Industry standards

August 20, 2022
 by Paul McGowan

When it comes to the recording industry there are no standards.

There are, however, personal standards for each company.

Take Octave Records for example. Working with DSD is always going to be a challenge with compromises, but so too is analog. In fact, the only analog process I know of without compromise is a direct-to-disc live recording. Everything else is to some degree compromised through process.

DSD can too be direct (but rarely is). One could record live in single rate DSD and then transfer that recorded file to an SACD without any conversions or processing. That’s rare (as are all single-take live recordings). I would estimate that 99% of all analog and DSD recordings have conversions back and forth in the process of becoming finished.

Take the process we at Octave Records (in our new studio) use. We record everything in 4XDSD (11.2mHz). Those files are then low pass filtered to 352.8PCM (DXD), mixed and mastered, then converted back to 4XDSD through the Pyramix DSD rendering process.

Compare this with how we started out working when we used the Sonoma system.

In the Sonoma 1XDSD system we recorded everything in 1XDSD, then ran each track through a DAC to produce analog, mixed on a vintage analog mix board (the Studer), then A/D converted back to 1XDSD. From there we low pass filtered the DSD master to get the PCM versions. The results were excellent.

That entire process was a real milestone of quality that we at Octave Records are proud of. It set a new standard that blows away most analog and PCM recordings.

But now, it’s time to take the next step upward on the journey.

Octave’s new Pyramix DSD system has raised the bar yet again.

Comparing the sound quality output of both methods—the original 1XDSD->DAC->analog->A/D->1XDSD to the new 4XDSD->DXD->4XDSD chain—the new all-digital Pyramix recordings are “light years” better. I can’t wait for people to make their own judgments.

But, this is all new ground and does not adhere to any industry standard.

While it is true there are no industry standards, there are certainly personal standards by which people like us, and Gus, and Cookie, adhere to because we hear the difference in poor choices of technique and abhor compromise simply for the sake of saying something is this way or that.

Each of our techniques may be different, but at the end of the day, it’s the results that matter.

Openness of process is critical. We want you to know how the “sausages are made” because when you hear how extraordinary the results are you’ll want to seek out more of it.

Subscribe to Paul's Posts

38 comments on “Industry standards”

  1. Paul, what will happen to the Studer mixing-console, now that you have your all-digital signal path?

    Why didn't you start first all-digital?

    Best regards,

    1. We will likely sell the Studer as it no longer has much use for us. For now, we'll keep it in case.

      When we first started Octave we did so as an outgrowth of our relationship with Gus who owns the Sonoma system and all its technology. And, we fell in love with the sound and magic of music captured in DSD. Time and again we would demonstrate on the stereo system to musicians (who had never heard of it) their own music as captured in traditional PCM vs. how it sounded in DSD. They were always stunned.

      Once the label started and we got into the nitty gritty of making recordings that we wanted to make, the limitations of the process began to emerge. Mostly it was the difference between what we heard in the mix room vs. what we heard in the stereo room. Countless marches between the two and many frustrated mix engineers later, we realized we needed something different.

      Plus, as my knowledge grew I started to hear what the Studer was adding to the sound. Subtle, but there always. There had to be a cleaner, better way.

      Pyramix had always been available to us, but the investment was huge, the learning curve even larger, and we were already immersed in Sonoma.

      Then, the guys at Four Square productions recorded Zuill Bailey on Pyramix. That tipped the scale. I was in. A million dollars later, we're all Pyramix.


      1. Paul, here are two excerpts of what you said about the various processes planned in Pyramix and why you’d never decided for Pyramix (because in standard it doesn’t sound so good), wouldn’t you use the Zephir process.

        Can you tell about the Zephir process integration and what ADC and format (1 or several bits) you currently use for DSD recording?

        By the way, in this post you in my understanding mistakingly said, Cookie uses DXD mastering, although she seems to be an analog proponent there from what I got.


        Our new plan is to use the Pyramix system to record in 2X DSD (with the option to go to 4X if our ears tell us that’s best), then, using Murrison’s proprietary Zephir low pass filter in post production, convert to DXD for mixing in a DAW that we’ve designed to handle those higher frequencies. The final mix is then either run through an SDM for conversion back to DSD (if we can find or design one we like) or a final stereo D/A, A/D.

