HD-DVD 30GB/Movie Length Facts [Text View]

tsd2005

12-20-06, 12:46 AM

If you mean a particular 4 hour movie you (or whoever makes the claim) should say that and be careful not to imply that any 4 hour movie would work. Some movies tend to compress better ("Lord of the Rings" might be in the group that does) and even things like whether a movie is 1.85:1 or 2.35:1 can affect the kind of bitrate they need (much like length can, but in a different way). One 4 hour movie might work great and then a shorter movie like "Saving Private Ryan" (1.78:1 and grainy) might be a problem. And even some scenes could be a problem since the bitrate ceiling limitations of HD DVD (and Blu-ray too) can show up where the size of the disc isn't a problem at all. As somebody said before, the bandwidth is like a hose and the size of the disc like the size of a pool. The pool could have plenty of space and the size of the hose still be a problem.

If an insider wants to give some example bitrates here, how about telling us the bitrate that a partially grainy movie like "Miami Vice" ended up at for average and peak.

--Darin

Darin, I know all the differences that can affect whether or not something will compress well, but the arguement may be ... "What movie will not fit on a 30GB disc with those extras?"

There is a lot of disinformation going on. It needs to be nipped in the bud. There is no reason for anyone to result to misinformation on either side.

xbdestroya

12-20-06, 01:01 AM

Here is my issue with Ben's assertions on the whole LOTR thing. He says that average bitrates would come to about 12.5Mbps, and says that there are current titles even below that, but the thing is... which titles are those? The MS crew keeps that info very close to their chest, and IMO would be useful to know before actually determining if 12.5 is enough for what the viewer would demand from such a flagship title.

I'm not against the idea that HD DVD is up to the task, but until there is more info on what bitrates and compression levels correlate to what in real world terms as they apply to released titles, I'm just more comfortable hoping for BD50, AVC at a high bitrate, a lossless compression scheme, and some good LOTR-esque extras thrown in for good measure.

PS - Actually I have an idea on where to get some real-world VC-1 bitrate info... I'll check it out and return.

xbdestroya

12-20-06, 01:43 AM

PS - Actually I have an idea on where to get some real-world VC-1 bitrate info... I'll check it out and return.

Ok here we go on some VC-1 cross-format titles.

Blazing Saddles - length 93 min - movie file 18.6GB - video bitrate 25.5 Mbps

Corpse Bride - length 77 min - movie file 13GB - video bitrate 19.9 Mbps

House of Wax - length 113 min - movie file 18GB - video bitrate 19 Mbps

Lethal Weapon - length 110 min - movie file 21GB - video bitrate 25 Mbps

Phantom of the Opera - length 141 min - movie file 19GB - video bitrate 16.5 Mbps

Superman Returns - length 154 min - movie file 20.5GB - video bitrate 16 Mbps

Unforgiven - length 131 min - movie file 18GB - video bitrate 17Mbps

...Ok so yeah, I *definitely* do not want LOTR to be at 12.5Mbps, unless one of the MS crew is willing to share a stunning title with the audience that has been encoded at similar levels.

Credit goes to Benes, who has been busy compiling a list of BD titles, their size used, codecs, and bitrates.

In this instance, it provides a clear window into the secretive world of VC-1 as well, since these specific BD titles are direct video ports of their HD DVD cousins.

Rio

12-20-06, 02:14 AM

I'd like to add this:

How many VC-1 titles released are grainy looks, as same level as Tomb Raider, MI3 or Xmen3? If it is grainy but still keep fine details, what is its video bit rate?

benwaggoner

12-20-06, 02:33 AM

xbdestroya

12-20-06, 02:38 AM

Unfortunately, your numbers are incorrect for the titles I know the real numbers for. How were they calculated?

FWIW, there are plenty of titles highly touted for quality out with <13 Mbps ABR.

And, no, sorry, I can say which are which...

The numbers are from here: http://www.avsforum.com/avs-vb/showthread.php?t=760714

And I'm going to have to trust them to be honest with you, as they're taken in the best way I know of; right off of a PC. IMO if you feel these numbers to be incorrect, it is incumbent on you to indicate how - and IMO, offer up a title as a sacrifice for the knowledge base. There is otherwise no reason to doubt something as unbiased as a PC bitrate readout.

'Highly touted' could mean anything. Let me ask you this - are any of the VC-1 titles out on BD coming in at under 13 Mbps in bitrate?

I own Superman Returns on BD, and it's VC-1. The PS3 allows for bitrate tracking during playback, so I myself will run a confirmation test on that title tomorrow to see if there are discrepencies.

hmurchison

12-20-06, 05:34 AM

I see no reason why AVC or VC-1 cannot look excellent at 12.5 Mbps avg bitrate. The codecs were designed to do well at these levels.

I imagine that if a confluence of factors happen we'll have no problem getting 4hrs of excellent PQ and AQ plus interactivity on a single disc.

1. Encoding improvements. Looking forward to seeing if Dynamic Muxing yields a significant benefit.

2. Moving PiP video to VC-1

3. Utilization of Dolby Digital Plus 1.5Mbps. If you can achieve nigh transparancy with a 24/48 audio master then I don't see much of a need for TrueHD for longer playing movies.

4. Newer movies with cleaner masters. The Panavision Genesis has been used in 3 high profile movies. Apocalypto is probaby the best looking one I've seen thus far over Click and Superman Returns.

What'sHD

12-20-06, 06:14 AM

As BD offers more space and b/w, any advice to stick with lower space and b/w is drivel, considering costs are on par (unless you think the PS3 is junk).

It's about Great High Def. As High Def fans on AVS, the extra space and higher b/w (lets not talk content cos it is not Inherent to BD as a technical format) is worth the extra cost of BD stand-alones people complain about. Or, considering the PS3, no extra cost and better specs.

Neutrals, please do not settle. Buy the format with the better specs and more content support.

When people who support HD say "BD is all about future promises while HD delivers now", I shudder at the thought that plenty of people prob thus buy HD without realizing that a huge %age of the most important part of High Def, the content is wayy in that same future for HD or maybe even never.

Yes, BD buyers lack Universal's content but which group of movies would you as a neutral person rather miss out on? Something to think about..

hmurchison

12-20-06, 07:05 AM

I would be a fervent Blu-ray support if:

1. Sony had not tried to cram 2 extra layers of Copy Protection down my throat in BD+ and ROM Mark. That's an appeal to studios but I'm not a studio and when my warranty runs out and the player begins to act wonky I don't want to have to sit and wonder which DRM layer is acting up.

2. The players were more full featured. Networking in only 2 BD products?? Lack of mandatory audio and video features means a wide dynamic range of possible hardware configs.

3. Reliance on MPEG2. Sony has tried to flog crusty technology for too long while pricing it like it is the new hotness. MPEG2 come on. Java for interactivity?? That's so 90's ...high efficiency codecs and XML based interactivity are distinctly more in tune with what's going on today

Thus we have the many of you on these boards asking us to appeal to our paranoia that somehow 30GB won't be enough, that under 30Mbps is limiting. When does it end? You already have the HVD proponents that want to fantasize about 300Gb discs.

If we're talking computers then I'm all for "going large" computing power is always eventually usurped. However the trend for video and audio codecs over the years has been "reducing" the datarate required to maintain quality reproduction.

The problem with Blu-ray fans is that you cannot be persuasive because no data exists that shows BD50 having any appreciable impact on Audio or Video quality. The current BD50 titles today (even the high bitrate versions) offer no quantifiable advantage over a 30GB VC-1 title. Maybe one day it may but right now none of you can back up your assertions with empirical evidence. Thus we HD DVD'ers aren't willing to support a format that is more expensive because of "potential"

When I walk into a Target or Walmart I see DVDs for $5 for decent movies. I think to myself "we need the format that is most likely to get us close to this type of pricing" right now my horse is on HD DVD being that format. 50GB discs are a solution looking for a problem today.

ILJG

12-20-06, 07:34 AM

Ok here we go on some VC-1 cross-format titles.

Blazing Saddles - length 93 min - movie file 18.6GB - video bitrate 25.5 Mbps

Corpse Bride - length 77 min - movie file 13GB - video bitrate 19.9 Mbps

House of Wax - length 113 min - movie file 18GB - video bitrate 19 Mbps

Lethal Weapon - length 110 min - movie file 21GB - video bitrate 25 Mbps

Phantom of the Opera - length 141 min - movie file 19GB - video bitrate 16.5 Mbps

Superman Returns - length 154 min - movie file 20.5GB - video bitrate 16 Mbps

Unforgiven - length 131 min - movie file 18GB - video bitrate 17Mbps

...Ok so yeah, I *definitely* do not want LOTR to be at 12.5Mbps, unless one of the MS crew is willing to share a stunning title with the audience that has been encoded at similar levels.

Credit goes to Benes, who has been busy compiling a list of BD titles, their size used, codecs, and bitrates.

In this instance, it provides a clear window into the secretive world of VC-1 as well, since these specific BD titles are direct video ports of their HD DVD cousins.

Are these video bitrates, as they're referred to here, peak or avg? Let's be honest.

And let's ask the flip side question. Which movies have had their PQ suffer because of bandwidth? (Spare me the "TrueHD was left off this title", or "IME was left off that title", because I don't think anyone who understands the audio would consider a 1.5 Mbps DD+ track "suffering"... and interactivity on the other format is severely lagging behind HD-DVD if not practically non-existent) Please give timestamps of specific titles to prove it was a bandwidth issue. Also, name a title that's had more bits thrown at it and therefore the exact problems in the above question have gone away.

I'm not saying the limitations aren't there. It's just that the people claiming the severity of the limitations (disc size, and much more importantly, bandwidth) haven't really proven their point.

What'sHD

12-20-06, 08:04 AM

namechamps

12-20-06, 08:28 AM

Hey, the chaps saying that sony's -ive public image is working against BD might be right. I feel pissed off at HD right now even though I know its crappy design by MS. Pissed off..

Well that explains a lot. Thanks.

The reason people have a problem with Sony is the constant manipulation. Why is Sony pushing titles in mpeg2? Cause mpeg2 needs 50GB. A movie that looks good w/ 50GB mpeg2 and LPCM sound will look equally good with 30GB VC1 and TrueHD.

Without mpeg2 the need for 50GB is substantially reduced. Without the need for 50GB the BD is a solution looking for a problem. HD 30GB has cost parity with BD 25GB now. Nobody knows how much 50GB really cost and how much of a subsidy Sony is kicking in during authoring.

Space only matters if it is needed. 30GB works fine. 50GB is nice but not at the added cost it bring. I mean where does it end. Why not wait for holographic playback at 300GB? You could use mpeg2, and have room for 6 tracks of LPCM.

avshaman

12-20-06, 08:32 AM

Look, if capacity does not matter then we would never have moved beyond DVD in the first place. HD movies could be spread out over just a couple DVDs for example, but why didn't we do it? Because we don't like the inconvenience of switching discs and because DVD's maximum bitrate of 9.8Mb/s would put limits on picture and audio quality. Does it really take that much imagination to envision scenarios where the extra 20 GB of capacity and higher bandwidth that BD affords could not be put to great use?

So what if you can prove to me that a HD DVD disc can fit a 4 hour movie while maintaining quality? It is easy to think of scenarios in which that will not be enough. Not to mention the fact that we cannot always anticipate uses that may come in the future in which the greater capacity would be of great benefit.

Studios can always switch codecs or make codecs more efficient but we cannot, if the war is decided in HD DVD's favor, change the physical limitations of the format.

gljvd

12-20-06, 08:41 AM

Anamorphiac

12-20-06, 08:56 AM

I think this is the first time in history that people are actually arguing for LESS storage capacity. And its mostly because of their irrational hatred of one company.

There can be MANY rational arguments and examples...just because others choose to ignore or overlook them, do not make them irrational. You can have your opinion, yes and overlook the issues at hand...and that is fine. It is because you choose to place your desire for the product above these issues. Others of us believe these company issues overshadow the current and future support and success of the format. You do not have to agree...but that doesn't make it irrational.

TomsHT

12-20-06, 08:56 AM

As someone that favors HD I would just say that I think HD has done a great job with fitting longer length movies on a 30g disk but I would never say any specific size is enough for all future content.

Now give me VC-1 along with a BD50 and/or TL45 and I think we will end up with even more content per disk then were already getting

spawnbadboy

12-20-06, 09:05 AM

I think this is the first time in history that people are actually arguing for LESS storage capacity. And its mostly because of their irrational hatred of one company.

if you knew anything about the formats you would know size isnt in the question here.
Your like a blind man looking at numbers. Its like saying a 320gb WD drive is better than the 150 WD raptor drive just beacuse it has more space... :rolleyes:

yes HD-DVD has less space than blu-ray but has has MUCH better features and is the better format

avshaman

12-20-06, 09:06 AM

Grubert

12-20-06, 09:12 AM

Cheering for the home-town company then huh?

You can say that again. We should probably have a "Washington State forum" for them alone. ;)

Only problem is, poor darin wouldn't stand a chance. :D

avshaman

12-20-06, 09:15 AM

TomsHT

12-20-06, 09:23 AM

Well, you're not everybody are you!? I look forward to the special features almost as much as the film itself and I don't like switching discs.
And again, are you really unable to think beyond the current moment to imagine uses which would require or benefit from the larger capacity?

I also enjoy the extras and although I would not want to force the extra on the same disk at the cost of reduced PQ/AQ I do enjoy them when they are available.

Reading your post though I have to wonder if you see the same problem I do, so far I have seen far less extras on my BR disks even including some of the BD50 disks released and coming soon.

So even with the extra space we seem to still lose out unless BR also decids to start using a more efficient codec for all there future releases

avshaman

12-20-06, 09:33 AM

There can be MANY rational arguments and examples...just because others choose to ignore or overlook them, do not make them irrational. You can have your opinion, yes and overlook the issues at hand...and that is fine. It is because you choose to place your desire for the product above these issues. Others of us believe these company issues overshadow the current and future support and success of the format. You do not have to agree...but that doesn't make it irrational.

Just because some of you have that opinion does not make it rational either. You must have a good argument and so far I have seen only one sided, biased attacks againsts Sony and uncritical, hypocritical support for Microsoft.

I know, I know, its cool to bash Sony right now. But that doesn't make it right.

avshaman

12-20-06, 09:40 AM

I also enjoy the extras and although I would not want to force the extra on the same disk at the cost of reduced PQ/AQ I do enjoy them when they are available.

Reading your post though I have to wonder if you see the same problem I do, so far I have seen far less extras on my BR disks even including some of the BD50 disks released and coming soon.

So even with the extra space we seem to still lose out unless BR also decids to start using a more efficient codec for all there future releases

I don't think there is as great a disparity between the two formats as you allude to but I do think you have a valid point.
However, there is every reason to believe that market forces will work that problem out, while the difference in capacity is inherent in the formats themselves.

