Hearing inside

September 29, 2022
 by Paul McGowan

Using only words, one of the more difficult concepts to explain is the idea of hearing into the music.

It is easy to hear but hard to convey.

When in Octave’s studio I can stand in front of the drum kit and hear every last nuance of detail in the cymbals as the musician strokes its metal with brushes. If I close my eyes I notice even more inner details.

I set the microphone at the same point I am sitting and go back into the control room to listen. If I’ve used the right microphone setup remarkably I hear even more detail in those brushed cymbals though now it is without feeling as if I am in the room. (to remedy that I must then add in a touch of the far away room microphones but that’s a different discussion).

Still, I can hear inside those brush strokes. If I then record what I am hearing through those microphones and play it back through the same monitoring system, there are no discernable differences as long as the recording medium is DSD256. If, instead, I set the system up to record in 192kHz PCM something is lost on playback.

That something is inner detail: the individual brushes hitting the cymbal. Live or as recorded with DSD there is no difference. Switching to PCM capture and all the sharp details remain but no longer can I hear into those brush strokes. I want to write that the information is blurred but that would be incorrect. Blurred is softened and the sound is definitely not softened.

Perhaps a better word is muddled or Chaotic as in trying to hear into a conversation in a crowded room where there is no loss of focus. Instead, there’s a loss of intelligibility.

It’s a hard one to describe.

Subscribe to Paul's Posts

54 comments on “Hearing inside”

  1. Even if the control room is absolutely identical with the recording studio while in the control room the studio’s drum set is replaced by a perfect (!) loudspeaker you will never have the same perception of sound due to the fact that drums/cymbals have radiation patterns most different from that loudspeaker. Thus maybe the direct sound might be identical, the reflected sound waves will be most different. Thus shouldn’t you first decide your goal: a) I am there! or b) The drums/cymbals are here or c) a most artificial but great sound conserving the essence of music and the passion of the musicians? Depending on your goals there will be different strategies for the recording set-up and the final mixing.

  2. Yes cymbals are always the easiest part to find limitations of digital (PCM as you say). That’s what also convinced me of some Mofi DSD256 transferred LP’s compared to other PCM processed ones. Although the AAA examples I have for comparison usually still do a little more in terms of quality there (then seems to be a different reason than DSD limitations), the quality of the cymbal sound of the few great Mofi’s has a new dimension compared to the digital sourced vinyl I heard so far on my system. Looking forward to how my DS develops with the Airlens when I can play DSD256 on the DAC then.

    Just too bad that such sound from digital then will just affect something like 0,0001% of a collection or generally available digitally processed music ;-(

    Hopefully our children once when stretching beyond their MP3 streaming, will have a majority of DSD256 albums available from major labels 😉

  3. Paul literally banging on about DSD256 again (sorry – couldn’t resist).

    I find it ironic that dCS, who pretty much invented DSD recording and playback, do not output DSD256. Their latest APEX hardware, entry at $30,000 for the Rossini player, will play DSD256 files but downscales to DSD128.

      1. “Hopefully our children once when stretching beyond their MP3 streaming, will have a majority of DSD256 albums available from major labels”

        You will first have to get wireless headphones to play native DSD256.

        NativeDSD list 262 titles recorded and available in DSD256. Could probably buy the whole lot for 5k, so a 60k player probably doesn’t make a lot of sense.

        Fortunately shimmering cymbals aren’t at the top of my list of priorities. Next outing on Saturday is The Sixteen doing Bach – great recordings on their own award-winning label, Coro, mastered in humble 24/96 PCM.

        1. Yes, although I’m listening to a lot of classical and other stuff, too, as a jazznut, cymbal sound jumps on me in most every recording in this genre since my first stereo 😉

          It was unbearable in early digital, too soft and less open or still artificial in better digital and absolutely satisfying since the DS, but still not open as used from analog.

          Those pure DSD transfers to LP were the first, providing nearly the openness of pure analog processing, combined with good cymbal body and shape. I’m willing to believe it’s possible to have the full performance there also from DSD256 in other cases, although I don’t expect it directly from my DAC when jumping on the DSD256 train with Airlens then. It may need a much higher investment in digital.

