Yes. You can not edit DSD in its current form. It can be edited when converted to analog, then back to DSD.
Thanks, Ted. I appreciate your sharing this information.
You are thinking of mixing. Engineers mix DSD on an analog mixing console, or they have to convert it to PCM (DXD) to mix it in the computer, which defeats the purpose to some extent.
The actual editing that happens in digital workstations converts either the entire file or just the edit points to either 8 bit DSD (Sonoma) or 32/352khz PCM (Pyramix).
I don’t know enough about how Sonoma works, but in Pyramix you can make it so only a millisecond here or there is converted to PCM, just for edit points and brief fades, and everything else is still DSD. IMO, the effect is inaudible. I had a DSD128 project where a 15 second digital fade out was added in Pyramix, and you would be hard pressed to notice any difference in audio quality. Although if the entire track was converted to PCM, you could notice.
So what people like Cookie Morenco do, and I assume Gus, is mix on analog, but do targeted editing in the digital workstation, converting only the edit points, and leaving the rest of the track as native 1 bit DSD.
1 Like
Interesting discussion
Over ten years ago I researched DSD editing and then posted this:
"DXD is how Pyramix workstations internally process DSD source files. DXD is what Merging Technologies and Sony/Phillips call 24bit/325.8kHz PCM. (Pyramix is a line of DSD workstation made by Merging Technologies.)
To mix and edit DSD, the Pyramix transcodes the DSD source file to 24bit/325.8kHz PCM for all editing. It transcodes it again back to DSD for output. That is, all DSD audio on a Pyramix goes through this conversion via internal PCM and sigma/delta converters.
SADiE (Sonoma) workstations do not convert the DSD source except at the edit crossfade point. During the length of the crossfade the DSD source is transcoded to 8bit/2.8Mhz PCM and back to DSD."
At this point I must admit I do not independently remember the specifics. 
I think this is mostly correct, although I’m not sure about SADiE since they haven’t been used for DSD in a long time.
Some time ago Pyramix developed a feature called “Render Mode” which converts only the actual edit point/fade audio to PCM and back. What I was told (although I can’t find a source) is that they use 352.8khz for DSD64, 705.6khz for DSD128, and 1411.2khz for DSD256.
To mix on Pyramix, you do have to convert to PCM, but if you use an analog console for mixing, only the most minimal bits of audio have to be converted to PCM. I think these days, Pyramix projects that want to remain “in the box” are mostly recording at 24/352khz to begin with.
1 Like
From Paul’s blog in 2013:
"A quick note correcting what I wrote yesterday about editing DSD. While it is true you cannot directly edit DSD, I reported that Cookie Marenco of Blue Coast Records recorded in DSD then converted to PCM, made her edits and then went back again to DSD. Not!
While this is what some DSD editing systems do, Blue Coast converts to analog, makes any adjustments required then back again to DSD. Duh. That’s a much better solution and an obvious one as well. I am sure thankful there’s plenty of people smarter than me. "
The day before, Paul said;
"Now let’s understand DSD isn’t all that friendly to recording engineers and studios. For one thing, you cannot edit DSD. You cannot digitally EQ DSD. When a recording studio wants to make edits or EQ the raw data they must first convert it to PCM, edit it, then convert it back to DSD. So that’s a problem – but it turns out this is ok if you start with DSD, go to PCM and then back again.
Modern recording equipment that can handle DSD, like the Sonoma workstation used by our friends Gus Skinas and Cookie Marenco, can go from DSD to PCM and back again with nary a hitch in the get along. There doesn’t seem to be a way to go from an original PCM recording to DSD and gain any of the advantages of DSD (which kind of makes sense), but going from DSD to PCM and back again is apparently ok. "
This all makes sense. Thanks.
We need to ask Cookie, but I suspect what she is doing is playing the DSD recordings, running this through an analog mixing console for mixing tracks, adding EQ, whatever - and recording the analog mixer’s output in DSD. This way the DSD file is never PCM.
The only downsides are the extra D/A and A/D conversions and whatever noise/artifacts the console adds to the sound.
That was my understanding, she is using the DSD workstation basically like a tape machine with limited edit capability.
DSD does have an ultrasonic noise problem that can overload some mixers with multiple tracks, but that is why DSD256 is handy: The noise buildup is so far out of the audio range that you can have as many tracks as you want. Every time you double the sample rate of DSD the noise moves farther away and out the bandwidth of any equipment you are using.
Using an analog mixer though, is not a problem at all IMO. I would argue that it is a good thing. There are some advanced tools that only exist in the digital realm, but most of the time analog mixing, and processing still sounds better. By a lot. I would not think of it as a loss. With the exception of some classical music and electronic, just about everything you hear has analog processing at some stage, if not during tracking or mixing, than often during mastering (compression, EQ)
With DSD the AD/DA conversion is also so much better than PCM that it is also not that big of a deal.
2 Likes
Agreed, an analog excursion is not a bad thing.
1 Like
It would be interesting, albeit likely impossible, to find out if Tidal used the same recording source as the DSD version. If the source lineage is different, then the comparison results get skewed a bit.
I have over 400 DSD albums. Many do sound amazing. But I have PCM versions that sound just as amazing, and PCM versions that are lesser. I also have multiple DSD versions of some recordings, and they also sound a bit different than each other. This makes me wonder how much the perceived superiority of DSD has to do with the mastering engineer for the DSD version to be more attentive to using a pure source, and it’s that pure source that makes a bigger difference than whether the final release comes out in DSD vs PCM.
I’m reaching no conclusion one way or the other, just questioning.
On my ultra A+ high quality recording playlist, DSD and PCM tracks are equally represented and they all sound fantastic.
My baseless hunch is attention to the purity of the source, and the skill of the mastering engineer, has a much greater influence than whether or not it’s DSD or PCM.
And DSD recordings are a very lonely club. Only a relative few even exist.
1 Like
As I have said many times before, the single most important factor affecting sound quality of an album is the mastering. A CD can sound way better than DSD256 or hires PCM.
However, I have several SHM CDs and SHM SACDs with the exact same mastering. In all cases, I much prefer the SHM SACD. They invariably have more detail and sound just a bit smoother at the same sound pressure. I can play the SHM SCADs at higher volumes with less fatigue than their SHM CD counterparts.
If the SHM SACDs were not available, I would be perfectly happy with the SHM CDs. But picking the SHM SACDs is an obvious and easy choice.
1 Like
I take this as the main insight for me out of this interesting discussion. The way most DSD is edited/mixed is (slightly) flawed, implications are audible for folks with good ears and gear. Going analog is the better way inspite of the process and format crack.
For me this makes obvious how close DSD is to analog, how superior analog still is (which we know from pure analog processing and playback) at least in comparison to PCM and inspite of its own limitations and how limited most of what’s produced digitally seems to be (talking of smaller differences regarding this topic however) due to PCM format involvement.
Mastering is of second importance. The quality of the initial recording is by far the most important.
Edit: Upon additional thought, the mix is often more important than mastering as well. Individual tracks can be compressed (which can be good or bad), Eq’d, etc.
2 Likes
Wonderful guys! So much knowledge, and such interesting dialogue. This is what makes the forum so enjoyable. Keep informing . . . I so prefer these intelligent, insightful perspectives. Very educational . . . Made for a good read while I enjoyed my morning coffee. Cheers!!
Dude, I am talking about the quality of the recording we get as consumers. Normally, all masterings are sourced from a master tape or a generational copy of that master tape that is already mixed. Rarely is anyone given access to the original multitrack tapes to make a new mix. Also, the quality of the original multitrack recordings used to mix and make the master tape are fixed. They are what they are and there is no way to improve them. Now, the quality of the master tape copies, if made and used, can be an issue.
The bottom line is that for the consumer the masterings are what make the difference as they are what are unique. We can’t usually go out and find CDs using different recordings, mixes, or master tapes. The masterings are where the differences lie…
All of which was not in your original post, but that’s OK, we can go with this, It is your post after all.
All you had to do was look at the context of my response (look at the post I was responding to) to see what I was talking about. Of course, you don’t let us include context so you, of all people, should darn well pay attention to the context of the post being responding to…
When do we, as consumers, have any choice in the quality of the recording of a particular album? The answer is never. The recording and mix is what it is. The only options we typically have are the format and the mastering.
If you go looking for the best digital recording of Steely Dan’s “Can’t Buy a Thrill”, there are not multiple recordings to chose from. There are not multiple mixes to choose from. But there are multiple masterings to choose from. In this case, the only digital format I am aware of CD. But, other albums may have hires PCM or DSD releases to choice too. Even then, there are not multiple recordings to chose from not are there multiple mixes to choose from.
Sometimes you get lucky and new mixes and masterings are released that actually sound better. This was the case with the Steven Wilson Jethro Tull releases. But this is quite uncommon.
So, when we are talking about buying specific albums, we basically NEVER to get a choice when it comes to recording quality and RARELY get a choice in mixes. In other words, since mastering matters more than format, the mastering is the most important factor when you want the best sound quality of a particular album.
No one is arguing with you. If this is what you meant, this is what you meant.
Enjoy your Saturday.