What'sHD

12-20-06, 09:46 AM

Well that explains a lot. Thanks.
I dont get what you mean but just to clarify, my 360 conked out 2 hours ago. Before this, I was hot for GoW and anticipating my add-on like the blazes.

Such is life.. ;)

scaesare

12-20-06, 09:58 AM

avshaman

12-20-06, 10:09 AM

I dont get what you mean but just to clarify, my 360 conked out 2 hours ago. Before this, I was hot for GoW and anticipating my add-on like the blazes.

Such is life.. ;)

I know what you mean. My first Xbox 360 that I bought at launch died on me within a few hours and had to be replaced twice. The second one also died after a few days of use. The third one finally worked but after I started playing Gears of War it started freezing on me every time I play that game.

My PS3 hasn't had one problem since I got it.

I really do like both systems but I hear so much anti-Sony drivel claiming the PS3 is crap I just feel the need to give credit where credit is due.

It is widely agreed upon that Microsoft had a far greater failure rate with the Xbox 360 launch than Sony did with the launch of the PS3.

Grubert

12-20-06, 10:09 AM

Firstly, I think you are quoting the mux for all streams on the disc, not avg video bitrate.

Secondly, I believe Batman Begins was in the 12 range.

Firstly, you obviously didn't follow the link so xbdestroya nicely gave us, which gave us movie and disc sizes, and total and (approximate) video bitrates.

Just an example:

Blazing Saddles
Codec VC1
Length 1:32:51
Movie size 18,601,672,704
Disc size 22,981,967,872
Total bitrate [all streams] 26.7121489197
[Approx.] Video bitrate 25.4961489197

Secondly, xbdestroya indicated they were cross-format titles. Which Batman Begins is not.

spawnbadboy

12-20-06, 10:15 AM

Art Sonneborn

12-20-06, 10:17 AM

Does anyone know the maximum space used even using MPEG2 on a BD ?

Art

namechamps

12-20-06, 10:19 AM

My question is , what movies have used close to the 50 gig capacity of bluray ?

Wouldn't we better served with what could be a cheaper 30 gig disc + second 15 gig disc for extras ? If the two disc set up comes out to be cheaper why not go for it ?

If what others in this thread are saying are true and a 30 gig hd-dvd disc is price close to a 25 gig bluray disc the 15 gig disc should be much cheaper and most likely the two together would be less than the bluray 50 gig .

I have no problems putting in a second disc for extras . The only things i really use on the current DVDs are the commentaries and on few discs (lord of the rings ) the extras

My understanding is that there is a negligible price premium between HD30 and HD15. I don't have a link now but last time i checked it was about $0.20 per disc.

HD30 will become the 100% standard just like dual layer DVD is used. Your point is still valid. One HD30 disc for the movie and one HD30 disc for extras. 60GB for less than the price of BD50. Not only that the studios love double discs. It has been proven over and over that the "ultra special edition" version of a movie on 2 discs sells more and generates more profit.

If/when HD DVD becomes mainstream expect the same thing as DVD now. One version with single disc for $12-$18 and another special version on 2 discs for $20-$25.

Combine that with HD30/9 combos allowing DVD and HD DVD without a loss of quality and it is easy to see that HD DVD is the low cost mainstream option. If HD DVD wins we will likely see it replace DVD much faster than BD. If BD wins it likely will end up a niche product along the likes of LD and ultimately be replaced by something more consumer friendly (IPTV, VOD, internet download, etc). So while BD has more of a chance of winning the HD DVD vs BD battle, I think HD DVD has more of a chance of winning the HD media vs DVD war.

-KEK-

12-20-06, 10:46 AM

I think this is the first time in history that people are actually arguing for LESS storage capacity. And its mostly because of their irrational hatred of one company.

We are not arguing for less storage, we are arguing that 30GB is enough. More is definately better, but at what cost? I would be all for a 100GB disc, but not if the movies cost $50 because of it. There is a reason why Sony is being so quiet about how much it costs to manufacture the BD50 discs.

scaesare

12-20-06, 11:08 AM

Firstly, you obviously didn't follow the link so xbdestroya nicely gave us, which gave us movie and disc sizes, and total and (approximate) video bitrates.

Just an example:

Blazing Saddles
Codec VC1
Length 1:32:51
Movie size 18,601,672,704
Disc size 22,981,967,872
Total bitrate [all streams] 26.7121489197
[Approx.] Video bitrate 25.4961489197

Secondly, xbdestroya indicated they were cross-format titles. Which Batman Begins is not.

Firstly, you are right grubert, he does indeed quote approx. video bitrate seperately.. indeed my bad. Thanks for pointing that out.

However, ben seems to take issue with these, and cjplay has given some bitrate indication for HD DVD/VC1 in the past that seems to support the notion that 12-13 is certainly doable. Given these, It's not entirely clear to me that a seperately encoded BR VC1 stream automatically means that that another VC1 encode could not look good at a lower rate. After all, the title of the thread is regarding HD DVD.

As for the Batman Begins comment: I'm answering providing a reponse to this request:

Ok so yeah, I *definitely* do not want LOTR to be at 12.5Mbps, unless one of the MS crew is willing to share a stunning title with the audience that has been encoded at similar levels.

Just because some cross-platform streams are being used as a point of comparison, that doesn't invalidate what has been deomonstrated as possible with HD DVD (which, once again, is the title of this thread)

xbdestroya

12-20-06, 11:25 AM

Of course lower bitrates are possible... and would probably still look 'good.'

But people, this is Lord of the Rings we're talking about, so let's be honest with ourselves. Good isn't what we're looking for; it has to be the best that it can be. And indeed, as Benes, the guy behind the study himself pointed out in this thread... these bitrates are average, not peak.

Again I don't have a problem with VC-1 *or* HD DVD... I think VC-1 is great and efficient, and I think HD DVD is, well... it gets the job done. But at the same time, it's not as if they are forever extensible into infinity as something that can preserve at four hours what we've come to expect in the two hour range. I'm just saying, Batman Begins isn't at the top of the PQ pile, and until Ben or Amir or someone else starts giving up the identities of some of these other sub-13 titles, I frankly have to consider playing flexibly with wordings on the part of MS.

benwaggoner

12-20-06, 11:29 AM

And I'm going to have to trust them to be honest with you, as they're taken in the best way I know of; right off of a PC. IMO if you feel these numbers to be incorrect, it is incumbent on you to indicate how - and IMO, offer up a title as a sacrifice for the knowledge base. There is otherwise no reason to doubt something as unbiased as a PC bitrate readout.
Sorry, I'm under strict instructions not to share the bitrates for any commerical title. Not my call to make (although I'd love to share the whole .xls if I could).

I know how the file size was found, but how was the video data rate determined? The file size contains a whole lot of other stuff, so just dividing file size by runtime isn't going to be particularly accurate. I presume there's some correction for audio as well, since some numbers were too high, and others were too low.

'Highly touted' could mean anything. Let me ask you this - are any of the VC-1 titles out on BD coming in at under 13 Mbps in bitrate?
I believe wso, yes.

I own Superman Returns on BD, and it's VC-1. The PS3 allows for bitrate tracking during playback, so I myself will run a confirmation test on that title tomorrow to see if there are discrepencies.
Again, do you know if it's tracking total bitrate, just audio bitrate, or what (I honestly don't know in this case).

Grubert

12-20-06, 11:37 AM

Just because some cross-platform streams are being used as a point of comparison, that doesn't invalidate what has been deomonstrated as possible with HD DVD (which, once again, is the title of this thread)

True. But when I look at a sample of 14 items with the following values:
15, 16, 16, 17, 19, 20, 20, 22, 22, 23, 24, 25, 25, 25

Now I'm no statistician, but it seems to me that the median is 21. ;)

What'sHD

12-20-06, 11:44 AM

True. But when I look at a sample of 14 items with the following values:
15, 16, 16, 17, 19, 20, 20, 22, 22, 23, 24, 25, 25, 25

Now I'm no statistician, but it seems to me that the median is 21. ;)
..and the mode is 25 while the mean is > 21.

What'sHD

12-20-06, 12:07 PM

30GB works fine. 50GB is nice but not at the added cost it bring.
That is, with all due respect, drivel unless you are a content producer who is not getting subsidies from Sony for BD50 video content.

If not, be glad that Sony is making BD50s as cheap as HD30s for us consumers. Since it does not cost you, its highly disingenuous to complain about it. Whether this comes from hatred of Sony or love of hD-DVD, it doesnt matter cos it is helping spread misinformation.

BuGsArEtAsTy

12-20-06, 12:07 PM

Check my sig for the HD DVD disc sizes thread, if you're interested.

bobgpsr

12-20-06, 12:14 PM

Does anyone know the maximum space used even using MPEG2 on a BD ? The data compiled by Benes, http://www.avsforum.com/avs-vb/showthread.php?p=9044310&&#post9044310
shows:
BHD at 48,703,078,400 bytes total on disc with movie size at 37.2 mb. It used linear PCM for sound.

KOH was 45,228,883,968 bytes on disc but with a movie size of 41.9 mb. It used dts-HD MA for sound.

xbdestroya

12-20-06, 12:18 PM

Again, do you know if it's tracking total bitrate, just audio bitrate, or what (I honestly don't know in this case).

The PS3 tracks audio and video bitrates seperately, and as for what Benes has done, the program he's using provides a straight average. So I'll leave it to him to provide the averages... but after playing Superman Returns for myself using the bitrate tracking, I've found that the bitrate varies *wildly* throughout the film... which makes sense of course. But seems to be above 20 as often as it's below 12.

Here are some pics I took (it's a CRT RPTV, so excuse the scanning), of two of the highest peak bitrate frames I found.

So those are peaks of 29.7Mbps and 31.3Mbps for those two scenes respectively.

http://i6.photobucket.com/albums/y206/xbdestroya/Pilots.jpg

http://i6.photobucket.com/albums/y206/xbdestroya/Leaves.jpg

DPowers

12-20-06, 12:38 PM

I know what you mean. My first Xbox 360 that I bought at launch died on me within a few hours and had to be replaced twice. The second one also died after a few days of use. The third one finally worked but after I started playing Gears of War it started freezing on me every time I play that game.

My PS3 hasn't had one problem since I got it.

I really do like both systems but I hear so much anti-Sony drivel claiming the PS3 is crap I just feel the need to give credit where credit is due.

It is widely agreed upon that Microsoft had a far greater failure rate with the Xbox 360 launch than Sony did with the launch of the PS3.

This is where you show your true bias. For everyone that posts that they had issues with their 360 there is more than a 10 to 1 ratio that will post that they didn't. And that is over estimating the percentage of failed 360s. You post like you are the majority when, in fact you are part of the vast minority of people that had issues.

I can understand your disappointment, but I am one of the lucky majority that had zero issues with my 360 from launch day. In fact I never had any issues with any console I've ever owned. I also never had any issues with my 360 drive and it works like a charm to this day. I also beleive that I will have zreo issues with my PS3.

I will never talk down a company because I had an issue with one of their products. That product is not the sum of their parts. I research and take each product on it's own merit. Just because I dislike the Sony Viao, doesn't mean I hate their displays...I own one.

Just remember that you are posting from your prospective and that is not the prospective of the average consumer, because they never had the anger inducing issues that you have experienced with certain products.

Johnsteph10

12-20-06, 02:15 PM

gljvd

12-20-06, 02:30 PM

Well, you're not everybody are you!? I look forward to the special features almost as much as the film itself and I don't like switching discs.
And again, are you really unable to think beyond the current moment to imagine uses which would require or benefit from the larger capacity?

I also don't want to pay an arm and a leg for something.

I can look foward into the future and if I see sony leading something I rather not have it happen .

I can settle for lower space with advance codecs if its priced properly. I don't believe bluray is worth the price

darinp2

12-20-06, 02:42 PM

The PS3 allows for bitrate tracking during playback, so I myself will run a confirmation test on that title tomorrow to see if there are discrepencies.An XBOX360 add-on plugged into a PC with the right software can also reported statistics like this. I asked a friend with this kind of setup to check the scene in "Batman Begins" where Bruce Wayne goes down into the bat cave for the first time (as an adult) and the bats are flying all around. He said that the audio (TrueHD) and video rates both stayed peaked there for a little while.

--Darin

xbdestroya

12-20-06, 02:45 PM

Darin... yeah I know on the add-on front, just seems that there's no one here on this forum willing to do the job. But that's great that your frend is running his own tests; you say it peaked there for a bit, but did he mention what that peak bitrate was?

By the way Batman Begins in general is a darker film with a number of slower 'drama' scenes... so I expect this is a large part of how MS has been able to get this film down into their much-touted sub-13Mbps range.

DPowers

12-20-06, 03:07 PM

People are much more likely to post negative experiences with their 360/PS3 than post positive ones (ie: working perfectly). That's the way life is...we don't remember/pay attention to the status quo.

For instance, when gambling, notice how all of the lights/bells go off when you win? What about when you lose? The casinos deliberately set you up to remember the one winning instances out of the 100 failures..so you don't think about/remember the normal circumstances (putting a quarter in and losing).

There are plenty of other circumstances in life that fit into this model.

Oh, the other thing to remember is that anyone can write anything on the internet. :D

John

Do you think you are letting us in on some kind of secret? Obviously people are more vocal about their negative experiences with a product and they have every right to be, but to base your whole outlook on a company because of issues with one product is just close minded.

I never implied that he shouldn't post what ever he wants, when ever he wants. I will imply that it is a good idea for you to post something that can remotely relate to the topic in stead of lecturing us on psycological trends in marketing of slot machines. Niether statement even remotely relates to my post.

darinp2

12-20-06, 03:20 PM

But that's great that your frend is running his own tests; you say it peaked there for a bit, but did he mention what that peak bitrate was?I believe he said the tool was reporting 20Mbps for the peak. I had understood it would be limited to about 19Mbps because of the bandwidth taken for the audio, IME, etc. and HD DVD's limitation of about 30Mbps for everything. As far as the audio bitrate, I don't recall. Just based on 16/48 TrueHD I think it would have been over 3Mbps and under 4Mbps, but I don't remember the number.

It seems that the bitrates are not to be shared and it wouldn't surprise me at all if the number of person hours spent to get the "Batman Begins" encode looking like it does at its average rate is something they wouldn't want people finding out.

--Darin

wreckshop

12-20-06, 03:25 PM

I think this is the first time in history that people are actually arguing for LESS storage capacity. And its mostly because of their irrational hatred of one company.

weird isn't it? if only hd dvd had VC-1, then I could see their point. but both sides have the exact same codecs, so theoretically should provide the exact same picture if encoded at the same bitrate. and all this conspiracy talk of "sony forcing everyone to use mpeg" is silly too. isn't codec choice up to the studio?

Its like saying a 320gb WD drive is better than the 150 WD raptor drive just beacuse it has more space...

