A new channel

November 23, 2021
 by Paul McGowan

Octave Records is growing. From the new studio we’re building, to the creation of new recording technologies that have never been used before, we’re forging ahead to help shape the future of the recording arts.

Because we’ve committed to DSD as our recording medium, we have a number of technological challenges to meet. Among them is the ability to mix DSD which is currently not possible. But, we’re working on that (stay tuned for more info on this subject).

Our chief engineer, Bob Stadtherr, and our amazing hardware guru, Darren Myers, have teamed up to design a state-of-the-art 8-channel A/D converter, the likes of which the world hasn’t yet seen. The A/D converter is the heart of a modern recording studio. The pro versions available today—the few that handle DSD—are just not up to the sonic standards we would expect. So, we built our own. It’s now in the testing stage and I’ll report on its completion when we’re ready.

We also have launched a new YouTube channel specific to Octave Records.

On the channel, we’ll be posting videos of the studio build-out, live recordings, mix sessions, interviews with musicians and engineers, music videos, film showcasing the new equipment designs, and launching some Master Classes on mixing and recording for those wishing to get into the weeds with us on how all this works.

If you subscribe to the new channel, you’ll get an update each time we launch a new video.

Have fun and see you there.

Subscribe to Paul's Posts

50 comments on “A new channel”

    1. Listening to those two encouraged me to check out the original CSN version. I was surprised to find I have two on different versions of the CD, both remastered. Couldn’t find a date on the first, except the original release year of 1969, on Atlantic 7567-82651-2 and remastered by Joe Gastwirt at Ocean View Digital who also did some of the rated ‘Yes’ remasters. The second a 2006 release on Atlantic released through Rhino 8122-73290-2 and with four bonus tracks. This one mastered by Bernie Grundman and Stanley Johnston at BGM, Hollywood and in HDCD. This is the better one, more detail.

  1. Great activity, fantastic seeing anyone pushing recording technology!

    Regarding 8 track: I already heard that „not more than 8 tracks“ is another limitation of DSD recording. Other recording engineers say, this often shows, which SACD‘s can’t be native DSD.

    But what does this limitation exactly mean in terms of possible complexity or concept of a recording and the workarounds?

    1. Not sure where you heard that there can only be 8 tracks. That’s not true. 8-tracks happens to be the industry standard for equipment “chunks” so that A/D and D/A converters are generally built with multiple of 8.

      8 is how we would up with the first 16 channel recorders (2 x 8) and next was 24 channels, and then 32, and so on.

      Octave will be running a 32 channel system of pure DSD.

      1. Ok…a recording/mastering engineer who does both, PCM and DSD recordings told me, that more than 8 channel recordings on DSD are only feasible with a PCM step inbetween. But as I have no know how myself, I can‘t provide any more arguments unfortunately.

          1. Congrats on the new U-Tube channel Paul…Octave Records coming along very nicely (Recordings with 32-channel editing/mixing pure DSD capability)…Amazing!!!

  2. Most existing news, Paul. Best wishes & good luck. Seems to be a strategy (now coming from the other end) as followed by Rupert Neve who designed top tier studio equipment first and later home audio products. Hopefully we finally hear a sound quality matching high end requirements!

  3. I am fully supportive of this. Of course I subscribed to the channel and I look forward to the new knowledge that will come my way. We honestly need more specialized/high performance recordings. I feel that the convenience of technology has allowed many recording artists to work away from real, salt of the earth sounding music. Sorry fellas, auto tune just doesn’t do it for me nor does over processed compressed garbage either.

  4. Ironically, before dCS ever made a consumer audio product, in the late 1980s they used their avionics experience to design professional A/D converters for the likes of Bob Ludwig and others. When SACD was being developed, they designed the A/D converters (the dCS 905 – DSD and DXD) that Bob Ludwig used to master the first SACD in 2000. According the timeline on the dCS website, it was by Guano Apes and called “Don’t Give Me Names”, which was news to me. He also did a Miles Davis remaster for Sony and a live DSD recording called “Django”, for which Bob Ludwig had an assistant engineer called Gus Skinas. Apparently there was 5.1 DSD mix as well. God bless Discogs for this information.