          I use an AMT tweeter but I think such differences are apparent also to many others with more conventional ones.

          1. All this said, I’m aware that…measured, at least in the audible frequency range…analog/tape has deficiencies (not only) at the top end against digita. Why listening to the different final media is a different thing, I don’t know.

          2. Having optimized my stereo system over a decade finally including XTC and loudspeaker/room correction from Trinnov most of those. unlistenable digital tracks now Sound most acceptable. And the differences to vinyl record Sound is minimized. 🙂 No giant gap anymore – however the analog source was optimized too recently adding a DS Audio optical cartridge. Wouldn’t it be nice if the Stellar phono preamp had an input option for this optical cartridge? It could motivate me to replace my PS Audio GCPH.

        2. I am in frequent discussion with many sound engineers and manufacturers of Top Tier DACs and most have indeed an audiophile stereo system. – no Yamaha Mini-Monitors. The majority claims that DSD64 doesn’t sound better than 24/88. Thus as always: Trust your own ears! I would encourage Paul to make available downloads of different formats allowing us to seriously check our hearing abilities and (!) the quality of our stereo systems. The download should have tracks of some 30 secs and a random switching between formats. Wouldn’t it be interesting for Paul to get to know the votes of the end users? I fear the majority would need DACs in the Mega Buck range as from Merging Technologies, dcs, Playback Designs, EMMlabs, GRIMM, MolaMola, T+A, CH Precision, NAGRA, Soulution and PS Audio DS DAC of course. 😉

    1. Steven literally banging on about dCS again (sorry – couldn’t resist).

      Steven, I realize that technical things are not your strong suit, but you are very wrong about dCS. They did NOT invent DSD in a way. They did invent the ring DAC which is a low bit, high sampling rate PCM. They did ( along with other companies ) invent DoP.

      I just did a review of the Rossini APEX player specifications. It is a ring DAC, it converts DSD to PCM and accepts DSD64 or DSD128 through various digital inputs.

      There was no mention that it accepts DSD256.

      1. Invented/developed – whatever. They made the hardware for Bob Ludwig to make the first SACD recordings, made the first DDC, pioneered a lot of things in the upsampling and DSD side of professional and latterly home audio. Their RingDAC goes back to the 1980s. I appreciate Meitner were also significantly involved. Besides being involved in this stuff throughout its history, for many people it is the reference for sound quality (just look at how many dealers and other manufacturers use it). It also produced the best sound I’ve heard from a stereo system. If only I could afford it.

        I read a review of the Rossini, the reviewer fed DSD256 and sound came out. He assumed it was downsampled as it can only output DSD128.

        If you look at the RME DAC manual https://www.rme-audio.de/adi-2-dac.html it gives an explanation from page 32 of how DACs output DSD. It seems far more complex than might be thought. I like the comment “While most DACs, even ones seen as ‘Hi-End’, leave the user clueless during DSD operation, …”.

        1. Steven, First, the Rossini ONLY outputs analogue. There is no digital output from the Rossini.

          I read the review that you did and found it very confusing. I talked about the Rossini up sampling to DXD ( which is 24 / 352.8 kHz PCM ), DSD64 and DSD128.

          I went back to the dCS’s website and in the specification for the Rossini found:

          Conversions – User selectable PCM upsampling to DXD, DSD or DSDx2. I am not sure what this means.

          I went to the Stereophile review of the Rossini by JV Serinus ( not my favorite reviewer ) which had measurements done by John Atkinson ( who I trust ) which showed data from reconstructions done for both DXD and DSD.

          The specifications for the Rossini clearly state that the only DAC in the unit is the ring DAC. To do a DXD reconstruction you need a more complex DAC than the ring DAC and to do a DSD reconstruction you need a simpler DAC than the ring DAC.

          I have no idea what dCS is doing in the Rossini. Either their specifications are very cryptic or very incomplete.

          If anyone here understands how the Rossini operates please let us know.

