Industry Insiders Q&A MASTER THREAD [separate thread for Xbox/Add On & PS3]

[alfbinet]

Is the XA2 firmware upgradeable to decode DTS HD MA?

[amirm]

Originally Posted by RobertR1

Amir,

Would you please check with Toshiba regarding the XA2 on one thing. Since it has HDMI1.3, will it pass on the DTHD and DTS HD MA streams untouched to a HDMI1.3 receiver and let it handle the decoding?

Thanks,
Robert.

I have been told (second hand) that it does but will confirm.

[bobgpsr]

Originally Posted by amirm

I have been told (second hand) that it does but will confirm.

This is surprising -- or will the XA2 only do this for HD DVDs authored in Basic mode ? -- good news anyway if so . I thought we had it explained to us many times that discs authored in Advanced mode required player decoding, mixing with button sounds and commentary audio and then output. Or is there a secret user option easter egg buried button to pass bitstream if the player sees a HDMI 1.3 receiver downstream that advertises dts-HD MA, etc capability?

[Andy Pennell]

Originally Posted by awp

Amir,

This may have already been asked, but there's 144 pages in this thread now so...

Will this update for the 360 add-on that's coming fix problems with 3X DVDs (HD DVD content on a standard DVD) created by video editing software? Pinnacle, Magix and Ulead all now support this format with their latest versions. However, none of them work directly (you can burn them separately with Nero, but I still have some issues with this).

On the Pinnacle forums, one of the tech support staffers said Microsoft was working on a fix for this. And this was months ago, so I was hoping this would be part of the upcoming fix. Is it?

Thanks!

Yes. The problems had nothing to do with the type of the media, but everything to do with the Standard Content that these tools created.

[amirm]

Originally Posted by bobgpsr

This is surprising -- or will the XA2 only do this for HD DVDs authored in Basic mode ? -- good news anyway if so . I thought we had it explained to us many times that discs authored in Advanced mode required player decoding, mixing with button sounds and commentary audio and then output. Or is there a secret user option easter egg buried button to pass bitstream if the player sees a HDMI 1.3 receiver downstream that advertises dts-HD MA, etc capability?

Isn't there a "bitstream" output format for HDMI audio already? If so, that should be all you need to set. If the receiver supports 1.3 and does MA decoding, then it should "just work." (says he, holding his fingers crossed, given the mess that HDMI is as far as compatibility).

But yes, you would lose interactivity sound plus picture in picture, etc. Same is true of BD format btw. I seem to recall that on the Samsung, one of its outputs has interactivity sounds, and the other does not! The same would happen here in that your analog/SPDIF output will have interactivity and HDMI not. As a work-around, you could hook up both connections and switch between them, say, to listen to Picture in Picture sound. This would be a good test to see how motivated people really are to get that last bit of quality they think they are not getting .

Also, always hearing the interactivity sound is not a mandatory thing. The player must support them, yes. But they can also put in optional features that defeat it such as the above. Content owners are sure to complain, but if the feature is reserved for high-end solutions, maybe it won't be as bad....

Finally, please hang in there until I confirm this is really there.

[bobgpsr]

Originally Posted by amirm

Isn't there a "bitstream" output format for HDMI audio already? If so, that should be all you need to set. If the receiver supports 1.3 and does MA decoding, then it should "just work."
...
Finally, please hang in there until I confirm this is really there.

Here is hoping that it is so! Thanks!

[AlexBC]

Originally Posted by paidgeek

Sorry for the very delayed response.

I understand the interest in having multiple versions on one disc, but there is not a lot of interest in doing this at the studio. I adds significantly to the complexity of preparing the title, in part because of audio splice issues that can arise. We will use this feature on a title this year, but not on "The Patriot". It will go out as the extended version.

Thanks Paidgeek

can we at least expect the TC down the road? I love it, but can't stand the EE.

