Here is the english translation of my (french) thread found here.
I started few threads on DIYaudio in the past years about the blind tests I conducted, such as that one and that's basically my overall conclusions. As simple and straightforward as I could express them. It might even be considered blunt by some... So be it.
In any ways, let's discuss 😉
-------------
It had been a long time since I had plunged back into the design, fabrication and adjustment of sound systems. I've been doing it since the age of 14 and I don't remember having stopped that long since. But, frankly, the last blind tests I made disgusted me a little.
Disgusted by the truth, that our ears and our brain make us hallucinate differences when there is nothing identifiable, therefore nothing real, therefore nothing that justifies investing hours and hours in certain parts of our hobby (or this job, for some). In addition to the tons of money that some pieces of equipment can cost.
But this is not a reason to throw the baby out with the bathwater, just focus on the important and avoid wasting time and money on what makes little or no difference. ..
Here, then, from the experience of around thirty years and after several long and tedious blind tests, my summary of the situation in 2020, in our modern era. As science and technology progress, we will discover other things, but for now that's it:
I started few threads on DIYaudio in the past years about the blind tests I conducted, such as that one and that's basically my overall conclusions. As simple and straightforward as I could express them. It might even be considered blunt by some... So be it.
In any ways, let's discuss 😉
-------------
It had been a long time since I had plunged back into the design, fabrication and adjustment of sound systems. I've been doing it since the age of 14 and I don't remember having stopped that long since. But, frankly, the last blind tests I made disgusted me a little.
Disgusted by the truth, that our ears and our brain make us hallucinate differences when there is nothing identifiable, therefore nothing real, therefore nothing that justifies investing hours and hours in certain parts of our hobby (or this job, for some). In addition to the tons of money that some pieces of equipment can cost.
But this is not a reason to throw the baby out with the bathwater, just focus on the important and avoid wasting time and money on what makes little or no difference. ..
Here, then, from the experience of around thirty years and after several long and tedious blind tests, my summary of the situation in 2020, in our modern era. As science and technology progress, we will discover other things, but for now that's it:
- The difference in sound quality between formats: vinyls, CDs, digital MP3 and AAC files as well as streaming does not exist.
Yes of course, you can have a file or streaming that is too compressed and the poor quality will be easily identifiable, but in 2020 the overwhelming majority will be delivered with a level of quality that exceeds what the human ear could detect as a difference to the uncompressed master file from the studio.
For vinyl, it's an urban legend propelled by a surge of nostalgia, quite simply. This cake tastes better because mom made it, kind of thing. The audible differences sometimes heard (and very real) are due to a difference in mastering, so not the same source at all, but when comparing oranges with oranges, there is no audible difference. An easy test is to take a turntable and digitize the playback of a vinyl record, then compare both in a blind test of the ABX type.
The human ear simply does not have the capacity to detect differences, however technically real on paper. We often talk about hearing skills limits, on different aspects, such as: 0.4-0.6db on the amplitude, 1/6 to 1/3 octave on the frequencies, and where an MP3 file as compressed as 160kbps can manage to go unnoticed in a comparison to its high-definition uncompressed counterpart. As an ''audio tool'' the ear-brain combo is unprecise, unreliable and downright weak. No doubt about it anymore.
- Cables, amplifiers, pre-amplifiers, digital / analog converters, all of this has virtually no impact on sound quality.
Either it works, or it doesn't... or the difference is in the simplest specifications, such as power and level of distortion (amplifiers, etc.). Everything else is esotericism. Nothing is audible to humans, in these esoteric differences. Bottomline, just follow the basic prerequisites for XYZ project, in terms of specifications, electrical requirements and common sense, then everything will be OK.
Fortunately or unfortunately, even the speakers do not escape this decapitation of fabulations.
A ''driver'', this component of loudspeakers, it produces frequencies between (x)Hz and (x)Hz at a certain amplitude (decibel) with a certain capacity of acoustical energy diffusion according to the axis (power response), that's about it. Generally speaking, the quality of sound for the human ear is obtained by achieving a linear frequency range which faithfully covers what we hear, 20Hz to 20,000Hz, for young ears and a little less for older ones. Nowadays the tools exist (DSP) to correct and achieve precise results, and it does not cost very much, especially compared to the so-called ''Hi-Fi'' high-end products... The rest is to find recipes for drivers that work well together: very linear frequency range (flat curve) + maximum extensions at the bottom and the top of the spectrum + amplitude capacity (power, decibels) which avoids to fall into their zones of distortions and mechanical limits. My most recent blind test made a very clear demonstration that a 5$ driver can ''beat'' a 2000$ driver when EQ-corrected and SPL-adjusted, within the same bandwith for both. In other words: a DSP can work magic and save you a lot of money.
