Industry Insiders Q&A MASTER THREAD [separate thread for Xbox/Add On & PS3] - Page 85 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 
Thread Tools
post #2521 of 4841 Old 03-02-2007, 01:26 AM
AVS Forum Special Member
 
Talkstr8t's Avatar
 
Join Date: Jul 2001
Location: Los Altos, CA, USA
Posts: 2,761
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by paidgeek View Post

No persistent storage is required for these players.

Paidgeek, I believe you are incorrect. There are two types of non-volatile memory ("local storage") defined in the Blu-ray spec - the Application Data Area (known as "Persistent Storage" in MHP) for preferences, bookmarks, and such, and the Binding Unit Data Area for storage of A/V clips. All Blu-ray players must support the Application Data Area, albeit not a huge amount. All BD-Video 1.1 and BD-Live players must also support the Binding Unit Data Area (256MB for BD-Video and 1GB for BD-Live).

- Talk

Blu-ray Insider
Speaking solely for myself, not the BDA
Talkstr8t is offline  
Sponsored Links
Advertisement
 
post #2522 of 4841 Old 03-02-2007, 01:37 AM
AVS Forum Special Member
 
Talkstr8t's Avatar
 
Join Date: Jul 2001
Location: Los Altos, CA, USA
Posts: 2,761
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by paidgeek View Post

Sony Pictures participated in both of these events and my recollection of the final results is different. First off, the tests were based primarily on sub 10Mbps encoding, the total amount of encoded content was on the order of minutes and there were no direct view full resolution panels present.

In addition, as I recall the event didn't include AVC high profile as one of the tested codecs, only main profile.

- Talk

Blu-ray Insider
Speaking solely for myself, not the BDA
Talkstr8t is offline  
post #2523 of 4841 Old 03-02-2007, 02:29 AM
AVS Forum Special Member
 
patrick99's Avatar
 
Join Date: Oct 2006
Posts: 3,778
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 11
Quote:
Originally Posted by amirm View Post

Unfortunately, there is no way to make such generalizations. Every movie will be subjected to so many production steps and mastering that it is hard to know what shape they wind up in by the time the land in a VC-1 encode pipeline. GDMX (Warner's post house) has stellar expertise in VC-1 encoding, going back two years and with wide open support channel to us. So little fault can be put in that part of the process as opposed to what Uni uses.

Amir, perhaps then you might consider my original question:

"Originally Posted by patrick99
Amir, in light of your extensive comments on Batman Begins and on VC-1 as compared to AVC, I wonder if you have seen the BD of The Prestige? If so, do you think that the differences that appear to exist in the PQ are attributable to differences in the masters? Since the director and the DP are the same on both, and since both have similar colorations, one would have expected both to look very similar. Thanks"

I think that Universal has demonstrated repeatedly that VC-1 is capable of producing outstanding results. But I assume you would agree that using VC-1 doesn't guarantee outstanding results.

It seemed to me that the comparison of Batman Begins versus The Prestige might give some insight into the question of whether the softness in BB is attributable to softness in the original material or is instead an effect introduced during the transfer process.

So is it possible that you might address my original question?

Thanks
patrick99 is offline  
Sponsored Links
Advertisement
 
post #2524 of 4841 Old 03-02-2007, 03:30 AM
AVS Forum Special Member
 
trbarry's Avatar
 
Join Date: Jan 2000
Location: Gainesville FL USA
Posts: 9,884
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by paidgeek View Post

Amir,

Sony Pictures participated in both of these events and my recollection of the final results is different. First off, the tests were based primarily on sub 10Mbps encoding, the total amount of encoded content was on the order of minutes and there were no direct view full resolution panels present. If you are claiming that these events proved anything accept that a reasonable picture could be produced as judged on 5 year old display technology, then I beg to differ.

paidgeek -

Interesting. Are you willing to suggest rules for a new shootout? Would Sony participate?

