Nuh-uh
Boy, I am getting thirsty. I can understand this post, and I will pour myself a glass of nice Napa white.
nuh-huh
ps–when we going to hear another tune by the uber talented, the legend of a lunchtime, Mark Mal Beuf?
Oh man, I’m cryin’ over that one.
Good one. Awesome.
re: the music - long road. Hope to sort it out, but I’m old, man. The fingers and pipes don’t do what they once did. At least I’ve recently sorted out the monitoring and recording end chez Beef. But that doesn’t guarantee anything.
Did I ever tell y’all about the time Ron sold me his castoff…Castoffs, mind you!! Longass Icono speaker cables? I still have them on my Main System. And I used them to inculcate by lending a (koff, koff) friend of mine into the wonders of Galen’s stuff.
I’ll reshare this one from a while back before I moved to Boulder.
Having last month been to a JT show (as has Ron), I can now say with some confidence that for the first time in my life, I can sort of sing as well as him. Which is maybe not at all kind, but hey. Paid the most for the tix than any show in my life. Love the guy though. Happy to contribute.
He wrote it for Joni🥰
Looks almost empty! Enjoy.
¾ full at that point, but not now, at 4:30pm Amurican Mountain Time.
DSD and DXD have been around for over 20 years. Not exactly new.
Wow, that was excellent, Mark. Didn’t even need auto-tune. Too bad JT didn’t have it at Tanglewood.
That’s the thing, whether DSD, high res PCM, the vinyl and CD version of these recordings sounded so good too that I stopped worrying about the digital format of the medium. Music is so much more enjoyable without thinking of it.
@tedsmith , can you tell, what sense it could make soundwise to convert a DXD master (coming from a DSD recording) back to DSD for listening, which format would be the better choice for sound reasons (even if hardly a difference) and if this would depend on anything like DAC choice etc. (assumed a DAC supports both formats as the DS does)?
Could you also explain if all DSD benefits, present on recording level, make it through the DXD/PCM mastering format…and if so…why are there differences between PCM and DSD then on recording level?
I remember you once said, IF not much is edited/tweaked during DXD mastering, it should be transparent…anyway I don’t understand, why then as an unedited recording format, both are not identical, too. DXD in my eyes is just a higher res PCM format and PCM was told so far as very inferior (not depending on sampling rate, but by the nature of its concept).
Thanks!
[I have not read any of this thread: this post is only referring to jazznut’s questions.]
Going from DSD to DXD and back to DSD is reasonably transparent, but not completely. Which system is used to make edit does make a difference. There’s also going from DSD to analog and back to DSD which can be more transparent. Of course, any particular DAC choice makes a difference and can tip the scales either way.
The fact that editing on a Sonoma can give a superior result to editing in DXD (and ultimately back to DSD) shows that DXD doesn’t quite capture everything that DSD does.
On the other hand, if an editor/masterer had a particular goal in mind, it could easily be that do that in DXD is better than the choices available with DSD to wide DSD to DSD or DSD to analog to DSD.
I don’t believe I would have said that categorically. I usually use weasel words in such statements 
I know that I’ve tried to point out to people that doing a crossfade (a simple splice) in- DXD isn’t very harmful, editing systems can keep the original DSD on both sides of an edit. And it’s rare that 10ms (or whatever) of a reasonably good edit will standout when listening. But that isn’t what I took your quote of my quote to mean.
When you get to DSD sample rates and you use gentle filtering PCM can be technically better (and sound better) than DSD at the same sample rate, certainly if much processing is done. These days there’s no technical reason that tools can’t do all of their work at DSD sample rates, even with 32 or more bits of resolution, except for intermediate disk space.
A way of thinking about this is that doing all of your editing with 44.1k 16 bit PCM isn’t a great idea. But doing your work in higher resolution and then converting to 44.1k 16 for delivery to the ultimate consumer can produce great results. Similarly doing your editing in high rate DSD (or very high rate PCM) and then (at the end) converting to DSD64 can produce great results. Still the customer can enjoy the music more if they are delivered the higher resolution final product, say 24/96 instead of 44.1/16 or double or quad rate DSD instead of single rate DSD.
I never used the words like “very inferior” when comparing different formats. There’s what’s theoretically possible, what practical, what someone like the masterer may want or what the ultimate product to the customer is and they all have differing sets of features and problems which may have different answers in these different scenarios.