        The standard conversion to DXD via Pyramix sounds (to our ears) not so good (I was going to write dreadful but in a kind mood today). This is one of the reasons we didn’t go with Pyramix in the first place. Secondly, their SDM isn’t much better. So, what they get right is the recording process. What we’ve done is (through the Bit Perfect folks) to find a lossless means of getting to DXD (let’s just call it PCM because DXD is a kind word for folks alergic to PCM). Had it not been for the BP Zephir zero phase filter we would never have chosen Pyramix. The problem with the Zephir filter is it cannot be used in real time (it’s a look forward scheme).

        1. I will retract the last part of my comment about the SDM sounding dreadful as well the DXD conversion in Pyramix. Turns out that observation was based on my own ignorance as to how to set up the system, And, I will tell you, setting up that system has whitened my hair far more than my years. 🙂

          (I will also add that some of that negativity came from others not so enamored with Pyramix - lots of prejudice and strong opinions in the pro world - based on a much older version of Pyramix. Merging Technologies continues to evolve the program and we're now "in bed" with the designers who are working with our engineering team. Fascinating stuff and ever evolving.)

          Once setup properly, the system is excellent sounding in all respects.

          We typically us Zephiir to make our conversions to PCM from DSD. Zephiir is a wonderfully transparent sounding low pass filter and converter made by Bit Perfect. When we need to take the 256 (or any DSD) to PCM, this is the filter/processor of choice. Indistinguishable from the original DSD master.

          For the Octave process we continue to lean more on staying within Pyramix through the whole process. The more we learn about settings and technique, the easier/better sounding it all gets. More on this later.

          I am experimenting with the SDM built into HQ Player Pro. It is quite versatile and excellent sounding. Expensive too. Ugh (isn't everything?).

          Lots to learn and plenty of changes afoot.

          1. Ok thanks, I see.

            Saying you use Zephiir WHEN you “need to go DSD -> PCM”, I understood, in your newly established process over Pyramix mixing, you ALWAYS need to.

            Will you still compare with maybe also further improved analog mixing stages than you initially tried, or did you decide to stay in a purely digital process and improve it without further comparisons?

            I could imagine you do, because I guess a parallel sophisticated analog investment could be too much.

            I think for the reason that labels have to focus on one path for money reasons, we might never have a chance to directly compare both sides on equally sophisticated level. Few studios only might have a state of the art analog and DXD path to really compare what’s possible with each.

          2. I’m a believer in DSD (256 only so far) since the Mofi debacle.

            No other DSD recording could achieve this until then because all SACD’s produced in the same way as the corresponding records were clearly inferior so far, no other DSD based records or its source quality had the potential to really show it for me. But although the Mofi’s so far didn’t have the very highest initial recording quality to evaluate if ambiance is as good as in best analog, they showed that a DSD based record on this level can be more or less as good as an all anlog based one.

            The reason why I still limit this to DSD256 as a pure recording format is, because in Mofi’s case for the records it was only used this way. All the rest was analog without further digital conversions or production processes. The SACD’s or final media files with their production process were of a lower quality again. DSD based records of other labels, too. Mofi must have a very good mastering chain (and engineers).

            I know you work on the rest on the digital side. Good luck!

  2. Thanks for this!

    Do you also record on the Pyramix and does the equipment you use record in 1bit DSD or is there another conversion step?

    Could you describe where the Zephir process is integrated into the Pyramix mixing/mastering?

    By the way, I’d not even say an analog direct to disc recording is without compromises. The remaining compromises still lie in the mechanical cutting and the playback process at least. I think you might confuse “compromises” with “transfer or conversion steps”. On both sides. Also in digital in general, even in DSD there’d still be further compromises I’d say, but in case of a direct to DSD recording played back like that, there’d at least be no conversion steps anymore (given the recorder really records in 1bit).

    So far the best DSD I heard was a DSD256 transfer over the analog way on a Mofi 1step or 3step record (beats their SACD”s from the same source by far). I still have to make experience with DSD256 playback on pure digital level.