- 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  
post #2525 of 4841 Old 03-02-2007, 04:29 AM
AVS Forum Special Member
 
RBFilms's Avatar
 
Join Date: May 2006
Location: New York / California
Posts: 1,248
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 46 Post(s)
Liked: 29
We are currently performing an Encode Test using several different types of content and several different labs. I am open to anyone who wants to contribute time, services, or content to the cause.

We would be using the following content:

1 - 100% Animation - Animation with a great deal of fine detail and precise motion. This is challenging content and we have used it in the past for standard DVD Compression tests. None of the encoders could produce results that we were happy with, but we did find one that was better than any of the others.

2 - 50/50 Hybrid - A mix of 1080i footage with CGI Overlays with a few very tricky elements that are encoding challenges.

3 - 65mm Film - Clips from a variety of IMAX films ranging from excellent to poor condition with and without pre-encode processing.

In general, we feel that we know the strengths and weaknesses of each Codec well enough to choose content designed to show their strengths and reveal their weaknesses.

It is easy for any encoder to look good if the film is squeaky clean and heavily processed to remove all dirt, noise, grain, artifacts, gate weave, flicker, and other artifacts prior to encoding. I do not consider encoding pristine prints ad true test of encoding capability.

For the record, I offered this to Microsoft.

Richard J. Casey



Disney WOW - World of Wonder


Producers Guild of America, New Media Council
(BD Industry Insider)
RBFilms is offline  
post #2526 of 4841 Old 03-02-2007, 04:40 AM
AVS Forum Special Member
 
TomsHT's Avatar
 
Join Date: May 2006
Location: Douglassville, PA
Posts: 4,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Posting screenshots of paused high def titles currently seems to be the most effective way for members to make visual comparisons of the same scenes and then discuss the advantages and disadvantages of the specific titles and the differences between them.

As with most public comparison tests more effort is usually put in by some members to debate how the test was done in the first place rather then the results of the test. Example in how the different devices are calibrated, what type of camera was used, how many megapixels the camera has, size & resolution of the posted pictures, whether the player displays differently while paused etc

I'd like to ask reps from both formats what you would consider to be fair as far as taking these screenshots for posting and making public comparisons by us members?

This way we can have a bit more uniformity and credibility with taking and posting these pictures for comparisons and the results wont be unfair towards either format.

Thank you in advance
TomsHT is offline  
post #2527 of 4841 Old 03-02-2007, 04:45 AM
AVS Forum Special Member
 
trbarry's Avatar
 
Join Date: Jan 2000
Location: Gainesville FL USA
Posts: 9,884
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by RBFilms View Post

We are currently performing an Encode Test using several different types of content and several different labs. I am open to anyone who wants to contribute time, services, or content to the cause.

We would be using the following content:

1 - 100% Animation – Animation with a great deal of fine detail and precise motion. This is challenging content and we have used it in the past for standard DVD Compression tests. None of the encoders could produce results that we were happy with, but we did find one that was better than any of the others.

2 – 50/50 Hybrid – A mix of 1080i footage with CGI Overlays with a few very tricky elements that are encoding “challenges.”

3 – 65mm Film – Clips from a variety of IMAX films ranging from excellent to poor condition with and without pre-encode processing.

In general, we feel that we know the strengths and weaknesses of each Codec well enough to choose content designed to show their strengths and reveal their weaknesses.

It is easy for any encoder to look good if the film is squeaky clean and heavily processed to remove all dirt, noise, grain, artifacts, gate weave, flicker, and other artifacts prior to encoding. I do not consider encoding “pristine” prints ad true test of encoding capability.

For the record, I offered this to Microsoft.

Richard -

Electronic cameras are increasingly able to offer 1080p/24 (or faster) material that might on good days approach 'squeaky clean'. Is that also a criterion?

Dunno how to phrase this as a question (what was that old quiz show?) but it seems the very existance of profitable electronic media distribution will exert a force to create movies that look good after being electronically coded. So wouldn't you also want to test some of these?

