''High Resolution Audio''. An Interesting Article.

[JimC]

I was pointed towards this the other day;

http://people.xiph.org/~xiphmont/demo/neil-young.html

A very interesting read as are all the links mentioned too.

James.

[AndrewR]

I just had a quick scan, but to quote from the article:

The ear hears via hair cells that sit on the resonant basilar membrane in the cochlea. Each hair cell is effectively tuned to a narrow frequency band determined by its position on the membrane. Sensitivity peaks in the middle of the band and falls off to either side in a lopsided cone shape overlapping the bands of other nearby hair cells. A sound is inaudible if there are no hair cells tuned to hear it.

The guy is fixated with the frequency domain and single test tones, using it as some form of proof we do not need higher sample rates. He does not even begin to consider the start-stop timing of the hair cell's response.

He does not even begin to look at Fourier analysis and how higher sample rates lead to better signal accuracy in the time domain - something that those hair cells *may* pick up on.

Andrew

[Quicky82]

Sorry to resurrect an old thread, but I hadn't seen this one and posted a link to the same article as the OP in a new thread..

AndrewR: Your reply is interesting. I remember reading a paper where a researcher had conducted tests to try and gauge how sensitive the human ear/brain system was to time-domain errors. I can't remember where I found this but he was arguing for higher sample rates on the basis that our hearing is much more sensitive to changes in the time domain as opposed to the frequency domain.

I'm wondering how this might correlate with some of the studies cited in the article above that seem to suggest that people can't reliably discriminate between material in 16/44.1 downsampled from 24/192 and native 24/192.

This bit specifically:

This paper presented listeners with a choice between high-rate DVD-A/SACD content, chosen by high-definition audio advocates to show off high-def's superiority, and that same content resampled on the spot down to 16-bit / 44.1kHz Compact Disc rate. The listeners were challenged to identify any difference whatsoever between the two using an ABX methodology. BAS conducted the test using high-end professional equipment in noise-isolated studio listening environments with both amateur and trained professional listeners.

In 554 trials, listeners chose correctly 49.8% of the time. In other words, they were guessing. Not one listener throughout the entire test was able to identify which was 16/44.1 and which was high rate [15], and the 16-bit signal wasn't even dithered!

If human hearing is able to discriminate differences in either way, how come these trials always seem to suggest people can't tell the difference?

In the meantime I'm going to track down some sample hi-res audio and try to find that article..

[walpurgis]

I think its all down to who's doing the listening in tests.

To appreciate variations in sound and sound quality takes 'education'. Most people would have difficulty discerning good sound quality, as opposed to something sounding pretty ordinary, let alone slight changes in characteristics. You need to be used to listening to varieties of good Hi-Fi equipment before your ear becomes adequately educated enough to discriminate.

For example, most teenagers would dismiss anything that does not have booming bass and very prominent treble and many older people like a warm less detailed sound. My wife for instance finds Hi-Fi brash and harsh.

[bobbasrah]

I'm pretty sure it has been raised before that the behaviour of the ear is not so simple as vibrating hairs, and do not act in isolation as a sensor for frequencies. There was a link on one of the prior threads on this to various studies which demonstrated that the traditional frequency range historically accepted was at best, flawed.

[nat8808]

I think its all down to who's doing the listening in tests.

To appreciate variations in sound and sound quality takes 'education'. Most people would have difficulty discerning good sound quality, as opposed to something sounding pretty ordinary, let alone slight changes in characteristics. You need to be used to listening to varieties of good Hi-Fi equipment before your ear becomes adequately educated enough to discriminate.

For example, most teenagers would dismiss anything that does not have booming bass and very prominent treble and many older people like a warm less detailed sound. My wife for instance finds Hi-Fi brash and harsh.

Unfortunately, that then begs the question why bother with 24/192 if you then have to be trained to hear any difference..

Efforts would be much better spent making better recordings and mastering for good sound in the first place.

[WOStantonCS100]

Unfortunately, that then begs the question why bother with 24/192 if you then have to be trained to hear any difference..

Efforts would be much better spent making better recordings and mastering for good sound in the first place.

Therein lies the problem with all subjective tests and why I give zero credence to A/B tests. All you'll get is preference and a percentage close to 50%. This is why all such arguments for or against high def digital (or anything else for that matter) never end; but, repeat themselves in endless debate with neither side willing to give an inch, not realizing it is absolutely pointless and a waste of time.

However, if the question is "why bother with 24/192" (or high def digital in general) and dutiful comparisons of A/B against a constant are made, then you can actually get somewhere. Then there is a reference. The question becomes more akin to, "In comparison to the luxury of a Rolls Royce Phantom, is a Cadillac DeVille closer to the Rolls or is a Toyota Prius." Now, there is a meaningful test because once again if you just ask someone if they prefer the Prius or the Cadillac, for whatever personal reasons and lifestyle, some will prefer the Prius and others, the Cadillac. No harm there. But, include the reference point, and only those who don't accept the commonly agreed upon definition of the word "luxury" would choose the Prius.

Such is the same with high def digital. "In a cloud", CD or high def makes no difference. Measured up against the master tape or a mic feed... totally different story, not to mention how much more of the actual analog waveform was retained in storage at the time of recording. For those that don't believe me, the maximum storage space on a CD is 700MB/80 mins (less if you count the TOC, right?). The average album I've record to DVD at 24/192 is about 2 GB for approx. 45-50 mins of music. There is no question that more of the original analog waveform/sound has been retained in the recording. This can be heard and I dare say one does not need to be trained to hear the difference; just exposed to it.

There's seemingly plenty of talk about what high def can capture beyond the typical limits of human hearing (ie. "bass you only feel" and "highs that raise the hairs on the back of your neck". All well and good. For me, it's the quality of the sound within human hearing that I am most concerned about. It's bringing details into sharper focus between 20Hz and 20kHz, in general, that is the most appealing aspect of the high def digital experience and moving further and further away from trying to fool the ear into thinking it's hearing what isn't actually there or charging the DSP chips with the task of reconstructing the waveform with less information to go on, as opposed to having more information to work with from the outset. In that regard, we're not done. 384kHz ADC's are here.

As an aside: The argument against high def because of "space" issues and/or download time seems quite comical to me. It would take a serious disconnect with history to believe that this is a real issue. The ratio of bytes to physical space to cost per byte over the past 20 years should quell most fears as there is a definite trend there. Fiber optic (or other higher speed connections) straight to the house anyone? Things that aren't affordable or practical now always seem to have a knack for becoming so down the road. Just have to be patient and wait for it.

[sq225917]

I have lots of hi-def, from traceable sources like 2L.No. It sounds great but I cannot reliably tell it apart from the same files down sampled. Mastering is very probably more important that sample rate and word length.

[Ali Tait]

Yep, tend to agree.

[Daniel75]

Very interesting chat about high resolution audio - quite long but worth watching.