testing because they are developed based on all the research we know about as they themselves publish such findings! They conduct double blind testing of speakers prior to release and measure and optimize them meticulously. So it should be no wonder that if you follow science when you evaluate speakers, that many Revel speakers wind up with positive reviews.

Harman has many competitors and I don’t hesitate to give them high marks when they do well. Genelec and Neumann for example compete head and head with JBL in studio monitors and have gotten some of the highest ratings I can give them, better than JBL in many cases.

On home front, KEF does as well. Read this review of KEF R11 Meta I just posted:

Conclusions
We expect excellence, objectively optimized response from KEF speakers and we have that in R11 Meta. My experience with budget coaxial designs is that they give up power handling which to me is a poor trade off. Not here. The R11 Meta has excellent bass handling with very low distortion allowing me to EQ it with no degradation as far as distortion of playback ability. There is a bit of room left here in there for enthusiasts who want the optimal performance to get there with EQ. Result was that even in our living room with many hard surfaces and large space to boot, a single R11 Meta roared to action, delivering optimal and super enjoyable response on every reference track I threw at it. Science and excellent engineering works!

I am happy to recommend the KEF R11 Meta speaker. Not only does it perform well, it is well priced as well.

Bottom line, I “sell nothing.” To wit, there are no ads, promotions or links to any Revel speakers on ASR to go and buy things from Madrona. It is an independent venture with highest standards of ethical conduct and transparency. Objective data is provided that cannot be gamed as can a subjectivist review elsewhere. All of this is known to membership at ASR and anyone who spends a bit of time reading the site without a specific agenda.

14 Likes
579

Thing is, I was auditioning used speakers due to budget constraints, and only had a handful I was interested in to choose from. In the end I did end up with what I considered the most neutral and easiest to listen to off-axis (Audio Physics Compact Classics) - having to be nailed to a ‘sweet spot’ wasn’t in the cards (my listening spaces have been anything but ‘controlled conditions’). I just don’t think, esp with speakers, it’s wise to purchase anything without actually listening, and listening to in your own space. I would not plop down $5k for anything based on either measurements or other listeners subjective experience and/or being already familiar with the brand (in sound synergy but also in reliability, after service etc).

1 Like
580

Indeed, but I actually think we can see the acceleration of cultural change in certain areas where exchanges like this take place, areas where there is collective mythology like religion and here in audiophile-land. These kinds of exchanges can serve as feedback “pumps” that are, for all their messiness, ways of refining our thinking through discourse. I don’t even mind the trolls all that much and occasionally jump on Audiogon to try new discourse approaches with cable fanatics…er, dealers.

In any case, ASR is a great addition to all the other resources in sorting out facts, fancies, opinions, preferences, value systems, and great audio reproduction equipment!

6 Likes
581

I hear you. It makes total intuitive sense. And matches what many audiophiles still think. The catch? It is the wrong position and dismisses decades of research into what makes a good sounding speaker.

It started with Dr. Toole joining National Research Council in Canada in late 1970s. He was curious if there were any similarity between speakers when it came to listener preference for them. He was surprised that not only there were similarity but that there were measurements that highly correlated with listener preference for speakers. No longer did we have to live in the wild west of anything goes when talking about sound reproduction.

Decades have gone by and many, many other studies have been conducted, confirming the same. That a specific set of measurements called “spinorama” or CEA-2034 standard, can predict with high degree of reliability whether a speaker will be liked by someone. And it doesn’t matter who they are. Most will prefer the same thing. See this research from Dr. Sean Olive and crew:

These are four speakers designated as P I B and M. The horizontal axis is the group of listeners and vertical axis is how much they liked the speaker. While rankings of how good a speaker sounds varies, ranked preference order of the four speakers remains essentially the same no matter what the listener group is. Audio Reviewers ranked them the same as college students.

