DVD-A Stereo Blunders

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Cai Campbell

In Remembrance
Since 2002/2003
Joined
Mar 4, 2002
Messages
2,970
Location
Seattle, WA
Okay Quadfather, please don't yell at me for talking about stereo here... :p

I've been listening to my DVD-A discs in stereo lately, more for an academic exercise than anything else. I feel a lot of "audiophiles" are dismissing DVD-A because of the format's emphasis on multi-channel and I think they are really missing out. I mean, the DVD-A hi-res stereo mixes sound phenomenal, and in my opinion, are better than SACD.

Unfortunately, it can often be a chore to get DVD-A discs to play in stereo. In every disc I've checked, the default play mode is multi-channel, making it easy to get a DVD-A "up and running" in multi-channel mode without having to use the on-screen menu. However, stereo guys have a bit more to contend with. Because of the lack of standardization in the DVD-A menus, you really need to use the on-screen menu to select the stereo tracks. The only alternative to avoiding the on-screen menu is to memorize the menu beforehand and perform the task “blind”.

This brings me to a fact that has me perplexed. Virtually all DVD-A titles claim that they can be played in hi-res surround or stereo, but many of them (roughly 50% of the 80 or so titles I own) do not contain separate stereo play lists. I can adjust my player for stereo or multi-channel but what does this really achieve in a case where there are not separate play lists?

To test this, I compared many DVD-A titles, containing a single play list, in both stereo and multi-channel mode. I paid very close attention to see if I could detect something “missing” or glaringly different moving from multi-channel playback to stereo playback.

Basically, I found three different types of behavior:

1. Stereo playback outright missing information found in multi-channel playback.
2. Some instrumentation in stereo playback sounds too low in mix compared to multi-channel playback.
3. Equal balance of all instrumentation in stereo playback compared to multi-channel playback, but lacking presence found in a dedicated stereo mix.

To expand on this explanation, here are three DVD-A titles that illustrate these results clearly and concisely. All three of these titles have a single play list (i.e. no separate stereo play list) and claim to support playback in hi-res stereo and multi-channel.

1. Grover Washington Jr. – Winelight.

This title offers stunning, discrete multi-channel performance, with rhythm guitar and supporting percussion mixed strictly to the back channels. After switching the player to stereo mode, the instruments directed to the back channels simply disappear!

2. Natalie Merchant – Tigerlily

Although not 100% discrete, this is still a great mix. Basically, it seems to be a stereo mix in the front channels and a stereo mix in the rear channels. However, the balance of instrumentation is different between the two. The piano and guitar are mixed more towards the back channels. After switching to stereo mode, the piano and guitar are still there, but they are buried further down in the mix. This results in stereo performance that is very unbalanced compared to multi-channel playback.

3. Bela Fleck – Bluegrass Sessions

The multi-channel mix here is very discrete, with some (but not much) commonality between the fronts and rears. After switching the player to stereo mode, all information previously heard in the fronts and backs are maintained at the same levels. At first it would seem we finally have an instance where stereo and multi-channel exists correctly together on a DVD-A with a single play list. However, upon further listening, something does not sound quite right. The sound stage collapses and the mix sounds too thick. The individual instruments do not stand out as they should and a muddy pall seems to hang in the air. I have come to the conclusion that this is simply “fold down” stereo, blindly mixing the fronts and the rears in lieu of a “proper” separate stereo mix.

In cases 1 and 2, the player is simply tossing out the rear channels. In case 3, the rear is blended to the front. Neither result is satisfactory and stereo listeners are being short-changed. Big time.

The bottom line is that DVD-A must offer a separate stereo mix for it to live up to its promise of stunning hi-res performance for stereo listeners. Not having a separate stereo mix is most likely a cost-cutting measure by the companies that do not offer them. This is a huge mistake, since it can only hurt the format and ostracize an entirely separate market (i.e. stereophiles).

Even if all DVD-A titles carried a separate and proper stereo mix, this market of stereophiles might still dismiss DVD-A as a viable format because of the lack of standards in navigation and the virtual requirement that a television set be employed. The BEST method I have seen that addresses all concerns is a double-sided DVD-A disc with a stereo mix on one side and a multi-channel mix on the other. This should have become THE standard for DVD-Audio, and if it had been adopted across the board, I think the DVD-A format would be embraced by just about everyone and give SACD an even bigger fight.