I'd take the 320 any day over the 150.

scaesare

12-20-06, 03:34 PM

Anamorphiac

12-20-06, 03:43 PM

...and so far I have seen only one sided, biased attacks againsts Sony...

Oh, I forgot...your opinions are considered fact and all other opinions are considered one sided and biased...

With that type of attitude, it is no surprise my ignore list just keeps getting bigger...one at a time.

xbdestroya

12-20-06, 03:44 PM

I believe he said the tool was reporting 20Mbps for the peak. I had understood it would be limited to about 19Mbps because of the bandwidth taken for the audio, IME, etc. and HD DVD's limitation of about 30Mbps for everything. As far as the audio bitrate, I don't recall. Just based on 16/48 TrueHD I think it would have been over 3Mbps and under 4Mbps, but I don't remember the number.

Thanks for the insights - that would definitely seem in line with the peaking I saw during the Superman playback.

It seems that the bitrates are not to be shared and it wouldn't surprise me at all if the number of person hours spent to get the "Batman Begins" encode looking like it does at its average rate is something they wouldn't want people finding out.

Yeah, Microsoft is definitely viewing this whole bitrate thing as a near state secret.

BIG ED

12-20-06, 03:52 PM

I think this is the first time in history that people are actually arguing for LESS storage capacity. And its mostly because of their irrational hatred of one company.
No.
It's becauce of their "rational" hatred of one company!
;)

Grubert

12-20-06, 04:06 PM

And where are those numbers coming from?

benes' table. Listed VC-1 video bitrates, rounded off to the nearest integer.

BIG ED

12-20-06, 04:06 PM

All you BD 50 bigger is better / bigger is the future pushers should be pushing for massive future formats, not little old BD 50.
(what REALLY cracks me up is how when BD was only 25 GB, they were the "size DOESN'T matter" pushers!!!)

And you should do it on another thread, not the "HD-DVD 30GB/Movie Length Facts" thread.
What in the heck does anything to do w/BD 50, have to do w/HD-DVD 30GB "facts"???
Maybe start the "BD 50GB/Movie Length Facts" thread?

xbdestroya

12-20-06, 04:17 PM

And you should do it on another thread, not the "HD-DVD 30GB/Movie Length Facts" thread.
What in the heck does anything to do w/BD 50, have to do w/HD-DVD 30GB "facts"???
Maybe start the "BD 50GB/Movie Length Facts" thread?

Well, the thread started out as a 'facts' thread for HD DVD with the notion of dispelling the misinformation that 30GB isn't enough for Lord of the Rings to look great on at 12.5Mbps. And then if you've been following... a lot has happened after that. And it's very relevent to exactly this scenario that the thread discusses. Information that IMO points to, indeed, a reduced PQ scenario vs what could be achieved on a disc greater than 30GB in size.

raaj

12-20-06, 05:17 PM

...

Yeah, Microsoft is definitely viewing this whole bitrate thing as a near state secret.
In the same breath, could you please ask Sony to enlighten us with how much they are subsidizing BD50 production and authoring costs for gaining the favor of certain studios? You would probably consider that a "trade secret".

Grubert

12-20-06, 05:22 PM

In the same breath, could you please ask Sony to enlighten us with how much they are subsidizing BD50 production and authoring costs for gaining the favor of certain studios? You would probably consider that a "trade secret".

Good question. Feel free to open a thread asking that.

In the meantime, we remain seized of the issue at hand, to wit movie file sizes and video bitrates.

Rob Zuber

12-20-06, 05:31 PM

As always, the BD side is seeking facts and objective measurements while the HD-DVD side aggressively pushes MS propaganda and expects everyone to take it on faith.

raaj

12-20-06, 05:36 PM

Good question. Feel free to open a thread asking that.

In the meantime, we remain seized of the issue at hand, to wit movie file sizes and video bitrates.

You would also do well to keep in mind that the title says "HD-DVD 30GB/Movie Length Facts" - and all I see here is ""HD-DVD 30GB/Movie Length Speculation". Until the process of arriving at these numbers has been blessed as accurate by MS or the post house or the like, all this is just an "educated guess" at best - no disrespect to Benes.

raaj

12-20-06, 05:37 PM

As always, the BD side is seeking facts and objective measurements while the HD-DVD side aggressively pushes MS propaganda and expects everyone to take it on faith.

Ain't that the truth, brotha !! *pat your back*

Grubert

12-20-06, 05:50 PM

You would also do well to keep in mind that the title says "HD-DVD 30GB/Movie Length Facts" - and all I see here is ""HD-DVD 30GB/Movie Length Speculation". Until the process of arriving at these numbers has been blessed as accurate by MS or the post house or the like, all this is just an "educated guess" at best - no disrespect to Benes.

We don't need a Microsoft representative to tell us an hour has 60 minutes.

xbdestroya

12-20-06, 05:52 PM

Until the process of arriving at these numbers has been blessed as accurate by MS or the post house or the like, all this is just an "educated guess" at best - no disrespect to Benes.

Well and there's the irony. If the method is sound, I wouldn't expect it to *ever* be blessed. Or even mentioned more than it has to be.

Speaking of which, there was an MS insider posting in this thread, but he has seemingly stopped (though he has continued to post elsewhere). If he returns, I've got a question for him. ;)

raaj

12-20-06, 05:57 PM

We don't need a Microsoft representative to tell us an hour has 60 minutes.

We have not limited the discussion to the "Movie Length" as in the "movie duration in hours, minutes and seconds" - have we? The issue is that the movie length is being used as a basis to make an "educated guess" on the "average bit rate" of the movie encode.

Though Benes' methodology is reasonable, I wouldn't try to spread it as gospel until its accuracy has been blessed by compressionists [like BenWaggonner, CJPlay etc.] who know exactly how it has to be calculated.

The fact that Ben has questioned the methodology makes me skeptical of the accuracy of Benes' claims. Now go ahead and question the motives of Ben and MS all you want, but that is not what this thread is about - as per your post in quotes. ;)

Good question. Feel free to open a thread asking that.

In the meantime, we remain seized of the issue at hand, to wit movie file sizes and video bitrates.

Cjplay

12-20-06, 06:24 PM

Ok here we go on some VC-1 cross-format titles.
Blazing Saddles - length 93 min - movie file 18.6GB - video bitrate 25.5 Mbps
Corpse Bride - length 77 min - movie file 13GB - video bitrate 19.9 Mbps
House of Wax - length 113 min - movie file 18GB - video bitrate 19 Mbps
Lethal Weapon - length 110 min - movie file 21GB - video bitrate 25 Mbps
Phantom of the Opera - length 141 min - movie file 19GB - video bitrate 16.5 Mbps
Superman Returns - length 154 min - movie file 20.5GB - video bitrate 16 Mbps
Unforgiven - length 131 min - movie file 18GB - video bitrate 17Mbps

I can back Ben up and say several of these are wrong. Note 2 things missing from Benes' list. Corpse Bride had 3 audio tracks, right? And its subtitle tracks take up ~32k a pop (average of 3 on Domestic titles). I'd say on every bitrate on that person's list you have to drop at least 1Mbit on every single title... Maybe 2.

We've got at least 5 titles in the field below 13Mbps. All getting 4-4.5 stars from most reviewers.

LOTR:ROTK EE's numbers are what I usually do (use search to find my last one), but I'm going to do a theoretical Hamlet (1996) full-length version at 4 hours and 3 minutes. The feature alone can handle 14.2Mbps ABR with 3 DD+ 5.1 tracks or 1 TrueHD track with no trailer. Drop about 1Mbit off the ABR and you could add a couple extras like a trailer or SD documentaries at 2-3Mbps (we have it working well in VC-1, btw). However, I'd be reluctant to put Ben Hur on one disc or even Last Emporer and GWTWind. Titanic would be a breeze on an HD30.

Cjplay.

raaj

12-20-06, 06:28 PM

Cjplay

12-20-06, 06:30 PM

As always, the BD side is seeking facts and objective measurements while the HD-DVD side aggressively pushes MS propaganda and expects everyone to take it on faith.

I'll not disagree with the "sides" part, as HD DVD has two big things going for it in my book - disc manufacturing cost (coming down for BD, I know) and almost feature completeness of its spec in the Gen1 players (Internet connectivity was missing from my scoresheet).

However, for some facts, feel free to pose questions. I hate propaganda and will give as close to the straight dirt as I can.

Cjplay.

Rob Zuber

12-20-06, 06:53 PM

However, I'd be reluctant to put Ben Hur on one disc or even Last Emporer and GWTWind.However, for some facts, feel free to pose questions.Thanks. How about a list of all Warner movies that you think would not properly fit on a 30 GB HD-DVD (assuming you want good PQ, of course)?

Rio

12-20-06, 07:06 PM

johnu

12-20-06, 07:10 PM

bobgpsr

12-20-06, 07:29 PM

Based on my record, Corpse Bride BD version has total 4 DD 640kbps trakcs. As to the overhead, usually I am assuming 5% overhead, and my calculation figured out that average video bit rate of Corpse Bride is around 18.8Mbps.

Rio,

Big question. Just who the heck are you? An "experts and professionals in the field with actual experience in HD encoding". A Panasonic employee? It seems that you use insider info to attack posts but refuse to answer direct questions or who you are.

Rio

12-20-06, 07:34 PM

The numbers shown are calculated based on very simple and inerrable theory, but just includes some minor margin of error. But for some people, the person who did it suddenly turns into "zero credibility" person by their definition.

Umm, looks like "science" is no more here at AV "Science" Forum... :(

xbdestroya

12-20-06, 07:45 PM

I can back Ben up and say several of these are wrong. Note 2 things missing from Benes' list. Corpse Bride had 3 audio tracks, right? And its subtitle tracks take up ~32k a pop (average of 3 on Domestic titles). I'd say on every bitrate on that person's list you have to drop at least 1Mbit on every single title... Maybe 2.

We've got at least 5 titles in the field below 13Mbps. All getting 4-4.5 stars from most reviewers.

LOTR:ROTK EE's numbers are what I usually do (use search to find my last one), but I'm going to do a theoretical Hamlet (1996) full-length version at 4 hours and 3 minutes. The feature alone can handle 14.2Mbps ABR with 3 DD+ 5.1 tracks or 1 TrueHD track with no trailer. Drop about 1Mbit off the ABR and you could add a couple extras like a trailer or SD documentaries at 2-3Mbps (we have it working well in VC-1, btw). However, I'd be reluctant to put Ben Hur on one disc or even Last Emporer and GWTWind. Titanic would be a breeze on an HD30.

Cjplay.

Cj, thanks for responding. You'll understand though my own personal position on the matter, and why the vagaries out there on the issue of VC-1 bitrates, coupled with viable bitrate meters, leads to this sort of situation where best guesses have to sub in for hard facts...

Any and all real-world examples you are comfortable talking about I think would be greatly appreciated... and I don't think just by us 'VC-1 bitrate skeptics' either, but by *all* parties on either side of the fence. Certainly I prefer when talk is of numbers rather than ideas, and it goes a ways towards soothing my angst to hear you speak concretely in terms of, 'drop 1Mbps, add a documentary... stuff like that. That's the world I operate in.

I want to finish by noting though that when judging the bitrates of the explored titles, you said that they would have to drop by 1Mbps on the video - maybe 2. Well, I think 1Mbps was already acknowledged in a way just by default... so even at 2 I think Benes deserves some credit for pursuing this in a very methodical manner. For the tech-heads on the encoding side vs those of us on the 'digging' side, I'd like to think there's mutual respect for the desire to understand the technology.

Cjplay

12-20-06, 08:01 PM

Cj, thanks for responding. You'll understand though my own personal position on the matter, and why the vagaries out there on the issue of VC-1 bitrates, coupled with viable bitrate meters, leads to this sort of situation where best guesses have to sub in for hard facts...

Any and all real-world examples you are comfortable talking about I think would be greatly appreciated... and I don't think just by us 'VC-1 bitrate skeptics' either, but by *all* parties on either side of the fence. Certainly I prefer when talk is of numbers rather than ideas, and it goes a ways towards soothing my angst to hear you speak concretely in terms of, 'drop 1Mbps, add a documentary... stuff like that. That's the world I operate in.

I want to finish by noting though that when judging the bitrates of the explored titles, you said that they would have to drop by 1Mbps on the video - maybe 2. Well, I think 1Mbps was already acknowledged in a way just by default... so even at 2 I think Benes deserves some credit for pursuing this in a very methodical manner. For the tech-heads on the encoding side vs those of us on the 'digging' side, I'd like to think there's mutual respect for the desire to understand the technology.

No argument on any point. However, for those insiders who watch rather than post, they may think something that isn't the truth. Benes deserves thanks for doing what I can't blatantly do, but I'd hope that he/she (no assumptions) would not use "main audio" but perhaps these numbers? For Warner, assume a 640Kbps BR or lower for DD+ and AC3. For TrueHD, assume something in the 1500k range, but Cyberlink should say what it actually is. For Subtitles, use ~32Kbps per stream. The IME will be about 1-1.5Mbps ABR. Therefore, use Benes' numbers for filesize, but please subtract properly so at least better numbers are being used. Most of Uni's DD+ 5.1's are 1.5Mbps, btw. And everyone knows the LPCM math, right? Bits times hz/samples? A lot of people will split the 1Mbps hair at the level I work at.

Thank you for your support and search for truth. Anything else needs smoothing, please let me know.

Cjplay.

mobius

12-20-06, 08:01 PM

scaesare

12-20-06, 08:22 PM

benes' table. Listed VC-1 video bitrates, rounded off to the nearest integer.

That's what I thought. Hence my comment:

It's not entirely clear to me that a seperately encoded BR VC1 stream automatically means that that another VC1 encode could not look good at a lower rate. After all, the title of the thread is regarding HD DVD.

Just because those bitrates exist on BR, doesn't automatically mean that a sperate encode on HD DVD at a lower rate (perhaps with new tools, or a third tuning pass, etc...) isn't just as good. And I provided BB as an example, as the OP was asking about VC1 on HD DVD.

scaesare

12-20-06, 08:27 PM

Well, the thread started out as a 'facts' thread for HD DVD with the notion of dispelling the misinformation that 30GB isn't enough for Lord of the Rings to look great on at 12.5Mbps. And then if you've been following... a lot has happened after that. And it's very relevent to exactly this scenario that the thread discusses. Information that IMO points to, indeed, a reduced PQ scenario vs what could be achieved on a disc greater than 30GB in size.

Sure.. if that were evidenced. So far we have evidence that the tools used for somthing like Batman Begins have produced a stellar encode at those rates.

Do you really think I couldn't find 2 mpeg encodes produced with different toolsets or work flows couldn't differ dramatically in bitrate? Heck I've seen high bitrate MPEG2 look worse than lower bitrate MPEG2 with similar footage.

scaesare

12-20-06, 08:31 PM

Ben if you actually read the thread (http://www.avsforum.com/avs-vb/showthread.php?t=760714) it tells you exactly how the numbers were calculated. None of these discs have any interactive features so there is not much other data in the stream to account for. Its very likely the numbers are slightly inflated and thats why I say they are not 100% accurate. But the audio tracks were accounted for and thats the best I can do.

