Industry Insiders Q&A Thread: only Questions to insiders please - Page 16 - AVS Forum
Forum Jump: 
 
Thread Tools
post #451 of 4623 Old 07-17-2006, 07:37 PM
AVS Club Gold
 
rdjam's Avatar
 
Join Date: Mar 2005
Location: Miami, FL
Posts: 9,736
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by Talkstr8t
Microsoft chose to apply the "iHD" label to both the HD-DVD standard and advanced content modes. The BDA labels their standard mode as "HDMV" and the advanced mode as "BD-J". This semantic allows Amir to correctly note that all current US HD-DVD titles make use of iHD. However, only two titles (to my knowledge), "Constantine" and "Bourne Supremacy", make use of iHD advanced content (in order to provide PiP).
Quote:
Originally Posted by Richard Paul
Amir, is this accurate?
I think you're missing the point here, Richard. As Bill stated above, all HD DVD players will handle both the advanced content, as well as the standard, which is why the "advanced discs" that Bill mentions will work on these 1G HD DVD players. However, 1G BD players only support BD-J "standard" profile, but cannot do the BD-J "advanced" profile stuff, even if someone released a BD movie with BD-J advanced content.


Quote:
rdjam, where did you hear this rumor from?
Doesn't matter, but it seems that it is true that 1G BD players do not support the full BD-J specs. How about saving the questions for the insiders?
rdjam is offline  
Sponsored Links
Advertisement
 
post #452 of 4623 Old 07-17-2006, 07:54 PM
AVS Special Member
 
Talkstr8t's Avatar
 
Join Date: Jul 2001
Location: Los Altos, CA, USA
Posts: 2,753
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
rdjam, rather than trying to define specific BD-J API's or subsets, I'll lay it out out very clearly:

BD-Video profile does not require PiP, network, and enhanced persistent storage support, though any of those features may be present if a manufacturer chooses to include them. BD-Live profile requires all those features. All other capabilities are present in all BD players regardless of profile. BD-J content not requiring PiP, network, or enhanced persistant storage facilities will operate on all BD players. BD-J content requiring those facilities will work on all BD-Live players. Content requiring one or more of those capabilities on a BD-Video player without support for that capability will either provide a suitable error message, provide reduced capability, or not be available, depending on how the content developer chooses to author the content.

I can't get any more clear than this, and the concept really should not be as difficult as you're making it sound.

- Talk

Blu-ray Insider
Speaking solely for myself, not the BDA
Talkstr8t is offline  
post #453 of 4623 Old 07-17-2006, 08:08 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,228
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 869 Post(s)
Liked: 517
Quote:
Originally Posted by RobertR1
Amir,

What aspects do you feel that HD DVD needs to concentrate and improve on to come out ahead in the long run? I'd just like your personal take on this.

Thanks,
Robert.
Tough question to answer here with all of our competitors looking on :).

But at high level, our main concern is meeting demand across the board. We have more companies wanting to build players now. More post houses want to author/encode content in VC-1/iHD. And more studio content to publish (like Studio Canal). And of course, making sure the right message is communicated to the public and hence my past time here :).

If we execute well on above, the rest takes care of itself. There will be more players (at various price points). More and better content. And proper consumer prespective on great value of HD DVD and its A/V quality and interactivity.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #454 of 4623 Old 07-17-2006, 08:23 PM
AVS Addicted Member
 
AnthonyP's Avatar
 
Join Date: Apr 2001
Location: Montreal, Canada
Posts: 17,632
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
just a general question on audio.

can audio be divided into tracks and then played back together. What I mean is let's say we have English, French and Spanish on a disk. One obvious way would be to have English, French and Spanish tracks. But could one have a generic track (songs, special effects......) that don't need dubbing and English dialog, French dialog and Spanish dialog tracks, so when you pick a language it is combining in real time the generic and appropriate dialog tracks? I am thinking more about lossless. According to Dolby lossless has a max of 18mbps and at that level none of these formats have enough for more then one lossless track and good video if they are both peeking at the same time (and that tends to happen often :) ). But if they can split it then generic could need as high as 18 but the dialogs most likely would not need anything near there.

PS would appreciate an answer for HD DVD and BD

PPS it would be interesting, if possible, to get your opinion on if studios will use it.
AnthonyP is offline  
post #455 of 4623 Old 07-17-2006, 08:23 PM
AVS Club Gold
 
rdjam's Avatar
 
Join Date: Mar 2005
Location: Miami, FL
Posts: 9,736
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by Talkstr8t
rdjam, rather than trying to define specific BD-J API's or subsets, I'll lay it out out very clearly:

BD-Video profile does not require PiP, network, and enhanced persistent storage support, though any of those features may be present if a manufacturer chooses to include them. BD-Live profile requires all those features. All other capabilities are present in all BD players regardless of profile. BD-J content not requiring PiP, network, or enhanced persistant storage facilities will operate on all BD players. BD-J content requiring those facilities will work on all BD-Live players. Content requiring one or more of those capabilities on a BD-Video player without support for that capability will either provide a suitable error message, provide reduced capability, or not be available, depending on how the content developer chooses to author the content.

I can't get any more clear than this, and the concept really should not be as difficult as you're making it sound.

- Talk
Thanks Bill - that's very helpful.

What "category" does PiP during a movie fall into, "BD-Video" or "BD-Live"?

I ask because you mentioned that the 1G BD players would not be able to perform this, but that you wouldn't rule out the possibility of a firmware update.