          1. I don’t see why it’s a difficult concept:

            The Rossini can be set to upsample incoming PCM data to DXD, DSD or DSDx2, your preference.

            The result in then sent to the DAC.

            In the Vivaldi it’s done by a separate upsampler box, in the Rossini it’s internal.

            From the owner’s manual:

            “DSD: Set to DSD, the DAC inserts a DSD upsampling stage towards the end of the PCM oversampling sequence, before conversion to analogue.

            DSDx2: Set to DSDx2, the DAC inserts a DSDx2 upsampling stage towards the end of the PCM oversampling sequence, before conversion to analogue.

            DXD: Set to DXD, the DAC uses its standard PCM oversampling sequence for PCM data.

            This feature does not apply to DSD data, which is processed differently.”

            The review in question is indeed confusing and IMHO poorly worded.

            1. The processing of these data format seems anything but simple, so I wouldn’t be so hard on the reviewers. Then you have processing of the data output, another layer of complication (see my reference to the RME manual). Then you have the choices of analogue or digital mastering.

              There are so many possibilities as to what happens digitally to the original microphone signal before it becomes analogue again that it just seems foolhardy to think the consumer will ever know or can make informed objective decisions about these formats. Just sit and listen. If those cymbals shimmer, then everyone’s a winner.

              The good news is that Octave now let you download the format of your choice and the premiums for DSD 128 and DSD256 are OK (still quite a bit more than nativeDSD), so everyone can make their own mind up. I bought a few when you had to buy three formats at once and my system can only play DSD64, which it converts to DXD, so I had three formats to compare (SD PCM, HD PCM and DSD).

      1. In terms of developing systems for professional processing of DSD and then consumer use.

        My understanding is that in the early days a lot of the kit was almost bespoke, dCS made stuff for Bob Ludwig, Meitner made stuff for Channel Classics etc.

        I wonder what Linn Records used, as they made DSD masters for several years. When they opened their download store they converted them to 24/96 PCM as next to no one could play DSD files.

  4. Paul if you record that to a SACD Disk I can hear that inner detail? Has that inner detail ever been captured on an analog recording? Or is it only possible on DSD 256? Did you A/B it to verify what you’re hearing?

        1. I assume besides the non availability of DSD256 on discs, even online streaming DSD256 might not make too much sense (independent of the bandwidth topic), because disadvantages of online streaming vs. streaming from own storage might mask DSD256‘s advantages against lower rates.

          But who would expect DSD256 will get a mass format. We could be happy if more recordings are done in this format and later downsampled for typical media availability up to max. DSD64.

          1. Ever look at the size of DSD256 file for a typical 10-12 track album? In my experience it very seldom fits on a single disc at 2x DSD. (Over 9 gig for an hour stereo playback at 4xDSD?)

        1. Quite the opposite. There was a hint of sarcasm, given Joe’s concern that SACD doesn’t do DSD256. I’ve never played an SACD, but I’m sure DSD64 is just fine.

  5. Paul,

    Very nice test.

    Is it possible that de delay of the input filter before the A/D for PCM is not lineair in the frequenty domain. And for DSD this problem is much smaller.

    kind regards Maarten

  6. This love for DSD256 is a classic example of not seeing the forest for the trees. The whole point of hi-fi is to hear music, not hear details. The analogy is preferring a high-resolution digital photograph of Yosemite Park to an Ansel Adams photograph of the same scene in the park because it has more “detail”.

  7. My initial thought is to agree with hrboucher….

    I guess it comes down to a matter of perspective, and level of detail one wants in any given recording.

    If we go by todays premise only, then no one with digital recordings below a DSD 256 sample rate has ever heard the ‘inner details’ of the cymbal’s being brushed except for the one doing the brushing.

    1. Soo happy to read hrboucher’s and Mike’s responses. The two of you seem to realize that you don’t want to go down the rabbit hole to find Alice. Most of us live in the real world, not inside a bubble. I would agree that what I’m listening for is my perception of what music sounds like in a normal every day listening environment and not a sterile laboratory. If I were working in a bio lab, I would use a microscope to identify miccrobes, The more powerful the microscope the more detail revealed . Those are laboratory conditions.

