So is Dual Disc making a splash or sinking? (1 Viewer)

[Kris Deering]

I'm surprised to hear you say that Lee. Since you work in the recording industry you of all people should know that the benefits of 24 bits far outweigh the higher sampling rate. Since 48Khz allows for 24Khz of usable information we're getting far more then we can hear already, but 24 bits adds A LOT of added dynamic range and is FAR BETTER then redbook in this respect. Remember that simply adding one bit doubles the resolution. Try adding 8.

I agree that 96/24 should be the goal, but depending on the master used, it can be a waste as well. Upsampling doesn't achieve anything in these applications except takes up space. Universal has proven time and again with both their DualDisc releases and DVD-As that they are only going to use what the master requires. I imagine that the majority of their SA-CD catalog is incredibly oversampled as they no doubt use the same masters.

 

[Kris Deering]

Oh, I forgot. I picked up a few of the Elton John SA-CDs that were just released. Incredible to say the least. Definately better then Yellow Brick Road so far.

 

[Thomas Newton]

Shouldn't that be, "if it isn't 88.2k or 96k or higher, your cat or dog isn't hearing anything more than Red Book"?

 

[PaulDA]

That's the sample rate, not the sound frequency. And Kris is right, in PCM 24 bits is far more important. Even HDCDs, at 20 bits, are an improvement in resolution. The problem is comparing apples to oranges, as DSD works on a 1-bit foundation, that's extremely highly sampled. My (limited) understanding is the extreme sampling rate of DSD is required to give hi-res from 1 bit, whereas, at 24 bit, hi-res is achieved at a much lower sample rate. I know I'm leaving out massive amounts of technical info, but the basic logic applies. Just because DSD requires such a high sample rate doesn't mean PCM requires 96khz and above to be considered "HI-RES". They are not the same, nor should they be directly compared on these variables.

 

[ChristopherDAC]

Take it from me, sampling rate and dynamic range are not direct tradeoffs. Close, but no cigar. I don't think DSD is a great idea, for this reason, and I wouldn't support 1 kHz 256 bit sampling either! Suffice to say that 16:48, while sufficient for most practical purposes, is not perfectly transparent, and only slightly more so than 16:44.1 -- as I recall, we would need to bump it out to about 20:60 and for reasons dealing with equipment limitations the recording should be, if possible, sourced higher than that. It's always best to sample the master at the highest possible quality, even if [due to the limitations of the master itself] the records are going to be pressed at a lower quality. So Mr. Scoggins has a point.

 

[Danny Tse]

http://biz.yahoo.com/prnews/050126/nyw057_1.html

I wonder if the certification is for Simple Plan's album in general, or just the DualDisc version of the album.

 

[Lee Scoggins]

Kris,

That is simply not true. In PCM and DSD sampling rate is incredibly important. This is why Chesky Records records in 24/96 or higher. I would like to have more word length also, but the sonic improvement on that is one tenth of what going from 48k to 96k does in terms of realism.

 

[Rachael B]

I think it's incredulously impossible that Simple Plan has sold 1 million copies of a Dualdisc as of now. Some baloney is phony or badly written or both.

 

[Lee Scoggins]

By the way, the reason I know this is that I was present during the recording of albums where the word length improved and later when the sampling rate went from 48k to 96k. No comparison my friend.

I use an Alessis deck to make local recordings at 88.2. But I sometimes record in 44.1 to show people the difference. When played back on my Maggies, you can hear easily the added tonality on the instruments, particularly any acoustic or small ensemble work we are doing.

Buy some of the Red Book Cheskys and compare these with their DVDA or DSD discs of the same title and you will "hear the difference" as we say.

 

[Lee Scoggins]

Rachael,

This is hilarious and then the Warner guy talks about how strong DualDiscs sold during teh holiday season. He must be trying hard to defend against all the negative publicity and press saying the exact opposite and the stories about consumers not seeing the discs anywhere.

Classic.

 

[Kris Deering]

Where is this negative press and publicity?? I have heard negative comments on these forums from members, but this is far from press and publicity.

DualDisc is making the same mistake DVD-A and SA-CD made, TOTAL lack of publicity and consumer awareness. Since it was posed to replace CD (not DVD-A or SA-CD mind you), people need to know what the hell it is bringing to the table. I think if the studios wanted to get REALLY serious about DualDisc, they would just skip a standard CD release all together and release major titles in DualDisc only. If you did this with really big titles people would become more aware and decide for themselves if this is worth getting or not. Personally I think Sony and company should have done this from the start with SA-CD, similar to DSOTM.

Lee,

You can't compare a Chesky redbook to a DVD-A or SA-CD. Because you are not on a level playing field in bit depth. If the CDs were 44.1/24 then that would be different. Bit depth directly correlates to dynamic range so the benefits are HUGE. The only way to properly compare higher sampling rates is to ensure that they are both at the same bit depth. I would wager that a 96/16 recording wouldn't sound much better then a 44.1/16 recording. And I don't even want to bring SA-CD into this, its dynamic range is only good in the lower frequency band anyways. Anything above the audible frequency band is more noise then anything else. This is why the SA-CD standard specifies that you have to do a brick wall filter at 20 kHz whenever you are doing measurements of DSD. It is also why they recommend a brickwall filter for any system that uses more then 100 watts of amplification at 50kHz. I know Denon has a toggle for just this in their SA-CD setup. Many of the popular "budget" players from Pioneer and other companies have this as a hard filter with no adjustment so that people don't have to worry about frying their amp or speakers.