Are there any other features like PiP that you didn't mention?
rdjam is offline  
post #456 of 4623 Old 07-17-2006, 08:25 PM
AVS Club Gold
 
rdjam's Avatar
 
Join Date: Mar 2005
Location: Miami, FL
Posts: 9,736
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Never mind the last question - it was answered already :D
rdjam is offline  
post #457 of 4623 Old 07-17-2006, 08:27 PM
AVS Club Gold
 
rdjam's Avatar
 
Join Date: Mar 2005
Location: Miami, FL
Posts: 9,736
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by Talkstr8t
rdjam, rather than trying to define specific BD-J API's or subsets, I'll lay it out out very clearly:

BD-Video profile does not require PiP, network, and enhanced persistent storage support, though any of those features may be present if a manufacturer chooses to include them. BD-Live profile requires all those features. All other capabilities are present in all BD players regardless of profile. BD-J content not requiring PiP, network, or enhanced persistant storage facilities will operate on all BD players. BD-J content requiring those facilities will work on all BD-Live players. Content requiring one or more of those capabilities on a BD-Video player without support for that capability will either provide a suitable error message, provide reduced capability, or not be available, depending on how the content developer chooses to author the content.

I can't get any more clear than this, and the concept really should not be as difficult as you're making it sound.

- Talk
Thanks for the answer - I agree it is quite simple to understand now.
rdjam is offline  
post #458 of 4623 Old 07-17-2006, 08:27 PM
AVS Special Member
 
HorrorScope's Avatar
 
Join Date: Oct 2003
Location: NoCal
Posts: 4,251
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
I asked a few pages back. Could iHD be used on a HD DVD player to act like a internet browser? If so could PiP play a part in this as well? Thanks.
HorrorScope is offline  
post #459 of 4623 Old 07-17-2006, 08:46 PM
AVS Special Member
 
Richard Paul's Avatar
 
Join Date: Sep 2004
Posts: 6,959
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 29
Quote:
Originally Posted by rdjam
Doesn't matter,
rdjam, why not tell us where/who you heard it from?
Richard Paul is offline  
post #460 of 4623 Old 07-17-2006, 08:50 PM
AVS Special Member
 
skogan's Avatar
 
Join Date: Feb 2004
Location: Koror, Palau
Posts: 4,415
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Richard Paul
rdjam, why not tell us where/who you heard it from?
With respect, you and rdjam have the whole rest of the forum to argue on. Please don't dilute this thread with your cross-talk.

...and finally,


Thank you to the insiders who made this the most interesting place on the web today.
skogan is offline  
post #461 of 4623 Old 07-17-2006, 09:03 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,228
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 869 Post(s)
Liked: 517
Quote:
Originally Posted by HorrorScope
I asked a few pages back. Could iHD be used on a HD DVD player to act like a internet browser? If so could PiP play a part in this as well? Thanks.
Sorry, I must have missed it. The core components in iHD can be re-used to support a browser. And if by PiP you mean fetching data remotely and displaying it in a Window, I don't know why not.

Of course, the above requires more software above and beyond what is already in HD DVD spec....

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #462 of 4623 Old 07-17-2006, 09:38 PM
AVS Addicted Member
 
Ken H's Avatar
 
Join Date: Nov 1999
Location: Metro Detroit
Posts: 45,876
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by Richard Paul

I will try, but I think that if we can't debate anything that is said by the insiders that the thread could easily become a one sided propaganda campaign.
Let us be the judge of that. Hopefully, all insiders can say what they want and readers can judge for themselves.

I can suggest starting other topics, that can be referenced here, for your type of rebuttal.

'Better Living Through Modern, Expensive, Electronic Devices'

Ken H is offline  
post #463 of 4623 Old 07-17-2006, 10:01 PM
AVS Addicted Member
 
Ken H's Avatar
 
Join Date: Nov 1999
Location: Metro Detroit
Posts: 45,876
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
This is going very nicely, I think.

Thanks to all who are participating, and perhaps even more so, to everyone who is complying with this topic's really specific rules.

'Better Living Through Modern, Expensive, Electronic Devices'

Ken H is offline  
post #464 of 4623 Old 07-17-2006, 10:19 PM
AVS Special Member
 
rover2002's Avatar
 
Join Date: Sep 2005
Location: Hong Kong
Posts: 2,690
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Amir,
There has been to date no animated release on either format.Is this due to a code problem or is there just no rush from studios to get 1 or 2 out? You would think that the quality of a CGI movie coded into HD/BR would be breathtaking and a huge bonus in either format.
Will.
rover2002 is offline  
post #465 of 4623 Old 07-17-2006, 10:22 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,228
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 869 Post(s)
Liked: 517
Quote:
Originally Posted by rover2002
Amir,
There has been to date no animated release on either format.Is this due to a code problem or is there just no rush from studios to get 1 or 2 out? You would think that the quality of a CGI movie coded into HD/BR would be breathtaking and a huge bonus in either format.
Will.
There is no technical reason. The studios pick their titles and I am not privy to their logic. They don't tell us how to design a codec, we don't tell them which movie to release first :).

But I agree they should look magnificent given the ease of encoding. Next in line though is movies for which DI is available. We are going to see some great examples of that soon...

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #466 of 4623 Old 07-17-2006, 10:28 PM
AVS Special Member
 
Talkstr8t's Avatar
 
Join Date: Jul 2001
Location: Los Altos, CA, USA
Posts: 2,753
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by amirm
Quote:
Originally Posted by HorrorScope
Could iHD be used on a HD DVD player to act like a internet browser? If so could PiP play a part in this as well?
The core components in iHD can be re-used to support a browser. And if by PiP you mean fetching data remotely and displaying it in a Window, I don't know why not.