[RobertR1]

Originally Posted by amirm

Isn't there a "bitstream" output format for HDMI audio already? If so, that should be all you need to set. If the receiver supports 1.3 and does MA decoding, then it should "just work." (says he, holding his fingers crossed, given the mess that HDMI is as far as compatibility).

But yes, you would lose interactivity sound plus picture in picture, etc. Same is true of BD format btw. I seem to recall that on the Samsung, one of its outputs has interactivity sounds, and the other does not! The same would happen here in that your analog/SPDIF output will have interactivity and HDMI not. As a work-around, you could hook up both connections and switch between them, say, to listen to Picture in Picture sound. This would be a good test to see how motivated people really are to get that last bit of quality they think they are not getting .

Also, always hearing the interactivity sound is not a mandatory thing. The player must support them, yes. But they can also put in optional features that defeat it such as the above. Content owners are sure to complain, but if the feature is reserved for high-end solutions, maybe it won't be as bad....

Finally, please hang in there until I confirm this is really there.

Talk about confusing So let me get this right:

Player A, which is HDMI1.3 and is able to pass on the encoded audio stream to the receiver, can only play the main audio track IF the receiver is doing the decoding. If you want the interactivity, the audio must be decoded internally, mixed with the interactivity audio and then output as PCM to the receiver.

So, if you want the full package, wating for a HDMI1.3 receiver that can handle DTS HD and TruHD decoding is pointless unless you only want the main audio track?

[xbdestroya]

Originally Posted by paidgeek

...numerous PS3-related responses...

Hi Paidgeek, first of all I of course want to say I personally appreciate your continued presence on this forum; you're of great benefit in getting questions answered and providing insights.

In terms of Playstation 3-related queries though, I have to wonder why you don't gracefully recuse yourself from that field. It's not a knock against you - believe me - but as someone that remains in close touch with several people over at SCEI, I get the sense sometimes that some of the PS3-related questions are as much an unknown to you as they are to some of the people asking. I just feel that there is a *lot* of drama wrapped up in this device, and personally in your shoes I would just say: "That's a question for SCEI, not for Sony Pictures."

A lot of folk around here seem to either forget or not to know in the first place that SCEI and Sony electronics... let alone Sony pictures... are two entirely different branches of the company that don't talk/deal/coordinate that much with each other save on a very high level.

I guess my specific question is... why even field that line of questioning when you know the black/white answers folk truly want are beyond all of our present abilities to acquire? I mean you've *got* to be tired of dealing with DTS-MA/PS3 questioning by this point; hell this is just the latest in a long legacy on that one aspect alone.

[FilmMixer]

Originally Posted by RobertR1

So, if you want the full package, wating for a HDMI1.3 receiver that can handle DTS HD and TruHD decoding is pointless unless you only want the main audio track?

In my opinion, yes.

[RobertR1]

Originally Posted by FilmMixer

In my opinion, yes.

Thank you. I'll just get a Marantz 7001 to take care of my needs. Now I'm upset I waited this long

The full package is important to me as I have grown a liking to IME and extras.

[paidgeek]

Originally Posted by xbdestroya

Hi Paidgeek, first of all I of course want to say I personally appreciate your continued presence on this forum; you're of great benefit in getting questions answered and providing insights.

In terms of Playstation 3-related queries though, I have to wonder why you don't gracefully recuse yourself from that field. It's not a knock against you - believe me - but as someone that remains in close touch with several people over at SCEI, I get the sense sometimes that some of the PS3-related questions are as much an unknown to you as they are to some of the people asking. I just feel that there is a *lot* of drama wrapped up in this device, and personally in your shoes I would just say: "That's a question for SCEI, not for Sony Pictures."

A lot of folk around here seem to either forget or not to know in the first place that SCEI and Sony electronics... let alone Sony pictures... are two entirely different branches of the company that don't talk/deal/coordinate that much with each other save on a very high level.