Would you mind explaining to me how it is mathematically possible that some of the numbers are LOWER than the real ones? If I take the file size and divide it by the runtime how in the world can it be lower than the real bitrate?

And what exactly is the big secret about bitrates? Why can't you divulge them? Surely you were aware that tools would eventually be available to people to at least get a good estimate on their own. I mean Fox puts the bitrates right on their covers. It would be nice if everyone did that instead of keeping it some big industry secret. If you're so proud of how low your bitrates are then why not tell us?

Microsoft produces a tool that a cusotmer uses to produce work product. They also may consult/support that toolset in a customer's proprietary environment. Do you honestly think that gives them liberty to disclose details about that customer's work product without explicit permission?

You think MS would be at liberty give you data about a customer's source code just because MS supplied Visual C++ to them? :rolleyes:

xbdestroya

12-20-06, 08:36 PM

Sure.. if that were evidenced. So far we have evidence that the tools used for somthing like Batman Begins have produced a stellar encode at those rates.

Firstly we don't know exactly the rates for Batman Begins, until we run it through the 'Numba Cruncha.' But yes, assuming it is 12Mbps on average, I agree it looks very good. Now, mind you I've given specific reasons why in my opinion this title more than others can look good there at that bitrate: it's dark, and there are a lot of slower 'drama' scenes. Plenty has come up here to, in my mind, question just how exciting we should take any 12.5Mbps developments to be.

Do you really think I couldn't find 2 mpeg encodes produced with different toolsets or work flows couldn't differ dramatically in bitrate? Heck I've seen high bitrate MPEG2 look worse than lower bitrate MPEG2 with similar footage.

Scaesare I absolutely agree with you. As the tools for VC-1 advance, I recognize that the bitrates can go down without an associated linear drop in quality. Linear drop... but in terms of absolute drops I think we're just going to need to do more testing as bitrate spreads increase.

Microsoft produces a tool that a cusotmer uses to produce work product. They also may consult/support that toolset in a customer's proprietary environment. Do you honestly think that gives them liberty to disclose details about that customer's work product without explicit permission?

Right... no. But that doesn't mean the MS smokescreen that often goes up on this forum is appreciated by all of us. If a question can't be answered, just state as such without the directional/perceptual feints.

I have a lot of respect however for Cj coming into this thread, and not only pointing out the possible errors in final bitrate estimates, but providing also the insights needed to reach better estimates. Now that's the kind of answer anyone can appreciate; whether they're interested in this stuff or not, the candor comes through.

BIG ED

12-20-06, 08:54 PM

Well, the thread started out as a 'facts' thread for HD DVD
Correct!
with the notion of dispelling the misinformation that 30GB isn't enough for Lord of the Rings to look great on at 12.5Mbps.
No where in the original or the first followup post by the OP, is LotR mentioned. And then if you've been following... a lot has happened after that. And it's very relevent to exactly this scenario that the thread discusses.
It's not my fault the thread got Off Topic. It is the fault of marauding fanboys. Information that IMO points to, indeed, a reduced PQ scenario vs what could be achieved on a disc greater than 30GB in size.
So what!
It's a "facts" about 30GB discs, not a "facts" about >30GB discs.
BD 50GB discs will not be able to hold multi seasons of HD TV shows.
So should I post on every BD 50 thread that BD 50 isn't "BIG" enough?
Don't answer that, I'll just do it! :p

If you read the OP, you would have seen;
Obviously this could be helpful for both HD-DVD and BD. Perhaps an official facts thread for each that can debunk the horrible misinformation?
Any "facts" you would like to add about HD DVD 30GB discs? :D

scaesare

12-20-06, 09:05 PM

Firstly we don't know exactly the rates for Batman Begins, until we run it through the 'Numba Cruncha.' But yes, assuming it is 12Mbps on average, I agree it looks very good. Now, mind you I've given specific reasons why in my opinion this title more than others can look good there at that bitrate: it's dark, and there are a lot of slower 'drama' scenes. Plenty has come up here to, in my mind, question just how exciting we should take any 12.5Mbps developments to be.

Do you have this tool for HD DVD? Until you can dispute it, Amir and Ben have referred to this rate for this title, and insiders such a CJPlay (I believe) have confirmed it. At this point I think it's a pretty safe bet.

Scaesare I absolutely agree with you. As the tools for VC-1 advance, I recognize that the bitrates can go down without an associated linear drop in quality. Linear drop... but in terms of absolute drops I think we're just going to need to do more testing as bitrate spreads increase.

Do you have some evidence that a drop from high 15's or 16's to 12 isn't possible? I've seen a _SINGLE_ encoding parameter change do similar on a different encoder. Or an additional pass. Or the decision to do scene re-encoding.

Are you catagorically stating that an absolute drop of several Mb is impossible with toolset revisions? If your answer is yes, would you tell me if you've ever laid eyes on these tools?

Right... no. But that doesn't mean the MS smokescreen that often goes up on this forum is appreciated by all of us. If a question can't be answered, just state as such without the directional/perceptual feints.

I have a lot of respect however for Cj coming into this thread, and not only pointing out the possible errors in final bitrate estimates, but providing also the insights needed to reach better estimates. Now that's the kind of answer anyone can appreciate; whether they're interested in this stuff or not, the candor comes through.

I too respect CJ. Especially because even tho he's taken some heat for posting here, he continues to do so. But the "smokescreen" as you say is not necessarily what people here are making it out to be. Amir and Ben have flat out said that they cannot comment on a release that IS NOT THEIR'S. As soon as the studio has divulged that info, or it has otherwise become public, they do so. I don't know how much more clear you want them to make that.

Do you really expect them to jeopardize their relationship with a studio to satisfy some internet forum jockeys (including yours truly)?

darinp2

12-20-06, 09:13 PM

However, for some facts, feel free to pose questions. I hate propaganda and will give as close to the straight dirt as I can.A recent title on both listed there is "Superman Returns". The calculations show 15.9Mbps for the Blu-ray version. Can you tell us if that is accurate and if not, how far it is off? Also, when you quote average bitrates, are you including the credits also? Is there a standard range that normal credits get?

Thanks,
Darin

BIG ED

12-20-06, 09:17 PM

As always, the BD side is seeking facts and objective measurements while the HD-DVD side aggressively pushes MS propaganda and expects everyone to take it on faith.
Why in the world is "the BD side" seeking facts and objective measurements about HD-DVD so aggressively?
Are you guys thinking of switching? :eek:
And you do realize BD uses "MS propaganda" (VC-1), don't you? :rolleyes:

The one & only "fact" that I was seeking when I tripped upon this thread (Ugh! What a mistake that was.), was; was the reason a TrueHD track was (is that a record for "was" in one sentence?) left off a KK a studio decision or a space issue.
Just the facts. Did in fact Universal leave TrueHD off of KK because they are not pushing for TrueHD at this time?
Thanks.

And I know fanboys, there is "no" room for lossless on Kong.
That's because they didn't make room for it.
Anyone have proof to dispell that 'fact'?

xbdestroya

12-20-06, 09:47 PM

Do you have this tool for HD DVD? Until you can dispute it, Amir and Ben have referred to this rate for this title, and insiders such a CJPlay (I believe) have confirmed it. At this point I think it's a pretty safe bet.

The tool is available for HD DVD, yes. Do I have it? Only insomuch as I don't have the HD DVD add-on. But I believe it is directly referenced earlier in this thread; if not this one then the other '30GB ' thread. So you can read about some of the results on Batman Begins there.

Do you have some evidence that a drop from high 15's or 16's to 12 isn't possible? I've seen a _SINGLE_ encoding parameter change do similar on a different encoder. Or an additional pass. Or the decision to do scene re-encoding.

Are you catagorically stating that an absolute drop of several Mb is impossible with toolset revisions? If your answer is yes, would you tell me if you've ever laid eyes on these tools?

You're misunderstanding me. Of course I think a drop in bitrate is possible; my concern is that a drop from 16 to 12 is not transparent; aka, there *is* some quality loss in that drop... however minor or not, we do not truly know. So, when I say linear vs absolute, what I was trying to convey is that... of course I do not believe that a linear reduction in bitrate, due to tool improvement, results in a linear decrease in PQ. Because that's just common sense and what better tools are about. But what I am saying, is just because there's not a linear drop... doesn't mean that there is no drop at all.

I too respect CJ. Especially because even tho he's taken some heat for posting here, he continues to do so. But the "smokescreen" as you say is not necessarily what people here are making it out to be. Amir and Ben have flat out said that they cannot comment on a release that IS NOT THEIR'S. As soon as the studio has divulged that info, or it has otherwise become public, they do so. I don't know how much more clear you want them to make that.

Ben has said he can't speak to it - and it would be fine if he stopped there. But before understanding how the numbers were reached, he tries to imply that they are bogus. When in fact, they end up pretty close to the truth. It's just that reflexsive sort of defense that relies on their 'street cred' alone that I have a problem with. When valid questions are raised, take the time to post valid answers. Saying 'I can't say anything' is fine... just so long as innuendos aren't inserted along with it.

Do you really expect them to jeopardize their relationship with a studio to satisfy some internet forum jockeys (including yours truly)?

Of course not.

scaesare

12-20-06, 11:25 PM

The tool is available for HD DVD, yes. Do I have it? Only insomuch as I don't have the HD DVD add-on. But I believe it is directly referenced earlier in this thread; if not this one then the other '30GB ' thread. So you can read about some of the results on Batman Begins there.

Are there numbers we can glean from it?

Given that VC1 has had significant more hands-on time and attention from the HD-DVD side of the fence, it may not necessarily be the case that we can draw absolute conclusions on what VC1 is capable of today with data based on encodes on a format that hasn't had the "tuning time". It would seem the entire argument in this "HD-DVD 30GB/Movie Length Facts" thread is based on data that is based on neither HD DVD nor a 30 GB disc.

You're misunderstanding me. Of course I think a drop in bitrate is possible; my concern is that a drop from 16 to 12 is not transparent; aka, there *is* some quality loss in that drop... however minor or not, we do not truly know. So, when I say linear vs absolute, what I was trying to convey is that... of course I do not believe that a linear reduction in bitrate, due to tool improvement, results in a linear decrease in PQ. Because that's just common sense and what better tools are about. But what I am saying, is just because there's not a linear drop... doesn't mean that there is no drop at all.

Certainly true and a valid point. However 2 things I think we should keep in mind:

1) There tends to be a poit of diminishing returns for a given Q setting. That is if I set a target Q of .9 for a given video, and the encoder only needs 12 to reach it, then even if I were to set the encoder for a CBR of 16, that would dictate an increase of 4Mbps on the disc with no perceptible increase in quality. Conversely, just becuse an encode initially started at 16 does not mean that it could not be re-encoded at 12 without a perceptable decrease in quality. Tis the nature of an encoder having the bits it needs for a given target, and throwing more at it doesn't necessarily help.

This is espicially the case if an early encode was rushed (or the folks were not as familiar with all the knobs of a new codec) or was encoded with an earlier tool revision, or they were time-contrained and did not do scene re-encoding, etc...

2) If the target medium has additional room, there may be no incentive to spend additional time tuning the encode. Therefore extrapolating what has been done to what could be done is not necssarily a valid argument. Especially when the formats differ physically, and have very different histories and the folks goin in to it have different mindsets.

As an example, one Sony representative made much ado about making sure they had a bit-rate meter in their BR player, saying something to the effect of "people are going to be impressed with some mind-blowing bitrates".

Ben has said he can't speak to it - and it would be fine if he stopped there. But before understanding how the numbers were reached, he tries to imply that they are bogus. When in fact, they end up pretty close to the truth. It's just that reflexsive sort of defense that relies on their 'street cred' alone that I have a problem with. When valid questions are raised, take the time to post valid answers. Saying 'I can't say anything' is fine... just so long as innuendos aren't inserted along with it.

Yeah, I've gone back and double-checked Ben's quotes, and I hope he'll clarify. For some reason I suspect he MAY be assuming those were HD DVD #'s, given the thread title, but I don't want to speak for him.

I will say this though, if Ben claims we are seeing ~12Mb titles on HD DVD (which his comments have all been regarding in the past, hence my suspicion about his assumptions), and as Cj mentions here there are other HD DVD's released in the same range:

We've got at least 5 titles in the field below 13Mbps. All getting 4-4.5 stars from most reviewers.

Then the fact that earlier encodes didn't hit this target on a different medium really are moot, no?

If a car can do 0-60 on the dragstrip in 3 seconds, then it can do it. It doesn't really matter if it can't do it on gravel.

So, it would seem that there are real-world examples of VC1 hitting ~12Mbps bit rates and producing excellent pictures.

Now, if we want to argue that the content of those encodes made them special cases, then I'm all for that, and perhaps Cj or others can specify what titles they are so we can argue about that for a while. :) But I think that using the BluRay #'s to determine VC1's ultimate capability is not going to really help get the truth, other than to say they were either poor, early, or "good enough to fit" encodes.

darinp2

12-20-06, 11:47 PM

It would seem the entire argument in this "HD-DVD 30GB/Movie Length Facts" thread is based on data that is based on neither HD DVD nor a 30 GB disc.It seems like you are ignoring that some (or maybe it was all) of those VC-1 releases for Blu-ray were encoded with HD DVD and 30GB in mind and then ported.

--Darin

xbdestroya

12-20-06, 11:49 PM

Are there numbers we can glean from it?

Well, essentially the numbers gleaned were that in some scenes, even Batman Begins was indeed maxing out the HD DVD bitrate. Not much else to report beyond that unless 'fleshier' numbers come out on that tangent.

Given that VC1 has had significant more hands-on time and attention from the HD-DVD side of the fence, it may not necessarily be the case that we can draw absolute conclusions on what VC1 is capable of today with data based on encodes on a format that hasn't had the "tuning time". It would seem the entire argument in this "HD-DVD 30GB/Movie Length Facts" thread is based on data that is based on neither HD DVD nor a 30 GB disc.

Well, but the VC-1 encodes for these cross-format titles are the same though, which is why they are useful. In fact, even more useful in that the way these titles are currently structured, it is not that difficult to islolate the data/bitrates of the video stream from the rest of it. It hasn't been explicitly stated as yet - and maybe we should ask for that - but I truly do believe that anything Warner is putting out on both BD and HD DVD using VC-1 is the exact same VC-1 encode. So I don't think it is inappropriate to use them as direct stand-ins for their HD DVD cousins in that regard.