Thanks mich Ted! What you said about a simple splice etc. was exactly what I meant. I just remembered it as I expressed. And that PCM was told to be (in my words) very inferior to DSD was nothing I got out of your context, but Paul‘s somewhere (who will also have used other words, but with a similar meaning).
It’s difficult now to connect the streams of Paul and you I referred to, as I understand, you may not always be exactly on the same track regarding each statement and opinion.
So I just put two questions again a bit differently, of which I felt you didn’t clearly answer them:
Is there rather a benefit or rather a disadvantage of externally (I mean not inside your DAC) converting a DXD mastering format back to DSD for playback?
I didn’t quite get if you say or don’t, that the quality of DSD and PCM at very high sampling rates isn’t much different (do you?), so my question remains … in case they are different…how can an assumed clear superiority of DSD vs. DXD/PCM be carried through this „inferior“ mastering format?
My simple,logic somehow says:
either a format isn’t much worse than the other, OR a noteworthy editing/mastering stage is as lossy as the existing difference between those formats. Hereby I assume that mastering a DSD recording in DXD not always or just rarely means, just making extremely simple splices.
Ted, can you elaborate on this? On what basis is that a fact? Has there been some objective measure that shows the Sonoma system being superior to editing in DXD on say a Pyramix?
Your questions are a little confusing to me. You seem to be asking me to reply to statement that others have made, but you aren’t quoting them or even fully hinting what they might have said.
Extremely high rate PCM (not just 352.8 or 705.6k) can carry everything DSD does, but with more resolution. I’ve not been paying attention to tools of editing or mastering for the last decade, but an editor that used wide DSD (say, 16 bits, 32 bits, etc. everywhere and only at the very end converted down to one bit) would almost certainly be more transparent than editing at 352.8k or editing say single, double or quad rate DSD but rounding back to single bit DSD for intermediate results (just like rounding back to 44.1/16 for intermediate results isn’t a good PCM editing strategy.
There are many ways of building a quality, transparent DAC. Using a single bit output guarantees a perfect linear response, but at the cost of some noise. Multibit DACs can get around the noise problems but need heroic efforts to be extremely linear (at least at high sample widths.)
In the simple case of DXD to DSD, you get the benefit of a linear output but perhaps at the expense of noise.
Anytime you convert from one format to another with two different devices you can have diffent results, simply because there’s no standard let alone a mathematically perfect conversion possible. Different vendors will make different trade-offs…
It’s the experience of the editors and other people who work with DSD as related to me. I’ve also heard the differences in edits with good mastering tools… I fully grant that this doesn’t quality my statement as a fact. I was using “The fact that” more loosely, like one might in an informal conversation with someone you know. I’m simply relating my personal experience and the experiences of people who seem to hear as I hear. I’m certainly not going to get into an “objective measure” argument (or even discussion.)
I agree, unfortunately I had no other chance, as the questions arose out of posts from Paul, but I needed you to answer them, sorry.
Does this mean for you, DSD (just 4x or even 1x) as a recording format still has advantages against 352.8 PCM or not?
This is why I assumed, a reconversion of the 352.8 mastering format to DSD can just have disadvantages (at least when the 352.8 format can be played on a DAC like the DS which converts it to DSD anyway).
It depends on what equipment you are using… It’s always been my experience that much higher sample rates are suppleir to more bits in your samples. This isn’t suprising, each bit added to a sample caries correspingly less information that the previous ones. Doubling the sample rate decreases the noise floor more than adding a bit to the sample width.
It’s my experience that even single rate DSD is more transparent than DXD if you are comparing analog to DSD to analog vs analog to DXD to analog with great equpment.
It’s also my experience that for people that want to do more than simple editing DXD gives much more flexibility and will lead to better results from people that want to get creative.
As to conversions, converting from DSD to analog or to PCM is much better defined and much less intrusive than going the other way. A low pass filter is simple math, sigma delta modulators are far from it.
I guess I’d like to add that these differences can be darned subtle. (Tho sometimes they are more obvious.) I have a lot of very good DXD recordings, both in DXD and DSD and enjoy them in either format.
Editing DSD on a Sonoma is different than editing in DXD (high rate PCM).
Editing DSD on a Sonoma is accomplished by widening the DSD to 8-bits at the DSD rate. After editing, the datastream is modulated back to 1-bit DSD. The sample rate is not changed, the data remains noise shaped, and the data does not spend any time as PCM.
This can sound better than editing in DXD because the data remains DSD-like when edited.
Perhaps this helps.