I guess my specific question is... why even field that line of questioning when you know the black/white answers folk truly want are beyond all of our present abilities to acquire? I mean you've *got* to be tired of dealing with DTS-MA/PS3 questioning by this point; hell this is just the latest in a long legacy on that one aspect alone.

Thanks for your comments and empathy.

It is accurate to say that I don't have much concrete information on the timing of feature releases from SCEI. Even so, I encourage members to continue asking questions on PS3 matters as I can forward them to the appropriate persons in SCEI.

[Dan Hitchman]

In a perfect world it would be nice if different branches of the same company would communicate more with each other... kind of like the left hand knowing what the right hand is doing.

Seeing as the PS3 has become, whether warranted or not, the player in the forefront of the Blu-ray charge, I can see why people (like myself) would want movie related issues with the PS3 to be resolved quickly and efficiently (DTS MA decoding and/or HDMI 1.3 audio passthrough, black and white level clipping, 1080p/24 output, random Blu-ray disc freezing when the network connection is engaged, quality upscaling, etc.) and am very glad that someone like paidgeek is around and is willing to forward our requests along.

Thank you, once again! Kudos.

[wickedbob]

Does anybody know whether the mpeg4 and h264 support in the new xbox360 update will allow me to view material encoded in 720p and 1080p resolutions?

[Andy Pennell]

Originally Posted by wickedbob

Does anybody know whether the mpeg4 and h264 support in the new xbox360 update will allow me to view material encoded in 720p and 1080p resolutions?

Are you asking about the HD DVD update, or the Dashboard update? The HD DVD player has always been able to decode those resolutions of all supported codecs. (I know nothing about the Dashboard update).

[TomsHT]

Originally Posted by paidgeek

Thanks for your comments and empathy.

It is accurate to say that I don't have much concrete information on the timing of feature releases from SCEI. Even so, I encourage members to continue asking questions on PS3 matters as I can forward them to the appropriate persons in SCEI.

They turn fast on ya when you post an answer that they dont want to hear, all of a sudden your not credible.

If you can add to the whole DTS-HD MA coming to the PS3 it would be helpful. Currently (as with most things hi def here) semantics is running wild on your past statements at to the exact meaning of whether or not the update is coming or not.

If it can be confirmed anyhow that would be great if not it would probably be better to put the whole issue to rest now so we can move on to the next subject.

Thanks Paid

[wickedbob]

Originally Posted by Andy Pennell

Are you asking about the HD DVD update, or the Dashboard update? The HD DVD player has always been able to decode those resolutions of all supported codecs. (I know nothing about the Dashboard update).

Sorry Andy, I thought I made this quite clear.
I mean the mpeg4 and h264 codec playback that is being made available in the new dashboard update.

Ps.. what exactly do you develop at Microsoft?

[John Haghighi]

Originally Posted by amirm

As I have written extensively in the past, detecting compression artifacts is a very challenging thing. Unless one is trained in it, one cannot hear these artifacts after some point. Indeed, even audiophiles have trouble in this regard. So I am not surprised that Peter did not hear improvement in quality. By his ears, DD at 640kbps had already achieved transparency.

I know that he has angst over lack of “high resolution audio” and hence the 4 star rating. But our goal in life is to achieve transparency. And if he can’t hear the difference beyond 640bkps, I submit that he would not hear the difference with lossless either, in a blind test. So the rule of automatically giving one less start due to lack of lossless audio, doesn’t do justice to the fidelity presented. This is especially so if the reviewer has no way of hearing the original to know what could have been there, but was not.


This is absolutely incorrect (it is pretty much non-sense but don’t want to say that and insult someone ). DD encoder when told to product 1.5 mbit/sec as its output, most definitely uses those extra bits to preserve the nuances of transients. Dolby would be the laughing stock of the audio compression community to have done what is claimed above to say nothing of the fact that there is no reason whatsoever to pad the output this way. Whoever thinks DD+ at 640bks and 1.5 mbit/sec sound the same, has no business arguing about lossless audio having merit. As there is far more difference between DD at 640k and 1.5 mbit/sec, than there is between 1.5 mbit/sec and lossless.

At 1.5mbit/sec, the lossy codec is running near the data rate of lossless codec. As such, it is applying so little quantization to the audio samples as to have no detectable effect on them. So the output is going to sound pretty much the same as lossless case. At 640kbps though, we are dealing with 2.5X higher compression ratio and skilled listeners will be able to hear the difference on transients. Non-skilled listeners though, will find the audio to be far better than anything DVD had and enjoy it just the same .



It is really simple. DD has a limitation in how it packetizes its output. To understand this, one needs to keep in mind that even though we use audio codecs in constant bit rate mode (i.e. we specify a fixed data rate like 640bksp), the codecs operate in variable bitrate internally. On steady-state signals (think of a constant tone), the efficiency is quite high and the codec outputs far fewer bits than what it inputs. When transients come, like guitar strings, then the codec must spend considerably more bits to avoid distortions such as pre-echo. So the data rate peaks.

To deal with these variations and output a constant bitrate, the codec uses an output buffer which packs the bits (i.e. averaging peaks and valleys). With right feedback loop from the output buffer, one can assure the target bit rate is hit, even with high and low variations. Turns out this is pretty tricky code to write in the generic sense. The codec has certain boundaries which must be obeyed even with this output buffering as sending incomplete fragments could cause problems if some data is lost (i.e. the receiver cannot recover properly). And this is the problem with DD bitstream syntax. When used in the MPEG transport stream in HD DVD, DD is not able to support data rates above 440kbps because that breaks its packetization logic. The fault here though, is NOT with HD DVD but with the way the codec output buffer system was designed.

DD+ fixes the above restriction, allowing any data rate to be used. As it should be because competing codecs such as our WMA Pro, also can output their data stream at any rate into MEPG transport stream without trouble.

Now, why BD stays with the older and more restrictive syntax is beyond me. Yes, by not using transport stream, they can get around some of the limitations in the DD packetization but they still wind up with a kludgey system when it comes to supporting higher channels which DD+ does natively.

If the above is the reason for hypothesis #2, then I can tell you that people are barking at the wrong tree . Packetization just determines how you pack the bits to be stored. The encoder works the same in both scenarios except that in the case of 1.5 mbit/sec, is able to “breath” a lot better and produce much cleaner transients. Per above though, the reduction in distortion may not be audible to most people. But the encoder definitely is working a lot less in compressing bits when given the larger pipe.

Amir, thanks for your continued detailed explanation on the audio front.
I foresee this coming up again and again, and what are we to take away from this?

Your opinion of what someone can hear is subjective at best correct?
It's fine to make sweeping statements about the masses, but many folks here spend lots of $$$ on equipment and some even have trained musical ears! Do you believe by discounting the possibility of various factors involved in a playback environment you are going to change opinions on this topic?

In the early days of HD DVD and Blu-ray, before it was even released, all the talk by industry insiders was lossless audio and 1080P video because there was plenty space, now that we've got real market products we are seeing detailed explanations of why we shouldn't care for lossless audio because we can't hear the differences. Do you understand why this topic keeps coming up again and again?

If the space is available, give us lossless, allow us to decode and transport it natively, and let the consumer decide what they can and can't hear, and what's marketing fluff...don't we do the same for video codecs (avc vs. vc-1; mpeg2 vs. vc-1)?

[benwaggoner]

Originally Posted by John Haghighi

If the space is available, give us lossless, allow us to decode and transport it natively, and let the consumer decide what they can and can't hear, and what's marketing fluff...don't we do the same for video codecs (avc vs. vc-1; mpeg2 vs. vc-1)?

There's never been a lossless video codec targeted for consumers. Heck, most video content production goes through at least one lossy phase.

[John Haghighi]

Originally Posted by benwaggoner

There's never been a lossless video codec targeted for consumers. Heck, most video content production goes through at least one lossy phase.

Ben to clarify, my point was with video we have cases were we can choose between an MPEG-2 and VC-1 titles and make are own decisions about which we prefer, key point being regardless of what the detailed explanations of why one codec is better than another with certain material, the consumer has a choice, perhaps that's not going to be case in the future as WB seems to be settling on VC-1...

All I'm saying is the studios should give consumers the choice when it comes to audio...if the space is there provide lossless or uncompressed audio...let us decide what to listen too "."

[Andy Pennell]

Originally Posted by wickedbob

Ps.. what exactly do you develop at Microsoft?

I'm on the HD DVD team (ie AmirM is my VP). I've been work mostly on the Xbox player in recent times, though before that I worked on iHDSim (which is the PC "HD DVD simulator" included in the HDi Jumpstart kit for HD DVD authors). Before that I owned the Visual Studio debugger. Before that, well, that was a long time ago.

[amirm]

Originally Posted by dr1394

It's not a DD audio issue, it's a DVD and HD-DVD MPEG-2 Program Stream muxing issue. Dolby added variable blocks per frame not to just have higher bitrates, but to get around the 448 kbps maximum bitrate issue of DVD and HD-DVD Program Streams.

HD DVD has no “448kbps maximum bitrate issue” as otherwise DTS would not be able to stuff 1.5 mbit/sec in there . As we do with WMA Pro just the same. Yes, given the way DD is designed, it suffers from that limiation. But that is why it was changed in the process of creating DD+. And hence my comment on where the blame resides.

Look, I am not here to defend mpeg specifications – we didn’t invent the darn thing . We use our own system stream for Windows Media. The point here is that we and others have no problem putting the output of our audio codecs in the same enclosure in HD DVD/DVD. So to condemn the enclosure because one codec can’t fully utilize it, seems “off base to me.” Not the other way around.

BTW, I'm not an audio expert, but I am an MPEG-2 Systems Transport Stream expert.
Ron

I suspect your expertise here does not explain why BD format seems to limit DD+ to 640kbps for 5.1 tracks in that format. Or why the limited the max rate for DD+ extension to a much lower value than HD DVD.

[Andy Pennell]

Originally Posted by Aitor Ibarra

...is it possible that Microsoft could spend more developer time improving the PC experience?

If you prefer to leave it to others to fill in the gaps, can you / are you doing something to help them? PowerDVD Ultra and the other players are still pretty buggy, the hardware situation is pretty dire too, especially if you're looking for hi-def audio over hdmi.

Yes, we are working with 3rd parties to help them with their PC HD DVD players. We have a special program called the Experience Initiative where we share information with 3rd parties and have "Plug-Fests" to check on title compatibility between the various players. For more info see http://www.microsoft.com/emea/itssho...px?videoid=175 . On the hardware front I know the Windows OEM team are working on the HDMI issues.

[amirm]

Originally Posted by John Haghighi

Your opinion of what someone can hear is subjective at best correct?

Well, it is and it isn't. We have tested hundreds of "subjects" in the last 10 years that we have been doing audio codec development. We have run large scale tests of listeners at Microsoft. We have tested the common man. We have tested audiophiles with golden ears. And we have tested "expert listeners." The only people who can reliably detect codec artifacts in small amounts are the last category (i.e. people who are trained specifically to detect codec artifacts and know what to listen for). This fact has remained constant in all of our testing.

It's fine to make sweeping statements about the masses, but many folks here spend lots of $$$ on equipment and some even have trained musical ears! Do you believe by discounting the possibility of various factors involved in a playback environment you are going to change opinions on this topic?

No I am not and am not attempting that. For enjoyment of music, people can chase the last bit of quality, real or not. I will never engage in that kind of argument. I know better . My beef is with film content. I simply remain unconvinced that people enjoy movies more if they are given that last bit of quality they think they are missing. There is a lot more to a movie enjoyment, than the purist audio experience. I mean when was the last time you saw a movie review that talked about "sound stage" and where the instruments were placed in front him/her? Yet that is common terminology in music reviews.

Movie sound is high synthetic anyway so talking about “lossless” makes one wonder, what is really meant by that. The sound has gone through so many mixing stages and other changes that the notion that it is lossless has little meaning. Lossless relative to what? In music, one can strive for very pure reproduction of what one would hear in the real world. Such does not exist for movies.

In the early days of HD DVD and Blu-ray, before it was even released, all the talk by industry insiders was lossless audio and 1080P video because there was plenty space, now that we've got real market products we are seeing detailed explanations of why we shouldn't care for lossless audio because we can't hear the differences. Do you understand why this topic keeps coming up again and again?

I do but I think it is more of an emotional thing, than a scientific thing. Most people watch movies at 448kbps DD on DVD today. And until now, there were no riots in the streets, with people complaining about audio quality. Yet the kind of people you talk about with $$$ equipment, really, really complained about CD being harsh and such as compared to LP. Why the contrast? The reason must be what I stated. Music and film are two different animals. As is the psychoacoustical aspects of 5.1 film content relative to stereo music.

If the space is available, give us lossless, allow us to decode and transport it natively, and let the consumer decide what they can and can't hear, and what's marketing fluff...don't we do the same for video codecs (avc vs. vc-1; mpeg2 vs. vc-1)?

I can’t argue with this . Studios might, in positioning some titles higher than others. But I won’t argue with it…

[wickedbob]

Originally Posted by Andy Pennell

I'm on the HD DVD team (ie AmirM is my VP). I've been work mostly on the Xbox player in recent times, though before that I worked on iHDSim (which is the PC "HD DVD simulator" included in the HDi Jumpstart kit for HD DVD authors). Before that I owned the Visual Studio debugger. Before that, well, that was a long time ago.

Cool. Thanks for the clarification and your hard work on the xbox player - it is the best way to view HD content in my house.

Do you have access to knowing if Microsoft has plans/ambitions to release a consumer iHD authoring application? Something a bit more "for dummies" than the jumpstart kit.

[TomsHT]

Originally Posted by wickedbob

Cool. Thanks for the clarification and your hard work on the xbox player - it is the best way to view HD content in my house.

Do you have access to knowing if Microsoft has plans/ambitions to release a consumer iHD authoring application? Something a bit more "for dummies" than the jumpstart kit.

Also can iHD be programmed in the VS.Net studio environment?

[Andy Pennell]

Originally Posted by wickedbob

Do you have access to knowing if Microsoft has plans/ambitions to release a consumer iHD authoring application? Something a bit more "for dummies" than the jumpstart kit.

I'm just an engineer, an engineer that values his job, so you won't see any info from me about future plans on anything unless someone else goes officially public with it first. It's no fun I know

At CES I did see Nero claiming HD DVD authoring coming soon as a free update.

[Andy Pennell]

Originally Posted by TomsHT

Also can iHD be programmed in the VS.Net studio environment?

Absolutely, that's what we use! VS has a reasonable XML editor and a Javascript one too, which is all you need for the Advanced content. There's no specific iHD suppport in VS, but as an editor (and a script debugger if using ihdsim) its very usable for HDi (yeah I know, I can't get used to the new name either). It can also handle the big-endian unicode js files that are required, which many other editors can choke on.

[ryanmcv]

Originally Posted by ryanmcv

Hi Amir,

Hopefully you can give some insight on my problem with the Elite.

I have my new Elite hooked up via HDMI to my Samsung 32" LCD HDTV (model LN-S3251D). The dashboard, DVDs, HD DVDs, and downloaded videos appear perfectly. But whenever I start up a game, I begin to have ghosting and after-image issues. Black/dark areas on the screen become a bit "snowy" and I can see the grey outlines of images that were previously there. Colors also appear faded and washed-out. Again, this ONLY happens when I play games. Everything else appears just fine.