Certainly true and a valid point. However 2 things I think we should keep in mind:

1) There tends to be a poit of diminishing returns for a given Q setting. That is if I set a target Q of .9 for a given video, and the encoder only needs 13 to reach it, then even if I were to set the encoder for a CBR of 16, that would dictate an increase of 4Mbps on the disc with no perceptible increase in quality. Conversely, just becuse an encode initially started at 18 does not mean that it could not be re-encoded at 12 without a perceptable decrease in quality. Tis the nature of an encoder having the bits it needs for a given target, and throwing more at it doesn't necessarily help.

Yeah, I agree with you there. Whatever the general case, I think that at least Batman Begins fit the profile of a specific case that would come in at a favorable PQ:bitrate ratio. As tiem goes on and more titles are documented, we'll get a better idea of just how broad these new encoder tools are in their power, and what titles benefit from what bitrates where.

This is espicially the case if an early encode was rushed (or the folks were not as familiar with all the knobs of a new codec) or was encoded with an earlier tool revision, or they were time-contrained and did not do scene re-encoding, etc...

2) If the target medium has additional room, there may be no incentive to spend additional time tuning the encode. Therefore extrapolating what has been done to what could be done is not necssarily a valid argument. Especially when the formats differ physically, and have very different histories and the folks goin in to it have different mindsets.

And I agree with you here as well. It's just as I previously said, I don't think that the BD encodes actually change anything over from the HD DVD encodes. When it comes to the PQ associated with these formats, it is all encode-side. A Warner VC-1 title is the same PQ on both HD DVD and Blu-ray; any differences are via the playback device itself.

As an example, one Sony representative made much ado about making sure they had a bit-rate meter in their BR player, saying something to the effect of "people are going to be impressed with some mind-blowing bitrates".

Well... much ado about nothing sure enough, but I have to tell you, when you are able to lok at these bitrates go by in real-time, it *is* awesome to watch. :)

Yeah, I've gone back and double-checked Ben's quotes, and I hope he'll clarify. For some reason I suspect he MAY be assuming those were HD DVD #'s, given the thread title, but I don't want to speak for him.

Well again, we are assuming they in fact *are* HD DVDs numbers, as we take the BD and HD DVD VC-1 titles to be identical.

I will say this though, if Ben claims we are seeing ~12Mb titles on HD DVD (which his comments have all been regarding in the past, hence my suspicion about his assumptions), and as Cj mentions here there are other HD DVD's released in the same range:

I mean I would personally want to know what those titles are (for curiosity), but yeah Batman Begins at least shows that its possible, and that it can still hold the line on a high PQ standard. So I don't dount that there are several other high-rating titles in a similar bit-range.

Then the fact that earlier encodes didn't hit this target on a different medium really are moot, no?

Well again, the VC-1 BD:HDDVD transitive property. :)

If a car can do 0-60 on the dragstrip in 3 seconds, then it can do it. It doesn't really matter if it can't do it on gravel.

So, it would seem that there are real-world examples of VC1 hitting ~12Mbps bit rates and producing excellent pictures.

Now, if we want to argue that the content of those encodes made them special cases, then I'm all for that, and perhaps Cj or others can specify what titles they are so we can argue about that for a while. :) But I think that using the BluRay #'s to determine VC1's ultimate capability is not going to really help get the truth, other than to say they were either poor, early, or "good enough to fit" encodes.

Yes, and I'm *all* for a discussion centering around the content itself to tell you the truth. That would be great, and I think it would be educational for everyone involved. As for Blu-ray #'s though, again I really do think they are just as useful to have as HD DVD #'s themselves; they're the same for all intents and purposes. The only difference of course, is that this provides a much narrower sample size. Now... if someone on the HD DVD side with their add-on drive was willing to get down and dirty with some HD DVD discs, it could greatly expand the knowledge base on this subject. I for one would love to know the King Kong average bitrate, among others. But until then, we just have to work within these limited bounds.

scaesare

12-21-06, 12:08 AM

It seems like you are ignoring that some (or maybe it was all) of those VC-1 releases for Blu-ray were encoded with HD DVD and 30GB in mind and then ported.

--Darin

Not really. My point is that if an encode exists that shows something can be done, then it can be done, no?

I had thought a number of my comments spoke to reasons why (toolset version, etc...) these encodes on a different medium are not necessarily representative of what VC1 can do today.

scaesare

12-21-06, 12:16 AM

Well, essentially the numbers gleaned were that in some scenes, even Batman Begins was indeed maxing out the HD DVD bitrate. Not much else to report beyond that unless 'fleshier' numbers come out on that tangent.

Well, but the VC-1 encodes for these cross-format titles are the same though, which is why they are useful. In fact, even more useful in that the way these titles are currently structured, it is not that difficult to islolate the data/bitrates of the video stream from the rest of it. It hasn't been explicitly stated as yet - and maybe we should ask for that - but I truly do believe that anything Warner is putting out on both BD and HD DVD using VC-1 is the exact same VC-1 encode. So I don't think it is inappropriate to use them as direct stand-ins for their HD DVD cousins in that regard.

Yeah, I agree with you there. Whatever the general case, I think that at least Batman Begins fit the profile of a specific case that would come in at a favorable PQ:bitrate ratio. As tiem goes on and more titles are documented, we'll get a better idea of just how broad these new encoder tools are in their power, and what titles benefit from what bitrates where.

And I agree with you here as well. It's just as I previously said, I don't think that the BD encodes actually change anything over from the HD DVD encodes. When it comes to the PQ associated with these formats, it is all encode-side. A Warner VC-1 title is the same PQ on both HD DVD and Blu-ray; any differences are via the playback device itself.

Well... much ado about nothing sure enough, but I have to tell you, when you are able to lok at these bitrates go by in real-time, it *is* awesome to watch. :)

Well again, we are assuming they in fact *are* HD DVDs numbers, as we take the BD and HD DVD VC-1 titles to be identical.

I mean I would personally want to know what those titles are (for curiosity), but yeah Batman Begins at least shows that its possible, and that it can still hold the line on a high PQ standard. So I don't dount that there are several other high-rating titles in a similar bit-range.

Well again, the VC-1 BD:HDDVD transitive property. :)

Yes, and I'm *all* for a discussion centering around the content itself to tell you the truth. That would be great, and I think it would be educational for everyone involved. As for Blu-ray #'s though, again I really do think they are just as useful to have as HD DVD #'s themselves; they're the same for all intents and purposes. The only difference of course, is that this provides a much narrower sample size. Now... if someone on the HD DVD side with their add-on drive was willing to get down and dirty with some HD DVD discs, it could greatly expand the knowledge base on this subject. I for one would love to know the King Kong average bitrate, among others. But until then, we just have to work within these limited bounds.

Cool... and I appreciate your work on this. Do we know for FACT that the encodes for these BR discs are the same as on HD DVD? I see you first say "Well, but the VC-1 encodes for these cross-format titles are the same though", and then in a later paragraph, "we are assuming they in fact *are* HD DVDs numbers".

And let me throw one stupid idea out there: It's been shown that at least one PS3 game padded out some files to almost 21 GB in size, simply to to move the data to the outer layer of the disc for faster load times. I'll assume that, given the nature of video not needing to be loaded any faster than it's display rate, that there's nothing like this going on...?

scaesare

12-21-06, 12:20 AM

Oh, and thanks for not dinging me on my not fixing up my math when I decided to revise my example #'s.

It's late here :o

xbdestroya

12-21-06, 12:28 AM

Cool... and I appreciate your work on this. Do we know for FACT that the encodes for these BR discs are the same as on HD DVD? I see you first say "Well, but the VC-1 encodes for these cross-format titles are the same though", and then in a later paragraph, "we are assuming they in fact *are* HD DVDs numbers".

No, we don't know for a fact (if you mean has it been explicitly stated). But... well, it's just sort of taken for granted that this is the case, just because it would make the most economic sense. It's been discussed many a time in these terms as well and not been denied by anyone that would be in a position to do so, so... well, anyway that's where that stands.

And let me throw one stupid idea out there: It's been shown that at least one PS3 game padded out some files to almost 21 GB in size, simply to to move the data to the outer layer of the disc for faster load times. I'll assume that, given the nature of video not needing to be loaded any faster than it's display rate, that there's nothing like this going on...?

Yeah, not here for the video, but I do know what you're refering to in terms of the PS3 games.

Oh, and thanks for not dinging me on my not fixing up my math when I decided to revise my example #'s.

It's late here :o

I know how you feel - I see you're in Ashburn; I'm in Arlington. :)

darinp2

12-21-06, 12:32 AM

Not really. My point is that if an encode exists that shows something can be done, then it can be done, no?Different point, but yes, if something shows that it can be done with a particular title, then it has shown that it can be done with that particular title :) (and probably ones with similar difficulty). I think we agree that 12.5Mbps average can be done with some titles with them still looking very good. We have well over 100 releases at this point. And it feels like the Microsoft camp is very careful to pick ones out on the low end when they give out numbers. I understand that, but people should understand when they are using words like "a bunch" it might only be 10% of titles. Just like when they tell us that some took only a couple of days to encode, that again might be less than 10% of titles (since I don't seem to get answers when they do that and I ask how many, it is hard to say what the real percentages are).

As I said much earlier, if they want to give out numbers, I would like to hear the numbers for "Miami Vice". That is a pretty recent title (probably used a fairly recent version of the encoder) and one where Universal dropped the TrueHD track, and it is pretty grainy at times. When I and a friend get some time, maybe we will look at the running bitrate for some scenes for that one.

If we are asking about the range of things that can be done, then the difficult ones are relevant. Which is why I've asked about "Saving Private Ryan" with lossless audio and IME. Parts of "Miami Vice" seem like they are grainy kind of like "Saving Private Ryan", but MV has the advantage of being 2.35:1 (so always has black areas that are easy to compress). SPR was film and I believe MV was video, so that is another difference though.

One thing I doubt we will see is New Line releasing a branched version of the LOTR trilogy, like the last release of those they did on DVD that used seamless branching. At least I hope they won't do that if they are targetting HD DVD.

--Darin

benwaggoner

12-21-06, 01:13 AM

Ben has said he can't speak to it - and it would be fine if he stopped there. But before understanding how the numbers were reached, he tries to imply that they are bogus. When in fact, they end up pretty close to the truth. It's just that reflexsive sort of defense that relies on their 'street cred' alone that I have a problem with. When valid questions are raised, take the time to post valid answers. Saying 'I can't say anything' is fine... just so long as innuendos aren't inserted along with it.

I don't think they were bogus - I'm sure it was a good faith effort that just wasn't completely accurate.

But I've got a database of the real numbers, and they simply didn't match :). I share the info I can, and why I can't speak about individual titles, like Cjplay, I'm happy to give the hints I can get away with.

xbdestroya

12-21-06, 01:20 AM

I don't think they were bogus - I'm sure it was a good faith effort that just wasn't completely accurate.

But I've got a database of the real numbers, and they simply didn't match :). I share the info I can, and why I can't speak about individual titles, like Cjplay, I'm happy to give the hints I can get away with.

Well... verrrrryyyy forthcoming of you there Ben. ;) So hint me this - would you agree that Cj's recommended filtering of the derived numbers would get us next to the actual averages?

By the way while you're here, Scaesare raised a question that might as well be decisively addressed. Cross-platform BD titles using VC-1 for the most part use a direct port of the HD DVD encode, correct? And please Ben, just say 'yes,' 'no,' or 'I can't say one way or the other' on that last question rather than throwing up a bunch of sand... because the last thing anyone needs is more confusion where there was none before. Even though I know you guys love it... por favor humor me on this. :)

PS - For you to have responded to the quote of mine you did would likely indicate that you read the posts preceeding it. Well, some of those posts asked direct questions (not all of them were mine), and I can't help but notice the lack of response to those questions thus far. Of note, Benes pointing out that that in some cases, you said the estimates were in fact *low.* How is that even possible? The files can't be outputting at a greater bitrate than they are outputting after all... it's science.

Wendell R. Breland

12-21-06, 02:26 AM

But I've got a database of the real numbers, and they simply didn't match :). Well 1,000,000 does not match 1,000,001 but most of us would call it the same in the context they are being used.

tsd2005

12-21-06, 02:27 AM

CJ, BEN:

Can you guys work on a sticky that opposing fanboys can't touch on facts to do away with their misinformation campaigns. I mean seriously we've had one fake poster already .... I'm of the mind about a couple others.

30GB is enough for a nice movie, trailer, audio tracks. That's still good in PQ.

An extra disc can be used for some extras or.... the internet.

This thread got hijacked for no real purpose but to ... i dont know defeat the message?

Grubert

12-21-06, 04:24 AM

30GB is enough for a nice movie, trailer, audio tracks. That's still good in PQ.

An extra disc can be used for some extras or.... the internet.

This thread got hijacked for no real purpose but to ... i dont know defeat the message?

This thread is not being hijacked at all.

We are all trying to answer your question by using science. Allow me to repost your OP:

It is my assumption from talking to people who make the VC-1 encodings that you can fit:

A nearly four hour movie
Dolby True HD
Audio Commentary track
Trailer

all on one 30GB HD-DVD disc very easily, meaning that compression is not a problem and a very beautiful print can be possible.

BTW I notice your requirements have loosened somewhat from the beginning. What was then "a nearly four hour movie" is now "a nice movie". Dolby True HD and audio commentary have been replaced by a generic "audio tracks". But nevermind.

The question you asked is whether a nearly four-hour movie, with Dolby True HD, audio commentary track and trailer, will fit a 30GB HD DVD disc. We are trying to answer your question, and you complain about that? :confused:

To the point:

Giving a plausible answer to that question requires looking at the history of VC-1 titles released and examining their relative file sizes and video bitrates ballpark estimates.

Unfortunately, there are no tools to know the video bitrate on HD DVD titles. benes has a PC with a BD drive and has had the initiative of working backwards from the file sizes, subtracting the audio bitrates (which are a given for lossy codecs). So now we have an estimate for VC-1 video bitrates on titles released on BD.

cjplay (who we must thank for giving a straight answer - a rarity for insiders) has said in the past that Warner is using the same VC-1 encodes for both formats, and that when BD25 wasn't enough, bonus features were recompressed, or the title was put on the backburner for BD50.

So, if we have an estimate for the video bitrate of a VC-1 title on BD, it'll apply to to the HD DVD release of the same title.

So, let's look again at benes' list and look for a modern action movie that they have been able to encode at a low bitrate. Superman Returns stands out. (Yes, I know it was shot digital...)

The movie file size is: 20,626,040,832 (and note no lossless track)

The movie length is: 2h 34' 12"

So, if Superman Returns were to run four hours, the movie file size (estimate) would be:

20626040832 * 14400 / 9252 =

32,102,787,287

Grubert

12-21-06, 08:30 AM

And as luck would have it, forum member bfdtv has posted an excellent post (http://www.avsforum.com/avs-vb/showthread.php?p=9229342&&#post9229342) on the BD section regarding bandwidth and space taken up by lossless audio.

As said above, [lossless tracks] are identical in quality, but not bandwidth or space.