- 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  
post #2528 of 4841 Old 03-02-2007, 06:13 AM
Advanced Member
 
paidgeek's Avatar
 
Join Date: Nov 2006
Location: So. Cal
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
Quote:
Originally Posted by Talkstr8t View Post

Paidgeek, I believe you are incorrect. There are two types of non-volatile memory ("local storage") defined in the Blu-ray spec - the Application Data Area (known as "Persistent Storage" in MHP) for preferences, bookmarks, and such, and the Binding Unit Data Area for storage of A/V clips. All Blu-ray players must support the Application Data Area, albeit not a huge amount. All BD-Video 1.1 and BD-Live players must also support the Binding Unit Data Area (256MB for BD-Video and 1GB for BD-Live).

- Talk

Talk,

Strictly speaking, you are correct, but since the question was with regard to version 1.0 players and only 64KB is required, I rounded to zero....

Sony Pictures BD Insider
paidgeek is offline  
post #2529 of 4841 Old 03-02-2007, 06:59 AM
AVS Forum Special Member
 
Frode's Avatar
 
Join Date: Mar 2001
Location: Asker, Norway
Posts: 1,679
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
RBFilms - I suggest that in addition to those you already mentioned you try film content that's been cleaned up and edited using [email protected] equipment, as was done early on before 24psf/50i became the standard. Those will typically have difficult cadence breaks causing challenges for the equipment. For the animation test try both CG and cel based animation, as they're difficult to handle in different ways.

/frode
Frode is offline  
post #2530 of 4841 Old 03-02-2007, 07:17 AM
Advanced Member
 
paidgeek's Avatar
 
Join Date: Nov 2006
Location: So. Cal
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
Quote:
Originally Posted by trbarry View Post

paidgeek -

Interesting. Are you willing to suggest rules for a new shootout? Would Sony participate?

- Tom

Tom,

There is not much upside for Sony to participate in a codec shootout at this point. The obvious problem is the the possibility of endless second guessing as to whether such a shootout has been conducted properly. If an impartial shootout can be arranged, SPE can contribute some samples for encoding that are known to be difficult for all of the encoders in current use.

Another issue that should be considered is the characteristics that a codec imparts on the results. For instance, in a shootout, one may compare the results of each codec side by side and a preference may be shown for the sample which looks "clean". Herein comes the question of whether codecs should be chosen based on how they potentially alter the look of the master versus how close an approximation they achieve of the master. For my part, any shootout, should allow side by side comparisons of the master to the encoded samples using identical, full resolution, properly calibrated displays.

Regardless of my own feeling about color, focus or film grain on any particular title, we have spent enough time with film makers to know they don't want their work altered according to the "taste" of a particular market, compressionist or studio. The agreement is, we are provided an approved master and our job is to reproduce it as accurately as possible on whatever the distribution format is: Television, DVD, BD and so on.

Sorry for the longish answer to your simple question...

Sony Pictures BD Insider
paidgeek is offline  
post #2531 of 4841 Old 03-02-2007, 07:39 AM
Advanced Member
 
paidgeek's Avatar
 
Join Date: Nov 2006
Location: So. Cal
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
Quote:
Originally Posted by TomsHT View Post

Posting screenshots of paused high def titles currently seems to be the most effective way for members to make visual comparisons of the same scenes and then discuss the advantages and disadvantages of the specific titles and the differences between them.

As with most public comparison tests more effort is usually put in by some members to debate how the test was done in the first place rather then the results of the test. Example in how the different devices are calibrated, what type of camera was used, how many megapixels the camera has, size & resolution of the posted pictures, whether the player displays differently while paused etc

I'd like to ask reps from both formats what you would consider to be fair as far as taking these screenshots for posting and making public comparisons by us members?

This way we can have a bit more uniformity and credibility with taking and posting these pictures for comparisons and the results wont be unfair towards either format.

