CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS

  • December 31, 2015, 11:19:44 AM
  • Welcome, Guest
Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 2 3 4 5 [6] 7 8 9 10 11

Author Topic: DSD vs CD vs High-Res PCM  (Read 14946 times)

0 Members and 1 Guest are viewing this topic.

xnor

  • Pirate
  • **
  • Brownie Points: +39/-50
  • Offline Offline
  • Posts: 154
Re: DSD vs CD vs High-Res PCM
« Reply #50 on: December 16, 2013, 10:53:30 PM »
Quote from: DaveBSC on December 16, 2013, 08:44:02 PM
There was a live demonstration at one of the audio shows where somebody used a chart like this to point out the positives of DSD - supposedly that it's more "analog like" because there's no hard cutoff as with PCM, and that "you still have 30dB of noise floor in the UHF range where PCM has none."

As I said before, if you don't filter the noise that is in the *ultrasound* range you can get in very serious trouble. Imagine the amplifier amplifying all that unfiltered noise and passing it to the speakers.

The 30 dB SNR is completely pointless, since music doesn't contain that much energy above the audible sound range. So everything up there is getting completely drowned out by the noise, before being attenuated by the filter (if there is one, hopefully).

The cutoff, even with CD audio, is easily above 20 kHz. As such there are no audible problems. To my knowledge there hasn't been a positive blind test on >20 kHz low-pass filters.
Logged
"I'm on a whole new adventure." - "Growing a mustache?"
"No. Bigger than that." - "A beard?!?"

Original_Ken

  • Thread Autocrat
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +19/-1647
  • Offline Offline
  • Posts: 254
  • uberFrost is the best deal in audio today
Re: DSD vs CD vs High-Res PCM
« Reply #51 on: December 19, 2013, 04:38:51 AM »
Quote from: DaveBSC on December 16, 2013, 06:28:01 PM
Here's the same sort of idea put a different way...

So how would one describe the audible effect of the difference between that graph and ideal response ?
Logged

DaveBSC

  • Best Korean Sympathizer
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +222/-50
  • Offline Offline
  • Posts: 2092
Re: DSD vs CD vs High-Res PCM
« Reply #52 on: December 19, 2013, 06:26:16 AM »
Found it.

The chart I posted is at 21:38.
Logged

ultrabike

  • Burritous Supremus (and Mexican Ewok)
  • Master
  • Pirate
  • *****
  • Brownie Points: +4226/-2
  • Offline Offline
  • Posts: 2384
  • I consider myself "normal"
Re: DSD vs CD vs High-Res PCM
« Reply #53 on: December 19, 2013, 07:03:18 AM »
That plot shows the spectral difference between the input and output signals of the delta sigma, i.e. noise PSD.

The audibility of it depends on how prone analog components down the chain are to generate audible non-linear distortion due to ultrasonic signals.

Below are the window dependent PSDs (instead of FR) of an input signal, unfiltered delta sigma output, and delta sigma noise (like the plot above) of a first (red) and second (black) order simple delta sigma (64 OSR, and using band-limited white noise instead of 1 kHz tone.).



There is quite a bit of ultrasonic noise (relative to input signal). Note that higher order delta sigmas trade off lower pass-band noise for higher ultrasonic noise. So it can get a bit worse for a 7th order delta sigma. Some noise psd shaping can be arranged though through internal gains and other stuff.

EDIT: Here is the same scenario above, but using log scale in the x-axis and a different window, which might make it more comparable to the Playback Designs DSD Noise PSD plot...

« Last Edit: December 19, 2013, 10:24:26 AM by ultrabike »
Logged

Willakan

  • Pirate
  • **
  • Brownie Points: +20/-17
  • Offline Offline
  • Posts: 113
  • I'm quite reasonable really.
Re: DSD vs CD vs High-Res PCM
« Reply #54 on: December 22, 2013, 06:07:58 PM »
I may be mistaken, but I believe the chart shown by DaveBSC is straight out of Sony's original publicity materials for the format. I particularly like how it employs an FFT for the DSD noise floor, then just draws a line for PCM at the RMS value for noise for the entire signal.

