Philips DVD-963SA: SACD, DVD, and 24/192 upsampling of CD, for $400?! (1 Viewer)

[KeithH]

Stacey, thanks for the explanation. I am not one of those PCM bashers, but since Sony and Philips are trying to distinguish SACD by the DSD encoding scheme, I would think that there would be impetus to develop a DSD-based bass management scheme. Could one be developed? In any event, I suppose 32/88.2 PCM isn't bad.
I agree that the James Taylor Hourglass SACD sounds great. The SACD is definitely better than the CD. As a result, I do not dismiss SACDs of PCM recordings out of hand, though some folks do.
As for lossy compression, I guess we might see future SACD players with some sort of decoding chip. I certainly hope that lossy compression does not catch on with SACD. Oh well. Just give me a stereo SACD track (full resolution), and I could care less what else is on the disc.

 

[John Kotches]

Keith,

The real question to be asked about JTs Hourglass is what would it sound like with no PCM --> DSD conversion. I don't know with any certainty the recordings bit rate, but I get the feeling it's either 20/48 or 24/48.

Regards,

 

[Eujin]

I agree that PCM based SACDs aren't necessarily bad at all. I've never listened to Hourglass, but I really enjoy the Patricia Barber SACDs I have in my collection, which are also PCM based.

 

[AustinKW]

Keith,

Pioneer says there is currently no DSP capable of dealing directly DSD for bass management/time alignment. The 47Ai feeds DSD over the iLink to the 49TXi which then converts it to 32/88.2 PCM then applies BM/TA and so on.

Austin

 

[Phil Nichols]

Stacey,

You are saying two different things in your post above:

"The 2310 has the ability to offer aspect ratio control."

"Before a DVD forum licensee can release a DVD player that outputs anything >480p, the CSS agreement needs to be updated with these new resolutions. I will warn you know that when these new resolutions come, it will be over DVI(HDCP)/HDMI only. There will not be any analog >480p that is allowed."

Never mind the ">480p" definition of scaling, since video information not on the DVD disk can't be created by DVD players or external to DVD player outputs - even by the mighty S&W, Faroudja, or Teranex video processors.

How about the "other" definition of scaling that it appears a whole lot of HT DVD player owners want implemented - the "resizing" of 480p images to either fit non-anamorphics to 16X9 screens or cropping of 2.35:1 anamorphics to better fit 16X9 screens. I don't think the CSS agreement cares about this. I "think" the S&W, Faroudja 3000/4000/5000, and the Teranex processors do this resizing and the FLI23XX chip family may permit this too(?) ....... but inexpensively in integrated circuit form.

A) Do you know if the FLI23XX IC family does this, and if yes,

B) Will the Philips 963a and other players due out with the FLI23XX chip(s) have this functionality finally implemented simultaneous with their great F deinterlacing?

What the world needs is the Mediamatics chip with it's no chroma upsampling error and superb digital-taps resizing ("scaling"), but with DVDO or Faroudja deinterlacing!

How close does the FLI23XX family get us to the 480p "resizing" market place requirement?

 

[StaceyS]

How about the "other" definition of scaling that it appears a whole lot of HT DVD player owners want implemented - the "resizing" of 480p images to either fit non-anamorphics to 16X9 screens or cropping of 2.35:1 anamorphics to better fit 16X9 screens.

That is called "Aspect ratio control" and I mentioned it above.

A. Yes.

B. Don't know.

The FLI23xx has 2x the digital taps as the Mediamatics.

 

[John Kotches]

Frankly, this seems like a suboptimal setup to me. Let's look at the signal processing chain:

DSD to PCM Conversion ---> Bass Mgmt + Time Alignment ---> PCM to DSD Conversion ---> D/A Conversion

Once you've done the DSD to PCM conversion what's the point of the last PCM to DSD conversion, other than to call it SACD?

It ceased to be SACD/DSD the moment a DSD to PCM conversion was applied.