Thank you in advance

The results of screenshots can be very misleading because, as I have mentioned in my previous posts, all encoded movies have flaws. Consider this, every encoder benefits from customization (a human looking at the encoded stream and adjusting bit rate, or other parameters). Since customization is useful, it stands to reason that all encoders produce artifacts (as they must given between a 50:1 and 100:1 reduction in data from the master). If you accept the above, then the question becomes, what can the codecs do with the least opportunity for human error (in requiring a great deal of customization), with reasonable bit rates and on difficult titles (those that have high entropy etc.). All this is further complicated by the fact that film grain is not an artifact and in the absence of a master to reference, the encoders success at capturing it can't really be judged.

To summarize, screenshots will tell you less about the codecs and more about whether or not the compressionist was paying close attention or had the right equipment (e.g. displays) to find faults during encoding.


PS.
Even though screenshots can be misleading, if you have the opportunity to view the same movie, side by side from two codecs, it can be useful and revealing. Full rez direct view LCD's are getting inexpensive right? just borrow the one from the bedroom and put it beside the matching unit in the family room. Kidding aside, this type of direct comparison is most telling...

Sony Pictures BD Insider
paidgeek is offline  
post #2532 of 4841 Old 03-02-2007, 08:29 AM
AVS Forum Special Member
 
eq_shadimar's Avatar
 
Join Date: Jul 2001
Location: Austin, TX USA
Posts: 1,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 52
Question about 2.35 subtitles and menus to all insiders -

Many pages ago there was discussion on how to accommodate users with constant height systems by adding some code (HDi and BD-Java??)that would automatically allow the menus and subtitles to fit in the 2.35 frame vs the letterbox or black bar portion of the 1.78 frame.

Have any studios expressed interest in doing this or should we just give up hope on this front? If there was interest are there any US titles that are released or soon to be released that will have this feature?

Thanks for your time,
Jeff

...wake up to find out that you are the eyes of the world...
My HT Construction Thread - Updated Pictures 3/15/07
My 2.35:1 Discussion Thread - Updated Pictures 3/15/07
XBL = eqshadimar
eq_shadimar is offline  
post #2533 of 4841 Old 03-02-2007, 09:35 AM
AVS Forum Special Member
 
Frode's Avatar
 
Join Date: Mar 2001
Location: Asker, Norway
Posts: 1,679
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I don't know about menus, but for subtitles I imagine that kind of positioning trick would easily be possible if the studios used scripted subtitles instead of bitmaps.

Since we're on the subject, one thing I would like to see is the possibility of user-added subtitles in some way, or at least support for downloading official subtitles for titles that were created/made after the disc was released. That way you could buy the disc for your local region, and then get extra subtitles for your preferred language. I know it's not exactly a priority (most english native speakers probably don't even think about it), but it's one of those things that could provide extra value to customers so they don't have to import discs and players from other regions to get their preferred language.

/frode
Frode is offline  
post #2534 of 4841 Old 03-02-2007, 10:00 AM
AVS Forum Special Member
 
cyberbri's Avatar
 
Join Date: May 2004
Location: San Jose, CA
Posts: 8,103
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 54
Question for Amir or anyone else involved with MS:


I read the Toshiba HD-DVD player firmware upgrade details.
http://www.highdefdigest.com/news/sh..._Available/496
Some of the issues addressed appear to be similar to issues found on the HD-DVD add-on player for the Xbox 360.

Are these same "fixes" shared for the Xbox 360 and its upcoming update, as the player is made by Toshiba I believe?

This update is already available for the Toshiba players, although the Xbox 360 update appears to be coming with the spring system/Dashboard update.


Thanks.


-Brian
cyberbri is offline  
post #2535 of 4841 Old 03-02-2007, 11:25 AM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Quote:
Originally Posted by cyberbri View Post

Are these same "fixes" shared for the Xbox 360 and its upcoming update, as the player is made by Toshiba I believe?

Toshiba (really, TSST -- Toshiba/Samsung optical manufacturing arm) only provides the drive to us. All the software comes from us and not from Toshiba.