Of course, the above requires more software above and beyond what is already in HD DVD spec....
And not that you asked about Blu-ray, but BD-J can support it simply by loading a browser written in Java from the disc. This is a quite common method of authoring content for MHP, the broadcast standard which shares a common Java foundation with BD-J.

- Talk

Blu-ray Insider
Speaking solely for myself, not the BDA
Talkstr8t is offline  
post #467 of 4623 Old 07-17-2006, 10:30 PM
AVS Special Member
 
Robert D's Avatar
 
Join Date: Dec 2005
Posts: 4,083
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by amirm
There is no technical reason. The studios pick their titles and I am not privy to their logic. They don't tell us how to design a codec, we don't tell them which movie to release first :).

But I agree they should look magnificent given the ease of encoding. Next in line though is movies for which DI is available. We are going to see some great examples of that soon...
DI?
Robert D is offline  
post #468 of 4623 Old 07-17-2006, 10:31 PM
AVS Special Member
 
RobertR1's Avatar
 
Join Date: Nov 2005
Posts: 3,777
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 25
Amir,

DI = ???

Also, will you let us know when LOTR and Matrix trilogy are being worked on? No release dates, just a hint. For the record, they're on the lookand.....com so technically they have been announced in some fashion thus you really wouldn't be in breach on the nda's. :)

Thanks,
Robert.
RobertR1 is offline  
post #469 of 4623 Old 07-17-2006, 10:31 PM
AVS Special Member
 
Talkstr8t's Avatar
 
Join Date: Jul 2001
Location: Los Altos, CA, USA
Posts: 2,753
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I'll ask one of my own for the encoding experts here. My wife and I watched "Good Night and Good Luck" the other night, and I got to wondering if a black and white film requires dramatically less space when being encoded? I would assume so, given how much smaller a black and white jpeg is than the equivalent color one. Can anyone confirm whether this is the case?

- Talk

Blu-ray Insider
Speaking solely for myself, not the BDA
Talkstr8t is offline  
post #470 of 4623 Old 07-17-2006, 10:34 PM
AVS Special Member
 
tkmedia2's Avatar
 
Join Date: Apr 2002
Location: CA, WA, USA
Posts: 2,293
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
DI=Digital intermediate
A final digital version of the movie before being printed on release prints. manipulated digitally not analog methods. most commonly used currently for colour, colour adjustment.
tkmedia2 is offline  
post #471 of 4623 Old 07-17-2006, 10:42 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,228
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 869 Post(s)
Liked: 517
Quote:
Originally Posted by Talkstr8t
I'll ask one of my own for the encoding experts here. My wife and I watched "Good Night and Good Luck" the other night, and I got to wondering if a black and white film requires dramatically less space when being encoded? I would assume so, given how much smaller a black and white jpeg is than the equivalent color one. Can anyone confirm whether this is the case?

- Talk
There is savings but not as much as one would think. For digital distribution (broadcast, HD optical, internet, VOD, etc.) we severely "subsample chroma" meaning the color resolution is much less than black and white content (luma). The formal term for this is 4:2:0 which contrasts with 4:4:4 which is used on the computers you are using to view this post (different color space but you get the picture). In simple terms, for every pixel, we transmit 8 bits for luma (black and white) and only 4 bit for color, instead of 16.

The above is done because the eye is not very sensitive when it comes to detecting sharpness in the color channel. So black and white movies while easier to encode, there is not a ton of savings.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #472 of 4623 Old 07-17-2006, 11:39 PM
Senior Member
 
Ian_S's Avatar
 
Join Date: Mar 2006
Location: (Old) Hampshire, UK
Posts: 232
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Are there any plans for either format to support the output of movies at 24fps instead of saddling us with 3:2 pulldown or over here (UK) PAL speedup... Is there any particular reason why movie disc playback seems to be stuck in traditional 50 and 60 Hz modes?

It would be really nice to have only 24fps encoded discs and screens that would support their playback at this speed. It is in the HDMI spec as an optional timing.

Ian.
Ian_S is offline  
post #473 of 4623 Old 07-17-2006, 11:54 PM
Senior Member
 
Ian_S's Avatar
 
Join Date: Mar 2006
Location: (Old) Hampshire, UK
Posts: 232
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by amirm
There is savings but not as much as one would think. For digital distribution (broadcast, HD optical, internet, VOD, etc.) we severely "subsample chroma" meaning the color resolution is much less than black and white content (luma). The formal term for this is 4:2:0 which contrasts with 4:4:4 which is used on the computers you are using to view this post (different color space but you get the picture). In simple terms, for every pixel, we transmit 8 bits for luma (black and white) and only 4 bit for color, instead of 16.

The above is done because the eye is not very sensitive when it comes to detecting sharpness in the color channel. So black and white movies while easier to encode, there is not a ton of savings.
I thought the HDMI spec supports either 4:2:2 or 4:4:4?? Does this mean VC-1 pads out data to meet that requirement?

Ian.
Ian_S is offline  
post #474 of 4623 Old 07-17-2006, 11:59 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,228
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 869 Post(s)
Liked: 517
Quote:
Originally Posted by Ian_S
I thought the HDMI spec supports either 4:2:2 or 4:4:4?? Does this mean VC-1 pads out data to meet that requirement?