One group was highly critical though: trained listeners. These are people who have gone through formal training to detect speaker colorations. As such, they voted speakers much more critically than say, audio reviewers and college students. What this says is that even though every audiophile seems to think they have God’s gift to heave when it comes to hearing, are really not trained in properly evaluating speakers. Indeed further research showed that reviewers are not even consistent in determining fidelity of a speaker when repeated in a test! Trained listeners were some 10 times more reliable.

The measurements that predict the listener preference look like this:

The comments on the graph tell you what to look for in that measurement which is flat on-axis response and smooth off-axis – the very thing you said you value in your other post. If you get these two things right, then from tonality point of view, the chances of you not liking the speaker is extremely small.

But let’s say you fall in that exception. Well, you are here which means you have Roon player. You can use its EQ to tailor the tonality to your liking. If directivity is correct, then EQ is highly effective in customizing the sound to any taste.

Note that you don’t have to go all the way here. You can use the measurements to rule out speakers that are clearly colored and have low chance of being preferred. You then arrive at a short list from which you can then audition. This is a massive help as it seems everybody in the world has gotten into speaker making.

I have a tutorial video on understanding speaker measurements if you want to learn more:

Finally, there are countless audiophiles buying speaker completely based on my reviews and measurements and success rate is extremely high. So as a practical matter, the value is there as well.

17 Likes
582

What’s absurd is this comparison. Countries are many many orders of magnitude more complex than the very simple and well-understood practice of digital music reproduction.

Consider that there are only three functional parts to the digital music chain: the DAC, to turn the digital representation into analog, the amplifier to provide enough power to drive the speakers, and the speakers themselves. We know how to measure the fidelity and noise of the DAC; we know how to measure the distortion and power of an amplifier, and we know how to measure the anechoic spinorama of speakers. Forget all this other crap which salesmen want to sell you, and concentrate on this three, with relatively simple measurements, and you will have an excellent system. The desired characteristics have been tested over and over again in well-designed studies, studies conducted with real people’s ears, and reading Toole and Olive’s papers is a good start to understanding that.

There’s no mystery here, no hidden secrets waiting to be found out, just simple physics already well understood. To think otherwise is to wallow in ignorance, like an adult who believes in Santa Claus and the Easter Bunny.

7 Likes
584

No need to have a tantrum.

Understand what was written - speaker measurements help you create a shortlist of what are well engineered speakers (based on proper science) - then audition and pick whats best for you and your setup.

Then with Roon you can further tune with DSP, and if you so desire can go much further with software.

By the way, are there measurements for the Larsen 3 that you only liked with some genre’s?
Might be something there that could explain what you heard.

7 Likes
585

I don’t want to speak for Amir, but it seems to me he did listen to you and took quite some time to demonstrate his position, with some data to back it up. If your answer to that is “whatever”, I guess it’s you who is not interested in his point of view.

15 Likes
586

What point of view? You were dismissive of all measurements altogether with no justification, no proof points, no research. I have invested in $100,000 speaker measurement system that I explained. I have reviewed 250+ speakers using it to very good effect. Here comes you saying it is all useless and you wanted me to roll over and agree with you?

I am one of those listeners. Per above, I have listened critically to 250 speakers in the exact same situation and environment. Exactly what is your scope of experience in this “listening” domain?

What on earth you are talking about? I listen to music when testing speakers, not test tones. Unlike you though, I have specific tracks I consistently use that are revealing of different performance characteristics of speakers. Your random, ad-hoc listening is of little value in comparison.

How do you know you know how to listen? Did you take a test or just gave that medal to yourself and call it done?

A few years back Harman invited some of the top acousticians in US to visit their HQ to learn about their ARCOS room EQ system. I went there and attended a session where Dr. Sean Olive ran a blind test of tonality errors in a piece of music and asked the attendees to vote. Here is a picture I took:

All these experienced people failed to get past level 2 or 3. I managed to get to level 6. Dr. Olive sail past me even more. If these experienced acousticians can’t do well in such a test, what hope is there for you and fellow audiophiles you speak of?