[All modern films are mastered with 24/48 audio. Anything less isn't true master audio.]

On BD25 titles, Sony and Disney use LPCM with 16/48 fidelity because there isn't space for higher fidelity without seriously compromising the picture quality. On some titles, even this 16/48 LCPM results in compromised picture quality with MPEG-2. Over in the Industry Insiders thread, a representative from Sony has said they will not use 24/48 LPCM on BD25 titles due to the space required, but may offer 20/48 LPCM at some point in the future, possibly after the move to AVC (MPEG-4).

On its BD25 titles, FOX delivers true 24/48 master audio using DTS-HD MA; they can do this because 24/48 LPCM packed (i.e. "zipped") with lossless DTS-HD requires less disk space than the 16/48 LPCM on Sony and Disney titles. Moreover, FOX takes things one step further -- when they feel that dedicating several gigabytes to the DTS-HD MA lossless track will compromise picture quality, they use AVC (MPEG-4) instead of MPEG-2 or a BD50 disk. As a result, FOX has many of the better-looking titles on Blu-ray; once we get DTS-HD MA decoding, they'll be the best sounding Blu-ray titles too.

Refer to the table below:

LPCM
2 hours @ 5.1 16/48 LPCM = 4.14 Gbytes @ 4.6 Mbps CBR (what Sony uses)
2 hours @ 5.1 24/48 LPCM = 6.21 Gbytes @ 6.9 Mbps CBR
2 hours @ 7.1 24/48 LPCM = 8.28 Gbytes @ 9.2 Mbps CBR
2 hours @ 7.1 24/96 LPCM = 16.56 Gbytes @ 18.4 Mbps CBR

Dolby TrueHD
2 hours @ 5.1 16/48 TrueHD = 1.26 Gbytes @ 1.4 Mbps ABR with 3+Mbps peaks
2 hours @ 5.1 24/48 TrueHD = 3.06 Gbytes @ 3.4 Mbps ABR with 5+ Mbps peaks
2 hours @ 7.1 24/48 TrueHD = 4.23 Gbytes @ 4.7 Mbps ABR with 6.6+ Mbps peaks

DTS-HD MA bitrates are comparable to TrueHD bitrates. CBR stands for constant bit rate and ABR stands for average bit rate.

As is obvious from the above, you aren't likely to see 24/48 (or better) 7.1 LPCM on a 25Gb or 30Gb disk, but the storage required by the same track with lossless TrueHD or DTS-HD MA packing is much more reasonable.

gljvd

12-21-06, 09:07 AM

Well here is the thing .

When you have space to spare you don't make things smaller. the titles we are comparing all fit on the discs they were released on. Superman Returns fits on 1 30 gig hd-dvd and 1 50 gig bluray disc . Why bother trying to make it smaller ? I dont see the point . If you get an encode the works without much hassle.

I find that when you have to fit something in a certian amount of time you will actually work to make it work well.

Batman begins is a film that is close to 12Mbp/s for the video. It didn't need to be as it should have fit on a disc with higher bit rates adn it looks great.

xradman

12-21-06, 09:30 AM

It seems to me that people are making too much out of maximum capacity or bitrate for HD DVD vs BD. If BD had an order of magnitude greater capacity, then maybe it might matter. But when you consider any sort of lossy qualitative algorithm, difference between 30 and 50 will be transparent to just about everyone. There are few simple experiment that most of can perform with a PC.

Take a large high quality bitmap color image. In PhotoShop or any comparable imaging program, save as JPG file with varying levels of high quality so that the compressed file is transparent to the original bitmap image. I would bet that one could not tell the difference between the file that is half the size of the other.

This is because beyond a certain point, you reach a point of diminishing return with these lossy algorithm. Theoretical benefit of 50 vs. 30 will only be apparent if you are operating at ultra low quality level where there is more linear relationship between file size and quality.

scaesare

12-21-06, 09:59 AM

No, we don't know for a fact (if you mean has it been explicitly stated). But... well, it's just sort of taken for granted that this is the case, just because it would make the most economic sense. It's been discussed many a time in these terms as well and not been denied by anyone that would be in a position to do so, so... well, anyway that's where that stands.

Cool. Somehow I simply have a hard time believing Ben is flat out lying (or misleading to such a huge extent as to be nit-picking over the admited imprecesion in bene's methodology). So I would hope that we can get confirmation of what the actual HD DVD stream rates are.... jus tto make sure there is not some other factor we aren't aware of.

I know how you feel - I see you're in Ashburn; I'm in Arlington. :)

You're farther East than I am... it was even LATER there! :D

scaesare

12-21-06, 10:11 AM

Different point, but yes, if something shows that it can be done with a particular title, then it has shown that it can be done with that particular title :) (and probably ones with similar difficulty). I think we agree that 12.5Mbps average can be done with some titles with them still looking very good. We have well over 100 releases at this point. And it feels like the Microsoft camp is very careful to pick ones out on the low end when they give out numbers. I understand that, but people should understand when they are using words like "a bunch" it might only be 10% of titles. Just like when they tell us that some took only a couple of days to encode, that again might be less than 10% of titles (since I don't seem to get answers when they do that and I ask how many, it is hard to say what the real percentages are).

As I said much earlier, if they want to give out numbers, I would like to hear the numbers for "Miami Vice". That is a pretty recent title (probably used a fairly recent version of the encoder) and one where Universal dropped the TrueHD track, and it is pretty grainy at times. When I and a friend get some time, maybe we will look at the running bitrate for some scenes for that one.

If we are asking about the range of things that can be done, then the difficult ones are relevant. Which is why I've asked about "Saving Private Ryan" with lossless audio and IME. Parts of "Miami Vice" seem like they are grainy kind of like "Saving Private Ryan", but MV has the advantage of being 2.35:1 (so always has black areas that are easy to compress). SPR was film and I believe MV was video, so that is another difference though.

One thing I doubt we will see is New Line releasing a branched version of the LOTR trilogy, like the last release of those they did on DVD that used seamless branching. At least I hope they won't do that if they are targetting HD DVD.

--Darin

Yeah, I'm all for arguing about what specific content and types will encode at what rates.

Given the 100+ title catalog, it's probably reasonable to say that ony the minority of them are using the latest tools/methods. And Cj, whom everybody here seems to feel is factual in his posts, is mentioning 5 or so titles in the ~12Mb arena. So, as a general statement, I don't think it's too unreasonable to say that ABR's of 12 are indeed possible with the latest tools for some content. And 5 out of the latest releases with the latest encoder may indeed be "a bunch".

It's probably not fair to say that we'll hit ~12 for all content. It's also probably not necessarily true that some bitrates that are 60-80% higher on another format using an unknown vintage of tool are an absolute indicator that these target bitrates aren't possible. Indeed we know they are.

So, I think we are at the point where we can say "It appears that 12Mb ABR is possible for some content... what prevents us from reaching that for other content?". Undoubtedly the quality of the master will have impact... so if you are suggesting they'll have a harder time with a grainy master as MV apparently have, you'll get no argument from me. But I don't think anybody has ever stated that the master wasn't a (or THE) significant contibutor to the overall ability to hit a given bitrate.

Cjplay

12-21-06, 12:40 PM

Cool... and I appreciate your work on this. Do we know for FACT that the encodes for these BR discs are the same as on HD DVD?

For Warner titles, yes.

Edit: We remove the 3:2 Pulldown flagging first, but that's only metadata.

Cjplay.

Cjplay

12-21-06, 12:45 PM

Given the 100+ title catalog, it's probably reasonable to say that ony the minority of them are using the latest tools/methods. And Cj, whom everybody here seems to feel is factual in his posts, is mentioning 5 or so titles in the ~12Mb arena. So, as a general statement, I don't think it's too unreasonable to say that ABR's of 12 are indeed possible with the latest tools for some content. And 5 out of the latest releases with the latest encoder may indeed be "a bunch".

5 is actually a low (read: safe) number. And they were with VC-1 versions about 2-5 generations older than current tools. So we've been doing this ABR level for awhile.

Cjplay.

scaesare

12-21-06, 01:22 PM

5 is actually a low (read: safe) number. And they were with VC-1 versions about 2-5 generations older than current tools. So we've been doing this ABR level for awhile.

Cjplay.

Thanks CJ.

Now, let's get to speculating about what those titles are so we can quickly get to disagreeing about whether they are representative of what could be done with LotR! ;-)

Cjplay

12-21-06, 01:50 PM

Thanks CJ.

Now, let's get to speculating about what those titles are so we can quickly get to disagreeing about whether they are representative of what could be done with LotR! ;-)

Yeah. Figure on the IME/THD discs to be the lowest, but not all of them out there are the IME/THD's and at least one was a DnD with the SD. Hints over...

Cjplay.

tsd2005

12-21-06, 03:26 PM

This thread is not being hijacked at all.

We are all trying to answer your question by using science. Allow me to repost your OP:

Hhmmm.. lets see I ask in my original OP for insiders to participate not you, nor any BD fan, in fact I state if BD wants to join in on this conversation for their own format they should do so.

What I've learned is BD fanboys continue to hijack threads for fear of their misinformation being debunked.

It's stupid.

This site has totally lost it's ability to handle this section. The mods don't have the time to stop people from hijacking threads or attacking other people. Attack the contents not the people!

Basically I've had one BD fanboy decide this all meant LOTR, where it doesn't at all. I hate the movie and I hate Peter on a very personal level, we don't get along. I'm so HAPPY Sam is taking over that franchise... but that is all beside the point.

I've had other BD fanboys decide that BD discs and their content would be identical to the HD-DVD. They decide to put their 2 cents in.

I've also got people like you who decide you know as well.

If you're not an insider, don't post a reply made out for them and fill it up with garbage just so they can't see it.

The question of

IS 30GB ENOUGH?

is important, and they would know.

xbdestroya

12-21-06, 04:31 PM

DaveKennett

12-21-06, 04:36 PM

avshaman

12-21-06, 05:42 PM

This is where you show your true bias. For everyone that posts that they had issues with their 360 there is more than a 10 to 1 ratio that will post that they didn't. And that is over estimating the percentage of failed 360s. You post like you are the majority when, in fact you are part of the vast minority of people that had issues.

I can understand your disappointment, but I am one of the lucky majority that had zero issues with my 360 from launch day. In fact I never had any issues with any console I've ever owned. I also never had any issues with my 360 drive and it works like a charm to this day. I also beleive that I will have zreo issues with my PS3.

I will never talk down a company because I had an issue with one of their products. That product is not the sum of their parts. I research and take each product on it's own merit. Just because I dislike the Sony Viao, doesn't mean I hate their displays...I own one.

Just remember that you are posting from your prospective and that is not the prospective of the average consumer, because they never had the anger inducing issues that you have experienced with certain products.

I am sorry you misunderstood my post. I was not trying to bash the Xbox 360, I was merely empathizing with the previous poster who also had trouble with his Xbox 360. I really like my Xbox 360. In fact, in the early days of it's life I was constantly defending it on forums from all the negativity and bashing that was going on at the time. I did all this despite the fact that I had so much trouble with getting a working system because I know that all complex electronics products will have a certain number of defects and I knew that the failure rate was being exaggerated. And most importantly, I like the Xbox 360 and believe it had a lot of potential. I just happen to also feel that way about the PS3.

Now the Xbox 360 has gained popularity and its now cool to bash Sony and the PS3. I am just being honest in telling people on this forum that I have not had any problems with my PS3. I did not mean to imply that the PS3 is free from defects/failures or that the Xbox 360 is prone to failures in all cases.
However, from what I have seen reading various gaming sites and forums, and despite how popular it is for people to be negative toward the PS3, the general consensus seems to be that the PS3 has had fewer outright failures than the Xbox 360 did.

Of course the 360 launched with more than twice the number of units available but that does not change the overall percentages of failures/defects.

Incidentally, my original PS2 bought in October 2000 has not given me a single problem while my Xbox gives me disc read errors all the time. But do I form anti-Microsoft opinions based on this? No, because I know many people have perfectly working Xbox consoles and I just work around it. Both console makers have their strengths and weaknesses and I enjoy both systems.

And I never "talked down a company" as you implied I did. Microsoft and Sony are what they are. They have both done things that I wasn't quite happy with but I don't hold a grudge against their products as some here seem to do.

Cjplay

12-21-06, 05:56 PM

We all know that some MPEG2 encoders are better than others - and things keep inproving. I expect VC-1 and AVC to keep improving as well. In a few years, we'll be using MPEG 9, and a SL DVD-R is all we'll need.
Dave

I'll back this up. HD VOD is either 19Mbps muxed or 15Mbps muxed depending on your cable system. We've run clips at 14.1 that had me thinking twice about the MPEG-2/VC-1 debate. Problem is the same clip in VC-1 today does it at 11Mbps. It took 2 days on a dual-proc machine for the MPEG-2, though, but at 1920x1080, that's not terrible. MPEG-2 can do a good job with the right level of hard work.

Cjplay.

avshaman

12-21-06, 06:15 PM

I also don't want to pay an arm and a leg for something.

I can look foward into the future and if I see sony leading something I rather not have it happen .

I can settle for lower space with advance codecs if its priced properly. I don't believe bluray is worth the price

The first comment is BS and has been shown to be BS so many times it is not worth arguing anymore. Fact: I bought my A1 originally for $500 and I bought my PS3 for $600 (could have paid $500 but I opted for more features not necessary for Blu-ray playback). And despite the comparable prices my PS3 has better performance and does a heck of a lot more than my A1 to boot.
Furthermore, when I go to buy movies for both formats I don't feel like I am paying an "arm and a leg" for BDs. In fact, prices for movies in either format seem about equal as far as I can tell. So for me, and for many people like me, we don't see this huge price discrepancy that you fanatics are always referring to. Blu-ray costs "an arm and a leg" but HD DVD is "a great value" is the mantra that is repeated ad nauseum as if that will somehow make it true. It doesn't.

As to your second point, I guess you would rather see Microsoft with it's track record rule the market? I'm not saying that one should not support HD DVD because it is backed by Microsoft. In fact just the opposite. In my opinion that kind of thinking is frivolous. If I wanted to hold silly grudges I wouldn't be buying from any of these companies.

Your third comment is just a repeat of the same old propaganda that you spewed in your first comment.

avshaman

12-21-06, 06:42 PM

Oh, I forgot...your opinions are considered fact and all other opinions are considered one sided and biased...

With that type of attitude, it is no surprise my ignore list just keeps getting bigger...one at a time.

No Anamorphiac, I never said that my opinions are fact. I am wrong all the time and I freely admit to it. However, I don't believe my opinions or yours are worth much unless we can demonstrate some reasoning as to why we believe as we do. The reason I criticized the anti-Sony comments that I have seen is that they are based on a petty and hipocritical stance toward Sony. In other words, Sony is condemned for certain actions, actions of which Microsoft is similarly guilty, yet Microsoft is not likewise condemned. That is what I meant by biased and hipocritical.