    I suppose demand for these high quality converters (the 905 had a matching D/A converter and an eternal master clock) dried up along with SACD all those years ago, hence the need to design new ones.

    1. I just purchased a CD by the Chick Corea Acoustic Band which is not DSD but it is a quality recording. Chick’s long time recording engineer Bernie Kirsch and Bob Ludwig worked together on this project and these two guys know what they are doing. Bob Ludwig has been one of the top mastering geniuses in the business for so many years. Didn’t know that Gus Skinas worked on at least one project with him.

      RIP Chick

  5. Much needed venture Paul. I’m a huge Kandice Springs fan! Love her voice and music, however I can only stream her. I bought a CD and some of her vinyl and “holy crap!” it sounds honorable! Both! I thought it was my system and started making plans for upgrades only to find out its the recordings… It sounds like they mic her in a tunnel. No depth to the sound or range. I hate to playing either of them. So “getter done” Paul, I’m waiting 🙂 Maybe she should come over to Octave? Now that’s a thought? I’ll have my people contact her people and her people contact your people Paul< yeah that's the ticket! Hey Paul take a listen to her if you haven't already.

    https://www.youtube.com/watch?v=xbSTDLSCfsU&t=1464s
    I think this is the link. This is her with the WDR Big Band. Enjoy

    Keep listening 🙂

    1. Never heard of Kandice Springs, but love Kandace Springs, on vinyl and streaming. I think I have two of her albums and not had a problem with them.

  6. I was surprised to see your new YouTube network three days ago and watched all of the videos. They are extremely well done and I’m really happy that I’m finally starting to learn more about the digital recording process which my knowledge is is really limited.

    It’s been quite some time since Mark Levinson did a demo for me in his former NYC retail store. After that demonstration, I knew that this was a real advance in digital audio technology that was absolutely necessary but it never seemed to progress in the following years. It’s happening now and you are involved in it up to your eyeballs. I wish you good luck and much success in your new SOTA recording studio and the Octave Record brand.

  7. You cannot edit single bit DSD. The issue isn’t the fact that multi-bit DSD is the only way to retain it in the DSD format, and not that attempting to utilize high sample rates in the time domain, because that argument goes out the window with PCM formats like DXD, it is the conversion of formats that you lose the benefits of any process, whether DSD or PCM. (or better said DXD) Loss occurs at every conversion. Sonoma isn’t a new concept, and the development of a hardware device to replicate what has had to be done on a soundcard still doesn’t address the conversion issues. DXD is just as viable as Sonoma at that point. (DXD is ~17Mbps vs. DSD64 at 5.6Mbps for stereo)

    1. This isn’t entirely accurate. Here’s the deal.

      You can easily edit 1-bit DSD all day long. Cut and paste to your heart’s content.

      What you cannot do is mix DSD not change it’s volume. To do that you must either convert the small part you want to change to DXD (PCM with a high bit and sample rate) or convert the entire track to analog and do it all there.

    2. I don’t think “cut and paste” is what you had in mind as editing.

      “you must either convert the small part you want to change to DXD or convert the entire track to analog”. What if you want to change a lot, not a small part?

      This is of course the precise reason why studios did not invest in DSD recording and mastering equipment.

      1. Agreed, Steven, but cut and paste is exactly what I had in mind when I wrote it (or perhaps you meant Barsley’s comment).

        I distinguish “editing” as cut and paste. Snip this bit off or snip it and move it.

        Mixing is very different in my lexicon.

        1. I think he was talking about mixing. How do you deal with equalisation with DSD sources? And compression? The one DSD orchestral recording I bought was drastically lacking in compression and made it unlistenable.

