Paul's Posts — 09 June 2012


Here we go again

I finished reading yet another article by a self proclaimed expert about why 192kHz 24 bit doesn’t make any difference over standard CD’s and the folks who are promoting these standards are simply “hoodwinking” you into believing they do.

What’s instructive about this article is the author, Monty, is extremely knowledgeable about a lot of things – as are most self proclaimed experts.  He knows a lot about a little (or a little about a lot depending on the subject) and no doubt knows a lot more facts and figures than any of us – therefore he’s an expert and we are encouraged to doubt what we know to be true.  Experts have a way of doing this to us.

Problem is, the expert is trying hard to make his argument and he does this by selectively presenting factual evidence that is irrefutable and leaving out the part that doesn’t support his argument.  This is a classic technique used by politicians, pundits and people “in the know” the world over.

This method works really well because when reading the article there’s little if anything one could argue with Monty about – except he misses the point.  Take, for example, Reference Recording’s Symphonic Dances and grab a copy of both the HRx 176kHz 24 bit version and another copy of the redbook version that was originally released.

If you have the ability, play them both on your DAC and tell me if you can hear the difference – have a friend or spouse choose for you so you don’t know which is what.

The HRx disc itself needs either one of our PerfectWave transports to play directly or if your DAC has a 192kHz 24 bit asynchronous input or network player attached you can rip and listen.

I have performed this particular experiment dozens of times with 100% results.  Does that prove that Monty is incorrect?  No.  Monty is correct he just leaves out a lot of valuable information.  For example, he leaves out the fact that nearly every recording mastered at a modern recording studio starts out at a higher sample and bit rate then redbook and to distribute a redbook or vinyl version of the track the mastering engineer must downsample the media – and the downsampled version clearly sounds worse than the master.

Monty also left out the fact that many times mastering engineers, like Keith Johnson, go back and “let ‘er loose” when they can distribute their work in a master media format – where originally they had to throttle back the redbook version.

It’s clear that if you take a redbook CD and upsample it to 192kHz 24 bit you’ve wasted your time and your bandwidth and memory.  It’s equally clear that an original master recording first captured at a high sample rate and bit depth and then downsampled to meet the lower redbook standards will sound remarkably different than the original.

But Monty didn’t bother telling you that part of the story.

Forward to a friend and help us engage more readers Forward to a friend and help us engage more readers

Get new and fresh stories like this each morning by joining the folks reading Paul's Posts. Click here

Related Articles


About Author

Paul McGowan is the CEO and co-founder of PS Audio Inc. a Boulder Colorado design and manufacturing company of high-end audio products and services. McGowan has been designing and building high-end products for nearly 40 years. Hobbies include skiing, music, hiking, artisan bread baking, kick boxing and cooking. He lives in Boulder Colorado with his wife Terri and his 4 sons.

(7) Readers Comments

  1. Is Monty a Python? Sorry, you are not familiar with British culture maybe.
    I think he is a little bit deaf.
    It’s useless to dispute with people who do not hear the difference.

  2. To prove that RBCD standards are inadequate or at least inferior to an alternative theoretically you’d have to prove that the Nyquist theory is wrong.

    To prove it is inadequate in practice you’d have to devise a double blind experiment where the word length and frequency are the ONLY variables being changed. Two different A/D or D/A converters could have other performance differences that are the source of the perceived difference.

    While no one has disproved the Nyquest theory yet, it is known that the 22 KHZ brick wall filter can cause frequency and/or phase distortion in the audible passband. This is the rationale for frequency upshifting 44.1, it allows the filter to be designed to a much higher frequency where its effects are far beyond the audible passband.

    This is a common flaw in so called proofs that one technology in audio is better than another, major differences in many areas of design but focusing on just the one aspect that you want to prove or disprove. This for example was just one of the many major flaws in Cheever’s master’s degree thesis setting out a performance metric to demonstrate why tube amplifiers sound better than transistor amplifiers. He started with four entirely different amplifiers one of which was a tube unit that measured badly, he and his friends liked it better and so he concluded it was better. Then he contrived a long convoluted explanation to justify why it sounds better related to the relative proportion of different harmonic distortion products. One serious mistake, he never established the minimum threshold of harmonic distortion below which any proportion of the products would be inaudible. He was lucky. Had I been on the panel judging his work, he not only wouldn’t have gotten a masters degree in Electrical Engineering, he might have lost his Bachelors degree for it. Junk science may convince tyros but real proof must withstand the scrutiny of peer review by meeting the highest standards of a profession.

