Remaining Considerations on DSD

[xaviescacs]

That actually was the impression I got from your post. I think you and I are after different things when listening to music on our systems.

Maybe. I want to enjoy music, but I've been polluted by people here, and now I can't avoid trying to learn how a flat and true to the source sound reproduction sounds like. Because of this I'm enjoiyng music a bit less than before, so I have to stop this at some point.

 

[analogSurviver]

DSD looks good on paper, but the reality is that even Paul McGowen admitted in a recent video that when they do any editing (even simple gain changes) on their DSD masters, it has to be converted to PCM to do the edits and then re-converted to DSD. Kind of makes DSD a moot point.

Speaking from my own experience, the simple variances in adjusting the bias on an analog tape machine routinely done before recording sessions makes far more real and measurable difference in sound than there ever could be between DSD and PCM - yet nobody ever screamed about that.

DSD does not have to be converted to PCM for the editing - not in its entirety.

Back in the day, Sadie offered DAW where conversion of pure DSD into something called DSD Wide ( 8 bit PCM - IIRC ) only at the split so that it can be made at all. Not entire file as required by Merging Pyramix which works with DXD.
Thus, about 99% of entire edited track/album is/was possible in pure native DSD - all the way from the microphone to the file purchased by the end user.

There are pure DSD recordings out there which have not originated on any other medium ( eschewing any negative attributes of any of the respective mediums ) - but, admittedly, they are extremely rare.

 

[srrxr71]

View attachment 180361I remember becoming interested in hifi and I remember how impressive that the Sony SCD-1 seemed at the time. I don’t even own a single CD or SACD, but I would still like to have an SCD-1 in my life.

Products like that are exciting and I suspect made an impression on a lot of the people still buying SACDs today. There’s an understandable nostalgia for an era when the great products looked and felt great.

Now, technical supremacy can be found with a chain consisting of a raspberry pi, apple dongle, and a humble class D amp. Which is also great progress that high quality sound can be found for so little - just different.

I still have this. It takes 30 seconds to read the disc and start playing. Built like a tank though. It needed servicing after about 2 years. Maybe that ruby bearing. Really finicky machine.

 

[howdoesitsound]

Less than a year ago I started getting into home theater and as a part of that process I bought a used Sony UBP-x700. I then realized that I could finally try an SACD (and stream DSD files from my NAS). The Sony delivers the DSD files to my Denon 4520ci which decodes them. My personal collection is FLAC, plus I pay for Apple Music and listen to those tracks in lossless hi-res. So I really wasn’t expecting much but picked up a few orchestral SACDs anyway….

And they sounded fantastic. I especially like the multi-channel. I know, Atmos Music is getting better all the time, but….I still enjoy being able to load a disc, and being able to *not* need to stream. Feels wasteful sometimes.

Anyway, all that to say as each format has its moment in the spotlight (some very briefly) I still enjoy being able to experience them. I still have a 1939 Mills jukebox and the aesthetic experience of listening to those 78s brings me a lot of joy. When I upgrade my AVR to something that is Atmos-capable, I hope it will still be able to decode DSD.

 

[tuga]

Any leaving considerations on DSD?

 

[howdoesitsound]

Seems to me that Ed explained it all pretty well in 2004. One thing we’ve learned (again) since then is that what becomes a standard is not always the best.

meitnerinterview

 

[Head_Unit]

meitnerinterview

An interesting statement is "Meitner: You should also get the Tom Petty disc that MoFi did. Compare that to the one that was done with the DSD system and then down sampled with the Sony Direct, SBM direct versus the original. It is a world of difference." However, do we have any idea if these are straight transfers? Because if any EQ fiddling was done at all, then it's not a valid comparison. I do like an SBM Boston disc I have, which might simply be due to a better transfer from the original tape, nothing to do with SBM at all.

Anyway Meitner reiterates the bit (ha ha) about how the A/D is one-bit so why not just preserve that? That part makes sense. Once you start mixing at all, though, my understanding is this is impossible in one bit unlike Meitner's seeming assertion regarding the Sonoma workstation (See counterpoint at https://www.realhd-audio.com/?p=1235) and then it's not pure anymore, and becomes just marketing baloney to some degree.

Personally, I have some "just CDs" that I think sound great, so I have no idea what Ed Meitner is talking about with evils of PCM sound. Neither was John Eargle, who once opined to me that he felt 24/96 reproduced the sound of his microphone feeds just fine. Of course everyone has different hearing acuity in some absolute numeric sense, and also processes sound differently such that some people can hear things other don't, and may be bothered by things that others don't. I absolutely HATE hiss, and snap/crackle/pop, and wow-and-flutter. Others are not put off so much. I think it was the Mobile Fidelity guys (but might have been Telarc) who at a surround sound conference said they had set up parallel PCM and DSD recording chains and felt the DSD was somewhat closer to the live microphone feed (which ignores the mixing issue).