This is why we measure. Because you are not a reliable instrument.

I can’t find any measurements of Larson 3 but stereophile did measure Larson which is a few generations later:

It was a horror show with all those drivers firing every which way. JA’s measurements are not as accurate and extensive as mine but we can still get an idea:

That peak around 600 Hz followed by sharp dip is going to cause serious colorations, impacting the all important vocals. Company commented that this measurement was no good and JA should have done in-room measurements. Well, JA had done that too:

This is stunningly poor. Yes, there are drivers firing every which way so there is likely some kind of spatial, phasey sound going on there. But as far as all important tonality, this is a broken design.

It takes a ton of mistakes for JA to say the measurements indicate a bad design. Usually he is politically correct and waves over bad stuff. Not so here:

“In my considered opinion, however, the Larsen 8’s measured performance reveals its audio engineering to be flawed.”

See how powerful measurements are in figuring out if a speaker is performant? It is only so if one applies himself to learn instead of trusting ears that they have never tested.

18 Likes
587

You have the ability to skew results in posting your opinions or conclusions because people pay attention. Every single measurement apart from the very bad and very good can be judged differently. That is science too.

588

That “ability” is incredibly limited. People who read ASR quickly learn what the measurements are about. I see people commenting here as well who are familiar with them.

Remember, the measurement basis is research and in the case of speakers based on national ANSI standards. Others have read them and know how to interpret them.

We also have large number of technology experts including top designers and CTOs of companies on ASR. This is on top of the researchers of very science we are talking about. If I am saying slightly the wrong thing, I get called on it.

Sure, some people value my judgements and conclusions. That opinion is not for sale as it is in other places. I state it just like an experienced mechanic would if you took him to check out a used car. You are welcome to learn the science and engineering and be free from my interpretations.

I don’t know that you are talking about “science” there. But sure, I get to interpret them and bring my knowledge of science, engineering and testing of so many products. Some people value that. Others just ignore them and read the measurements. You are not force fed to believe my opinion as is the case with purely opinion based subjectivist reviews. Seeing how those reviews are almost always positive, you can be sure that the spin that you worry about is ever present there.

9 Likes
589

Another misunderstanding that I often see in the scientific or medical field is attributing authority to the person and not to the facts certified through the scientific method.

We have seen many examples of this during Covid-19. Doctors, sometimes Nobel Prize winners, disseminate views and opinions without any scientific evidence to support but only “Hey, I have a Nobel Prize in medicine! Believe me!!”

This isn’t about trusting what Amir says because Amir says it. Amir is a spokesperson for the science behind it and provides data that he detects and does not invent.
Precisely because we are talking about science, anyone who has the tools and knowledge can replicate these data (or refute them provided they provide a detailed analysis. And in this case we will try to understand the reason for the difference).

On the other side of the fence, there is the brain. The brain that is influenced by marketing words, by friends, by forums, by “Gurus”, by the eyes that see the object, by the psycho-physical state at the moment of listening, or by a thousand other variables that should not influence the process of judgment and choice.
“…Hey, I have €20,000 that I want to spend on a product that according to science is defective and non-performing, but it’s my money and I do what I want.”

Great!!!

But this only applies to you and has nothing to do with audio or whether a product is actually good or not.

I know many people (me first and foremost) who years ago wanted to approach the world of hi-fi and immediately turned back after being bombarded with “folkloristic” terms and adjectives and when they realized that they had to spend absurd amounts of money on a hi-fi system.
The ASR should help prevent this

6 Likes
590

@Amir_Majidimehr

I am sure you understood my words and the context very well. There are many different types of science and non all of them are measureable.

Nevertheless I am happy like many other people worldwide that there is ASR. It is a great value. We can clearly see that the audio community demanded transparency after decades of only bad advertising and meaningless tests in (online or paper) magazines.