      When it comes to listening to music in the real world (not in a laboratory), we hear details that we will never hear in real life situations. That’s how we should normally perceive sound or to be more specific, music. Perhaps the real question should be when do we stop delving deeper and deeper for more and more detail that we would never hear in real life situations like our own listening rooms. The word “sterile“ keeps popping up into my head lately when reading what the lab techs here believe is supposed to be a higher quality music experience.

      1. Conductors and musical directors have to make similar decisions. Karajan was famous because of his overall sound, although he was blessed with the best musicians. Some conductors focus on the details at the expense of the overall musical experience. It’s why we go and see or hear the same thing time an again, because it’s never the same.

        I once heard Paul Lewis play the last three Beethoven sonatas twice on the same evening and the performances were different.

        1. I never I had the pleasure of seeing Karajan live but I own many his DGG Beethoven Symphonies. He had a very powerful personality. I really enjoy Carlos Kleiber‘s reading of the Beethoven 5th even though it isn’t a correct interpretation by a longshot. It was ultra exciting though. Just thrilling!

          1. Kajaran voluntarily signed up to the Nazi party and his name was cleared after the war due to the rather tenuous fact that his wife had a Jewish grandparent. He never fully shook off his past. Although just about every major orchestra visited Israel, because of Karajan the Berlin Philharmonic did not until the year after his death, under Zubin Mehta, already long associated with the Israel Philharmonic. I went to the first concert in Jerusalem. Beethoven 3 and the Violin Concerto (Itzhak Perlman soloist). the only time I heard them. A very emotional evening and an overwhelming feeling of relief that HvK had finally dropped dead.

            1. Believe I mentioned Perlman yesterday because he was a customer of mine in New York City. Both he and his wife were so damn cheap. He was quite pleasant though.

              Didn’t American music lovers call Furtwangler
              “My Favorite Nazi“?

  8. I love what you are talking about here, Paul. I really understand what you are getting at. More importantly, I hear what you are talking about with the Carmen Sandim album, “Brazilian Beasts.” I really love the recording of that Octave release and to articulate your point of hearing I’d use this recording as an example to someone to understand how different cymbals /crashes can sound when there is proper microphone placement, more air, and detail around the instrument itself.
    I’m a pretty extreme headphone listener and detail is the name of the game with me. 🙂

    Thank you.

  9. Would it be possible to record and master in the same room at the same time.

    A Direct to Digital DSD 256?

    That eliminates a mixing room.

    I realize that’s impossible to do all the time. Because of time and environmental considerations.

    I suspect while the latest DAC may do DSD 256 within 8 years or so we will be considering a new DAC and everything will sound better at DSD 512.

    Some DACs can playback DSD 1024 natively now though I don’t know anyone that records that way.

  10. Its all relative…. If you are sitting ten feet away from the cymbals? You need to close mic them and sit ten feet away from the speakers.

    To sit ten feet away from your speakers, and record ten feet away from the cymbals? Gives you what you would hear twenty feet away from the cymbals when played. No?

  11. That sounds logical but I don’t believe it’s not the only story.

    For example the Direct to Disc recordings by Sheffield Labs. They were world class. If you’ve never heard one of the original Direct to Disc LPs, you can’t know what I’m referring to. The CD/SACD,FIM recordings don’t come close at all to the Direct to Discs.

    Of course they weren’t perfect.

    But if you tried your method and moved your chair closer to the speakers while playing a Sheffield CD you still would not have the same experience as Direct to Disc.

    I’m sure Paul has sone of those discs in his collection.I’m sure he knows what I’m taking about.

    1. My vinyl record „I‘ve got the music in me“ (Thelma Houston) has incredible PTaT while the Sheffield CD of this recording sounds sterile and uninvolved. I wonder if a todays remaster might recreate the original PRaT.

      1. Is your copy of I’ve Got the Music in Me from the Treasury series or the original Direct to Disc?

        I have the Treasury version of that album. I’ve always thought it was pretty good. I’ve always heard the original was quite remarkable. I’ve never heard the original.