Ahem...

Employing the latter method of calculating the 'noise floor', I will now perform a completely scientific analysis of PCM vs DSD.

Number for 8-bit PCM: ~48dB
Number for DSD: ~7dB

The conclusions to be drawn should be obvious. Am I doing this right?

Logged
Indecent lover of cheap opamps...

ultrabike

  • Burritous Supremus (and Mexican Ewok)
  • Master
  • Pirate
  • *****
  • Brownie Points: +4226/-2
  • Offline Offline
  • Posts: 2384
  • I consider myself "normal"
Re: DSD vs CD vs High-Res PCM
« Reply #55 on: December 23, 2013, 05:33:19 AM »
The idea is that if one filters the DSD output signal with a 20 kHz corner frequency low pass filter, one might get better than 7dB.
Logged

Willakan

  • Pirate
  • **
  • Brownie Points: +20/-17
  • Offline Offline
  • Posts: 113
  • I'm quite reasonable really.
Re: DSD vs CD vs High-Res PCM
« Reply #56 on: December 24, 2013, 11:21:41 AM »
Quote from: ultrabike on December 23, 2013, 05:33:19 AM
The idea is that if one filters the DSD output signal with a 20 kHz corner frequency low pass filter, one might get better than 7dB.

I realise - clearly needed more sarcasm/lulz tags on my earlier post (I would have thought using 8-bit PCM as an example gave it away!) - just laughing at the sheer arbitrariness of Sony's graph o' bollocks.
Logged
Indecent lover of cheap opamps...

DaveBSC

  • Best Korean Sympathizer
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +222/-50
  • Offline Offline
  • Posts: 2092
Re: DSD vs CD vs High-Res PCM
« Reply #57 on: December 26, 2013, 07:24:53 PM »
You guys may find this interesting: http://www.audiostream.com/content/ayres-pcm-dsd-comparison

I'm not the biggest Ayre fan, but it seems that Charlie Hansen of Ayre has about the same opinion of DSD as the gentlemen from Schiit... not that it stopped him from putting an ESS9016 in his DAC and adding it as a feature.

http://www.ayre.com/insights_dsdvspcm.htm
Logged

burnspbesq

  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +50/-23
  • Offline Offline
  • Posts: 640
Re: DSD vs CD vs High-Res PCM
« Reply #58 on: December 27, 2013, 05:31:52 PM »
Lavorgna pretty much nails the essential silliness of this conversation.

The simple fact that I've also enjoyed listening to music in all manner of PCM and DSD formats is simply a fact. Nothing I learn will be able to go back and undo that enjoyment. It just so happens that some of the most musically engaging experiences in the digital world I've had, have come about when listening to DSD formatted music files. Does this mean anything more than what I just said? I don't think so but people tend to like easy answers to complicated questions that don't really need to be asked.

http://www.audiostream.com/content/separating-fact-fiction

For all of its known "flaws," my experience has been that far more often than not, DSD gets the most important thing right.  It sounds like music.
« Last Edit: December 27, 2013, 06:09:26 PM by burnspbesq »
Logged

Anaxilus

  • Phallus Belligerantus Analmorticus
  • Pirate
  • **
  • Brownie Points: +65535/-65535
  • Offline Offline
  • Posts: 3493
  • TRS jacks must die
    • The Claw
Re: DSD vs CD vs High-Res PCM
« Reply #59 on: December 27, 2013, 05:47:12 PM »
Quote from: burnspbesq on December 27, 2013, 05:31:52 PM
For all of its known "flaws," DSD gets the most important thing right.  It sounds like music.

I'm not sure how you jump from the Lavorgna comment to your universal claim.
Logged
"If you do not change direction, you may end up where you are heading." - Lao Tzu

"The Claw is our master. The Claw chooses who will go or who will stay." - The LGM Community

"You're like a dull knife, just ain't cuttin'. Talking loud, saying nothing." - James Brown
Pages: 1 2 3 4 5 [6] 7 8 9 10 11
« previous next »