As we find bugs, if they are in the same HDi implementation we provide to Toshiba, we also give a drop to them. It is their option to provide that update to their customers. There is little the comes back in reverse though as Toshiba does not provide any code to us for Xbox 360.
amirm is offline  
post #2536 of 4841 Old 03-02-2007, 11:31 AM
AVS Forum Special Member
 
cyberbri's Avatar
 
Join Date: May 2004
Location: San Jose, CA
Posts: 8,103
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 54
cyberbri is offline  
post #2537 of 4841 Old 03-02-2007, 11:38 AM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Quote:
Originally Posted by RBFilms View Post

We are currently performing an Encode Test using several different types of content and several different labs. I am open to anyone who wants to contribute time, services, or content to the cause.

None of the encoders could produce results that we were happy with, but we did find one that was better than any of the others.

...

In general, we feel that we know the strengths and weaknesses of each Codec well enough to choose content designed to show their strengths and reveal their weaknesses.

For the record, I offered this to Microsoft.

Hmmm. I just checked with my team and none of the technical people seem to know anything about your project, let alone having the clips and such, or participated in your testing. They also have not had any contact with you one way or the other. The only thing I personally know about is you offering to send us the current retail version of your title in MPEG-2.

Maybe you have been working with Ben and given the fact that he has been sick, the content hasn't made it to the codec team. Please follow up again as we are happy to help out here in any way we can. Until then, please kindly note that until you work with us directly and we can bring you up to speed on ins and outs of VC-1 and our tools, I don't agree that you know the weaknesses and strength of VC-1 . Can you also clarify if you are doing your own in-house encodes and not using an outside shop as we think you currently are?
amirm is offline  
post #2538 of 4841 Old 03-02-2007, 12:00 PM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
On MI3 VC-1 vs MPEG-2.

Just heard back from the team that that the preliminary assessment is that the segment in question does indeed show some artifacts. It is our opinion that this could have been remedied easily as the segment is pretty easy to encode so one could easily fine tune it/bump the data rate.

What is puzzling though, is Benes' claim of he got those screen snapshots. He says he used Media Player Classic with a filter for VC-1. This doesn't compute. Media player (any version) lacks code to properly de-multiplex (separate the a/v streams) in HD DVD. HD DVD shipped some 8 years after the last update of Media Player Classic! While we use VC-1 in WMV encoded files for PC use, those files use our file format ("ASF"), which is wildly different than what is used in HD DVD. So the player has no way of playing HD DVD content, even if the copy protection is stripped away. We have provided specialized tools to play such files without access to an HD DVD player but I don't see how Benes' would have access to such a tool, without working for one of the companies with access to our professional encoder toolset.

In addition, our VC-1 decoder that he says he used, lacks post processing as that was too CPU intensive for the time period we release that code. So you are bound to see more artifacts in some cases as compared to real HD DVD hardware (sadly, I understand some software HD DVD players lack the same post processing).

Anyway, maybe adventurous hackers out there have reformatted those streams to get it to play in Media Player (but heaven forbid, have not actually re-encoded it). Or that there is a de-muxer I don't know about but Benes does but didn't list. Either way, it would have been good to have such data so that we know how to investigate. Otherwise, I am left wondering how really got those shots

As an aside, he should also rename the title of his thread. The MPEG-2 encode is probably running at 2X the data rate of VC-1. Remember, the HD DVD version actually has two video streams with a Picture-in-Picture on top of the move itself - BD version in MPEG-2 lacks that. So if that scene is the pinnacle of what is wrong with the VC-1 encode running at half the data rate of MPEG-2, while at the same time providing a secondary video stream, I still stand proud even if this were an issue with VC-1 .
amirm is offline  
post #2539 of 4841 Old 03-02-2007, 12:04 PM
Rio
Advanced Member
 
Rio's Avatar
 
Join Date: Feb 2003
Posts: 915
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by amirm View Post

Maybe you have been working with Ben and given the fact that he has been sick, the content hasn't made it to the codec team. Please follow up again as we are happy to help out here in any way we can. Until then, please kindly note that until you work with us directly and we can bring you up to speed on ins and outs of VC-1 and our tools, I don't agree that you know the weaknesses and strength of VC-1 . Can you also clarify if you are doing your own in-house encodes and not using an outside shop as we think you currently are?