- The Achilles heel of the audiophile domain is now the measures and the capacities/aptitudes of acting on those measures.
The vast majority of audiophiles, even old timers (especially old timers?) do not know how to measure and make adjustments on a DSP, and/or do not have the proper tools to do so. It takes expertise and experience, which is not given to everyone. So, I think that explains the popularity of sound system solutions that ''self-adjust/calibrate'' with built-in microphones, etc ... I have never heard of such a commercial solution that is really satisfying, but I'm sure it will come someday. It should also be noted that the tooling is also part of the Achilles heel. Indeed, what our ear hears is not the same as our neighbor's pair of ears, and it is very likely that it is far from what the measurement microphone itself picks up. In short, the absolute reference is difficult to obtain and perhaps even downright impossible. The quest for a reliable measurement kit and an correction procedure that satisfies his ears is therefore the key for the modern audiophile and DIY enthusiast.
Last edited:
In this forum you find endless circulating debates about audio questions that have been answered by knowledgable people decades ago. I share most of the opinions of the TO, except the effect of loudspeakers - their directivity pattern is important and cannot be corrected by a DSP.
The conclusion, that the technical aspects of hole hi-end audio theme had been worked out exhaustively in the past will make the value of many "subjective investigations" here questionable at least.
It is interesting to read the word "dared". Sounds a bit like the TO uttered some blasphemy in a religious community.
Perhaps we should try to debate the "frontier" between "objectivists" and "subjectivists".
Lots of nonsense here, especially on the vinyl vs cd and drivers. Vinyl sound "better" to some, because it lacks D/A conversion and, and it's frequency range and harmonics are set entirely direent (and distortion are higher) than cd. For human being clean sound does not sound good (even though by objective measurements it's better), hence for many vinyl sounds better. I would call it rather different, especially when everything is recorded in digital nowadays (some exotic places still record in full analog for vinyl btw).
For drivers... There is much much more than linear frequency range. Apart from stratospheric prices for some drivers, usually when you pay more you get more. But there is again - apart from drivers - much science behind speaker construction than drivers and crossover.
And I take full laugh from people listening wall sockets and fuses, as these have actually no impact besides one's psychics.
For drivers... There is much much more than linear frequency range. Apart from stratospheric prices for some drivers, usually when you pay more you get more. But there is again - apart from drivers - much science behind speaker construction than drivers and crossover.
And I take full laugh from people listening wall sockets and fuses, as these have actually no impact besides one's psychics.
Your results are your results but the fact stands that Mastering Engineers the
world over such as BoB Katz do properly conducted blind tests to evaluate all
manner of digital gear from converters to clocks etc. etc.
At the highest level of sound reproduction and with properly trained listeners,
there are differences.
TCD
world over such as BoB Katz do properly conducted blind tests to evaluate all
manner of digital gear from converters to clocks etc. etc.
At the highest level of sound reproduction and with properly trained listeners,
there are differences.
TCD
I think I can hear the difference between D/A converters, when I expect them - "in our era" - to all sound the same. I think it's more than my brain bending to some knowledge about the component - such as knowing this one converts a digital word directly to a pulse width. There's simply some I like and some I dont, however interesting it would be to know precisely why that is.
I prefer two way speakers crossed to the bass driver well below 1kHz. I've always preferred that - even when I didnt know what it was exactly about a speaker that I was liking. This preference is strong enough that I'd forego other important aspects of a speaker in order to have it, such as polar response versus frequency.
I think DSP certainly marks "our era" in allowing all tuning - what one perhaps did with the classic analog graphic EQ - to take place entirely within the recording format of the music itself, without becoming analog until the split up and individually tuned signals hit the speaker driver terminals.
Further out into the future, there may be opportunities acted upon to kick Mr Katz out of his chair - except for an initial "recommended settings" provided with a recording, which will be provided with all original tracks available (it's only bandwidth and we know that's on a relentless march toward infinity) where the consumer is allowed to mix down, eq, reverb and compress (or not) them all any way they damn well please.
When I worked at Amazon, I suggested this in an employee "next big thing" campaign, as I thought they may have the clout to make it happen with the recording industry - they certainly have the bandwidth. Fell on deaf ears, so maybe someday long after I'm gone will future kids get to play on their music in this way.
Currently enjoying an all digital signal path from the audio file right up to the amp output, 100Hz xover implemented in DSP and a DAC architecture my ears like listening to - for some reason. Not feeling a hint of stress, unless I turn it up way louder than I like normally, which has always been true since the thrill of loudness ("turn it up") went away along with my youth.
