Good information to know. I enjoy DSD and hi-res PCM. There are a
lot of DSD bashers and it's hard to tell what's fact vs. opinion.
I do check hi-res audio with Audacity and have noticed many high
sampling rate music rolls off sharply at 44.1 kHz (early digital
recording or CD-master conversion) or it rolls off gently at 50 to 60
kHz (tape). Does anyone know anything about Paul Simon's hi-res
albums from the 70's and 80's? All roll off at 44.1 kHz even though
they were recorded on tape and all are being sold as 96 kHz files.
On Wed, 26 Jul 2017 18:50:42 -0500, SuperAudioLabs <Super@udio.labs>
wrote:
>On Mon, 24 Jul 2017 07:39:39 -0500, Puffer Belly wrote:
>
>> You didn't understand hardly anything I wrote. When DSD is converted to
>> PCM, which is how people without DSD DACs listen to DSD, the equivalent
>> sampling rate (FOR PCM NOT DSD) is a multiple of 44.1 kHz. As I stated,
>> DSD64 (SACDs) are converted to 24 bits sampled at 88.2 kHz, DSD128 176.4
>> kHz. etc. When people convert DSD to PCM using sampling rates other
>> than what I just listed, such as a multiple of 48 kHz (96, 192 kHz,
>> etc.), that causes more distortion than using a multiple of 44.1 kHz.
>> Converting DSD to PCM or vise versa is not distortion free. If the math
>> confuses you, then learn something about converting DSD to PCM.
>>
>>
>> On Sun, 23 Jul 2017 17:23:18 -0500, SuperAudioLabs <Super@udio.labs>
>> wrote:
>>
>>>On Sun, 23 Jul 2017 11:31:11 -0500, Puffer Belly wrote:
>>>
>>>> I'd just like to point out that DSD sample rates are based on
>>>> multiples of 44.1 kHz, so DSD64 is equivalent to PCM sampled at 88.2
>>>> kHz, DSD128 176.4 kHz, etc.
>>>>
>>>>
>>>That is actually entirely incorrect. That is just some modern bullshit
>>>some anti-DSD guys came up with to bash the format.
>>>
>>>DSD is not the equivalent to anything PCM....the two are basically like
>>>gasoline and diesel...similar products with an entirely different level
>>>of operation.
>>>
>>>Equivalence in PCM is made up...much in the same way we make up a
>>>horizontal resolution for analog video back in the day when there really
>>>was no such idea.
>>>
>>>IF what you said had any kind of truth to it; than the CD players of
>>>yester-year which worked with 2x, 4x, or 16x oversampling would not have
>>>been able to produce the sound quality they did....as you yourself said
>>>DSD64 is the equivalence of 44.1khz.
>>>
>>>BTW, any time you see something like "oversampling", or "MASH", of
>>>"Pulse Edge Modulation" on a CD player...it means it's using a
>>>delta-sigma 1-bit modulator internally. It's taking the 16-bit data from
>>>CD and modulating it at 1-bit....at a much slower rate than DSD.
>>>
>>>DACs up till about 10 years ago were using 1-bit DS modulation; these
>>>days the higher end models are using multi-bit delta-sigma.
>>>
>>>No one has made a DAC that can handle more than 16-bit samples that's
>>>not an oversampling design.
>>>
>>>DSD64 just means it's sample rate is 64 times the normal CD rate. Why
>>>was the CD sample rate held to a standard? Because it was the first time
>>>they actually standardized a sample rate.
>>>
>>>Actual performance of any DSD rate depends on what created the DSD
>>>stream and what is playing back the DSD stream.
>
>I took what you wrote literally. I was never talking about conversion, you
>never mentioned conversion.
>
>I'll also add that the conversion technology has matured; 20 years ago
>resampling from say 48khz to 44.1khz was pretty bad. These days, the
>process is not a problem any any distortion you get is usually inaudible.
>
>DSD conversion itself has matured a lot. Originally the only sample rate
>ever used for DSD -> PCM was 352.8khz; which yes...does fold evenly in to
>44.1. But that was also close to 20 years ago (13 - 17). There has been
>development in the DSD world; used to be you couldn't really do *any*
>processing in DSD...now there is software out there that will let you do
>basic equalization and level adjustments while in 1-bit.
>
>I'm pretty sure the software we're using here actually converts the DSD to
>the standard "DXD" PCM rate...then further resamples using FFT based SSRC.
>
>In 2002/3 I would have been right there with you; but the technology and
>code has matured quite a bit.
>
>Plus...have you ever looked at some of the 192khz stuff from HDTracks?
>They claim they give us what they get "from the record label"...in which
>case, the record label is outputting stuff from DSD in the highest
>consumer rate they can.
>
>Even worse...you ever seen a SACD that actually states it was sourced from
>PCM? I seem to recall some of the SHM-SACD Queen releases even stated they
>were made from 96/24 PCM.
>
>It's easy to tell a DSD->PCM conversion, it's next to impossible to tell
>PCM->DSD. There's a SACD ISO of "Will The Circle Be Unbroken" floating
>around that we sourced from the 192/24 HDTracks.
>
>Then there's my colleague next door from AHUSL who has been playing around
>with feeding high-end ADCs with a direct feed from high-end DACs. But I
>think he's just looking for excuses to play with his new reflow station.
|
| Follow-ups: | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
|