    Apart from that, my guess so far is, the final quality (if we leave out the most important step, the recording quality itself for the moment), depends the most on the quality and compromises within the conversion phases, may they be analog or DXD. The Mofi analog mastering/cutting chain seems to be extremely good for example (however I still didn’t get from their description if they do the knob twisting when transferring the tape to DSD or during cutting the record respectively during an analog conversion process before making the final SACD format, I guess the former).

    I’d be also interested in the analog equipment and process differences e.g. you used in your previous releases vs. what others like Cookie or Mofi use. I guess there could be huge differences which may relate your DSD/analog/DSD vs. DSD/DXD/DSD experience strongly to your individual choices within those attempts.

    I just mention this because you’re quick with ranking things and partly implying your choices and experiences might be transferable to a more generalistic level, where I guess, you might even not have made valid comparisons that would only make those statements valid (like making a parallel all analog production on tape/vinyl using as good equipment as those do who put as much effort therein as you do for DSD). Some others in fact did this with rather different results than you mention (possibly without having done a real comparison).

    As you say, each of the techniques are different and only the result counts. I think there shouldn’t be claims, which is better by how much, unless a real apples to apples comparison was done by anyone on the best possible equipment (DSD/analog/DSD, DSD/DXD/DSD, all analog, all DXD).

  3. The second method is apparently the standard industry method used by most people recording in DSD, you can read about it in the Pyramix product manual.

    The exception referred to, Channel Classics, mix the recording live in analogue and then transfer to DSD through an amazingly impressive kit list.
    The reason being is that Channel Classics is the baby of Jared Sacks, an industry great, who after years as a professional classical musician became a recording engineer, which he still does 35 years later. So their mixed analogue is transferred to DSD using Pyramix, which is used to splice the final product. Their recordings of the likes of Rachel Podger are well known.

    Channel Classics have just joined OutThere, a Belgian aggregator and distributor of some of the finest smaller classical labels, some use PCM and some DSD. They distribute all their downloads via NativeDSD.

    It's a fairly typical marketing and distribution business model.

    1. A good question also is, what’s “native” DSD? To our knowledge meanwhile not more than something initially recorded in DSD (1bit or not) running through different conversion processes in the following. So it’s just a differentiation against SACD’s produced from 24/96 PCM recordings, against nothing else in my understanding.

      1. I think there is Native DSD from the recording perspective, which is a digital capture that remains as DSD (except when converted to PCM by Pyramix for mixing and editing). You can read it in the manual here:
        This manual is over a decade old and relates to processing for SACD. Whilst this included DSD, the primary purpose of SACD was multi-channel PCM.

        Then there is NativeDSD playback. My system can play DSD64, but not native, as it converts it to 40-bit PCM.

        The limitations come both ways - editing and mixing when recording, the inability to use DSP when playing back in DSD.

        1. Thanks that’s interesting.

          Given due to that manual, that it seems for SACD authoring there’s another conversion to DSDIFF necessary (which again is probably called “lossless”), there’s still so much equipment, cirquits and processes involved, that using any words as “direct”, “native”, “fully transparent”, “lossless” appear as an attempt to embellish and sell things.

          Finally the results and valid listening comparisons count. Unfortunately those seem mostly just done for snippets of the process structure so far, not for a start to end product.

          1. There doesn't seem to be a lot of "pure" about it. Ironically, I was looking at the booklet of the Mervine recording I purchased and on the back cover is a logo for "DSD Pure". I have no idea what that it.

            I suppose one thing you could do is a listening test of the 24/192 and DSD versions of Rachel Podger's award-winning Vivaldi recording on Channel Classics, both from the same extremely high quality analogue source with state of the art conversion (bespoke A/D made by Grimm). There's been no DXD editing in Pyramix. I can't do it as I don't have native DSD playback.

            1. I think there are quite some comparison formats available to compare DSD with PCM and I think DSD will be superior. To compare both in a DSD converting DAC like
              mine just doesn’t seem to be meaningful for me.

              But comparing an AAA and a DSD recording of the same event is another thing. Some talk about assumed results, but few ever made a comparison imo.

              1. I have a copy of the 45 rpm direct-to-disc cut made by Nitin Sawhney at Metropolis about 10 years ago the box also comes with a CD. It is 9 LPs all done in one take. It was fed to digital at the same time as it was cut, I have no idea what recording format. It was of course mixed live, I assume the digital was not post-processed.