Ian.
Output format is independent of the source. HD DVD/BD only support 4:2:0 sampling for all three codecs. The output link then upsamples it as needed to 4:2:2 or 4:4:4 (often incorrectly :)). But there is no data beyond 4:2:0 in the source.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #475 of 4623 Old 07-18-2006, 12:13 AM
AVS Special Member
 
dr1394's Avatar
 
Join Date: May 2002
Location: Mizar 5
Posts: 3,159
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 21
Quote:
Originally Posted by Talkstr8t
I'll ask one of my own for the encoding experts here. My wife and I watched "Good Night and Good Luck" the other night, and I got to wondering if a black and white film requires dramatically less space when being encoded? I would assume so, given how much smaller a black and white jpeg is than the equivalent color one. Can anyone confirm whether this is the case?

- Talk
A typical 4:2:0 MPEG-2 sequence expends on average about 25% of the total coded DCT coefficient bits on chroma. This makes sense because each macroblock contains 4 luma blocks and 2 chroma blocks (1 for Cb and 1 for Cr). Here's a dump of some typical video:
Code:
Interactive MPEG Decoder intd 0.72A
@ Statistics output turned on.
@  Found Sequence Start Code at byte position 0.
* --------  Sequence Start Code at 0        ---------
* --------       GOP Start Code at 22       ---------
* --------          Intra frame at 30       ---------
* Frame number: 2              Average mquant: 3.21 
* --- Number of bits used ---
*     High level headers :   8805  (0.79%)
*     Macroblock headers :  47948  (4.29%)
*         Motion vectors :      0  (0.00%)
*       DCT coefficients : 1060563  (94.92%)
*                  Luma  : 820629  (77.38%) 
*                Chroma  : 239934  (22.62%) 
* --- Macroblock types ---
*                  Intra :   8160  (100.00%) 
* --------  Bidirectional frame at 139989   ---------
* Frame number: 3              Average mquant: 4.55 
* --- Number of bits used ---
*     High level headers :   7444  (2.22%)
*     Macroblock headers : 112149  (33.50%)
*         Motion vectors :  95500  (28.53%)
*       DCT coefficients : 119689  (35.75%)
*                  Luma  :  88732  (74.14%) 
*                Chroma  :  30957  (25.86%) 
* --- Macroblock types ---
*                  Intra :     45  (0.55%) 
*          Frame Forward :   1845  (22.61%) 
*          Field Forward :     66  (0.81%) 
*         Frame Backward :   5872  (71.96%) 
*         Field Backward :     27  (0.33%) 
*    Frame Bidirectional :    154  (1.89%) 
*    Field Bidirectional :      8  (0.10%) 
*                Skipped :    143  (1.75%) 
* --------  Bidirectional frame at 182103   ---------
* Frame number: 4              Average mquant: 3.28 
* --- Number of bits used ---
*     High level headers :   7440  (1.67%)
*     Macroblock headers : 131419  (29.47%)
*         Motion vectors :  94157  (21.12%)
*       DCT coefficients : 212895  (47.74%)
*                  Luma  : 160392  (75.34%) 
*                Chroma  :  52503  (24.66%) 
* --- Macroblock types ---
*                  Intra :     35  (0.43%) 
*          Frame Forward :   3908  (47.89%) 
*          Field Forward :     41  (0.50%) 
*         Frame Backward :   3864  (47.35%) 
*         Field Backward :     52  (0.64%) 
*    Frame Bidirectional :    160  (1.96%) 
*    Field Bidirectional :      6  (0.07%) 
*                Skipped :     94  (1.15%) 
* --------      Predicted frame at 238201   ---------
* Frame number: 5              Average mquant: 2.22 
* --- Number of bits used ---
*     High level headers :   7444  (0.83%)
*     Macroblock headers : 126886  (14.19%)
*         Motion vectors : 103744  (11.60%)
*       DCT coefficients : 655924  (73.37%)
*                  Luma  : 494258  (75.35%) 
*                Chroma  : 161666  (24.65%) 
* --- Macroblock types ---
*                  Intra :     58  (0.71%) 
*          Frame Forward :   7721  (94.62%) 
*          Field Forward :    301  (3.69%) 
*                Skipped :     80  (0.98%) 
* --------  Bidirectional frame at 350230   ---------
* Frame number: 6              Average mquant: 4.51 
* --- Number of bits used ---
*     High level headers :   7444  (2.21%)
*     Macroblock headers : 114993  (34.17%)
*         Motion vectors :  93682  (27.84%)
*       DCT coefficients : 120419  (35.78%)
*                  Luma  :  83728  (69.53%) 
*                Chroma  :  36691  (30.47%) 
* --- Macroblock types ---
*                  Intra :     74  (0.91%) 
*          Frame Forward :   5426  (66.50%) 
*          Field Forward :     74  (0.91%) 
*         Frame Backward :   2267  (27.78%) 
*         Field Backward :     28  (0.34%) 
*    Frame Bidirectional :    106  (1.30%) 
*    Field Bidirectional :     11  (0.13%) 
*                Skipped :    174  (2.13%) 
* --------  Bidirectional frame at 392712   ---------
* Frame number: 7              Average mquant: 4.50 
* --- Number of bits used ---
*     High level headers :   7440  (2.22%)
*     Macroblock headers : 115117  (34.33%)
*         Motion vectors :  93362  (27.84%)
*       DCT coefficients : 119410  (35.61%)
*                  Luma  :  82104  (68.76%) 
*                Chroma  :  37306  (31.24%) 
* --- Macroblock types ---
*                  Intra :     80  (0.98%) 
*          Frame Forward :   3903  (47.83%) 
*          Field Forward :     44  (0.54%) 
*         Frame Backward :   3807  (46.65%) 
*         Field Backward :     55  (0.67%) 
*    Frame Bidirectional :     82  (1.00%) 
*    Field Bidirectional :      7  (0.09%) 
*                Skipped :    182  (2.23%) 
* --------      Predicted frame at 435010   ---------
* Frame number: 8              Average mquant: 3.19 
* --- Number of bits used ---
