Octave Records PS Audio

A little provocative: if I’m right, the fun thing is (although it can be very good), there was never a worse format recordings were made to than the CD format…since the Grammophone record. I just count the cassette as copy medium.

Not meant in its technical potential, but in its common practical performance for top level demand. There tape was better, direct to disc was better (all inspite of their shortcomings) and hires is better.

I’ve been doing a bit of research and the point you make is exactly correct. Bang on the money!

The consumer considers 16/44 to be lossless, because formats below that used by Spotify and others use lossy compression.

In the technical literature, codecs that allow the original file to be reconstructed are considered lossless compression. These include FLAC and ALAC. In both the letter “L” is for Lossless. ALAC does not reconstruct exactly the same file, but close enough.

The starting point is always the published data file, which for digital audio has always been assumed to be 16/44 PCM. So any file that is original 16/44 PCM (e.g. WAV) or can be reconstructed to the original (e.g. FLAC) is considered lossless.

So when 20 years ago Apple sold mp3 size files through the iTunes Store, they were lossy, because they could not be reconstructed to the published data file.

I agree with @elk that if a file was issued as sold mp3, it might be considered lossless, but mp3 is by definition a lossy codec. I’ll leave that conundrum to the philosophers. I’ve never seem an original release in mp3.

If anyone released music ONLY in 24/96, a 16/44 version would be considered lossy. Again, I’ve never seen one.

Likewise, a file released only in DSD64, if converted to 24/192 PCM it would be lossy, but probably a larger file. Again, I’ve never seen a DSD-only release.

So as long as 16/44 PCM is the default size for music releases, it is the benchmark for what is lossless, and any version that can be reconstructed to 16/44 would be considered lossless.

Where @Rushton seems to be misleading us is because his “lossless” is about conversions when the music is being recorded and processed. In the literature the terms “lossy” and “lossless” only relate to the published music file. In the processing stage the music could be encoded in DSD, PCM or on strings of spaghetti, if it’s published as 16/44 PCM, that’s lossless.

There is a disagreement on High Definition:

(1) The US definition marketed by RIAA is 20/48 or higher with this logo
Screenshot 2023-08-23 at 23.33.44

(2) The definition by the Japanese equivalent is 24/96 or higher with this logo.
Screenshot 2023-08-23 at 23.33.53

No, unless consumer definition is relevant; not so for recording engineers, sufficiently revealing systems, and/or discerning ears

Edit…note these comments in one of the articles Rushton shared,

Every conversion to any other digital format after this initial conversion of the analog signal to DSD results in some unintended changes in the audio integrity. It is mathematically impossible to make later conversions without also introducing some artifacts. It is also mathematically impossible to reconvert without introducing yet additional artifacts. (It is not a “lossless” process.)…By converting this initial PDM bitstream into DXD for post processing we unavoidably introduce greater artifact introducing opportunities…This PDM to PCM conversion is necessary for traditional direct binary computer math processing because a PDM bitstream contains no digital value information. Unfortunately, the conversion process of PDM to PCM is a lossy process, meaning that the resulting PCM samples have incurred losses such that the original PDM bitstream can not be recreated from it. Translated: this means there’s a loss of very fine sound resolution detail in the PDM to PCM conversion…the losses in the PDM-to-PCM conversion are audible to some listeners.

Octave converts from DSD/PDM to DXD/PCM and then back to DSD for SACD, DSD download, and vinyl sales.

If you read the literature, from researchers, music industry agencies, even government committees, music quality is only discussed with reference to the published music file provided to the consumer.

These DSD engineers usurp the word “lossless” for their beloved DSD data stream, but by their definition probably 99.9999% of all audio files released are “lossy”, so the term would be meaningless. They are wrong because you can look at every major music distributor, Apple, Amazon, Qobuz, etc, and hifi publication, they all have a pretty consistent idea what is lossless and what is lossy. These guys saying only their Pure DSD is “lossless” is just plain snobbery, implying anything else is a lesser product.

Pyramix is widely used in the audio industry almost always using DXD, often to produce music sold on 16/44 CD. Those recordings are not considered lossy by anyone except @Rushton and his mates. To his credit, @Paul is very realistic about the limitations and difficulties DSD processing presents and uses “impure” processes which are clearly good enough.

How about the new Gabriel Mervine? Wait am I posting in the right thread?


Indeed. I think I smell popcorn.


I initially had created a new thread for the recent Mervine remaster but that thread was removed by management and inserted here, an Octave Records thread

I started asking about Octave’s DSD conversions so I am to blame for the excursion

Thus, you are in the right place…if you dare

1 Like

Steven, you are so way off on a tangent grinding your own ax. Never have I discussed the “data stream” as lossy or lossless, nor do I recall anyone claiming this in the course of this conversation. It is the conversion from one format to another that is lossy or lossless. No one, but you, has talked about the DSD data stream itself as “lossless”. That is a nonsensical statement.

