Elliot Mazer on HD Digital, Analog, and Audiophiles (it's not all good)

I was following one of those interesting Internet threads based on and around Elliot Mazer. For those of you that don't about producer, executive, technologist, and project leader (according to Wikipedia) Elliot Mazer, he has been in the music business for decades as a record producer for many classics including Neil Young's Harvest, Janis Joplin's Cheap Thrills, and the music for Martin Scorcese's Last Waltz to name just three very heavy hitters. Mazer also helped develop AirCheck, a system for monitoring when songs are played on the radio and TV, which was eventually sold to Radio Computing Services, Inc (RCS). Lately, he's been consulting to among others, Orastream which is where the string started for me.

If you recall, Orastream was involved in the high rez streaming service for Neil Young's newest Psychedelic Pill. I went from Mr. Mazer's Wikipedia page to Google which lead me to this interview in Sound On Sound and this nice quote:

"With 192kHz, the recording medium is not there, which is slightly different than when you use the best analogue system. The 192kHz format makes listening to digital pleasurable, and it does not sound digital. It moves me more. I feel the same about DSD [Sony's one-bit, super-high sample rate technology used in the Super Audio CD format] but 192/24 PCM is my favourite, and 192 sounds better than 96 by a long shot. I'll bet that 384 will sound even better. The differences are in the width, depth and height of the recording." ~ Elliot Mazer
There are interesting parallels to comments made by Barry Diament and Bruce Brown in their respective Audiostream Q&As. Here's Barry:
"When I first heard properly done 24/192, it was a jaw dropper. For the first time in my experience, those reservations I have always had about digital, where I felt there were some things the best analog did better, simply evaporated. This is, to my ears, a bigger jump up in quality over 24/96 than that was over 16/44. It no longer feels like a great digital recorder or a great analog recorder. It feels like the recorder has been effectively removed from the equation and I am listening directly to the mic feed." ~ Barry Diament
And Bruce Brown:
"To fully capture the nuances of tape, I feel the only way to do that is into either DSD64fs via a Grimm AD1 or to DSD128fs. That is the closest I’ve heard digital to master tape. That is how good tape can be. It trumps everything else." ~ Bruce Brown
Interesting, no?

Then I came across this interview with Mr. Mazer from Tape Op magazine which is where things take a turn for the worse:

Does the audiophile still exist today, or is this a dying breed?
The audiophile breed is disappearing. They mostly don’t listen to music anyway - they listen to sound. There are a few record labels that specialize in that market. They produce “high fidelity” recordings of music that sells equipment.
Ouch! I quickly ran to the nearest mirror and was relieved to see that I am not, in fact, disappearing (if anything, the opposite is true). What's more is when I add up all of these quotes, I have to wonder who Mr. Mazer thinks the audience for 192kHz and DSD recordings is/are? Hint: Audiophiles. And what's more is we are driving the market for HD download outlets like HDtracks, Qobuz, and Linn Records for remastered HD versions of classic records, not just "high fidelity" pap. Harumph!

I'd like to track down Mr. Mazer and see if I can interest him in a Q&A for AudioStream and also let him know that audiophiles come in all shapes and sizes and do not compress so neatly into such a small and unattractive package (but I admit that some audiophiles fit his description to a beefy-T).

I'd also like to thank him for such wonderful records and ask—What's up with Pono?

hotsoup's picture

Can someone suggest a 24/192 download (rock or classical) that will blow my mind? I'd really like to do some more listening to hi-rez. I've heard samples but maybe the quality varies from recording to recording.. need something that really sells the format.

As for dying breeds, hasn't equipment and tweaking always been a part of the hobby? Is the whole hobby dying? Aren't vinyl sales good? Not sure where he's getting that idea..

bdiament's picture

Hi hotsoup,

In my experience (as I've said elsewhere), 90-95% of a recording's ultimate quality ceiling has already been determined by the time the signals are leaving the microphones.  This is before they get through the mic cables, much less get recorded to *any* format.

That's why I find an MP3 of a Keith Johnson recording can often have more "Life" than the original masters from some other engineers.