*     High level headers :   7444  (1.26%)
*     Macroblock headers : 123337  (20.91%)
*         Motion vectors : 103264  (17.50%)
*       DCT coefficients : 355928  (60.33%)
*                  Luma  : 267252  (75.09%) 
*                Chroma  :  88676  (24.91%) 
* --- Macroblock types ---
*                  Intra :     34  (0.42%) 
*          Frame Forward :   7753  (95.01%) 
*          Field Forward :    271  (3.32%) 
*                Skipped :    102  (1.25%) 
* --------  Bidirectional frame at 509135   ---------
* Frame number: 9              Average mquant: 3.74 
* --- Number of bits used ---
*     High level headers :   7444  (1.87%)
*     Macroblock headers : 124339  (31.29%)
*         Motion vectors :  94906  (23.88%)
*       DCT coefficients : 170685  (42.95%)
*                  Luma  : 127323  (74.60%) 
*                Chroma  :  43362  (25.40%) 
* --- Macroblock types ---
*                  Intra :     39  (0.48%) 
*          Frame Forward :   4332  (53.09%) 
*          Field Forward :     88  (1.08%) 
*         Frame Backward :   3421  (41.92%) 
*         Field Backward :     10  (0.12%) 
*    Frame Bidirectional :    129  (1.58%) 
*    Field Bidirectional :     10  (0.12%) 
*                Skipped :    131  (1.61%) 
* --------  Bidirectional frame at 559161   ---------
* Frame number: 10             Average mquant: 3.73 
* --- Number of bits used ---
*     High level headers :   7440  (1.87%)
*     Macroblock headers : 122835  (30.88%)
*         Motion vectors :  93724  (23.56%)
*       DCT coefficients : 173742  (43.68%)
*                  Luma  : 130223  (74.95%) 
*                Chroma  :  43519  (25.05%) 
* --- Macroblock types ---
*                  Intra :     38  (0.47%) 
*          Frame Forward :   2640  (32.35%) 
*          Field Forward :     23  (0.28%) 
*         Frame Backward :   5170  (63.36%) 
*         Field Backward :     39  (0.48%) 
*    Frame Bidirectional :    108  (1.32%) 
*    Field Bidirectional :     10  (0.12%) 
*                Skipped :    132  (1.62%) 
* --------      Predicted frame at 609187   ---------
* Frame number: 11             Average mquant: 2.60 
* --- Number of bits used ---
*     High level headers :   7444  (0.93%)
*     Macroblock headers : 125215  (15.69%)
*         Motion vectors : 103622  (12.98%)
*       DCT coefficients : 561859  (70.40%)
*                  Luma  : 428627  (76.29%) 
*                Chroma  : 133232  (23.71%) 
* --- Macroblock types ---
*                  Intra :     50  (0.61%) 
*          Frame Forward :   7754  (95.02%) 
*          Field Forward :    264  (3.24%) 
*                Skipped :     92  (1.13%) 
* --------  Bidirectional frame at 709256   ---------
* Frame number: 12             Average mquant: 3.73 
* --- Number of bits used ---
*     High level headers :   7444  (1.81%)
*     Macroblock headers : 126604  (30.84%)
*         Motion vectors :  94640  (23.06%)
*       DCT coefficients : 181798  (44.29%)
*                  Luma  : 136732  (75.21%) 
*                Chroma  :  45066  (24.79%) 
* --- Macroblock types ---
*                  Intra :     60  (0.74%) 
*          Frame Forward :   2799  (34.30%) 
*          Field Forward :     52  (0.64%) 
*         Frame Backward :   4963  (60.82%) 
*         Field Backward :     29  (0.36%) 
*    Frame Bidirectional :    122  (1.50%) 
*    Field Bidirectional :     13  (0.16%) 
*                Skipped :    122  (1.50%) 
* --------  Bidirectional frame at 760938   ---------
* Frame number: 13             Average mquant: 3.74 
* --- Number of bits used ---
*     High level headers :   7440  (1.85%)
*     Macroblock headers : 126246  (31.47%)
*         Motion vectors :  94213  (23.49%)
*       DCT coefficients : 173212  (43.18%)
*                  Luma  : 130903  (75.57%) 
*                Chroma  :  42309  (24.43%) 
* --- Macroblock types ---
*                  Intra :     44  (0.54%) 
*          Frame Forward :   2983  (36.56%) 
*          Field Forward :     13  (0.16%) 
*         Frame Backward :   4827  (59.15%) 
*         Field Backward :     58  (0.71%) 
*    Frame Bidirectional :    101  (1.24%) 
*    Field Bidirectional :      7  (0.09%) 
*                Skipped :    127  (1.56%) 
* --------      Predicted frame at 811332   ---------
* Frame number: 14             Average mquant: 2.64 
* --- Number of bits used ---
*     High level headers :   7444  (0.94%)
*     Macroblock headers : 125723  (15.88%)
*         Motion vectors : 102701  (12.97%)
*       DCT coefficients : 556075  (70.22%)
*                  Luma  : 409386  (73.62%) 
*                Chroma  : 146689  (26.38%) 
* --- Macroblock types ---
*                  Intra :     72  (0.88%) 
*          Frame Forward :   7717  (94.57%) 
*          Field Forward :    277  (3.39%) 
*                Skipped :     94  (1.15%) 
* --------  Bidirectional frame at 910665   ---------
* Frame number: 15             Average mquant: 3.71 
* --- Number of bits used ---
*     High level headers :   7444  (1.82%)
*     Macroblock headers : 124989  (30.61%)
*         Motion vectors :  94901  (23.24%)
*       DCT coefficients : 180975  (44.32%)
*                  Luma  : 135095  (74.65%) 
*                Chroma  :  45880  (25.35%) 
* --- Macroblock types ---
*                  Intra :     42  (0.51%) 
*          Frame Forward :   2912  (35.69%) 
*          Field Forward :     70  (0.86%) 
*         Frame Backward :   4875  (59.74%) 
*         Field Backward :     16  (0.20%) 
*    Frame Bidirectional :    120  (1.47%) 
*    Field Bidirectional :      9  (0.11%) 
*                Skipped :    116  (1.42%) 
* --------  Bidirectional frame at 961979   ---------
For black and white encoding, there's 20 to 30 percent more bits available for the the luma DCT coefficients.