I’ve only said that the conversion from DSD to DXD is lossy, meaning the converted file cannot be restored to the original data stream. When mastering engineers process a DSD tracking file in DXD and then output that edit master to DSD again, the file has changed. And it doesn’t sound as clean and transparent as the original DSD file, even though that original file may well have needed all the Pyramix post-processing that was applied to it in the Pyramix editing process. Have you listened to the file samples I’ve provided links to?

DXD is a great format in many respects. I have hundreds of albums in DXD. They can be of exceptionally high quality. Sheeze, this is not the point. Never has been.


Though recently issues I reported were confirmed by PSA and several releases were fixed…to be applauded

I c what you did thar :grin:

Woop, wrong thread again :sweat_smile:

1 Like

You said above:

  • Moving PDM into DXD is a lossy process (as noted in comments above), meaning you can’t go back out of DXD to the same signal. That original is lost. Sure, DXD in Pyramix, or other tools, makes a mastering engineer’s life easier. Sliders to your heart’s content. It just doesn’t sound as good once it goes into DXD in my opinion.

That’s what the discussion was about any my comment about mozzarella! It’s not about making an engineer’s life easier, it’s about making it possible to do things you have to do to make decent recordings in a wide range of subject matter.

Your articles had an entire section headed “ Format Conversions are Not Lossless.”. @akro quoted it above.

Further on, they say: “ This PDM to PCM conversion is necessary for traditional direct binary computer math processing because a PDM bitstream contains no digital value information. Unfortunately, the conversion process of PDM to PCM is a lossy process, meaning that the resulting PCM samples have incurred losses such that the original PDM bitstream can not be recreated from it.”

This is all old hat, because over here Linn did a lot of recordings in DSD (hundreds, I think) from about 20 years ago and gave up because of its limitations.

I have listened to the files. I used headphones and turned off upsampling. I couldn’t hear any difference at all. The tracks are too long to make valid comparison, so I switched between each one of the three pairs after 5 seconds, then 10 seconds, then 15, because I also wanted to lose track of which version I was listening to in case I did notice something different. I just couldn’t tell them apart.

I went through this in 2015 when I bought a DSD DAC and some DSD files. I then used PCM until buying a DSD-capable DAC this year. I have one DSD256 download, Rachel Podger, I’ve compared it to the 24/192 stream and can’t tell the difference with that one either. For exquisite solo violin playing and recording quality, this is up there close to the top:

I’ve long felt Mr. Steven Segal has always wanted to be “right”, even when he was not! This latest barrage hasn’t done anything to change my opinion of him.


One just needs to read a few of Steven’s posts on what he’s drinking, where he’s eating, what concert he’s been too, or where he’s been traveling to see that this statement is a hilarious hypocrisy.


I picked Paul out today for this kind of snobbery (which he did in a nice way) in his video post. He said at 4’ in that Octave is setting the standard and the majority of other recordings are mediocre or bad. This just isn’t true, and it seems to me pretty disrespectful of the music industry.

I’ll gladly admit to going to lots of shows, which is what music is all about - which we do on a budget. No $2,400 tickets like Paul mentioned for the band Chicago. I briefly had a step-brother who was head of Ticketmaster Asia and he was proud to run what seemed to me a criminal cartel. I swore never to buy through an agency again and I never have.

Our last concert on Sunday was $20 each after a fish and chips lunch on the beach. If I listen to something I like, I usually check to see if I can hear it live. Picked this one out of Jazzwise this week, it was their album of the month, got tickets to see them in September, $40 for a 2-hour set at Ronnie Scott’s. Next outing tomorrow is a Broadway musical, Guys and Dolls, proper classy entertainment for the masses.

Yet another dull, boring, uninteresting, drivel laden post about YOU - is leaving me in a state of ennui. You love to talk in the first person. Don’t you ever get tired of having yourself around. Because it sure is hell from where I’m sitting reading your monotonous, self-indulgent crapola.!


You’ve suddenly become a common bloke, well good for you! I won’t drag this out, no point. But … please drop your campaign against DSD, we really don’t care what you think of it after so many posts on the subject, just let it go. There are lots of folks here (and elsewhere) who like DSD and don’t need to be continuously reminded of our moral failing …




with butter?

Any flavoring you prefer, CBD is an up charge, but may be appropriate considering.

1 Like

Two hearts from me. Maybe 3 or 4.

1 Like