        But I do have several other of the Sheffield Direct to Disc recordings. Among them Amanda McBroom Growing Up in Hollywood Town, West of Oz, also Confederation, Harry James, both Eric Leinsdorf albums, Lincoln Mayorga and Friends and probably several others I can’t remember.

        While probably not perfect there is a liveness to the sound that other recordings just don’t have. I play these very infrequently as to not wear out the records.

        Of course being cut Direct to Disc there’s no opportunity to play with the Fidelity. The recordings speak for themselves warts and all.

        1. The Sheffield CD-2 is from 1991 and the booklet claims that “the CD was made from the analog master tape. The CD offers a dramatic increase in transparency and phase linearity due to our own custom electronics and microphones.” ???? “This Sheffield Lab CD has all the unique musical excitement and energy that can only be found in live, unedited performance recordings.” I cannot agree here with the booklet’s claims. The vinyl record is some 10 years older. Thus either the vinyl EQ process adds PRaT or the digitization robbed PRaT or all my DACs (Meitner BiDAT, MSB Analog DAC, PS Audio DS DAC, LessLoss R2R DAC) were of mediocre quality. 🙂

          1. Ok. The vinyl version you have, while good, is not the same as Direct to Disc. I have the same Lp. A tape recording was made at the same time the masters were cut. That’s what’s been digitized.

            I have many CD versions, and I believe an SACD, and even FIM. The FIM I believe was the best but even it does not compare to the original Direct to Disc albums. The digital versions just don’t sound correct at all. They were made from a tape. You’re right, no dynamics or prat they are flat sounding.

            The Treasury LPs that were put out sound much better than any digital copy. Whoever mastered them knew what they are doing.

            I have a Lampizator Pacific DAC soon to be upgraded to a Horizon. The Lampi DACs play DSD better than any DAC I’ve heard and they play DSD 512 natively.

          2. prat was always an issue with digital. DSD is better here, too. I first noticed from classical recordings of Esoteric. They sound nice but clearly worse than the LP reissues or originals but they have analog ease. Next thing was classical from HDTT tapes on DSD, which have ease, too.

        2. Early in my career as a studio recording violinist, I participated in Several Sheffield Direct to disk recordings. I was a techno geek at the time but not an audiophile-I could simply not afford great equipment. Several times, I sat in that recording booth attached to the MGM Scoring Stage and heard the Monitor feed of recordings That I was not playing on. Lincoln Mayorga and Doug sax really cared and even that minored sound was some of the best and realistic I have ever experienced. It started there. After more the 50 years and hundreds of times in a recording booth I frankly can say that there were only perhaps a dozen recording engineers who really captured what was in the room. If it ain’t there, then there is absolutely no way of capturing the magic-you can’t add magic in a mix. The Direct to disk recording process was exciting, stressful and exileratting. The closest to the live energy of a great live concert recording that I ever had.

  12. Paul finished his intro here searching for a descriptor word.
    He came up with muddled and chaotic.

    Whenever I Have made an upgrade change and it is a positive change, a word that fits is “tidier”. Pretty much what Paul says.

  13. The Sheffield recording chain (I can’t remember the details) was exemplary. It stretched and was optimized all the way from the microphones to the cutting lathes. Lincoln Mayorga and Doug Sax did everything possible to optimize their recording techniques.
    I believe they are both still with us. It would be interesting to find out what happened to their equipment.
    Perhaps it’s in some storage unit waiting to be rediscovered.
    Maybe Paul knows how to reach them and find out.
    The Fidelity of their records was unreal.
    It would be great if Octave could try recording the same way. Direct to DSD 256.
    I realize not all music could be recorded that way. Finding artists with the skills and chutzpah to attempt such a feat would be difficult.
    At one point a copy of “I’ve got the Music in Me” after it went out of print was worth around $1500.00. or more. I have never seen an original copy.
    I do posses several Sheffield Lab originals and they still sound amazing. The Excerpts from Prokofiev’s Romeo and Juliet with conductor Eric Leinsdorf is one of my favorites.

Leave a Reply

© 2022 PS Audio, Inc.

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram