A special case sounds pretty good to me. Sounds like where the magic might live. 
It does indeed.
Thank you, Badbeef! I loved working with Tony Furtadoā¦ and still do. Are you listening to the album with Buckethead and Kelly Jo Phelps? That was a killer album. Iāve been remixing it for funā¦ from the original 2" tape.
By the way, I always careful about quality ā¦ even before I knew I was an audiophile. 
When I started Blue Coast Records I had 20 years experience prior running a commercial studio and producing more than 400 albums (5 Grammy nominated and 2 were gold).
Back then I didnāt care if quality didnāt matter to othersā¦ I had to listen to the music for long periods of time and I wanted it to sound great for me.
Hereās a live solo album Tony cut in an afternoon at my studioā¦
Enjoy and thank you!
Cookie
Blue Coast Music
Yeah - that record.
Sheesh, 2" Analog uhhh, Woodā¦ from the recording perspective. Apologies if thatās Politically Incorrect nowadays, but I suspect you know what I mean.
Just a great sounding record.
Told you it breaks heads. Sorry.
You seem to think Iām making a point that Iām not making, though. I didnāt say they sound the same, and I didnāt say it doesnāt matter which format you use. But if you dig deeper youāll get a better understand of why they can sound different and why the mechanism of D-to-A conversion is relevant to that and why the DS sounds so uniquely good.
Remember that literally everything that comes out of the DS DAC is the sound of DSD, because the output stage is switching between fixed positive/negative voltages at a bit over 11MHz. But itās also the sound of ultra high resPCM because everything that was input got upsampled to 50+ bit ~56MHz during processing.
DSD is a unique case of PCM with particular characteristics chosen to suit audio encoding and playback for human ears. This is both the theory and the reality. 
1 Like
Despite what many here may think, my headās mostly not broken. I have my doubts on a daily basis about it, however ; ).
And "remember that literally everything that comes out of the DS DAC"s is a Decade and a half after the first DSD I used in a professional capacity. I get what is happeningā¦Way After the fact.
Iāve been around the block in that respect.
OK. So help me understand what youāre saying. What are you disagreeing with?
If one accepts this premise, I have no argument. No one could. Kinda like arguing with Soundmind.
It is one of those things that one cannot argue, essentially. I am just approaching it from an experiential perspective rather than a Mathematical one. I am sick to death of Nyquist, as I have lived probably more than half of my life with him (and have taught his theory at college) but do not feel his shit is an absolute. It is Ancient Digital History, so I donāt trust it. Been reading folks telling me heās the Frigging Einstein of Audio most of my life. Just donāt necessarily buy it. But Iām nobody in this respect.
I have my doubts. Can I argue it with a Mathematician? Nope.
Ah, OK. That does help a lot with understanding your perspective ā thanks. Thereās a totally valid set of criticisms to be made along the lines of how the sampling theory has been misapplied by proponents of various products/technologies to make claims of perfect audio that are belied by the reality of actually listening. I think that the mistake those people make is also the reason why DSD has been a breath of fresh air and again goes right to the heart of why the DS DAC is such a special machine.
Itās all about low-pass filters.
You know this: if you donāt filter out all the sub-Nyqvist frequencies before sampling you get noise thatās impossible to remove. And during reconstruction you have to also filter super-Nyqvist frequencies, especially those that might fall in the audio band. Because in the early days of digital the industry compromised on too-low a sampling frequency we put extra pressure on the implementation of low-pass filters, both analog and digital. Alongside jitter, the challenges with steep filters are IMO the biggest problems that digital audio has had to solve.
The enormous bandwidth of DSD lets us use much gentler filter slopes, which are easier to implement and have less side-effects in the passband. DSD lets us be more mindful of the constraints of the sampling theorem and get better results as a consequence.
1 Like
One reason the DS sounds good is I pay attention to Nyquist and my ears tell me that DSD sounds best, the DS is the bridge.
2 Likes
A related question which perhaps one of you can answer.
In my own recording, I have noted that increasing bit depth of PCM often sounds better than increasing sampling rate. That is, 44.1/24 sounds better than 88.2/16.
Why is this?
( Of course, as I can have both high resolution and bit depth, I record in 96/24. But it is fascinating to experiment.)
My first guess is that youāre enjoying the lower noise floor ā thatās the only thing that changes with sample depth.
Hereās an analog response to this from today:
Just getting started on this tune, and some new toys to make it happen.
Because, as a musician and Music Engineer, in many ways, I donāt understand what you guys are going on about. Not a judgement or what have you. Just seems like Theory vs. Sound or Musicā¦
2 Likes
Respect, Mark. Sounds great. 
Thanks, man.
This makes sense, but with 96 dB to work with and quiet digital equipment Iām not bumping my head on dynamic range. And, as a practical matter, when recording in a venue one rarely sees a noise floor below 40dB in any event. That is, there is always perceptible noise.
Good stuff
Until you apply dither, quantisation noise can be highly correlated with the audio signal. Thatās very audible and almost always unpleasant. Pushing that noise down another 48dB makes a big difference.
Iām not an expert on production processes but I have read that for this exact reason itās best practice to record and mix in 24-bit or higher, then apply dither as part of the final mastering/downsampling to 16-bit.
A good point, and yes we truncate bit depth and add dither as the last step.
Being pedantic: I assume that āincreasing bit depthā and āincreasing sample rateā refer to settings during recording rather than say processing 44.1/16 to 44.1/24 or 44.1/16 to 88.2/16? Things also might be different if you could record in DSD or, say, 24/176.4 or 24/192 and then down sample (and possibly dither) to 44.1/24 or 88.2/16. Empirically in the past 48/16 (or 50/16 that Telarc did) sounded noticeably better than 44.1/16 to a number of us, but perhaps that was the limits of the earlier recording equipment.
In any case I think it would take a set of experiments to determine if the recording equipment and/or the processing software or something else was causing the noticed differences between 44.1/24 and 88.2/16.