Meitner talks a lot about the importance of the analog parts-which I personally think are likely more important than the digital parts. In some resonance to HDCD, perhaps the great sound of some DSD stuff is due more to his EMM converters used, in the way many felt HDCD sounded better not because of the process but due to Pacific Microsonics' well-engineered A/D converters.

 

[howdoesitsound]

“the bit” good stuff.

Yeah in the interviews I’ve listened to and read, he explains that in order to edit you need to convert into the PCM world at a very high rate in order to preserve as much of the original as possible (hopefully to the degree that the audible spectrum is essentially unaltered). That does not in my opinion make it “marketing baloney.” DSD serves as a good base should anyone decide to do something with a master decades later.

Music needs to be archived in a format (as he puts it) that involves the “least” conversion. That way you can move into other formats from there if you need to. Alas, it’s pretty much water under the bridge for new music as we know how things played out since 2004 when he gave that interview. I have no idea what percentage of archived tape masters have been converted (and within that number, what formats were used) but unless I’m misunderstanding him, labels will be in pretty good shape if they come back to a DSD transfer decades from now. Here is something he said about some non-DSD archiving:

”What's sad about it is that a lot of the digital transfers on CDs were done before the advent of delta-sigma converters and have gone through some horrible, horrible processing. I have a feeling that this is part of history lost.”

 

[howdoesitsound]

oh…and I agree with you about the analog parts being so important, but I see that as hand in hand with the file that needs to be delivered to the ears, and for that matter, a great performance!

 

[RDoc]

An interesting statement is "Meitner: You should also get the Tom Petty disc that MoFi did. Compare that to the one that was done with the DSD system and then down sampled with the Sony Direct, SBM direct versus the original. It is a world of difference." However, do we have any idea if these are straight transfers? Because if any EQ fiddling was done at all, then it's not a valid comparison. I do like an SBM Boston disc I have, which might simply be due to a better transfer from the original tape, nothing to do with SBM at all.

Anyway Meitner reiterates the bit (ha ha) about how the A/D is one-bit so why not just preserve that? That part makes sense. Once you start mixing at all, though, my understanding is this is impossible in one bit unlike Meitner's seeming assertion regarding the Sonoma workstation (See counterpoint at https://www.realhd-audio.com/?p=1235) and then it's not pure anymore, and becomes just marketing baloney to some degree.

Personally, I have some "just CDs" that I think sound great, so I have no idea what Ed Meitner is talking about with evils of PCM sound. Neither was John Eargle, who once opined to me that he felt 24/96 reproduced the sound of his microphone feeds just fine. Of course everyone has different hearing acuity in some absolute numeric sense, and also processes sound differently such that some people can hear things other don't, and may be bothered by things that others don't. I absolutely HATE hiss, and snap/crackle/pop, and wow-and-flutter. Others are not put off so much. I think it was the Mobile Fidelity guys (but might have been Telarc) who at a surround sound conference said they had set up parallel PCM and DSD recording chains and felt the DSD was somewhat closer to the live microphone feed (which ignores the mixing issue).

Meitner talks a lot about the importance of the analog parts-which I personally think are likely more important than the digital parts. In some resonance to HDCD, perhaps the great sound of some DSD stuff is due more to his EMM converters used, in the way many felt HDCD sounded better not because of the process but due to Pacific Microsonics' well-engineered A/D converters.

In view of the fact that MoFi has now admitted using DSD for most of it's records, I wonder if that difference, assuming there actually is one, is just that: a difference, not particularly an improvement.

 

[TimF]

There's a good sized article in today's Washington post about the MoFi scandal. These record albums that were and are sold via fraudulent claims are like many consumer products if not nearly all consumer products. I suppose Bayer aspirin is still a genuine article. What is the dirties market? Perhaps gem stones? I suspect that about 25% or more of art works in major museums are copies or fakes and especially the old master paintings. It is when you find out your spouse conned you that you wake up in life.

 

[Head_Unit]

Once you start mixing at all, though, my understanding is this is impossible in one bit unlike Meitner's seeming assertion regarding the Sonoma workstation (See counterpoint at https://www.realhd-audio.com/?p=1235) and then it's not pure anymore, and becomes just marketing baloney to some degree.

Having stumbled on an interview with Ed Meitner, he says there that preserving the sample rate was a key thing. I suggested to Stereophile that and other topics would make a great interview. So perhaps it's not total marketing baloney. Anyway DSD or PCM matters a lot less than the skill of the engineers and mixers.