If I am wrong, show me I am wrong. Debate the point with me. But apparently I can't expect that of you. I guess it is just easier to go through life ignoring those who you don't agree with.

hdkhang

12-21-06, 08:07 PM

People are quick to jump to conclusions, but very slow to accept corrections...

Can people just accept the fact that as long as a movie fits on a disc that the studios will not spend more time (i.e. money) in trying to get it to use less space?

That just because those shorter movies can get away with higher bitrates does not equate to their not being able to be made to look good at lower rates.

Can we also agree that for important titles like LOTR, HP, Matrix trilogy etc. that these studios will not release a product that won't wow us like KK did? If it requires a future gen VC1 codec, then they'll wait... it's people like CJplay that give me confidence that I am right on this one.

Cheers...
Duy-Khang Hoang

avshaman

12-21-06, 08:27 PM

BIG ED

12-21-06, 09:16 PM

avshaman

12-21-06, 10:17 PM

hdkhang

12-21-06, 10:22 PM

Good points. But the salient point as I see it is that we have two comparably priced formats that seem to offer similar PQ and SQ yet one format offers a significant increase in capacity and bitrate. Why do HD DVD fanatics try make that advantage out to be a bad thing?
I ask you Duy-Khang Hoang why BD would not likewise benefit from improvement to encoding technology? Why is it that any time the subject of improved encoding technology is brought up it is always positioned as a unique benefit of HD DVD?

I ask you avshaman where in my post did I mention anything in favour of or against either format?

The point I was refuting was that people are suggesting that based on the bitrates they have gleaned, that VC1 is taxing out HD-DVD, when in actual fact, it's more a case of the studios intend to hit a certain size and if they hit it with the quality they are satisfied with, they aren't going to spend more time and money making adjustments. All these doomsday predictions of movie X is only Y minutes long and takes up Z gigabytes means all other movies of longer duration will suffer due to lower bitrates is just a show of ignorance.

Especially since this argument has been hashed here many times before.

In any case, these cases of benefits here and disadvantages there are all useless in one direction, the only direction worth watching is the space occupied by present BD only studios who designed their video encodes around BD limitations, if these same studios one day decide to release on HD-DVD, and capacity/peak bitrates exceed those of HD-DVD, it would only then be pertinent to compare the new encode vs the old. In these cases, the arguments would go along the line of if an improved encoder can close the gap or if the gap is there simply due to there being enough space and hence no further need to tweak. For now, neutral studios design for HD-DVD limitations in terms of video, and no studios that have released significant content in BD has announced plans to release in HD-DVD... so what's all the fuss?

Cheers...
Duy-Khang Hoang

benwaggoner

12-21-06, 11:17 PM

Well... verrrrryyyy forthcoming of you there Ben. ;) So hint me this - would you agree that Cj's recommended filtering of the derived numbers would get us next to the actual averages?
Absolutely. We both WANT you to know the real numbers, we just aren't allowed to tell you. I hope someone releases a simple parser that'd give the real numbers - we'd have much more useful conversations then.

And people would be much more comfortable with the LOTR:ROTK:EE if they had access to the full data.

By the way while you're here, Scaesare raised a question that might as well be decisively addressed. Cross-platform BD titles using VC-1 for the most part use a direct port of the HD DVD encode, correct? And please Ben, just say 'yes,' 'no,' or 'I can't say one way or the other' on that last question rather than throwing up a bunch of sand... because the last thing anyone needs is more confusion where there was none before. Even though I know you guys love it... por favor humor me on this. :)
How's "most do" for you? I don't know the percentage off the top of my head.

BIG ED

12-22-06, 12:24 AM

To EVERYONE,

Don't mind me! :o

I just "FREAK OUT" that I can't go on a HD DVD thread w/o getting "rayed"!
BD backers say the same thing happens to them on Blu-ray threads.
So I guess both sides, unfortunately, are equally to blame. :eek:

I don't like going to BD areas because of the negative impression I've gotten from day one. However, when I go to a BD forum, I just try to get info & get out clean.

Well, now I'm just as bad as Blu's (general term for the BD lurkers, nothing against BD fans), cause I'm spiting out venom against anything Sony or Blu. :mad:

Just wish the combo's would come & this war would be done with.
(Wish I actually believed that! :rolleyes: )

Best wishes to ALL in '07!
Hope we ALL get the HD Disc experience we are looking (ha!) for. :)
E.

xbdestroya

12-22-06, 12:36 AM

Absolutely. We both WANT you to know the real numbers, we just aren't allowed to tell you. I hope someone releases a simple parser that'd give the real numbers - we'd have much more useful conversations then.

And people would be much more comfortable with the LOTR:ROTK:EE if they had access to the full data.

I think we're close enough that we could have useful conversations if we wanted to Ben. ;)

How's "most do" for you? I don't know the percentage off the top of my head.

Well Cj beat you to it, but thanks anyway - that indeed would have been a valid answer. And of course, always appreciate the continuing dialogue on the subject. :)

darinp2

12-22-06, 01:05 AM

Batman begins is a film that is close to 12Mbp/s for the video. It didn't need to be as it should have fit on a disc with higher bit rates adn it looks great.One thing to keep in mind is that because of other things they included, BB's video maxed out at basically 19Mbps for HD DVD and would have even if they split it across two discs. So, even if they had made it constant bitrate and it had plenty of room, it wouldn't have gone above 19 (other than very short things using the buffer). Make it variable bit rate (as these are) and even if 25% of the movie stayed pegged at the peak bit rate, the average should have been lower than 19Mbps. So, in this case a fair part of the final average bit rate was based on the peak bitrate for HD DVD and the IME and audio they included, even if it didn't need to be that low. Amir mentioned a rule of thumb of PBRs at 50% higher than ABRs and that rule of thumb would result in a 19Mbps PBR title ending up at about 12.7Mbps ABR. And 12.7Mbps is a number that some would then tell you was 12Mbps.

--Darin

benwaggoner

12-22-06, 01:05 AM

I think we're close enough that we could have useful conversations if we wanted to Ben. ;)
Well, yes, I think we've adequately determined that 30 GB is enough for a spectacular presentation of LOTR:ROTK:EE on a single side of a single disc, right?

And that there's still some work to be done on figuring out current discs real-world video ABR numbers.

And that both CJplay and I are hungry for someone to release a tool that reveals the real numbers, since we think there will be good news in there for the above scenario.

xbdestroya

12-22-06, 01:13 AM

Well, yes, I think we've adequately determined that 30 GB is enough for a spectacular presentation of LOTR:ROTK:EE on a single side of a single disc, right?

And that there's still some work to be done on figuring out current discs real-world video ABR numbers.

And that both CJplay and I are hungry for someone to release a tool that reveals the real numbers, since we think there will be good news in there for the above scenario.

Ben don't turn into Amir here on me... ;)

I meant rather, that enough is known about the bitrates of certain titles, that perhaps we can start digging into what scenes VC-1 requires higher bitrates for relative to certain 'easier' scenes, and what those spreads look like relative to other codecs.

I'm talking about a purely technical discussion on the strengths of the VC-1 encode algorithms, something that hopefully should elicit both excitement from you and a desire to answer candidly considering your role in it all. I mean I honestly find it all very interesting, so long as talk remains technically focused.

As for Lord of the Rings... well, no I'm not convinced of 'spectacular' yet, but I'm not expecting sh*t either. I think that describes me as open to whatever gets crafted. :) (Though my own ideal preference stands on the record.)

And we can just use the straight and easy method Benes has used up until now; the variance is within ~1-2 Mbps and the results scale consistent across titles, even if shifted by said amount in bitrate.

raaj

12-22-06, 01:25 AM

...

As for Lord of the Rings... well, no I'm not convinced of 'spectacular' yet, but I'm not expecting sh*t either. I think that describes me as open to whatever gets crafted. :) (Though my own ideal preference stands on the record.)

And we can just use the straight and easy method Benes has used up until now; the variance is within ~1-2 Mbps and the results scale consistent across titles, even if shifted by said amount in bitrate.

Xbd,

Why don't you wait till you see the actual finished encode before declaring it to be "spectacular" or a wee bit less than so? What is the use of the endless arguments about this hypothetical point when probably none outside of MS and WB/NLC have seen this encode? Hell, do we even know if the final encode on BR would end up being VC-1 or MPEG-2 or AVC? Do you know if a <15Mbps VC-1 encode of this movie suffers obviously in comparison to a higher bit rate encode afforded to Blu-Ray? Without seeing the actual encodes and comparing both HD-DVD and Blu-Ray versions with possibly different bitrates, you and some others are repeatedly beating the dead horse that "HD-DVD might not have enough bandwidth to handle this movie on one 30GB disc".

This is turning out to be "arguing just for the heck of it" - when the subject of discussion does not even exist in anything other than "proof of concept" encode form.

benwaggoner

12-22-06, 01:37 AM

I meant rather, that enough is known about the bitrates of certain titles, that perhaps we can start digging into what scenes VC-1 requires higher bitrates for relative to certain 'easier' scenes, and what those spreads look like relative to other codecs.
High detail and high motion are generally the things that take relatively more bits. Foliage blowing in the wind, is a classic hard scene to encode.

LOTR is easier than some, since it's 2.35:1, so you get nice big black bars that take essentially no bandwidth, letting you spend more bits per pixel on the image area.

LOTR being new and not particularly grainy; there arent' any scenes I can think of off hand I'd be that worried about.

I'm talking about a purely technical discussion on the strengths of the VC-1 encode algorithms, something that hopefully should elicit both excitement from you and a desire to answer candidly considering your role in it all. I mean I honestly find it all very interesting, so long as talk remains technically focused.
It's kind of a hard thing to describe if you haven't seen the tool. You basically do an initial 2-pass encode, and then watch it to see what tweaking might be needed. You get a graph of the PQ (basically a measure of how lossy each frame was encoded), which is a good initial indicator of places to double check on quality.

If you find a scene or a shot that has some issues (like the texture in shadows looking a little off), you'd set that segment up for reencoding with some different paramters tuned to its particulars. You might try a few variants to hit the combo you need.

Once the operator things he/she's encoded it to their satisfaction, it gets sent off to a 3rd party who hasn't been staring at it for a review. They'll watch it, and the send a log sheet back with any points of concern. Then the compressionist either addresses those issues of concern, or determines they were present in the master.

The workflow is pretty similar for any codec. The biggest difference in VC-1 is the depth to which you can tweak things in the 3rd pass. Amir recently mentioned being able to change encoding parameters for a particular area in a particular frame, for example. And there are other broader functions, for example setting modes for how bits are distributed that apply to only certain degrees of texturing. The net effect is that, compared to other encoders, it's often possible to fix an issue on a scene without increasing the bitrate - most tools just let you add bits in the 3rd pass, but have dozens of things that can be done.

Bear in mind this is a product that gets used by full-time staff, so it's not something you'd be able to see a screen shot of and go "ah!" And it takes a few weeks to really get up to speed on how to adjust parameters.

And we can just use the straight and easy method Benes has used up until now; the variance is within ~1-2 Mbps and the results scale consistent across titles, even if shifted by said amount in bitrate.
But they're not consistant - some were too high, some were too low.

darinp2

12-22-06, 01:55 AM

LOTR is easier than some, since it's 2.35:1, so you get nice big black bars that take essentially no bandwidth, letting you spend more bits per pixel on the image area.Interesting you say that after your boss worked hard to try to keep people from understanding this after using one of those things along the lines of, "I'm the expert here and you're not." Should be considered one of the lower moments for Microsoft on this site IMO. Glad to see you are willing to admit the above (even though in this case it was helping your argument here).
LOTR being new and not particularly grainy; there arent' any scenes I can think of off hand I'd be that worried about.My mind goes to some of the battle scenes with lots of characters in them, but not sure if they would be the toughest. I think you are right about you guys being able to do these LOTR titles very well though, largely for the reasons you've stated.

You've mentioned 12.5Mbps average for LOTR. What kind of peak bitrate would you be comfortable with for it?

So, how about something like "Saving Private Ryan"? There's that 1.85:1 vs 2.35:1 thing where somebody trying to keep people from the truth could try to keep them from understanding why that matters. What kind of bitrates would you be comfortable with for SPR?

--Darin

xbdestroya

12-22-06, 02:14 AM

@Raaj: I'm not arguing anyone, and never was in the first place honestly. This discussion I am involved in is all about understanding VC-1. Originally from a standpoint of my septicism, then to wondering and wanting answers as they related to Benes bitrate finds (which still are remarkably accurate it should be noted), to finally just feeling that a lot of insight into this codec has been learned and just honestly wanting to go further on it while we're in the spirit of learning. Lord of the Rings is just a proxy case in all this - it's a movie a lot of us care about deeply, and at the same time it's a good test case for the 4hr+ film lengths we're talking about here. There's more to the BD vs HDDVD position on this movie than just PQ alone obviously, and I sated those earlier on, but ultimately this discussion has become about VC-1 through several pages of exploration into the subject.

And since this is the most discussed codec on the forum - for many reasons - frankly I think any and all should be glad it's been pursued as it has in this thread. Without questions, there can be no answers afterall. And thanks go to Cj and Ben for their input on these matters; it's been some of the most informative talk here in a long long time.

xbdestroya

12-22-06, 02:24 AM

High detail and high motion are generally the things that take relatively more bits. Foliage blowing in the wind, is a classic hard scene to encode.

Well right, I mean those are obvious cases. But I guess what I meant by 'spread' is, does VC-1 need to spike higher as a ratio of its baseline bitrate in order to handle these scenes relative to say, MPEG-2, or does it scale similarly up to handle them? I ask because as I mentioned, when watching Superman Returns the bitrate could fluctuate wildly, whereas it seems more level throughout in the other codec encodes I have. (Though that's something I'll do some further testing on.)

LOTR is easier than some, since it's 2.35:1, so you get nice big black bars that take essentially no bandwidth, letting you spend more bits per pixel on the image area.

That's a good point.

It's kind of a hard thing to describe if you haven't seen the tool. You basically do an initial 2-pass encode, and then watch it to see what tweaking might be needed. You get a graph of the PQ (basically a measure of how lossy each frame was encoded), which is a good initial indicator of places to double check on quality.

If you find a scene or a shot that has some issues (like the texture in shadows looking a little off), you'd set that segment up for reencoding with some different paramters tuned to its particulars. You might try a few variants to hit the combo you need.

Once the operator things he/she's encoded it to their satisfaction, it gets sent off to a 3rd party who hasn't been staring at it for a review. They'll watch it, and the send a log sheet back with any points of concern. Then the compressionist either addresses those issues of concern, or determines they were present in the master.