Ron

HD MPEG-2 Test Patterns http://www.w6rz.net
dr1394 is offline  
post #476 of 4623 Old 07-18-2006, 12:41 AM
AVS Special Member
 
dr1394's Avatar
 
Join Date: May 2002
Location: Mizar 5
Posts: 3,159
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 21
Quote:
Originally Posted by Talkstr8t
I'll ask one of my own for the encoding experts here. My wife and I watched "Good Night and Good Luck" the other night, and I got to wondering if a black and white film requires dramatically less space when being encoded? I would assume so, given how much smaller a black and white jpeg is than the equivalent color one. Can anyone confirm whether this is the case?

- Talk
A typical 4:2:0 MPEG-2 sequence expends on average about 25% of the total coded DCT coefficient bits on chroma. This makes sense because each macroblock contains 4 luma blocks and 2 chroma blocks (1 for Cb and 1 for Cr). Here's a dump of some typical video:
Code:
Interactive MPEG Decoder intd 0.72A
@ Statistics output turned on.
@  Found Sequence Start Code at byte position 0.
* --------  Sequence Start Code at 0        ---------
* --------       GOP Start Code at 22       ---------
* --------          Intra frame at 30       ---------
* Frame number: 2              Average mquant: 3.21 
* --- Number of bits used ---
*     High level headers :   8805  (0.79%)
*     Macroblock headers :  47948  (4.29%)
*         Motion vectors :      0  (0.00%)
*       DCT coefficients : 1060563  (94.92%)
*                  Luma  : 820629  (77.38%) 
*                Chroma  : 239934  (22.62%) 
* --- Macroblock types ---
*                  Intra :   8160  (100.00%) 
* --------  Bidirectional frame at 139989   ---------
* Frame number: 3              Average mquant: 4.55 
* --- Number of bits used ---
*     High level headers :   7444  (2.22%)
*     Macroblock headers : 112149  (33.50%)
*         Motion vectors :  95500  (28.53%)
*       DCT coefficients : 119689  (35.75%)
*                  Luma  :  88732  (74.14%) 
*                Chroma  :  30957  (25.86%) 
* --- Macroblock types ---
*                  Intra :     45  (0.55%) 
*          Frame Forward :   1845  (22.61%) 
*          Field Forward :     66  (0.81%) 
*         Frame Backward :   5872  (71.96%) 
*         Field Backward :     27  (0.33%) 
*    Frame Bidirectional :    154  (1.89%) 
*    Field Bidirectional :      8  (0.10%) 
*                Skipped :    143  (1.75%) 
* --------  Bidirectional frame at 182103   ---------
* Frame number: 4              Average mquant: 3.28 
* --- Number of bits used ---
*     High level headers :   7440  (1.67%)
*     Macroblock headers : 131419  (29.47%)
*         Motion vectors :  94157  (21.12%)
*       DCT coefficients : 212895  (47.74%)
*                  Luma  : 160392  (75.34%) 
*                Chroma  :  52503  (24.66%) 
* --- Macroblock types ---
*                  Intra :     35  (0.43%) 
*          Frame Forward :   3908  (47.89%) 
*          Field Forward :     41  (0.50%) 
*         Frame Backward :   3864  (47.35%) 
*         Field Backward :     52  (0.64%) 
*    Frame Bidirectional :    160  (1.96%) 
*    Field Bidirectional :      6  (0.07%) 
*                Skipped :     94  (1.15%) 
* --------      Predicted frame at 238201   ---------
* Frame number: 5              Average mquant: 2.22 
* --- Number of bits used ---
*     High level headers :   7444  (0.83%)
*     Macroblock headers : 126886  (14.19%)
*         Motion vectors : 103744  (11.60%)
*       DCT coefficients : 655924  (73.37%)
*                  Luma  : 494258  (75.35%) 
*                Chroma  : 161666  (24.65%) 
* --- Macroblock types ---
*                  Intra :     58  (0.71%) 
*          Frame Forward :   7721  (94.62%) 
*          Field Forward :    301  (3.69%) 
*                Skipped :     80  (0.98%) 
* --------  Bidirectional frame at 350230   ---------
* Frame number: 6              Average mquant: 4.51 
* --- Number of bits used ---
*     High level headers :   7444  (2.21%)
*     Macroblock headers : 114993  (34.17%)
*         Motion vectors :  93682  (27.84%)
*       DCT coefficients : 120419  (35.78%)
*                  Luma  :  83728  (69.53%) 
*                Chroma  :  36691  (30.47%) 
* --- Macroblock types ---
*                  Intra :     74  (0.91%) 
*          Frame Forward :   5426  (66.50%) 
*          Field Forward :     74  (0.91%) 
*         Frame Backward :   2267  (27.78%) 
*         Field Backward :     28  (0.34%) 
*    Frame Bidirectional :    106  (1.30%) 
*    Field Bidirectional :     11  (0.13%) 
*                Skipped :    174  (2.13%) 
* --------  Bidirectional frame at 392712   ---------
* Frame number: 7              Average mquant: 4.50 
* --- Number of bits used ---
*     High level headers :   7440  (2.22%)
*     Macroblock headers : 115117  (34.33%)
*         Motion vectors :  93362  (27.84%)
*       DCT coefficients : 119410  (35.61%)
*                  Luma  :  82104  (68.76%) 
*                Chroma  :  37306  (31.24%) 
* --- Macroblock types ---
*                  Intra :     80  (0.98%) 
*          Frame Forward :   3903  (47.83%) 
*          Field Forward :     44  (0.54%) 
*         Frame Backward :   3807  (46.65%) 
*         Field Backward :     55  (0.67%) 
*    Frame Bidirectional :     82  (1.00%) 
*    Field Bidirectional :      7  (0.09%) 
*                Skipped :    182  (2.23%) 
* --------      Predicted frame at 435010   ---------
* Frame number: 8              Average mquant: 3.19 
* --- Number of bits used ---
*     High level headers :   7444  (1.26%)
*     Macroblock headers : 123337  (20.91%)
*         Motion vectors : 103264  (17.50%)
*       DCT coefficients : 355928  (60.33%)