 

[Blumlein 88]

There's a good sized article in today's Washington post about the MoFi scandal. These record albums that were and are sold via fraudulent claims are like many consumer products if not nearly all consumer products. I suppose Bayer aspirin is still a genuine article. What is the dirties market? Perhaps gem stones? I suspect that about 25% or more of art works in major museums are copies or fakes and especially the old master paintings. It is when you find out your spouse conned you that you wake up in life.

You don't trust Ivory soap to be 99.44% pure soap? You are in a skeptical phase these days aren't you?

I suspect you might be near the mark on paintings in art museums. Technology is proving some very embarrassing revelations for some art collections. What is actually interesting to me in the things I've read about paintings is how ingenious some of the methods have been in making the fakes with rather low technology and lots of guile.

 

[Saidera]

A few years have passed, and by now I can see that ASR has heaps more DSD threads on top of what was already quite a lot. For me, DSD had ceased to exist once I began to understand it in greater detail thanks to ASR. I also began to realise the benefits of Bluetooth audio, of HE AAC v2 and OPUS files, and DSEE HX, and how convenient PCM is for editing and listening. I read this article https://www.washingtonpost.com/music/2022/08/05/mofi-records-analog-digital-scandal/ and was reminded of DSD. Why would MoFi not use high rate PCM?

 

[tmtomh]

A few years have passed, and by now I can see that ASR has heaps more DSD threads on top of what was already quite a lot. For me, DSD had ceased to exist once I began to understand it in greater detail thanks to ASR. I also began to realise the benefits of Bluetooth audio, of HE AAC v2 and OPUS files, and DSEE HX, and how convenient PCM is for editing and listening. I read this article https://www.washingtonpost.com/music/2022/08/05/mofi-records-analog-digital-scandal/ and was reminded of DSD. Why would MoFi not use high rate PCM?

That's a good question about MoFi - and I think it's gotten overlooked because of the analogue/digital aspect of the controversy. But I would guess they used DSD because it's viewed as more "pure," "audiophile," and/or "analogue-sounding" than PCM. To be clear, I don't believe this myself, and I agree that objectively there is no evidence to support the view that DSD is superior. I'm just saying I can see how MoFi would think DSD would be the "best, most purely audiophile" way to digitize analogue master tapes.

Along with that, DSD works in a way similar to modern Delta-Sigma DACs, so they might have felt that DSD would provide for the least conversion among different types of digital sampling when they ran the signal through the DAC hardware in their mastering chain. Again, I'm not saying I think that matters - but they probably do think it matters.

And finally, MoFi also puts out a lot of SACDs. So they probably have a DSD workflow for that, which makes more sense given that DSD is SACD's format. And so if they feel it's proper and appropriate for that workflow, it makes intuitive sense to use essentially the same digitization process for the production of LPs too.

One final, semi-related note: If you look online at interviews with mastering engineers - the guys at MoFi, Bernie Grundman, Robert Ludwig, and pretty much all the others - you will find that famous mastering engineers who have a track record of producing excellent-sounding masters and remasters, hold some technical beliefs that are just false (or at least not supported by any known evidence). But they still produce great-sounding results, which is what's important (at least to me). I think we are fortunate that most of the inaccurate technological beliefs these guys hold are about things that they think make a difference but actually don't. So while DSD didn't make those MoFi releases sound better than PCM would have, it also didn't make them sound worse - so no harm. Same deal with mastering engineers who go on about how mastering at 192k instead of 96k produces clearly superior results - it provides no audible sonic benefit, but it also does no audible sonic harm to the final result.

Even Bernie Grundman's attachment to tube gear probably makes very little if any audible difference in the masterings he ends up with - and regardless, it's still part of the production process rather than the playback chain, and so if tubes do add a smidge of euphonic warmth or take a little edge off a master tape originally mastered with harsh treble, and we enjoy the sonic result, then who cares if he believes tubes are higher-fi than solid state or that analogue is higher-fi than digital?

I mention this because I don't think it makes sense to publicly go after mastering engineers based on the idea that they are "helping spread false information that's harmful." Of course we should feel free to point out inaccurate things they might say. But I think it's pointless and often counterproductive to, for example, go after them directly about this stuff if they happen to log on to one's favorite music or hi-fi web forum, or to spout vitriol or snark about them. When hi-fi reviewers, YouTubers, and hi-fi gear makers spout false information, absolutely go after them with both barrels - they're on the playback/reproduction side of the industry and hobby, which is where fidelity is the guiding principle, and there's no excuse for BS in that realm. But mastering engineers are on the creative/production side - they might have misguided beliefs, but they're not trying to sell snake oil, and the gear and formats they prefer in their mastering chains has no impact on the end-consumer pricing of the CDs, SACDs, and LPs they've worked on.

So as long as Bernie Grundman keeps making amazing masterings and doesn't get into DAC and amplifier design, I'm grateful for his work and couldn't care less what he believes about digital sampling theory.

 

[JSmith]

thanks to ASR.