It sounds like only Microsoft can bring out VC-1's capability and posthouses that have been using Microsoft provided encoder can't do that. Is it what you meant? If so, "shootout" with Microsoft's hand tuned stream won't really simulate the quality of actual production, unless all VC-1 titles are going to be encoded by Microsoft?

Maybe compression shootout by one posthouse which has capability of all encoders and has been doing Hollywood titles with those systems may reflect more "real" productivity/quality of actual products?
Rio is offline  
post #2540 of 4841 Old 03-02-2007, 12:18 PM
Rio
Advanced Member
 
Rio's Avatar
 
Join Date: Feb 2003
Posts: 915
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by amirm View Post

Or that there is a de-muxer I don't know about but Benes does but didn't list. Either way, it would have been good to have such data so that we know how to investigate.

If you check the doom9 forum, you'll find EVO demultiplexer is there.

Quote:
Originally Posted by amirm View Post

The MPEG-2 encode is probably running at 2X the data rate of VC-1. Remember, the HD DVD version actually has two video streams with a Picture-in-Picture on top of the move itself ? BD version in MPEG-2 lacks that. So if that scene is the pinnacle of what is wrong with the VC-1 encode running at half the data rate of MPEG-2, while at the same time providing a secondary video stream, I still stand proud even if this were an issue with VC-1 .

I think bene can strip out both elementary stream of VC-1 and MPEG2 as well as secondary PiP VC-1 stream (above demultiplexer can do this), and would let us know the "real" bitrate of both, if he wanted.
Rio is offline  
post #2541 of 4841 Old 03-02-2007, 12:21 PM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Quote:
Originally Posted by Rio View Post

It sounds like only Microsoft can bring out VC-1's capability and posthouses that have been using Microsoft provided encoder can't do that.

Not at all. Go to GDMX, DLUX, Cinram, etc. and you can get top notch encoding done in VC-1. That is because their people have attended our training and know how to use the system. But more importantly, they have experience with it. One can't say they know VC-1's capabilities, not having done any encodings of their own.

VC-1 is an advanced codec with a lot of features. And our tools are designed with professionals in mind who go through training, rather than casual users (the level of documentation simply is not there for this kind of use). It is not like we charge for our training anyway. We are more than happy to help out, including such specialized and small companies such as RB's. Given that, there should be no excuse not to take us up on learning how to use our tools and technology.

BTW, I hear Panasonic now has a preference for doing the encoding for studios, rather than selling their AVC encoder to others. If so, that shows they are uncomfortable supporting users of their encoder. Or they think they can do a better job than others. So if are more confident in AVC being used by others, what's up with this position by Panasonic?

Quote:


Is it what you meant? If so, "shootout" with Microsoft's hand tuned stream won't really simulate the quality of actual production, unless all VC-1 titles are going to be encoded by Microsoft?

Per above no. But as people who invented VC-1, and not getting paid a fixed budget to do a project (or paid at all by the studios ), it is true that we can probably extract that last bit of quality out of an encode. But that is not synonymous with others not being able to do an excellent job. To date, every VC-1 encode is done by a post house, not by us. And that includes the many reference quality titles.

Quote:


Maybe compression shootout by one posthouse which has capability of all encoders and has been doing Hollywood titles with those systems may reflect more "real" productivity/quality of actual products?

Unless that one post house does the encoding for the entire industry, with the same operator who does the shoot out, and they spend equal time on the shoot out than they do on real title, with the test title being representative of all titles they encode, and him/her not having too much to drink on the day encoding, the test, maybe you have a point . Otherwise, read Paid's post for excellent reasons on why public shoot outs are not of much value.
amirm is offline  
post #2542 of 4841 Old 03-02-2007, 12:26 PM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Quote:
Originally Posted by Rio View Post

If you check the doom9 forum, you'll find EVO demultiplexer is there.