*                  Luma  : 267252  (75.09%) 
*                Chroma  :  88676  (24.91%) 
* --- Macroblock types ---
*                  Intra :     34  (0.42%) 
*          Frame Forward :   7753  (95.01%) 
*          Field Forward :    271  (3.32%) 
*                Skipped :    102  (1.25%) 
* --------  Bidirectional frame at 509135   ---------
* Frame number: 9              Average mquant: 3.74 
* --- Number of bits used ---
*     High level headers :   7444  (1.87%)
*     Macroblock headers : 124339  (31.29%)
*         Motion vectors :  94906  (23.88%)
*       DCT coefficients : 170685  (42.95%)
*                  Luma  : 127323  (74.60%) 
*                Chroma  :  43362  (25.40%) 
* --- Macroblock types ---
*                  Intra :     39  (0.48%) 
*          Frame Forward :   4332  (53.09%) 
*          Field Forward :     88  (1.08%) 
*         Frame Backward :   3421  (41.92%) 
*         Field Backward :     10  (0.12%) 
*    Frame Bidirectional :    129  (1.58%) 
*    Field Bidirectional :     10  (0.12%) 
*                Skipped :    131  (1.61%) 
* --------  Bidirectional frame at 559161   ---------
* Frame number: 10             Average mquant: 3.73 
* --- Number of bits used ---
*     High level headers :   7440  (1.87%)
*     Macroblock headers : 122835  (30.88%)
*         Motion vectors :  93724  (23.56%)
*       DCT coefficients : 173742  (43.68%)
*                  Luma  : 130223  (74.95%) 
*                Chroma  :  43519  (25.05%) 
* --- Macroblock types ---
*                  Intra :     38  (0.47%) 
*          Frame Forward :   2640  (32.35%) 
*          Field Forward :     23  (0.28%) 
*         Frame Backward :   5170  (63.36%) 
*         Field Backward :     39  (0.48%) 
*    Frame Bidirectional :    108  (1.32%) 
*    Field Bidirectional :     10  (0.12%) 
*                Skipped :    132  (1.62%) 
* --------      Predicted frame at 609187   ---------
* Frame number: 11             Average mquant: 2.60 
* --- Number of bits used ---
*     High level headers :   7444  (0.93%)
*     Macroblock headers : 125215  (15.69%)
*         Motion vectors : 103622  (12.98%)
*       DCT coefficients : 561859  (70.40%)
*                  Luma  : 428627  (76.29%) 
*                Chroma  : 133232  (23.71%) 
* --- Macroblock types ---
*                  Intra :     50  (0.61%) 
*          Frame Forward :   7754  (95.02%) 
*          Field Forward :    264  (3.24%) 
*                Skipped :     92  (1.13%) 
* --------  Bidirectional frame at 709256   ---------
* Frame number: 12             Average mquant: 3.73 
* --- Number of bits used ---
*     High level headers :   7444  (1.81%)
*     Macroblock headers : 126604  (30.84%)
*         Motion vectors :  94640  (23.06%)
*       DCT coefficients : 181798  (44.29%)
*                  Luma  : 136732  (75.21%) 
*                Chroma  :  45066  (24.79%) 
* --- Macroblock types ---
*                  Intra :     60  (0.74%) 
*          Frame Forward :   2799  (34.30%) 
*          Field Forward :     52  (0.64%) 
*         Frame Backward :   4963  (60.82%) 
*         Field Backward :     29  (0.36%) 
*    Frame Bidirectional :    122  (1.50%) 
*    Field Bidirectional :     13  (0.16%) 
*                Skipped :    122  (1.50%) 
* --------  Bidirectional frame at 760938   ---------
* Frame number: 13             Average mquant: 3.74 
* --- Number of bits used ---
*     High level headers :   7440  (1.85%)
*     Macroblock headers : 126246  (31.47%)
*         Motion vectors :  94213  (23.49%)
*       DCT coefficients : 173212  (43.18%)
*                  Luma  : 130903  (75.57%) 
*                Chroma  :  42309  (24.43%) 
* --- Macroblock types ---
*                  Intra :     44  (0.54%) 
*          Frame Forward :   2983  (36.56%) 
*          Field Forward :     13  (0.16%) 
*         Frame Backward :   4827  (59.15%) 
*         Field Backward :     58  (0.71%) 
*    Frame Bidirectional :    101  (1.24%) 
*    Field Bidirectional :      7  (0.09%) 
*                Skipped :    127  (1.56%) 
* --------      Predicted frame at 811332   ---------
* Frame number: 14             Average mquant: 2.64 
* --- Number of bits used ---
*     High level headers :   7444  (0.94%)
*     Macroblock headers : 125723  (15.88%)
*         Motion vectors : 102701  (12.97%)
*       DCT coefficients : 556075  (70.22%)
*                  Luma  : 409386  (73.62%) 
*                Chroma  : 146689  (26.38%) 
* --- Macroblock types ---
*                  Intra :     72  (0.88%) 
*          Frame Forward :   7717  (94.57%) 
*          Field Forward :    277  (3.39%) 
*                Skipped :     94  (1.15%) 
* --------  Bidirectional frame at 910665   ---------
* Frame number: 15             Average mquant: 3.71 
* --- Number of bits used ---
*     High level headers :   7444  (1.82%)
*     Macroblock headers : 124989  (30.61%)
*         Motion vectors :  94901  (23.24%)
*       DCT coefficients : 180975  (44.32%)
*                  Luma  : 135095  (74.65%) 
*                Chroma  :  45880  (25.35%) 
* --- Macroblock types ---
*                  Intra :     42  (0.51%) 
*          Frame Forward :   2912  (35.69%) 
*          Field Forward :     70  (0.86%) 
*         Frame Backward :   4875  (59.74%) 
*         Field Backward :     16  (0.20%) 
*    Frame Bidirectional :    120  (1.47%) 
*    Field Bidirectional :      9  (0.11%) 
*                Skipped :    116  (1.42%) 
* --------  Bidirectional frame at 961979   ---------
For black and white encoding, there's 8 (low end for B-frames) to 30 (high end for I-frames) percent (of the total bit budget per frame) more bits available for the the luma DCT coefficients.