To the subject at hand (24/192), the problem most listeners will encounter when they attempt to see what the format can do, is that different formats tend to come from different masterings.  Something might have been out on CD for years, then a "high res" version is released but it was done by another engineer, with different sensibilities, probably different equipment and definitely different ears than the engineer who mastered the CD.

The only fair way is to compare different formats that came from the same mastering.  I'm glad to say my label (Soundkeeper Recordings) provides some of these, which can be downloaded for free and compared at the listener's leisure.  The files are in raw PCM format (in this case, .wav files) so no additional processing or expansion is involved in the listening.  I hope you enjoy the comparisons.

The files can be downloaded from the Soundkeeper Recordings Format Comparison page.

Best regards,

hotsoup's picture

Thank you Barry and Michael !

I look forward to downloading these samples this evening.

Michael Lavorgna's picture

And 2L is a good place to start for classical - http://www.2l.no/

Go to their "Test Bench HD Audio Files" page and have at 'em.

steven2583's picture

Maybe I'm simplifying this but if you take a 192/24 and convert it down to 44/16 wouldn't that be the best way to compare the loss in quality between the two formats.  That would remove the engineer differences.  Or is there more to it and I'm missing something? If there is some difference it could be because a different converter is used but you can always try it with different converts. (ffmpeg, itune's, etc)

bdiament's picture

Hi steven2583,

What you're saying is true but in order to have a fair test, the tools used to do the conversion must be as transparent as possible, lest they color the results with their own flaws.

To date, I've tried several dozen sample rate conversion algorithms and an equal number of dither/noise shaping algorithms.  To my ears, only a few are reasonably transparent.  In my experience, most sample rate conversion algorithms tend to harden and brighten the sound and most dither/noise shaping algorithms tend to cloud the soundstage and also alter instrumental timbre.  

Happily, there are some of each type that don't "editorialize" to the same degree others do.  The best of each can create results that sound very much like the unprocessed original.

It is also important that the D-A converters can actually perform well at 192k.  My experience has been that many converters with "192" in their specs actually sound *worse* at this rate than they do at lower rates.  This, I attribute to the significantly increased demands on clocking accuracy made by the higher rates, as well as the demands on analog stages that can truly deliver at the wider bandwidths.  It would seem an easier thing to write a spec sheet than to design a real world device that can deliver on the promise.  Happily, some converters do.  In any comparison test, one must be sure the converters can truly deliver.

Best regards,

P.S. The above assumes the 24/192 original is in fact 24/192.  As we've found, not all "high res" is what it is touted to be, with all too many being in fact, low res in a high res "package".



Bruce B's picture

As Barry stated, many samplerate converters are not worth their weight in paper. Quite a few we have found there is a 15-20 degree phase shift in all freq above 10-12k.

The best test would be to record a master tape at each sample rate so you can hear what each bring to the table.




Dr. AIX's picture

There are also REAL HD-Audio files available at www.RealHD-Audio.com. Just fill out the form and I'll email you the credentials to get a file that won the 2003 CEA Demmy Award for Best High Resolution track. This is a native 96 kHz/24-bit recording and contains a lot more dynamics and high frequency information than any analog tape master. Making a digital track as good as "analog" would be a downgrading of quality not a goal to aspire to.

deckeda's picture

And so I couldn't read Mr. Mazer's comment in context; the link above doesn't have much for me. However, more often than not I've seen the word "audiophile" used as a shorthand description for someone who only cares about sound quality, which of course is only "half right."

What can get lost in converations when professionals hear music consumers complain about this-or-that sound or question a technological production choice is that it is only because we want the best sound for the music we love.

For example, we audiophiles often appreciate nuance just as much or more than dynamics and high frequency information, but hey, you can't get that just anywhere! ;)

We're invested emotionally and financially in ways that engineers and producers sometimes are not! If we didn't care about the music we wouldn't raise such a stink. If an engineer or producer raised a stick --- and lacked the power to affect change ---- they just might be out of a job or reassigned to another project.

The sound quality discussion isn't a distration from the music and should never be interpreted that way by anyone without first doing a double-blind test. (I kid, heh.)