          Thinking about the problems of mixing and even mastering DSD files, makes me think of when there were different mono and stereo mixes. Maybe DSD recordings should be mixed differently for DSD playback and PCM playback.

          If DSD recordings are going to be commercially viable they will be listened to mostly from PCM, so why not just mix the whole thing in DXD? My system is one of several that converts everything to DXD (40/384 in my case), including DSD64 files. I was told native DSD was in the pipeline, but it never happened.

          1. DXD is PCM. They can put whatever fancy names they want but it’s still PCM. That said, what’s critical is how DSD gets converted to PCM. Therein lies the problem.

            To decimate PDM to PCM one must use a digital IIR low pass filter. Those have phase issues. So, conversion ranges from good to awful.

            That’s what we’re working on now, a perfect IIR LP decimator.

            1. I am aware that DXD is PCM. dCS had been designing A/D converters for Bob Ludwig etc. for 10+ years at up to 24/384 PCM. They then developed A/D converters for DSD when SACD started. Because PCM and DSD were being used side-by-side by these mastering engineers, they also developed D/D converters, to convert DSD to 24/352 and 24/384 PCM. These were all devices and data processing developed for professional use before being put in consumer products.

              My own device does a high quality conversion of DSD64 to 40/384 PCM, called “MAT® DSD Core technology”. This was implemented about 5 or 6 years ago when DSD created some interest. The intention was getting the benefits of DSD with PCM. Like dCS 20 years earlier, the device does all its processing at 40/384 and by having a good converter it makes use of DSD files.

              In my unit DSD is enabled over usb, AES/EBU, Toslink and S/PDIF using DoP and by ethernet. I am listening to a DSD64 file at the moment over Roon RAAT.

              It just seems that high bitrate/frequency PCM (DXD) and DSD can complement each other, which is what appears to have been the case when DSD was first used.

  8. “I seem to enjoy thr female voice is much more than male as a general statement”
    Talking about voices in jazz music… I only buy cd’s from women.
    IMO female voices are so much more agile and suited for this genre than male voices.
    Jazz and male singers.. an absolute no-no for me.

      1. Stimpy2,
        I do have a view on politics, religion, masks, vaccination, social distancing.
        Not on music, that is “just” taste.
        And no, I don’t like the 2 men you mention either. Like I said, I can’t stand jazz sung by male voices.
        Male voices are much more attractive to me in rock. Especially the ones that are/were not really great singers, but have/had a voice with (for me) the X-factor, like Dylan, Cohen, Knopfler, Deville, Newman to name only a few.
        All not-so-great singers (understatement), but “something” in these voices attracts me.

        And the voice has to go together well with the kind of music.
        Pavarotti…great voice, divine voice. But don’t let him sing a rock song. For my ears the song is killed then.
        Some years ago, when he was one of the “friends” on “Pavarotti and friends”, Sting said it perfectly :
        “I am not Pavarotti of course, but he is not Sting” (or words to that effect).

        BTW., I don’t know which is more important : opinion or taste 🙂

    1. Thanks, Tony. That’s technically not possible because of the limitations on mixing but…and it’s a big but….we can do better than how it is currently done and, when I say better, I believe without any sonic penalty. Stay tuned.

  9. Paul, I just want to compliment you on the entire Octave endeavor. After a lifetime of upgrades and a system I finally found delightful, I came to the sad realization that the limiting factor in getting even sound was probably not more, expensive equipment but rather getting GOOD RECORDINGS! A great stereo reveals how bad commercial recordings can be, e.g., most CDs recorded in the last century.

    Also, good luck on advancing the state-of-the-art in A/D converters. A veteran of the Analog Devices audio team, I appreciate what a formidable challenge you are taking on! Best of luck.

  10. This is a false paradigm, because electronic mixing is distortion. So is panning, equalization, compression, gating, and added reverb. Every knob and DSP function is inherently distortion of time and space, no matter how low the THD and IMD.