Thanks. I don't hang out there anymore. But sure would have been nice for him to give the details as otherwise, we don't know how to replicate his results. Neither does anyone else.

Quote:


I think bene can strip out both elementary stream of VC-1 and MPEG2 as well as secondary PiP VC-1 stream (above demultiplexer can do this), and would let us know the "real" bitrate of both, if he wanted.

Measuring peak rate is difficult without specialized tools (hence the reason PS3 has trouble with it). Just because you can play it, doesn't mean you can measure the instantaneous peaks. I suspect you have the data already but just not saying it. Are you claiming that MI-3 uses HD DVD data rate for its MPEG-2 encode then?
amirm is offline  
post #2543 of 4841 Old 03-02-2007, 12:55 PM
Rio
Advanced Member
 
Rio's Avatar
 
Join Date: Feb 2003
Posts: 915
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by amirm View Post

Not at all. Go to GDMX, DLUX, Cinram, etc. and you can get top notch encoding done in VC-1. That is because their people have attended our training and know how to use the system. But more importantly, they have experience with it. One can't say they know VC-1's capabilities, not having done any encodings of their own.

So, if someone wants to know the strength and weakness of some codecs, asking those posthouses to do encoding supposes to work for getting quality in real world, right? It sounds little bit contradicting what you said "Until then, please kindly note that until you work with us directly and we can bring you up to speed on ins and outs of VC-1 and our tools, I don't agree that you know the weaknesses and strength of VC-1 ", anyway...

Quote:
Originally Posted by amirm View Post

Measuring peak rate is difficult without specialized tools (hence the reason PS3 has trouble with it). Just because you can play it, doesn't mean you can measure the instantaneous peaks.

It's not difficult job to measure the amount of bits certain picture is using, if someone knows bitstream syntax and has elementary data. And, anyone can download SMPTE 421M specification.

Quote:
Originally Posted by amirm View Post

I suspect you have the data already but just not saying it. Are you claiming that MI-3 uses HD DVD data rate for its MPEG-2 encode then?

I don't have those numbers, and no interest to do that. BTW, have you seen this bene's thread? If you are interested in ABR of BD titles, it's quite telling you something.
Rio is offline  
post #2544 of 4841 Old 03-02-2007, 01:21 PM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Quote:
Originally Posted by Rio View Post

So, if someone wants to know the strength and weakness of some codecs, asking those posthouses to do encoding supposes to work for getting quality in real world, right? It sounds little bit contradicting what you said "Until then, please kindly note that until you work with us directly and we can bring you up to speed on ins and outs of VC-1 and our tools, I don't agree that you know the weaknesses and strength of VC-1 ", anyway...

No it is not contractory. RB said he knew the difference between the codecs, not that his post house did. For him to have such personal experience, he would need to use to learn to use the tools and have done some work with it.

Quote:


It's not difficult job to measure the amount of bits certain picture is using, if someone knows bitstream syntax and has elementary data. And, anyone can download SMPTE 421M specification.

Have you done it? If so, then you would know it is not so easy. Even our tool approximates the rate. The only way to have 100% accurate data is to look at the encoding report. Otherwise, things like the buffer model and such, get in the way.
amirm is offline  
post #2545 of 4841 Old 03-02-2007, 01:23 PM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Rio, any comment on the Panasonic encoder business model? I am sure someone in your company has the answer .
amirm is offline  
post #2546 of 4841 Old 03-02-2007, 01:42 PM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Quote:
Originally Posted by Frode View Post

Amir "Media Player Classic" is a 3rd party media player that's coded from the ground up, and is not a Microsoft product. While it mimics the look of the old media player included in windows, that's merely a matter of cosmetics:

http://sourceforge.net/projects/guliverkli/

It has full support for customized filter graphs, and also some internal codec support. If someone says they used MPC to play back a HD-DVD my guess is they're using additional dshow filters to achieve that as the current version doesn't come with it out of the box so to speak.