  3. A listen is worth a thousand words.I happen to have the Malcolm Arnold overtures from HD tracks in three different formats 16/44, 96/24, and 192/24. I would like to invite Monty to hear this and then tell me there’s no differences.What’s his e-mail address?

  4. OK…..I have some training in digital design 25 years ago…..but I am no expert and probably forgot much of it anyways. What Monty presumes many think is true, but he says is not is the following blurb regarding sampling frequency:

    “And say, “Ugh!” It might appear that a sampled signal represents higher frequency analog waveforms badly. Or, that as audio frequency increases, the sampled quality falls and frequency response falls off, or becomes sensitive to input phase.

    Looks are deceiving. These beliefs are incorrect.

    All signals with content entirely below the Nyquist frequency (half the sampling rate) are captured perfectly and completely by sampling; an infinite sampling rate is not required. Sampling doesn’t affect frequency response or phase. The analog signal can be reconstructed losslessly, smoothly, and with the exact timing of the original analog signal.”

    Captured “perfectly”? Is this a correct interpretation of what Nyquist’s sampling theorem describes? If so, I was one of those people that “thought” (or remembered incorrectly) that all samples taken at less than 1/2 the frequency sampled are noise, and must be filtered out/removed. But I also thought that all samples taken at greater than 1/2 the frequency sampled are not noise but a valid approximation/sample correlated to the data being sampled. I erroneously that that taking more samples, at tighter intervals would produce a more valid approximation.

    So is that part of Monty’s article correct…….is the minimum sample rate, according to the math of Nyquist, just as accurate in reproducing the original signal than the maximum? Is digital “sampling” not a valid approximation like I thought, but an IDENTICAL representation of the analog waveform? Taking all the complexities of filter design out of this idea for now, because I think I (am most of us get that)…. I just want to make sure what Monty said is true about more bits not being better at representing the original analogu waveform. I look at those jaggedy graphs and my brain says it cannot. My ears definitely say it is not…..but the superior baseline sound maybe for other reasons you talk about in downsampling and losses associated with pre-production and mastering, not sample-rate.

    Thanks for helping set me straight on this point.

  5. “I personally don’t do any quality comparison tests during development, no matter how casual, without an ABX tool. Science is science, no slacking.”

    Said Monty Heisenberg.

    ABX is a profitable scam, a weapon in the hands of the ignorant and smug, masquerading as “science”.

    • “ABX is a profitable scam, a weapon in the hands of the ignorant and smug, masquerading as “science”.”

      It seems to me it’s the other way around. It’s the lack of any hard evidence at all that is the scam, the lack of real science. Testimonials as acceptable evidence went out with medicine shows selling cures for baldness except for audiophiles. Fortunately for manufacturers of audio equipment they are way below FTC’s radar because they are so small compared to the market for consumer goods as a whole. To be strictly compliant however they have to be very careful what claims they advertise. Their lawyers (if they have enough revenues to afford to pay a lawyer) will tell them just how far they can go.

  6. Is there more than one PS Audio Perfect Wave Transport? This excerpt from the above is confusing: “The HRx disc itself needs either one of our PerfectWave transports to play directly or if your DAC has a 192kHz 24 bit asynchronous input or network player attached you can rip and listen.”. If there’s only one PWT perhaps it would be better to have written, “…needs either our PWT…or for you to own a 192khz 24 bit asynchronous DAC with a network player attached so you can rip and listen”!!!

    Regarding it being a waste of time and bandwidth to upsample redbook files – some remarkable improvements in sound have been documented in both Stereophile and TAS doing this in the latest software. At the very least it should be mentioned that “IF” the data were originally 196/24 the higher res. filters are as preferable over lower res. filters as 196/24 is to Redbook and so the high res. filters sound better in upsampled Redbook even though the data “might not” depending on what software you use.

Leave a Reply