Okay, so why the hell do I care about all this? I mean, I listen to my DVD-A discs in multi-channel mode, so why should I be crying for the stereo guys? Well, I happen to like this format. A lot. I’ve done extensive listening to DVD-A and SACD and I much prefer DVD-A. If there are more people that like, and buy, these discs, then more titles will be produced. It just pains me to see the potential slipping away.

Meanwhile, SACD has a virtual stranglehold on the audiophile community. Sony’s marketing muscle and production resources promise to make SACD a standard. Titles by CCR, The Stones and The Who are all slated for SACD. Would I rather see these on DVD-A, even if they are not multi-channel? You bet.

End of rant. Hopefully you’ve found some insightful information here. For what it’s worth, I plan on generating a list of titles that are known to have a separate stereo mix so that we can point our stereo friends in the right direction. It’s damage control time for DVD-A short-sightedness.


 
Try the first example again but connect the stereo Left Right outputs instead of the 5.1 left right output. It could be a player 'feature'.
For example, if I select the stereo track of Queens ANATO I get no sound from the 5.1 left right output and I must connect the normal left right outputs on my Denon 3300...

Luckely most of the latest warner releases use a 24bit 192ks stereotrack and they are direct transfers from the stereo masters. (Carly Simon, Eagles, Randy Newman and America)
I think because the 192ks equipment and mastering tools are just recently becoming available.

I think this has nothing to do with 'DVD A short sightednes' but is the result of inexperienced mastering practices.
The titles in your examples are among the first titles that where produced for DVD A.

Luckely It seems that Sony is in for some damage control too.
It seems that a number of sacd releases where mastered from 16bit 44.1 pcm sources. They went for catalog quantity instead of quality.

Frank


 
When I play stereo, I switch my player/receiver to the dedicated stereo outputs/inputs. When I play multi-channel, I switch my player/receiver to the dedicatated 5.1 outputs/inputs. I peformed the proper setup for ALL my experiments, and as you see, I got different results from different discs. Does anyone else here have Winelight or the other discs I mentioned? Please repeat my experiment and tell me if you get different results. I can cite other examples of this erratic behavior if you want them...

I am absolutely convinced that a DVD-A without a dedicated stereo mix will always be compromised when played back in stereo. The only exception to this I can see is a surround mix that uses a derived center and/or rears from the full frontal stereo mix. If I dug a little deeper I could probably find an example of this (and it would most likely end up being a Silverline disc!) In this situation, it may be the stereo playback that sounds better and surround playback is the one that suffers!

Actually, the Winelight disc is one of the newer releases. The Shortsightedness I refer to is the lack of a well-defined standard. Having an open standard may seem appealing at first, but I am seeing too big of a downside here. Consumers should not have to suffer as companies learn how to do it right. The standard should have been defined correctly from the get-go. Also, the open standard will lead to different companies finding different "right" ways of doing things...

Yes, the 24/192 stereo mixes are absolutely stunning, but it is not the resolution that I am complaining about, it is simply the absolute lack of a dedicated stereo mix and the expectation that audio compromises in stereo playback are to be accepted by the consumer. Whether it be 24/192, 24/96, 24/88.2 or even 16/44.1, that would be fine, as long as there was SOMETHING mixed properly for stereo. Heck, look at DTS Entertainment. They offer (of all things) a DOLBY DIGITAL 2.0 track as their stereo solution! I think that's pretty stupid and is probably driven by their desire to make DD sound as bad as possible (most compression plus least channels equals worst sound translates to DD is bad and DTS is good). Whatever their reason, at least they are offering something in stereo.

If you want to talk about the MOST recent titles hitting the market, look no farther than Silverline to see how bad they can screw things up. NONE of their releases offer a separate stereo mix. This could almost be forgiven if their releases were worth half a damn, but as it is they are shooting themselves in both feet (crappy sound and lack of stereo mix).

 
I know what you mean. Because of the additional material on the Doors' "L.A. Woman" dvda, I made a cd of it to play at work and amaze friends and influence people. I had to make an analogue recording off the stereo outs, which is when I noticed that the stereo mix was incredibly muddy. At work I have a Sansui quad receiver, and decided to give the mix a try in QS - I have to say it is one of the best QS mixes I have ever heard. I recommend listening to the stereo mix this way.
 
Actually, L.A. Woman does have a separate stereo mix. It sounds pretty good, but has an inordinate amount of tape hiss compared to the surround mix. I'm guessing the surround mix was mastered from the original multi-track tapes and the stereo mix was just a straight transfer from existing stereo tapes.

