Some possible reasons why some people find Hi-Res to sound better:
There is a claim that hi-res intermodulation distortion can be euphonic.
There are experts who believe that the brickwall filter used in 44.1kHz CD production is very damaging to sound quality.
Not so sure for no. 1, but it should sound more transparent since the first cut off digital filter happens around 40kHz, much away from CD at 22.05kHz.
For no. 2, most of studio recordings today are mostly done 88.2kHz and above, so for the ADC, digital filter cut off happens above 40kHz. If they want to offer as 44.1kHz for CD and lossless streaming, it is simply down-sample it.
Hi @Marco_de_Jonge!
How can you call a subject well-explored scientifically if you think it’s difficult to arrive at clear scientific conclusions about it? From my point of view, that would be the perfect definition of the very opposite.
This is what @Brian wrote in his post:
I absolutely agree with Brian’s conclusion. I’m currently working on my postdoctoral thesis in this field of research (perceptibility of frequencies far above 20 kHz artificially added to CD-quality music tracks). I’m a member of an international research team. None of us (or any other researcher I Know) thinks this subject matter has been satisfactorily explored. That’s why we’re trying to construct and perform reliable/valid listening tests, which proves a very demanding task. However, such listenings tests are absolutely indispensable — together with well-founded technical explanations and accurate measurements.
In this sense, I’d also like to reply to @joel, who wrote:
Trying to find out what people can hear and how reliably they can hear it (e.g. with the help of scientifically conducted listening tests) isn’t the same as “settling for it”. On the contrary, it’s a prerequisite for both “a better understanding of the science” and “for making real improvements”.
As a researcher and professional musician I’m interested in how to record and reproduce music in the best way possible. We can’t afford to ignore the capabilities and limits of the human ear and brain. If music is “an art of sound in time that expresses ideas and emotions”, then trying to record and reproduce it as well as possible is a passionate attempt at trying to capture the ephemeral. We know it’s impossible and yet we keep trying!!!
I intentionally wrote ‘explored deeply’, not ‘explored well’.
Searching very deeply for evidence on something (audibility or inaudibility) and then not finding it, is not the same as proving its (in)existence. I guess we both agree.
Sorry for yet another analogy 
: People have been searching for proof on the existence of God, Allah, or aliens for ages.
Still, no definitive conclusions have ever been possible : Presence or absence was never proved.
Now, that clearly does not mean that any of both statements is true.
But is IS an indication on how likely it is, to ever find such proof. One could ask himself : does the answer to the question really matter to me, given how unlikely it is to find it ?
That’s the point I was trying to make.
(This should NEVER be read as an insult to religious people. It is not.)
I do not disagree with you, there must at least be differences.
But in my opinion (whatever that is worth) that difference is ‘hi-res can be barely noticably better’ at best.
And at worst, ‘hi-res can have negative side effects’ in the current state of audio equipment. To me, the latter is much more likely. And can be avoided, by just not doing it.
I know I’m likely to be accused of “hairsplitting” for this, but when you used the concept of “depth” in your last post, I considered this an even stronger (I’m almost tempted to say “wronger”) statement in the context of academic research (“deeply explored” > “well explored” IMHO). I’m sure we both agree “depth” (in its figurative sense) usually implies more than sheer quantity. Anyway, now I know what you mean(t).
In contrast to your “orange juice” and “Fiat” examples, which I enjoyed reading and thinking about, your “God” analogy is quite misleading — particularly (but not only) with regard to the subject matter discussed in this thread. “God” (in Arabic = Allah) is believed to be an “eternal” entity (i.e. one wihout a beginning and end) in both Christianity and Islam. By definition, science can’t make any sensible statements about the “existence” or “non-existence” of such an otherworldly entity, because in non-religious contexts the term “existence” necessarily implies “duration” and is usually, quite literally, considered “unthinkable” without it. Against this background, any speculations about the (im-)probability of God’s “existence” are doomed to failure and the overwhelming majority of both scientists and theologians know this (if they don’t, they can’t be taken seriously anyway). In other words, this is a clash of two entirely irreconcilable views of reality and has absolutely nothing to do with the question of whether or not human beings in the here and now of the perceptible world can or can’t hear a difference between HRA and non-HRA.
Among “experts” (understood as people involved in academic research on the subject matter we’re discussing here), it is widely, although not universally, accepted that HRA - with increased sampling rate and bit depth - actually does deliver improved sound quality. However, it is also widely accepted that the difference in sound quality is relatively “small” — in the sense that many people don’t hear it (despite intact hearing and specific listening training). Nevertheless, there’s hard and fast evidence that some people actually can discern the difference between HRA and non-HRA. One of the things my research team and I are trying to gain a better understanding of is how this variation in people’s perceptual abilities can be accounted for. (For example: I don’t hear the difference, although I have perfect pitch and my hearing is above average for a guy my age; on the other hand, a colleague of mine, who has average hearing and is older than me, has proved numerous times that he can hear it (double-blind ABX listening tests -> comparison of music playback with and without ultrasonic frequencies [score 75 to 80 %]).
To come back to your “God” analogy: In contrast to “God’s existence”, which - in the sense I outlined above - can neither be proved nor disproved with the help of research, it’s quite possible and also necessary to make scientifically valid/reliable statements about the (in-)audibility of a completely earthly phenomenon such as ultrasound.
That’s not hairsplitting to me. You are just discussing passionately, and trying to keep a nice tone. Well done I say :).
I don’t fully agree being ‘misleading’, but I’m very sorry that it’s coming across like that.
I understand your ‘God’ point. Not when applied to ‘aliens’. But I won’t make a problem out of that.
Honestly speaking, I should have known this before joining: I just do not care enough for this, to keep up with discussing the subject :(.
From my own perspective, there is nothing to gain by ‘winning’ (lacking a better word?) this discussion. Maybe I should say : by convincing people of my own opinion.
If you don’t mind, I will leave the discussion now. I wish you good luck in your explorations.
(Serious, I mean it. It’s a question that deserves a definitive answer. But make sure that you actually answering the right question : is this a difference, or is it an actual improvement? See the doubt I had with the collected studies above…).
Yes…but in order to sample a waveform as discrete integer values like that, an extra non-information-preserving step (quantization) is required.
Not assuming anything, just stating a fact. I was being very careul not to draw such conclusions.
http://boson.physics.sc.edu/~kunchur/papers/HIFI-Critic-article-by-George-Foster.pdf
This was an interesting read on whether timing is just “wiggle room” or not.
There is something that has been ignored IMO. The typical DAC’s behavior with a 44.1 file versus the same file upsampled to 24/96 or 24/192.
This is what makes the difference IME. The Nyquist analysis is interesting, but the user wants the best experience. The digital filters and analog filters seem to work better on the higher sample-rate track, and therefore it sounds better.
Even my own DAC, which I leave at 24/192 digital filter setting sound better. Must be the analog filters.
Steve N.
Empirical Audio
A really well-written comment.