    The only spatial cues in mixed multi-track recordings are panning and artificial reverb. Panning information is represented by one byte per track, as level differences are crudely sensed. Even this is patently artificial, because of cross-room reflections. Out of a 3 minute 16/44 pop track made from a 96 track master, the information content of panning is .0003% of the total information. The artificial reverb is almost always a statistical algorithm, and generally applied to all the tracks after mixing, as if they were in the same point in a theoretical but physically impossible room.

    I found the solution to the DSD problem in the CORRECT way in 2007, and verified the solution from 2012-2017 with ~750 live DSD recordings in my live recording studio. I realize that PS Audio engineers and customers are all sold on the propaganda of mixed and panned multi-mono track, which is FAKE STEREO, but I thought I was getting through to all of you when I laid out the “one ear test” over a decade ago.

    The reality is your hearing adapts to sounds you hear regularly, ‘breaking in’ to panned fake stereo and other deficiencies of the audio chains in your environment. If you did not hear acoustic music regularly as a child, your brain will lack the neural networks for decoding real 3D musical sound, which takes billions more neurons. With no stimulus to grow them, your body conserves energy and this results in a smaller brain. This makes it really hard to break the vicious circle.

    http://www.musicianbrain.com/papers/Hyde_MusicTraining_BrainPlasticity_nyas_04852.pdf

    Real stereo has three components: Inter-Aural Level Differences (IALD), Inter -Aural Timing Differences (IATD) and Head Related Transfer Function (HRTF). These represent about <1%, 10%, and 90% respectively of the 3D information content of real 3D sound waves. These have been well documented for speech and real world 'sound effects', including triangulation through the phase decoding of direct and reflected vectors. OTOH, audiological research since 1932 has shown that humans in post-industrial environments are phase deaf to music. This is significant because within 7 years of the Kellogg & Rice loudspeaker patent, more people in 'developed' economies were learning to hear to music through speakers than live performance. Notably, the participants in the Blumlein experiments were Decca engineers who listened to speakers for a living!

    The mechanisms of neurogenesis, synapse genesis, and synapse programming are stimulated by consistent and coherent sensory inputs. 3D hearing is the highest data rate sense, and highly under-rated because of the selection bias in audio research which has used radio and phonograph listeners for test subjects for the last 90 years.

    HRTF is different for every ear, so there is no way to encode it in a fixed channel recording. You can reproduce a soundfield or image by playing a multi-track recording at one speaker per track, assuming the speaker has the same polar radiation pattern as the target instrument. This is my ultimate solution. Obviously, this only applies to 'destination audio' in dedicated pro level listening roms. Consumers are not going to have a set of speakers to represent all the instruments they listen too. Piano fans have been known to purchase speakers that are piano centric, and so on, but so far I have dozens of speaker configurations for acoustic, electric and electronic music and have not addressed wind instruments.

    Timing differences allows triangulation between direct and reflected sounds in a real room – but there are no algorithms I am aware of to generate this information discretely for the different origination points for each track. Two musicians can't occupy the same place at the same time, so every track has a different set of reflection timings. Traditional near-coincident recording techniques went beyond Blumlein to separate the mics to capture IATD, including ORTF, NOP, spaced omnis, Jecklin disc and derivatives, Schoeps sphere, and binaural heads. This is the best you can do with two channels.

    So back to mixing. If you make a multi-track recording, then you can play it back in an appropriate room with one dedicated speaker per track and mix ACOUSTICALLY to a near coincident pair of microphones using acoustic equalization, acoustic compression, acoustic spatialization including acoustic reverb. This is the true audiophile method to convert a multi-track DSD recording to a quasi-stereo 2 channel recording with REAL IATD as well as REAL IALD for both direct and reflected vectors – because the reflections cross from one side to the other irl.