Ah! We use the term "classic" to also talk about our original OCX control . Thanks for clarifying that. My comments regarding our video decoder stands though.
amirm is offline  
post #2547 of 4841 Old 03-02-2007, 01:49 PM
AVS Forum Special Member
 
bobgpsr's Avatar
 
Join Date: Jun 2004
Location: Cedar Rapids, Iowa
Posts: 2,646
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by amirm View Post

In addition, our VC-1 decoder that he says he used, lacks post processing as that was too CPU intensive for the time period we release that code. So you are bound to see more artifacts in some cases as compared to real HD DVD hardware (sadly, I understand some software HD DVD players lack the same post processing).

Amir, Can you clarify that all of the Toshiba standalone players that use the Broadcom chip do in fact have the "post processing" for VC-1? It was "I imagined" elsewhere that it is missing also in those players.

Thanks!
bobgpsr is offline  
post #2548 of 4841 Old 03-02-2007, 02:57 PM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Quote:
Originally Posted by bobgpsr View Post

Amir, Can you clarify that all of the Toshiba standalone players that use the Broadcom chip do in fact have the "post processing" for VC-1? It was "I imagined" elsewhere that it is missing also in those players.

Thanks!

Yes it does have it. They are not constrained by the PC power of the yesterday.
amirm is offline  
post #2549 of 4841 Old 03-02-2007, 03:10 PM
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,829
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1334 Post(s)
Liked: 749
Quote:
Originally Posted by patrick99 View Post

Amir, perhaps then you might consider my original question:

"Originally Posted by patrick99
Amir, in light of your extensive comments on Batman Begins and on VC-1 as compared to AVC, I wonder if you have seen the BD of The Prestige? If so, do you think that the differences that appear to exist in the PQ are attributable to differences in the masters? Since the director and the DP are the same on both, and since both have similar colorations, one would have expected both to look very similar. Thanks"

I have not yet watched it. I was sick and now in Japan and way behind watching movies . But even if I did, I wouldn't be able to say anything useful here. I already mentioned that even within one movie, Batman Begins, there are variations in apparent sharpness. So what can you expect, comparing different movies? Really, there is nothing to be learned from these kinds of comparisons.

Quote:


I think that Universal has demonstrated repeatedly that VC-1 is capable of producing outstanding results. But I assume you would agree that using VC-1 doesn't guarantee outstanding results.

Actually, Universal does not do any encoding themselves. They use a post house that others also utilize. But I do agree with your point on VC-1. Feed us VHS, and we can only make it look like VHS .

Quote:


It seemed to me that the comparison of Batman Begins versus The Prestige might give some insight into the question of whether the softness in BB is attributable to softness in the original material or is instead an effect introduced during the transfer process.

It really doesn't. Per above, there is zero information to be gained here. Thousands of people work on a movie. Budgets vary. Equipment varies. Experience changes. Intent evolves. I know you all looking for some way of getting relative comparisons done but "this dog don't hunt. " If you don't trust me, as Paid to see what he thinks. I am confident he would agree.

Quote:


So is it possible that you might address my original question?

Thanks

Hopefully we can put this to rest and move on to other things...
amirm is offline  
post #2550 of 4841 Old 03-02-2007, 03:34 PM
AVS Forum Special Member
 
eightninesuited's Avatar
 
Join Date: Jun 2006
Location: Toronto, Republic of Canuckistan
Posts: 5,173
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 309 Post(s)
Liked: 306
Is there an inherent bandwidth issue with HD DVD. I ask this because I've yet to see a movie with True HD and a high VC-1 encode.

Proud Member of "Bass-a-holic Anonymous" - Dual Bottle of Choice: SVS PC-13 Ultra + SVS PC-12 NSD. Drink of Choice: Ready Player One Delorean Race
DaLite 0.9 Grey 110" screen. Epson 3000 1080P Projection. 7.2 SVS and Energy Reference + Parasound THX Ultra Amplification system.
eightninesuited is offline  
Sponsored Links
Advertisement
 
Closed Thread HDTV Software Media Discussion

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