 

[Rachael B]

Lee, I was just looking at the latest mailer from Elusive Disc. Now, Warner has a bunch of "DVD-A + bonus CD" releases lined up, including several R.E.M. discs. If Dualy-Disc is such a smashing hit, why are they back to DVD-A's + a CD? They want $24.95 for these sets too. No matter what the format/set-combination is, will be, might be, it cannot cost $24.95!:frowning: If Sony started making SA-CD's again and wanted $24.95, I'd say the same thing. Format roulette continues. I think consumers should boycott the majors at this point. That's what they deserve. They have walked on us customers a step too far.

It always comes back to the same thing. If consumers won't pay $5 to $10 more, they won't supply hi-rez, and mostly moldy oldies if they did.

I'm getting fed up...I'm beginning to rant...

 

[John Kotches]

ChristopherDAC says:


To be transparent to the end user, and across even humans with the highest tested acuity, the channel should be a minimum of 22 bits @ 65kHz. Keep in mind, this bit depth doesn't take into account any signal processing, and once you add in signal processing you get to 24-bits being necessary pretty quickly. Look up "Acoustic Renaissance for Audio" and you'll find much more information.

24-bits was chosen for a few reasons. 96K is the next logical stopping point from 48K.

Cheers,

 

[Kris Deering]

Warner already came out and said why they were doing this.

Warner's head of music stated directly that the choice of doing a 2-disc release or a DualDisc release for multi-channel releases is entirely up to the artists. If they want to do it as a DualDisc, Warner will support it. If they want the 2-disc route ala The Flaming Lips, they will support that. They said that some of the artists preferred the 2 disc approach with the CD as a bonus to the DVD-A. REM was one of those.

I think this is a nice consideration by Warner but I think the retail price of the 2-disc releases are a bit too high.

 

[John Kotches]

Kris D says:


I think Capt. Fantastic sounds the best, but others probably have a different opinion.

Cheers,

 

[Rachael B]

Kris, " ...a bit too high." How 'bout $10 too much?...!

 

[John Kotches]

Lee says:


They're both equally important.

Sampling depth in PCM runs into theoretical noise issues across the 1-4K region below about 20-21 bits of delivered sampling depth. Sampling rate in PCM gets into issues with input filters, pre/post ringing etc etc at lower sampling rates.

Your discussion of sampling/depth rate is quite the red herring when talking about DSD. DSD is a one-dimensional sampling technique, all that can vary is the rate. To get something resembling high fidelity, you have to sample at 64fs. That's barely adequate to the task.

256fs or 512fs gets a bit more interesting as these could maintain essentially linear performance out to comparable performance with ~20bit/96K and ~22bit/192K. The
negative is that it's relatively data intensive.

Philips seems to be taking a different approach with DXD (Digital eXtreme Definition). 24bit/352kHz is the approach. This figure is vaguely familiar -- it's the same as the Pyramix mixing stations run internally.

Cheers,

 

[Lee Scoggins]

Kris, you should know better as an editor of an online magazine. Sampling rate is incredibly important. Every engineer I know accepts this as truth and also true is that sampling rate is more important.

Look, the comparison with Chesky is an accurate comparison because they are using 24 bits in much of the chain. Going to 24/96 is mostly showing the impact of the sampling rate. The extra definition and transient capture is almost entirely due to the extra data being recorded.

John, you are also wrong. They are not "equally important" - sampling rate wins the race.

Now as far as DSD, I brought that up as an example of how superfast sampling rate leads to great sound. Unfortunately you cannot compare the two since the recording techniques are entirely different since DSD looks at CHANGES in height and PCM looks at AMPLITUDE. The bit comparisons make no sense. By the way John, DSD is not one dimensional-it records two dimensional waveforms by including changes in height and time on the x and y axis, this is no less robust than PCM recording amplitude and time as the x and y axis.

If you doubt me, find a local engineer with an Alessis deck and record two inputs, one at 24/48k and the other at 24/96k. You will be surprised at the difference.

 

[John Kotches]

Lee says:


Can you modify the sampling depth of DSD? Nope. It is by its very nature a fixed depth of 1-bit. Since the only thing that can change is the rate, you have a one-dimensional sampling function. If you would rather that I say you can only vary one dimension (time/frequency) that's a valid point.

The bit comparison makes perfect sense, if you look at the technical performance parameters of DSD vs linear PCM.

Equivalent PCM performance for 64fs DSD in the 10-20K octave is ~15 bits, 5-10K ~18 bits and below this, it's in the same range as PCM with between 20 and 22 effective bits delivered. Because of the noise incursion, you have output that really should be filtered @ 25kHz which is roughly equivalent to PCM @ 50kHz, which is conveniently close to 48K.

If you push out to 128fs, it becomes comparable performance to 24/96K, since the filter can get pushed up into the 50K range. Unfortunately, this necessitates a data rate that is literally 2x the 24/96K PCM rate.

Once you go out to 256fs, you finally reach 24/192K performance, but with a per channel data rate of 10 Mbits/second, it's about 2x that of 24/192K PCM.

Philips, Sony's SA-CD partner, is pushing DXD as their next generation digital technology. It is 24bit/352K PCM. If DSD is so great, why is Philips' strategy to go away from DSD for archiving/recording? Conveniently, the data rate for 24bit/352K PCM is only about 75% of 256fs DSD, which means a nice savings in terms of mass storage.

So one half of the "dynamic duo" of SA-CD is "going back" to PCM. Here's just one link of many describing DXD.

Pretty damning to me that Philips doesn't want to capture all the noise at the start of the chain.

Cheers,

 

[Marc Colella]

Thanks for the info on DXD John. I've never even heard of it until now.

It sounds like it deserves it's own thread.