    I have been recording DSD to near-coincident pair, and using dedicated speakers to amplify instruments to achieve level balance and signal to noise ratio. A live acoustic mix of the speakers and/or acoustic instruments in a room designed as a performance space instead of studio acoustics captures reality instead of 2.1 dimensional cartoon distortions. The people who know BEST what music sounds like – conservatory trained musicians from around the world – say these are the best speakers and recordings.

    Further, concerts using my speakers generate a higher level of euphoria than PA speaker amplification or even 'high end" stereo speakers, which are all designed to be one size fits all, and none well. Rather, PA and Stereo speakers are evolved to be spatially vague. If you can hear where the speakers are in the room the FAKE stereo 'image' collapses. OTOH, real instruments in the room are easy to locate aurally. That alone tells you that stereo speakers are inherently inaccurate spatially.

    One other possible solution: Set up a soundstage with a full band, and anchor a near coincident pair to the floor. Record all your tracks and overdubs in NCP stereo using the room compression and reverb. Then you can mix the tracks, and each instrument will have the reverb signature of the instrument location in the room. This is called "Acusonic process", and was invented by Bruce Swedien who used it on "Thriller" and "Bad" – so what I am talking about is not esoteric, the consuming public responded quite positively. Also note that Michael Jackson and Quincy Jones gave Bruce equal artistic credit for creating those successes.

    I talked to Bruce about my system of speakers that match the spatial parameters of acoustic instruments, and he thought it was a good extension of his technology.

    As an example, here is a recording of my acoustic guitar speaker for an audience of conservatory trained Classical guitarists, Jeremey Andrew Bass playing Daniel Pesca's "Intermezzo":

    http://www.dropbox.com/s/cgi3ctc0nvp3otc/3%20Pesca%20Intermezzo.wav?dl=0

      1. Can you give me an example of a guitar recording you do like?

        One of my references is the 1994 Chesky album “Solo” by Badi Assad, which was a landmark for both audiophile and guitar playing critics.

        I also love Derek Gripper’s “Libraries on Fire”, a little harder to come by.

        1. Let me give you several.

          Al Di Meola, Paco DeLucia and John McGloghlin ‘ Friday night in San Francisco. Best to try to get it from analogue productions for incredible quality.

          Nils Lofgrin ‘Acoustic Live’ CD

          Philip Hi ‘Bach, New transcriptions for guitar’. GSP 1012CD

          I hope that out of the three you will enjoy at least one of these recordings.

        2. Let me give you several.

          Al Di Meola, Paco DeLucia I am John McGloghlin ‘ Friday night in San Francisco’. Best to try to get it from analogue productions for incredible quality.

          Nils Lofgrin ‘Acoustic Live’ CD

          Philip Hi ‘Bach, New transcriptions for guitar’. GSP 1012CD

          I hope that out of the three you will enjoy at least one of these recordings.

        3. I want to give you one other recording from my favorite label… MapleShade Productions. Paul says they use minimalist recording techniques and I think they are spectacular but this is not just a guitar it’s a Delta Blues Trio

          ‘Early Morning Blues’.

          In my opinion this is one of the finest recordings in my collection but it’s all a matter of taste. So if you’re not into the blues don’t purchasing it. You can’t stream it.

  11. Live in the moment native DSD recordings? That’s one way to get around the DSD mixing issue. Wasn’t playing in real time the approach many recording engineers took back in the 50s and early 60s that yielded so many musical sounding recordings?

    Straight to 2 track, no pro tools, no plug ins, magnificent talent and yes indeed, in that era, tubes!

    1. Hello Dr.
      I have a few recordings from the era you talk about that have been remastered from the original tape to DSD. One take, live in a studio, no over dubs or mix. Those recordings catch an intangible presence you don’t hear very often now a days. The equipment may not have been technically superior in this days, but in many respects the recorded sound is.

      1. I have some old blues records ( vinyl ) that were done that way and they are wonderful. They were done with three or four piece bands, setup in one room with one mike and done in one take. Of course, the instruments all bled together, but that’s what makes the recordings so wonderful.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

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