Rerurn to Romy the Cat's Site

Didital Things
Topic: DVD-A should have been great but...

Page 1 of 1 (5 items)


Posted by el`Ol on 02-14-2008
Hello!

I hadn´t cared about the topic for years, but recently I informed myself about SACDs and DVDs at Amazon. The product range is still ridiculous, yet I find it has at least "started to start". But it seems to have happened what I expected at the very beginning of the talk about the new formats: PCM has become the standard for pop, bitstream for classical productions. The only serious label that had supported the DVD (MDG) has also switched to the SACD.

Has anyone observed how this happened?

Oliver

Posted by Romy the Cat on 02-14-2008

The product range is unfortunately growing. I have seen quite a few recordings that I was interested but they were recorded in SACD and therefore were compromised by the format. I would not give a lot of credit to association between format and the musical genre – if some recording labels go with one or other format then it has very little to do with consideration  of actual sonic result but it is driver by other, mostly business, factors.

Here is my very brief observation of formats. There are well documented expert analyses about the SACD vs. DVD with measurements, knowledge and evidences – my observation is nothing like that. My observation is just a user experience and reflects what I personally have witnessed.

SACD – very bad format. The original, phenomenally sounding (I heard it), brilliantly conceptualized, 4-bit DSD conceived by Ed Meitner in 90s was trashed out and replaced by a cheap commercial adaptation of 1-bit SACD that is just “not enough” for good sound. There are no good sounding SACD players, not good SACD DACs, no good SACD recorders - the entire SACD world is a huge unfortunate lead astray. The PCM 16/44 layer that is typically released on the SACD disks usually is horrible.

DVD-A - strangely bad format. Any single DVD-A disk I head sounded wrong – I have no idea why. If I have a good 24/94 files and put in DVD-A disk then the quality of sound go down to drain.

CD – very bad format. 16-bit is very low resolution, it is just not enough. Still, the 16/44 file might do relatively acceptable but as soon I convert it into CD files then all quality evaporates.

Some generalizations

1)    The physical disks that require file transformation are evil

2)      Sampling rate or resolution conversion are evil

3)      Any DSP activation is evil

4)      The raw files (regardless of format) should be juts copied to disks, distributed and then just played as raw files.

The subject of change the quality of sound after a raw file was copied and re-saved is under investigation.

Rgs, Romy the Cat

Posted by Telstar on 02-14-2008
...But they are poorly implemented. Something to do with the 24/192 conversion. AFAIK nowadays most recordings are done 24/96, but it will eventually go up when higher resolution devices becomes the standard (I think about Prism, but theres more).

Out of my curiosity, with which spinner did you try DVD-Audio-s? I think that The Esoteric UX-01 is the current best (as transport).

I obviously agree with everything else that Romy said.

Posted by Telstar on 02-14-2008
http://www.iar-80.com/page48.html

That made me change my mind about getting one.

It's a very long and technical read, but it is a must-read.

I'm not sure that all double format CDs are horrible on the redbook side, but i may be wrong, as my chain is not near as revealing as yours.

Posted by el`Ol on 02-14-2008
No wonder 4-bit DSD sounded so good. 5-6bit ~3Mhz is how most of the ADCs and DACs work anyway. In principle I am in favour of not changing the sample rate. However the problem I see is the 7th order noise shaping. This is done by combining 3rd and 4th order noise shaping in ways that are even hard to analyze mathematically. What I find interesting is that even Sony has left the path of the true no-SRC-believers. Their new digital receivers work with 8bit ~350kHz PWM, same at TI/TacT/Lyngdorf. In my opinion this would also have been the best compromize for audio coding.

Page 1 of 1 (5 items)