I prefer two way speakers crossed to the bass driver well below 1kHz. I've always preferred that - even when I didnt know what it was exactly about a speaker that I was liking. This preference is strong enough that I'd forego other important aspects of a speaker in order to have it, such as polar response versus frequency.
I think DSP certainly marks "our era" in allowing all tuning - what one perhaps did with the classic analog graphic EQ - to take place entirely within the recording format of the music itself, without becoming analog until the split up and individually tuned signals hit the speaker driver terminals.
Further out into the future, there may be opportunities acted upon to kick Mr Katz out of his chair - except for an initial "recommended settings" provided with a recording, which will be provided with all original tracks available (it's only bandwidth and we know that's on a relentless march toward infinity) where the consumer is allowed to mix down, eq, reverb and compress (or not) them all any way they damn well please.
When I worked at Amazon, I suggested this in an employee "next big thing" campaign, as I thought they may have the clout to make it happen with the recording industry - they certainly have the bandwidth. Fell on deaf ears, so maybe someday long after I'm gone will future kids get to play on their music in this way.
Currently enjoying an all digital signal path from the audio file right up to the amp output, 100Hz xover implemented in DSP and a DAC architecture my ears like listening to - for some reason. Not feeling a hint of stress, unless I turn it up way louder than I like normally, which has always been true since the thrill of loudness ("turn it up") went away along with my youth.
I'm not sure how appealing it would be to the casual consumer, but it sure would be interesting (to me, anyway) to have a format that delivered unprocessed stems along with the chosen mastering engineer's attempt, and allow fans / listeners to share their own "masters" using some standardized bundle of mix / EQ / compression / reverb algorithms.
Once all those goals are achieved, should we expect to hear a presentation capable of fooling a listener it's live?
In a sense, this is exactly the case. It is at least my vision of this world, since I sailed there, as an amateur and as a professional many years ago. I say it without petty ulterior motives, it is only my observation.
Unfortunately, or fortunately, it depends of the viewpoint, the entirety of the results of each of my experiments points to the opposite of your assertion. No equipment or acoustic environment, as expensive and high-end as it is, nor any ear, as exercised as it's reasonably possible, can seem to demonstrate, even a little, that the results may somewhat differ.
With all due respect, I think you are completely missing the point.
The central concept of the blind test is precisely to extract the subjective factor from the evaluation. Let's call it the human factor. The human ear, and the brain that deciphers the electrical signals picked up by said ear, are clearly not instruments made for testing or evaluating sound systems.
No, they are rather instruments made to ensure the survival of a human being. Without going into biological or philosophical terms, I think it is imperative to understand that our ears and our brains can play tricks on us.
All of our perceptions are in some sense a 'trick' played on us. Evolution wouldn't allow it any other way. See here :
Imitation and Extinction: The Case Against Reality - Los Angeles Review of Books
Natural selection now has a precise formulation known as evolutionary game theory. Applying game theory to evolution means we can run simulations and prove theorems about its effects. The results are clear. Veridical perception does not assist with survival. In fact, an organism whose perceptions are veridical is less fit than its nonveridical twin whose perceptions are focused on what we call “fitness payoffs,” a central concept in evolution that we’ll explore next. In short, seeing the truth will make you extinct.
Substitute 'hearing' for 'seeing' in that last sentence.
Last edited:
Like almost all audiophiles on the planet, I have always relied on my senses, on my ears, to evaluate audio components. Until the day when I decided to seriously launch myself into blind testing. The revelation was total but also the disappointment and a certain shame. The shame of being fooled for so long. For quite a while, I was even frustrated with the industry, with the pros and the amateurs who spread these lies. But this is no longer the case at all now. I especially understand that everyone must live their experiences and the stage I have reached is not everyone's stage.
Do you consider you've reached the end stage? Or could there be stages beyond where you currently are?
Thank you for having the courage, energy, etc. to post this. I ran out of the energy long ago, but it is important to fight the exploitation of newcomers to the hobby.
Several stages or levels remain to be explored. But since technology (and the industry in general) has gone in a direction that is not taking into consideration physiological constraints, such as the endless high resolution race, I think that the majority of the levels to be explored are at the psycho-acoustic level.
Even as a hobbyst, we should always be mindful of finding a (technical) solution after identifying a specific problem, limit or constraint. And not the other way around. In our case, the bottleneck is clearly human.
I was talking about personal developmental levels in my question as you'd said that people must live their experiences.