Ron

HD MPEG-2 Test Patterns http://www.w6rz.net
dr1394 is offline  
post #477 of 4623 Old 07-18-2006, 01:13 AM
AVS Special Member
 
kjack's Avatar
 
Join Date: Jul 2001
Location: Milpitas, CA, USA
Posts: 2,458
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Ian_S
Are there any plans for either format to support the output of movies at 24fps...
Today we demo'd 1080p24 output on a BD reference design, playing off-the-shelf BD movies.

Keith Jack
Sigma Designs
BD, IPTV, HDTV decoder supplier
Blog: http://www.keithjack.net
kjack is offline  
post #478 of 4623 Old 07-18-2006, 01:17 AM
AVS Special Member
 
kjack's Avatar
 
Join Date: Jul 2001
Location: Milpitas, CA, USA
Posts: 2,458
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Talkstr8t
And not that you asked about Blu-ray, but BD-J can support it simply by loading a browser written in Java from the disc.
Personally, I think it would be easier to simply port an existing embedded browser and avoid all the potential compatibility issues. :)

Keith Jack
Sigma Designs
BD, IPTV, HDTV decoder supplier
Blog: http://www.keithjack.net
kjack is offline  
post #479 of 4623 Old 07-18-2006, 01:35 AM
Senior Member
 
Ian_S's Avatar
 
Join Date: Mar 2006
Location: (Old) Hampshire, UK
Posts: 232
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by amirm
Output format is independent of the source. HD DVD/BD only support 4:2:0 sampling for all three codecs. The output link then upsamples it as needed to 4:2:2 or 4:4:4 (often incorrectly :)). But there is no data beyond 4:2:0 in the source.
So does this mean that the increased bit depth touted in HDMI 1.3 for more colour is not realisable from the existing HD codecs?
Ian_S is offline  
post #480 of 4623 Old 07-18-2006, 03:14 AM
AVS Addicted Member
 
trbarry's Avatar
 
Join Date: Jan 2000
Location: Gainesville FL USA
Posts: 10,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by kjack
Today we demo'd 1080p24 output on a BD reference design, playing off-the-shelf BD movies.
Kjack -

Was that 24p all the way thru from decoding or deinterlaced using another Faroudja chip? And did the demo include handling things like overlays, PIP etc?

I'd sort of given up on seeing a 24p chain all the way from disc to display.

- Tom

Why don't we power our electric cars from greener, cheaper Liquid Fluoride Thorium Reactors?

Tom Barry - Find my video filters at www.trbarry.com
trbarry is offline  
Closed Thread HDTV Software Media Discussion

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off