I have tried using several different HDMI cables, and I've tried using both of my HDMI ports on the TV. Nothing seems to be helping. When I connect the 360 via component, I have no problems with ghosting. So it doesn't appear to be a problem with the video card on the console. I've also tried hooking it up via HDMI to a Sony 32" LCD and haven't had any problems with that set. I only get the ghosting on my Samsung LCD. I've tried adjusting all of the video settings on the Samsung, but nothing gets rid of the ghosting while playing games.

Any idea what could be causing this? Will the spring update next week address this? Thanks for all of your hard work! I really appreciate the Xbox team's commitment to their customers.

Originally Posted by ryanmcv

Amir,

Just to add to my above post about the ghosting issues I'm having with the Elite:

I have a friend who has the 40" model of that same TV and he is also having this ghosting issue with the HDMI output on his Elite. There is also a thread going on over at the Xbox.com forums where another user is having the same problem on a newer-model Samsung (link). And there are more postings by Samsung-owners on other sites that are reporting this same issue.

There definitely seems to be an issue with the Xbox 360 Elite and certain Samsung HDTVs. I have reverted back to component cables for now, as I simply cannot achieve a good gaming experience through the HDMI port due to the ghosting issues that I described in my above post.

Just wanted to bring this issue back to light, as it has been burried by all these posts about audio on HD DVD and Blu-ray.

[Roger Dressler]

Originally Posted by amirm

...And this is the problem with DD bitstream syntax. When used in the MPEG transport stream in HD DVD, DD is not able to support data rates above 440kbps because that breaks its packetization logic. The fault here though, is NOT with HD DVD but with the way the codec output buffer system was designed.

...HD DVD has no “448kbps maximum bitrate issue” as otherwise DTS would not be able to stuff 1.5 mbit/sec in there. As we do with WMA Pro just the same. Yes, given the way DD is designed, it suffers from that limitation.

The issue with DD bitrate in DVD and likewise on HD DVD has nothing to do with MPEG streams breaking the DD audio. One of the very first DVDs made, the Delos DVD Spectacular, has a hidden DD 640 track on it, which plays fine in Sony DVD players--maybe others, new and old. Then why can’t DVDs use 640? The DVD Forum establishes rules for how the format works. One of the rules requires every elementary stream packet to have a start header in it. It helps with trick play, branching, etc. But the format doesn't break if that rule is violated, as was shown with our secret test disc. (By the way, the Pink Floyd concert DVD "Pulse" also has a 640 track on it.)

In DVD, the audio packets are 2048 bytes, or 16 kbits. One frame of DD is 32 ms long, so 16kb/32ms = 500kb/s. That is the maximum allowed. The highest supported rate for DD is 448 kb/s. The next higher bitrate of 512 kbps would not fit.

So how is it that DTS can have 1.5 Mbps, 3x the 500 kbps rate we calculated? DTS uses a 10ms coding frame (1/3 the duration of the DD frame). Hence, three times the bitrate. (16kb/10ms = 1.6Mbps.)

What we did with DD+ was design the syntax to support frames of 32, 16, 10, and 5 ms, thus allowing up to 3 Mbps in the HD DVD structure.

Originally Posted by amirm

I suspect your [MPEG] expertise here does not explain why BD format seems to limit DD+ to 640kbps for 5.1 tracks in that format. Or why the limited the max rate for DD+ extension to a much lower value than HD DVD.

The BD format is not limiting DD+, per se, except that since DD+ is optional rather than mandatory, Dolby worked out a way to allow the disc to carry the benefits of DD+ (higher bitrates and more channels) while also ensuring the stream can be decoded by standard DD chips. Since standard DD is designed with a max bitrate of 640 kbps, that is the limit for the core 5.1 portion of the stream.

And just so no one thinks I’m seeking to disagree with Amir, I totally agree with his views in Post 4345. It mirrors my own findings. Well said.