                There is a detailed article of the recording process.

            2. “DSD Pure“ is the marketing name for PDM at 2.8 MHz with 6 (?) bits. It’s a special form of DSD64 whose bit depth has been widened to a couple of bits to allow for editing. It’s the format used in the Sonoma DAW.
              (I am not sure about the number of bits in DSD Pure though. Maybe someone here knows better.)

  4. It’s nice to know the process that will be used. For that I say many thanks. But hidden in a small legal ease font somewhere there’s probably a line that states “process subject to change without prior notice”. 😀

    While the content of todays post was informative and seems to be transparent I suspect it was written to avoid a MoFi type back lash.

    The process you describe has been used by a few of the entities I get some remastered files in DSD from. The difference being they are starting with tape. Some of the brand new recordings from brands like IPI are still recorded on highly modified reel to reel recorders before the conversion. What I’m trying to say is the process is not foreign.

    Of course I go right back to the music. No matter how technically great the recording, if the music doesn’t capture me then what is the purpose of me owning? A once in a blue moon ‘show off’ the system or recording exercise?

    Here’s one to for those looking for a direct to bits ( 😉 ) recording…. A brief synopsis…
    “The high level at which these musicians can play is shown by the fact that they played the titles live into the DSD converter via analog mix – direct-to-disc. Honest music without a mains or ground.”
    The bassface swing trio. Bossa ballads and blues
    - Recorded in single rate DSD

    If you’re interested you’ll easily find it in a search. Download or SACD and of course the direct to disc vinyl.

  5. I’d certainly be interested in trying one of your Octave Records DSD download recordings, however, like many even quite high end players I can only play SACD DSD64 or double DSD128.

    I see you now have DSD256 but it would be great to try DSD128 as I understand for actual end user listening there’s much more of a benefit going from 64 to 128 and much less so for 256?

    1. This touches another aspect of what I brought up yesterday. The DSD playback infrastructure is at best difficult. I have no doubt that DSD is superior. I’m sure there is an 8X DSD on the horizon that is going to be superior to 4X. Every time you question it, the answer is always “The format is at the theoretical limit of hearing.”, yet new unsupported sampling rate/formats keeping rippling out.

      When it outpaces the capacity to consume it, you are left in the lurch of needing to provide music in multitudes of the various levels of previously cutting edge sample rates.

    2. Alan - ADT U.K.
      August 20, 2022 at 5:37 am
      “I’d certainly be interested in trying one of your Octave Records DSD download recordings, however, like many even quite high end players I can only play SACD DSD64 or double DSD128.”

      Hi Alan,

      Most, but not all Octave recordings, come with an SACD and another DVD with the DSD and PCM formats.

      Check the specifics on each.
      And call PSA if you need more help.

      1. Yep, I’m aware of that John, I have a few Octave Records SACDs and DSD files at DSD64. My point was that they now offer DSD256 for some downloads as well as DSD64 but not the more widely playable DSD128. Hence my comment that Paul picked up on…

  6. I believe I’m starting to get the hang of this new and rapidly changing technology you laid out in today’s topic and the way you laid it out is making a do a 180. In the past I didn’t want to dip my toe in the water but now I am going to start downloading your new releases.

    You definitely don’t let the grass grow between your toes Paul. Great presentation!

  7. Congratulations, Paul! I am happy for you that you have achieved a recording, mixing and mastering process of which you are so proud that you are excited to tell us how the “sausage” is made!

  8. I'm still happy with all of my vintage Philips, Magnavox, and Conrad Johnson CD players. One bit, 14 bit and 16 bit and sound great.

    For stored and online music I do have two 24 bit audio engine DACs. One is a blue tooth and the other uses a usb cable with a headphone jack built in. Both disable any sound cards in the signal path.

    I just found a rare Philips CDB630 CD player. Dual 16 bit 1541A DA Converters, 4x sampling and 27 bit digital filter. With the awesome 4/19 transport laser mechanism. Has both fixed and variable outputs plus a digital output, all gold plated. This thing looks brand new. The remote looks unused. To find this unit is almost impossible let alone in mint condition. Wow does this thing sound good. Its a keeper.

