SACD flawed to begin with? S&V David Ranada's article:

JediFonger · 07-07-2006, 08:38 PM

P. 82 of the July/Aug Sound and Vision Magazine, David Ranada writes:

Quote:

... 4). a mathematically proven flaw in the 1-bit DSD (Direct Stream Digital) encoding system that requires it to produce noise modulation and distortion, albeit at low levels; and 5) the inability of conventional digital signal processing to efficiently handle the 1-bit encoding, which accounts for the continued lack of good bass management and speaker-distance compensation in most SACD-capable players.

those 2 points were news to me. although it's hard for me to hear the diff. between the CD & SACD, does this mean that SACD is still introducing noise or will that *always* be inherent in all digial sound?

the second point is that what does encoding have to do with decoding?

JeremyErwin · 07-07-2006, 11:32 PM

Quote:

the second point is that what does encoding have to do with decoding?

DSD is fundamentally different form PCM.

PCM: At this point in time, the signal has this particular value.

DSD: At this point in time, the signal is less than or greater than the last sample.

With a bit of creativity, you could argue that DSD was somehow more similar to vinyl records.

But PCM was a well established technology. Hundreds of thousands of mathematicians and engineers had published reams of papers on the manipulation of PCM and similar forms of data. It's relatively trivial to analyze the frequency components in a piece of PCM data. But DSD?

You could convert the DSD into PCM, apply the FFT, do some manipulations, apply an inverse FT, and then convert the result back into DSD. But Sony claimed that some benefits of DSD were lost during the conversion process. You could also develop a new algorithm to work on the DSD directly, but such development costs time and money.

The benefits of SACD, as far as I'm concerned, lies in the depth of the music library.

Kevin C Brown · 07-08-2006, 01:56 AM

I actually think the benefit of SACD, unlike DVD-A, is that you don't need a freaking video display to play the discs.

I have also seen the noise argument put forth for SACD. I see two problems with that: a) The noise is high enough in frequency that humans can't hear it, and, b) I have enough SACDs and DVD-A's to compare that if there is extra noise there, I don't hear it.

I have also heard the contention that the presence of that noise does make SACD sound more analog. I don't think I believe that either. Both SACD and DVD-A sound more "analog" simply because both are a much better digital representations of the original analog waveform than CD is.

I have heard that the noise is automatically filtered out if you convert to (high rate) PCM before going to analog. Most i.Link and HDMI connections result in this if you're going to use any extensive BM or TA in the receiver or pre/pro. And, players like the 3910 and 5910 that do extensive BM and TA themselves also do a DSD to PCM conversion too.

gene c · 07-08-2006, 08:17 AM

And to me, the biggest problem with SACD is the lack of video content! Oh well. Different strokes for....

Felix Martinez · 07-08-2006, 08:42 AM

Quote:

I actually think the benefit of SACD, unlike DVD-A, is that you don't need a freaking video display to play the discs.

I've had my projector for 5 years now, and rarely fire it up to play DVD-As - unless I want to see the video content. I just hit stop when the disk spins up and then press play. I can browse the Groups with the remote as well. Works this way with my Onkyo sp800 universal player and Panny RP-91.

But...I cannot toggle between the stereo and m/c layers of SACDs...I have to go to the Onkyo set up screen for that - and there I do need a monitor.

**HTF Ads**

ChristopherDAC · 07-08-2006, 09:16 AM

Mathematically, the dynamic range of a one-bit (delta) modulation system such as DSD falls off with increasing frequency. Simultaneously, because the "slew rate", or size of each up or down step, is fixed, there is a great deal of "hunting" involved in following the slope of the input waveform. It's this latter phenomenon which generates the noise inherent in the process. The noise is greatest at high frequencies, but it actually extends down to the very lowest frequencies. In essence, while the "image noise" produced by spurious frequency components in a PCM system is all above the sampling frequency [twice the maximum reporduced frequency] as long as a proper anti-alias filter is used, in a one-bit system a large fraction is actually within the passband. It is for this reason that SACD players incorporate a 50 kHz brick-wall filter [thus limiting the frequency band to about the same as 96 kHz PCM on DVD-A], to avoid the high frequencies damaging your amplifier.

There was a period when 1-bit analog-to-digital and digital-to-analog conversion circuits were widely used in recording and reproducing equipment, with internal conversion to or from PCM, because they were cheaper and easier to build than actual multibit ADCs and DACs which would achieve 16-, 20-, or 24-bit performance. Sony's papers and promotional materials tout the advantages which DSD does have when used in an environment of 1-bit devices, by avoiding multiple stages of PCM-delta conversion. To the best of my knowledge, however, while 1-bit ADCs and DACs are still used in the majority of low-end and general consumer products, they have fallen out of use in professional and audiophile devices, except those dedicated to SACD use!

Incidentally, for a quality comparison, I read an article about the remix of — I am fairly certain — Brothers in Arms, the Dire Straits album. The recording engineer described his experiences comparing the various audio formats, and it came out to something like : 24/96 PCM is best, then 1-inch 30ips analog tape, then DSD, then 15ips tape.

Manus · 07-08-2006, 12:32 PM

" It is for this reason that SACD players incorporate a 50 kHz brick-wall filter "

Funny thing about this was the dummy who worked on the Sacd of 'Dark side of the moon' showed his ignorance by actually claiming at the launch that this was a flaw of the Dvd-A format !

" I actually think the benefit of SACD, unlike DVD-A, is that you don't need a freaking video display to play the discs "

My Meridian player I just press play and the default mlp 5.1 track starts playing ; pressing the Audio button changes between the Audio tracks if required and Video is only really required if you want to view extras.

~M~

gene c · 07-08-2006, 01:18 PM

Different players, and different discs, handle the start-up differently. On my most played DVD-A's I remember which button or two to push so turning on the display isn't always needed. My new H/K 47 dvd player also has an audio button on the remote that switches between formats "on the fly", or so it says in the manual. Haven't tried it out yet though. BTW, with what you must have payed for that Meridian, it should bring you a cup of coffee and the newspaper in the morning too! Yes, that's envy talking.

LanceJ · 07-08-2006, 03:02 PM

From the pro audio discussion forums: supposedly because of the noise inherent in DSD, DSD>PCM>DSD conversions need to be limited to less than @5. Past that, then all that noise starts to reveal itself by giving the music a "veiled" quality, which is most apparent in the higher frequencies.

And on a personal note, I don't much trust a system that only uses data that says "change" or "no change" to describe a waveform.

BTW: dvd-audio discs can authored so they don't need ANY onscreen navigation to operate them or any video portion *period*. For some reason, except for that Elv1s dvd-audio (a BMG title, before Sony bought them), the labels seems to think everyone likes farting around with TVs to get to the stereo or bonus tracks on the disc. The audio button does usually work with most newer discs to do this, but AFAIK the "Group" button was really the button that was supposed to be used to change between sets of tracks. But for some reason only a handful of dvd-audio players include them, usually just Panasonics (who designed most of the dvd-audio format's hardware), JVCs and Denons.

Considering that the head of the Dvd-Audio Council is also the CEO of Silverline, IMO I'm totally not surprised the standards for dvd-audio are so effed up.