That's interesting that it would perform so well through a variomatrix decoder. I wonder if the original stereo release also decodes well?

 
No, it doesn't decode all that well on the standard stereo release. The stereo mix on the dvda is definitely encoded - it can't be an accident.
 
Wow, that's bizarre... so it must be an entirely new stereo mix as well! Your contention that the sound is muddy in stereo might be an indication that the stereo mix is a "fold-down" of the surround mix. This could also be the reason why it decodes well. It might be that the rear was folded out-of-phase to the fronts to create a mix that would maintain some semblance of depth, since a straight fold-down would result in a compressed soundstage. The out-of-phase information would then be picked up the quad decoder for decent quad playback. Anyway, just a theory...

But, if this is correct, it brings to light another disturbing idea. A separate stereo play list offers no guarentee that you will get decent DVD-A stereo performance...

 
I'm going to have to go back and listen, as when I first got the LA WOMAN DVD-A, I pressed play and it started in stereo as my receiver was set to "auto". At first, I moved back to reset my receiver to 6 Ch EXT, but then I stopped because the sound was amazing!!

It was one of the few times I actually noticed a stereo playback as being so defined and clear!?

Maybe my medication was wearing off!! :D j/k

:-jon


 
Me? yell at you? Oh no, not this little mouthed frog! In fact I think you have a good point. We definitely want stereophiles to be happy with DVD Audio discs. What I would like to see is one of these formats emerge as a standard and replace the CD. And I really don't care which one. (except that I would prefer DVD Audio, since I already have a bunch of them.) I just hope this format war doesn't kill 'em both like the quad format wars.

The Quadfather
 
Okay, slight correction to my DVD-A stereo rant a while back. The Grover Washington Jr. - "Winelight" disc DOES have a separate stereo play list, and it is 24/192 to boot! It sounds great! Anyway, sorry for the confusion.

In looking into this whole stereo thing a little further, I am finding that (generally) only the earlier releases on the major labels (e.g. Warner) are missing a separate stereo play list (e.g. Natalie Merchant - "Tigerlily").

The one label that consistently excludes a separate stereo play list (with only a couple of exceptions) is Silverline. Unfortunately, since they have released a veritable glut of DVD-A titles, they alone really skew the list of titles without stereo play lists. Quantity, not quality, is their creed and I still believe they are doing a disservice to the format with their horrible surround mixes and lack of a stereo play list.

 
Actually, the Carly Simon "No Secrets" DVD-A was mastered with a Sony Sonoma DSD system - not a PCM 24/192 master.

>>Luckely most of the latest warner releases use a 24bit 192ks stereotrack and they are direct transfers from the stereo masters. (Carly Simon, Eagles, Randy Newman and America)

I think because the 192ks equipment and mastering tools are just recently becoming available.<<


 
>>Actually, the Carly Simon "No Secrets" DVD-A was mastered with a Sony Sonoma DSD system - not a PCM 24/192 master.

Actually the Sonama DSD System operates at 32bit 8fs in the processing stages.
Only the last mastering step ,down converting to DSD was not executed because two 24bit 192ks tracks where needed for the disc.

So far I did not find any 'evidence' to confirm this statement about the title being mastered on the Sonoma system.

Frank


 
The discussion about the No Secrets DVD-A being mastered with the Sonoma DSD Editing system was in several of the trade magazines months ago. It was apparently discussed at some seminars regarding hi rez disc production.

More recently, there was an article in Billboard Magazine about the production team of the group Korn using the Sonoma and the Meitner DSD DACs. In this case, DSD was used to print the final 2 channel mix and master for their new album. (Check page 49, June 29th issue of Billboard - still available at some Tower Records stores).

By the way, this isn't too surprising since the engineer on the Carly Simon remix and the Korn project is the same guy - Frank Filipetti. Filipetti is a big fan of DSD and is also the producer/remix engineer who handled the excellent Multichannel SACD of "Hourglass" by James Taylor.

 
So the Sonoma system was used.

But that system is working in the pcm domain and only at the final mastering step there is a conversion to a DSD stream.

Now there are two possibilities:
The Sonoma can output 24/192 PCM as well
or
can only output DSD and that's convertered to 24/192.

Now it's clear to me why I had my doubts about the Carly Simon stereo track. It sounds edgy and not as transparant and smooth as the Randy Newman and the Eagles track.
I thought it was the bad state of the analog tape, now it's possible that the dsd to pcm conversion could also be the culprit.

Frank
 
Back
Top