It's good to have a place where logic is accepted, rather than negated... I won't name that other forum.


JSmith

 

[voodooless]

I read this article https://www.washingtonpost.com/music/2022/08/05/mofi-records-analog-digital-scandal/ and was reminded of DSD. Why would MoFi not use high rate PCM?

The article has the answer:

Syd Schwartz, Mobile Fidelity’s chief marketing officer, made an apology.

“Mobile Fidelity makes great records, the best-sounding records that you can buy,” he said. “There had been choices made over the years and choices in marketing that have led to confusion and anger and a lot of questions, and there were narratives that had been propagating for a while that were untrue or false or myths. We were wrong not to have addressed this sooner.”

So it’s purely a marketing ploy, nothing else.

 

[JSmith]

You don't trust Ivory soap to be 99.44% pure soap?

Well no... I expect it to be pure elephant ivory, of course. (totally against the ivory trade btw)


JSmith

 

[firedog]

A few years have passed, and by now I can see that ASR has heaps more DSD threads on top of what was already quite a lot. For me, DSD had ceased to exist once I began to understand it in greater detail thanks to ASR. I also began to realise the benefits of Bluetooth audio, of HE AAC v2 and OPUS files, and DSEE HX, and how convenient PCM is for editing and listening. I read this article https://www.washingtonpost.com/music/2022/08/05/mofi-records-analog-digital-scandal/ and was reminded of DSD. Why would MoFi not use high rate PCM?

It's pretty simple: they weren't allowed to take the analog masters off location to make the LPs, so they they tested various ways of transferring the analog tapes to digital, and decided DSD 256 gave the best sounding result.
They are only converting to digital and remastering, not remixing, so there's no reason not to use DSD.

MoFi President Jim Davis Addresses the Digital LP-Mastering Controversy - The Absolute Sound

MoFi President Jim Davis Addresses the Digital LP-Mastering Controversy - Jim Davis, president of Mofi, gives JV and RH insight into the company's mastering process in this exclusive interview.

www.theabsolutesound.com

note: (in other places I read they did try both analog and PCM transfer and preferred the DSD 256 as the best sounding version).

JV and RH: Why did you decide to master from DSD files rather than from analog mastertapes, as you used to do? What are the advantages of mastering from files vis-a-via mastering from tape, and what (if any) are the limitations?​

Jim Davis: Some record label tape vaults changed policy regarding shipment of mastertapes. At that point our only option for those recordings was to go to the mastertapes. Once we were able to access these masters, the dilemma was how can we best retrieve the information from the master? We experimented with making analog copies from the master. Various tape stocks (½**, 1**) and speeds (15ips, 30ips) were tried but rejected. There was no way to overcome the noise-floor disadvantages of copying from one analog tape to another. When we tried DSD, it was immediately clear this was a vastly superior method for maximizing information retrieval. Developed as an archival format, DSD is sonically transparent, with a very low noise floor. Combined with the painstaking transfer process described below, the capture is a virtual snapshot of the master, revealing detail and nuance at a level that conventional methods could not. Counterintuitively, this capture yields, in our evaluation, superior sonics compared to a cut that is direct from the analog tape to the lathe.

The process of achieving these captures at a remote studio location is expensive and time consuming. We ship our proprietary gear, including our Tim de Paravicini-modified Studer A80 tape machine, to the studio, rent studio time, and fly and lodge our engineer for several weeks at a time. The process of making a DSD capture using our techniques takes a day or more alone for each tape. These are long and exhausting days, and I’m proud of the hard work Mofi engineers put into each project and of the results they consistently achieve. I’m not aware of any other audiophile record label that puts that time and expense into each release.

Beyond the additional time, effort, and expense, I’m not aware of any sonic limitations of using this process.

 

[Saidera]

Thank you all for your detailed and insightful responses! It's fascinating to delve into the complexities of audio formats, mastering processes, and the beliefs that can shape these practices. The perspective on how MoFi might have chosen DSD due to its perceived "audiophile" qualities makes a lot of sense, even though objective evidence might not necessarily support that viewpoint. I also appreciate the detailed information you've provided regarding MoFi's decision to use DSD 256 for their mastering process. It's clear that their choice was a result of practical considerations and extensive experimentation to achieve the best possible sonic outcome.
I guess there is the sometimes subjective nature of audio perception and the fact that the end result matters more than the specific beliefs guiding the process. It's reassuring to hear that these beliefs don't necessarily translate to harmful sonic outcomes. Your distinction between playback/reproduction and creative/production sides of the industry also makes sense. While accuracy is crucial in the playback realm, creative choices on the production side can indeed lead to unique and enjoyable sonic experiences, even if they're based on beliefs that might not hold up under strict technical scrutiny.
I have heaps of respect for Bernie Grundman and so thank you for mentioning him!