The utilization of 32bit/88.2K for bass management indicates that (contrary to claims otherwise) SACDs resolution is no better than 2fs is for delivering content in the frequency domain.

Regards,

 

[AustinKW]

Huh? Where did the last PCM-->DSD conversion come from? The 49TXi converts once to PCM applies BM/TA then D/A's the PCM and onto the amps. Agree that DSD dies at the PCM border but that's the way of it until DSD DSPs are available. Of course, same thing's going on in the players that apply BM to SACD. No such thing as a free lunch!

Austin

 

[John Kotches]

Austin,

THis was in reference to the Philips schematic which shows the signal path I illustrated.

Regards,

 

[StaceyS]

John

Perhaps the reason for going back to DSD has to do with the analog filter section. This would make since if they converted the CD PCM to DSD but if not then you are correct.

 

[John Kotches]

Stacey,

An interesting experiment would be having someone implement an analog filter on Hi-Resolution PCM which mimics the filter on DSD.

Someone once wrote "In the future all we will be listening to is different digital filters."

 

[Dan Hitchman]

What is Sony/Philips thinking?? They need to have DSP chips that perform in DSD mode pronto!

If you leave the bass management off this player and also off in the Pioneer Elite 49TXi receiver, does the DSD stream stay that way to the DAC's (which also must be able to handle DSD correctly) or are we stuck with DSD to PCM converting all the time?

Dan

 

[KeithH]

That last PCM-to-DSD conversion step is interesting. It would be interesting to compare the sound quality with and without this last step.
John, perhaps this last PCM-to-DSD conversion is performed to introduce high-frequency noise, which is the sonic signature of SACD. O.K., where's Lee?

 

[KeithH]

Dan, good question. Perhaps disabling the bass management feature would be like a "DSD-to-Analog Direct" mode, whereby the PCM circuitry would be bypassed. In that case, I would be interested in comparing the sound quality when using the player's bass management circuitry (digital bass mangement) versus an ICBM (analog bass management).

 

[John Kotches]

Keith,

Andre Yew (I don't recall which forum this was) posted some information about SACD and its potential effect on tweeters.

He pointed out that the ultrasonic noise heats up the tweeter voice coil, which acts as an additional smoothing filter. I'd have to dig around to find it, but thought that was an interesting observation if accurate.

Regards,

 

[KeithH]

John said:

He pointed out that the ultrasonic noise heats up the tweeter voice coil, which acts as an additional smoothing filter.

Interesting. It's funny to think how these sorts of things might be interpreted. I suppose while some could interpret the "smoothing filter" as providing an improvement in the sound, others will probably interpret it as providing a coloration to the sound. To each his own.
John, by the way, I too would be interested in hearing Hourglass in its native PCM state, as CD is downsampled a bit. Alas, it ain't going to happen.

 

[Ron Newsome]

AustinKW
Is DSD->PCM also being done in the top of the line Sony ES SACD players?

 

[KeithH]

Ron, note that the Sony players do not offer an adjustable crossover or time alignment. All you can do with the Sony players is set the speaker configuration. I don't know if these settings would involve a DSD-to-PCM conversion.

 

[Ron Newsome]

But when you set the speaker configuration to large fronts and small everything else, plus a subwoofer, and you're rerouting all the bass from the small speakers to the sub, aren't you using a crossover? The crossover may not be adjustable but it's got to be one there. And since there are no DSD DSPs to perform Bass Management wouldn't it have to be converted to PCM first? Could Sony be doing DSD->Analog, then using analog crossovers to perform the bass management? Even in the cheaper models?

 

[KeithH]

Ron, I thought about this some more after entering my previous post. You obviously did too, as you make some excellent points. Time alignment aside, there is a crossover to consider when speakers are set to small. In the Sony players, the crossover is fixed at, I believe, 120 Hz (which is too high). I doubt the Sony players are doing bass management in the analog domain. Chances are, bass management is being done via a conversion to PCM. It's been said that bass management cannot be done in the DSD domain, but I am not sure why that should be.