    Probably going to sell one of my CDB582 and keep the modified one. Keep the one bit modified Philips Conrad Johnson and also the 14 bit Conrad Johnson and the two Magnavox 2041. These will be collectors items.

    I have two new Philips 4/19 laser transports in case I ever need them. Might buy two more. But the Philips 630 is the one that's going to play my CDs in my main system. The sound is absolutely gorgeous from top to bottom. Most of my CDs are from the old analog tapes so 96 db of dynamic range is all I will ever need to play my CDs.

  9. Oh, there are definitely industry standards - whoever sells the most records, no matter what recording quality, is producing the best music. I'm just glad there are a few who don't try to meet this standard.

  10. I am confused about the term "Native" DSD and what we learned ( or maybe I should say relearned ) about the ADC process. It seems that to record something in DSD or do an analog to DSD transfer involves an ADC that is actually using a "multi-bit" DSD digitization process. ( I would still love to find a real explanation of how multi-bit DSD works. )

    What is "Native" DSD? Is it DSD64? If it comes from a multi-bit DSD ADC is that Native DSD? Paul started this post with industry standards. Maybe some standard definitions would be helpful.

    1. We get deeper reaching answers very slowly.

      What I learned/understood by researching and information from others in this regard is the following:

      Ted and other pro’s like from plangent processes say, most DSD capable ADC’s record in few-bit PDM and internally convert this to 1bit DSD or any other desired format. Very few directly record in 1bit DSD (from my research Pyramix does).

      I assume both, 1 bit and few bit PDM is called “native DSD”. I don’t expect industry to be too chintzy with marketing terms.

      What we definitely learned is:

      - There’s no standard for DSD recording/production
      - There are very different opinions of pro’s about each method
      - There’s hardly any „pure, direct, native, lossless, fully transparent, one step etc.”
      - it’s still very good and even better than PCM without all that conversion

  11. This is what I found out so far how a complete production process of an SACD in the Octave ecosystem should look like from what I understood of what Paul described at various places and what Merging/Pyramix describes.

    Although this doesn’t read like a direct DSD recording just put on a disc, it’s probably a very usual process of producing digital media.

    - Recording in 1bit DSD
    (depending on ADC. Most seem to record in few bit PDM and then internally convert to 1bit DSD)

    - Zephir conversion to DXD

    - Pyramix Mixing/mastering in DXD

    - Pyramix conversion to DSD for further media production options (PCM, DSD download or SACD)

    - Authoring the DSD files

    - Creation of a cutting master file

    - Production of the SACD/transfer to disc

  12. As the digital recording equipment, software and techniques advance toward perfection, the limiting factor in recording may be the microphones used to capture the sound waves being recorded. Is there any recent progress in microphone technology that can render voices and instruments even more realistically? No matter how perfect the digital recording process is, the telltale difference between live and recorded to my ears is typically the signature of the microphone(s) used. Microphones just don't perfectly capture all the subtle nuances of live sound.

    1. You surly realize that it is only fanatics such as ourselves that care about this stuff. On the production side it is only Paul and a few others that care about DSD, DSD256, multibit DSD, Pyramix, etc. And do you think the average Joe listening on his music from an iphone into earbuds ever thinks about what microphone was used to to record the music? There probably a few fanatics working on better microphones, but they are few and far between.

      1. To me, I think more effort should go into microphone technology. I have never heard a recording that sounds like a real Steinway, and I blame it largely on the limitations of current microphones. Also, I have never heard a person's recorded voice that sounds exactly like their real voice.

  13. Dear Paul,

    Forgive me, please, if you have answered this question already.

    To moot any theoretical or sonic concerns about DXD, and for the excitement of attempting it, would you consider making a recording “direct-to-DSD” — with on-the-fly microphone mixing, Sheffield Lab style — and avoid the DXD step and conversion back to DSD?

    Thank you.

    1. Well, maybe. Here's the problem. Our best foot forward in recording is DSD256 but that would mean almost no one could play it back (since most people can't play 256). So, then we go down the rabbit hole of compromise.

      I suppose we could have 3 A/D converters going simultaneously each at a different rate. Might be fun. I'll think about that.

      Direct to DSD. What an interesting concept.

      Also, I would have to build or buy an analog mixing board.

Leave a Reply

© 2022 PS Audio, Inc.

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