The workflow is pretty similar for any codec. The biggest difference in VC-1 is the depth to which you can tweak things in the 3rd pass. Amir recently mentioned being able to change encoding parameters for a particular area in a particular frame, for example. And there are other broader functions, for example setting modes for how bits are distributed that apply to only certain degrees of texturing. The net effect is that, compared to other encoders, it's often possible to fix an issue on a scene without increasing the bitrate - most tools just let you add bits in the 3rd pass, but have dozens of things that can be done.

And it's that third pass tweaking that caught my attention when Amir alluded to it (as I mentioned in that thread as well); the sub-scene optimization as it were. So basically what I'm understanding is that in scenes that require a higher bitrate, or with some issue in the base encode, you are able to either a) increase the general bitrate but keep it low in segments of the scene that don't need the breathing room, or b) keep the bitrate steady but increase the bandwidth alloted to certain problem areas in said scenes if that'll solve the issue.

Is that more or less the deal? From a software perspective... or rather a functionality perspective... it's an impressive feature. But I imagine this is a very labor intensive process as well. I'll give you your due though on maximizing efficiency however. :)

Bear in mind this is a product that gets used by full-time staff, so it's not something you'd be able to see a screen shot of and go "ah!" And it takes a few weeks to really get up to speed on how to adjust parameters.

But they're not consistant - some were too high, some were too low.

This too low thing again... too high I understand, but how is too low possible? I mean that *was* the real-world readout, y'know?

tsd2005

12-22-06, 02:55 AM

The irony here TSD is that this thread has probably been *the* most educational thread as it applies to VC-1 in the last six months or so... and not only have you had nothing to do with contributing to that in a positive way, you just want to take it back to a grandstanding level.

Hey, by all means...

I don't mind CJ and Ben pointing out why VC1 works, etc.

I have a problem with the Grubert, etc. posting BS that has nothing to do with the thread. This has NOTHING to do with BD.

It has everything to do with HD-DVD and it's size.

Is 30GB enough?

So far everything said by the insiders is that it IS enough.

Grubert

12-22-06, 03:54 AM

I have a problem with the Grubert, etc. posting BS that has nothing to do with the thread. This has NOTHING to do with BD.

Thanks for the constructive criticism. :)

I only mentioned BD inasmuch as it provided us a window into VC-1 stats. Nothing more. If the HD DVD section had a comprehensive video bitrate thread, that'd be what I'd use. It's not personal.

It has everything to do with HD-DVD and it's size.

Is 30GB enough?

So far everything said by the insiders is that it IS enough.

King Kong disagrees. ;)

Supermans

12-22-06, 06:22 AM

Let's get this straight Blu-bloods. Compare BD 50 titles to their HD DVD counterparts:
SR: BD 1 disc & HD DVD 1 disc.
Sopranos: BD 4 discs & HD DVD 4 discs.
(Any others?)
So where is your "significant increase in capacity"???

And this thread, dang it, is not about: "HD DVD fanatics try make that advantage out to be a bad thing?". Did you read the thread title???

It's about HD DVD 30GB discs & only about HD DVD 30 GB discs!!!

Where do you see in the title or the OP, where it states ANYTHING about "HD DVD fanatics" stating anything, good or bad, about BD 50???
Where was that posted???

Your making stuff up, to put people down!
That's sick. :(

I bought Superman Returns on Blu-Ray and wonder myself how much better the title could have been had they re-encoded it to better use the extra space BD has available.

This thread is intersting to read. I can't agree that this thread has nothing to do with Blu-Ray since this entire thread is about space available on a disc vs how much compression and what codec a video gets.

We are at the point that movie's that are shorter in lengh and using VC-1 should look a whole lot better than much longer movie's being placed on the same size disc using the same codec. A 4 hour movie cannot compare in quality to an hour movie on the same amount of space unless the people doing the encoding purposely make the shorter film not look as good as it should or can.

If it was up to me, I'd use at the very least 2 HD-DVD discs that would equal 60GB of space for one Lord of the Rings movie. It will be a mess if you try and fit 4 hours in 30GB of space in comparison to what it can look like otherwise. I wouldn't trust someone who tells me they can fit 4 hours of movie and some extra's on a 30GB disc and make it look its best...

DigitalfreakNYC

12-22-06, 07:41 AM

King Kong disagrees. ;)

How so?

Cjplay

12-22-06, 07:52 AM

Cjplay

12-22-06, 08:01 AM

And that there's still some work to be done on figuring out current discs real-world video ABR numbers.

And that both CJplay and I are hungry for someone to release a tool that reveals the real numbers, since we think there will be good news in there for the above scenario.

Yes. Someone PLEASE come up with a decent way to figure this out!!!

Using Cyberlink's player, turn on the bitrate display. Record the audio track's bitrate (for DD+ and DD). Subtitles are ~32Kbps ABR, so just subtract the # of subs times 32Kbps. DTS Core is 1.5Mbps. The 2 lossless formats are normally in the 1.5/2.0Mbps ABR range if the source was 16/48. At this point assume Warner's have been all along 16/48. I'm chomping at the bit here, gang. You've got to do the homework...

Cjplay.

Grubert

12-22-06, 08:15 AM

How so?

Because right now there is an American version of KK with three languages and bonus materials, and a Europe/Japan version with six languages and no bonus materials. Same with Miami Vice for PiP, according to announcements.

However, the European release of F&F: Tokyo Drift gets all the languages and all the bonuses (including PiP). Same with Bourne Supremacy. So it isn't that they just were unwilling or unable to translate the PiP.

ILJG

12-22-06, 08:15 AM

I bought Superman Returns on Blu-Ray and wonder myself how much better the title could have been had they re-encoded it to better use the extra space BD has available.

This thread is intersting to read. I can't agree that this thread has nothing to do with Blu-Ray since this entire thread is about space available on a disc vs how much compression and what codec a video gets.

We are at the point that movie's that are shorter in lengh and using VC-1 should look a whole lot better than much longer movie's being placed on the same size disc using the same codec. A 4 hour movie cannot compare in quality to an hour movie on the same amount of space unless the people doing the encoding purposely make the shorter film not look as good as it should or can.

If it was up to me, I'd use at the very least 2 HD-DVD discs that would equal 60GB of space for one Lord of the Rings movie. It will be a mess if you try and fit 4 hours in 30GB of space in comparison to what it can look like otherwise. I wouldn't trust someone who tells me they can fit 4 hours of movie and some extra's on a 30GB disc and make it look its best...

More available space and bandwidth doesn't necessarily translate into better PQ when it's being "taken advantage of." Click is the perfect example. A shorter movie than say King Kong, HP4, and Batman Begins...yet these movies have excellent PQ, while Click...well....

You can choose not to believe people who do this for a living when they know how certain scenes and types of techniques can result in certain types of encodes, that's certainly your perogative. I certainly give them more credibility than those who don't do it for a living.

Rob Zuber

12-22-06, 10:41 AM

However, for some facts, feel free to pose questions. I hate propaganda and will give as close to the straight dirt as I can.I'm still waiting for the "straight dirt". Which Warner movies do you believe would not properly fit on a 30 GB HD-DVD (assuming you want good PQ)?

I'd be reluctant to put Ben Hur on one disc or even Last Emporer and GWTWind.What are the others? For completeness, which Warner movies do you believe would not properly fit on a 50 GB BD?

Cjplay

12-22-06, 12:13 PM

I'm still waiting for the "straight dirt". Which Warner movies do you believe would not properly fit on a 30 GB HD-DVD (assuming you want good PQ)?

What are the others? For completeness, which Warner movies do you believe would not properly fit on a 50 GB BD?

I meant those for HD30. I can't give a completel list as I don't know every single one. I would have a hard time with Giant, Cleopatra, and many other long and old films over ~3-3.5 hours as well. BD50? My only question there is "what exists over ~5 hours besides TV shows and documentaries?" BD50 has a space advantage if all the authoring tools could handle VC-1 and AVC. BTW, "The Cure for Insomia" is one title I don't see anyone releasing on anything smaller than 15 BD50's...

Cjplay.

majortom

12-22-06, 12:59 PM

I meant those for HD30. I can't give a completel list as I don't know every single one. I would have a hard time with Giant, Cleopatra, and many other long and old films over ~3-3.5 hours as well.

Just to be clear, you think there are a number of films that will not fit on an HD DVD 30GB disk.

BD50? My only question there is "what exists over ~5 hours besides TV shows and documentaries?" BD50 has a space advantage if all the authoring tools could handle VC-1 and AVC. BTW, "The Cure for Insomia" is one title I don't see anyone releasing on anything smaller than 15 BD50's...

Cjplay.

At 87 hours, The Cure for Insomia is not something I am likely to own, but thanks for letting me know how much space I should reserve on my shelves if I decide to buy it. :-)

Cjplay

12-22-06, 02:35 PM

Just to be clear, you think there are a number of films that will not fit on an HD DVD 30GB disk.

Yes, but the overwhelming percentage (about 98% of theatrically-released material) that will (We're averaging ~110 minutes by my book, IMDB needs stats like this!) makes HD DVD sufficient for many applications. Not that I'm downplaying that BD50 advantage, just that HD DVD does fit about 4-4.2 hours of a well-mastered and low grain title while it holds about 3-3.5 hours of severely grainy and not well-mastered 1080 material. I won't mention titles, but I'm sure the lower Tier 3's and all 4's know who they are... ;)

My search for the longest theatrical film lead me there, so I though it good humor to mention it.

Cjplay.

bobgpsr

12-22-06, 03:44 PM

I'm chomping at the bit here, gang. You've got to do the homework... I keep getting around 14 Mbps ABR for both BB and KK. Is there something about the mux that I'm missing? I thought the answer should be 12.5 Mbps?

Edited per feedback numbers:
Using 360 add on connected to a PC
BB
2 big EVO's at 19,857,584,128 bytes total * 8 bits/byte / 140 min / 60 sec = 18.912 Mbps
- 3 subs 96,000
- 5.1 THD 1,800,000
- 5.1 DD+ English 640,000
- 5.1 DD+ French 640,000
- IME 2,000,000
--------------------------------
gives 13.736 Mbps

KK
2 big EVO's at 28,912,744,448 total * 8 bits/byte / 188 min / 60 sec = 20.505 Mbps
- 3 subs 96,000
- 5.1 DD+ Spanish 768,000
- 5.1 DD+ English 1,500,000
- 5.1 DD+ French 768,000
- U-Control/with PiP 2,000,000

--------------------------------
gives 15.373 Mbps

tsd2005

12-22-06, 04:06 PM

Thanks for the constructive criticism. :)

I only mentioned BD inasmuch as it provided us a window into VC-1 stats. Nothing more. If the HD DVD section had a comprehensive video bitrate thread, that'd be what I'd use. It's not personal.

King Kong disagrees. ;)

They decided to not use TrueHD for reasons besides fitting everything on a disc. At least that is what I was told.

My assumption has always been that only extremely long old movies like the ones CJ listed: Giant, Cleopatra, etc. would have a problem with HD30. At that point a flipper disc or 2 discs is not a problem for me. Intermissions are just fine for such long movies, and in fact the movies themselves had intermissions when played at the Theater.

CJ has now stated what should be obvious to everyone:

30GB is more than enough for 98% of all titles to be treated RIGHT.

Some people don't understand that higher bitrates don't always mean higher PQ, it just means a larger size.

darinp2

12-22-06, 04:07 PM

I keep getting 14 Mbps ABR for both BB and KK. Is there something about the mux that I'm missing?Are the secondary languages on KK 1.5Mbps? I didn't think Universal was doing that for secondaries.

--Darin

majortom

12-22-06, 04:08 PM

Yes, but the overwhelming percentage (about 98% of theatrically-released material) that will (We're averaging ~110 minutes by my book, IMDB needs stats like this!) makes HD DVD sufficient for many applications.

I have a hard time arguing for a new format that has hit its limits at launch. If either format is successful, it will be around for at least 10 years. Cost of discs today is irrelevant over the life of either format. Physical limits of these formats do matter for their life span. When DVD was released, dual layer discs were too expensive, and it took several years before they became the norm for shipping movies. Now it adds a few cents to a disc and is common.

My search for the longest theatrical film lead me there, so I though it good humor to mention it.

Cjplay.

Add Andy Warhol's Empire to your list (485 minutes).

/carmi

namechamps

12-22-06, 06:17 PM

I keep getting 14 Mbps ABR for both BB and KK. Is there something about the mux that I'm missing? I thought the answer should be 12.5 Mbps?

Using 360 add on connected to a PC
BB
2 big EVO's at 19,392,172 KB total * 8 bits/byte / 140 min / 60 sec = 18.468 Mbps
- 3 subs 96,000
- 5.1 THD 1,500,000
- 5.1 DD+ English 640,000
- 5.1 DD+ French 640,000
- IME 1,500,000
--------------------------------
gives 14.09 Mbps

KK
2 big EVO's at 28,235,102 KB total * 8 bits/byte / 188 min / 60 sec = 20.02 Mbps
- 3 subs 96,000
- 5.1 DD+ Spanish 1,500,000
- 5.1 DD+ English 1,500,000
- 5.1 DD+ French 1,500,000
- U-Control/with PiP 1,500,000
--------------------------------
gives 13.92 Mbps

Where did you get your # for THD, IME and U-Control?

My understanding is that THD 5.1 16/48 is more like 1.8mbps and that is an average. Some soundtracks maybe more or less it depends on the "compressability" of that sound track. So BB THD track maybe anywhere from 1.4 to 2.1mbps.

IME and U-Control are likely more than 1.5mpbs.

If those # are wrong then that may be why you are getting the wrong bitrate.
If for BB the THD was 1.8mbps and the IME was 2.6 then you would end up with right around 12.5.

namechamps

12-22-06, 06:45 PM

He is using the numbers provided by cjplay. But I believe TrueHD uses a variable bitrate and IME probably does also. Its important that people realize that we cannot get exact numbers right now. Its just a good estimate. Maybe +- 2Mbps from the real thing. Usually our calculations will be a little Higher than the real thing and that is certainly the case here.

Do you have a link to the post? Just wondering what the context was.

I would have thought IME would take more than 1.5mbps as it contains video and audio.

TrueHD is variable but it will have an ABR so the 1.8 would be the ABR likely with peaks in the 3.0mbps range. LPCM for 6 channel 16 48 is 4.5mbps (6x16x48000).
TrueHD according to Dolby's whitepaper has an average compression ratio of 2.5:1

I understand it is just an estimate but we can strive to make it as accurate as possible. The more information we gain the more accurate we can get. Maybe someday someone will crack AACS and we can pull individual streams to get the exact stats.

xbdestroya

12-22-06, 07:17 PM

Do you have a link to the post? Just wondering what the context was.

I would have thought IME would take more than 1.5mbps as it contains video and audio.

TrueHD is variable but it will have an ABR so the 1.8 would be the ABR likely with peaks in the 3.0mbps range. LPCM for 6 channel 16 48 is 4.5mbps (6x16x48000).
TrueHD according to Dolby's whitepaper has an average compression ratio of 2.5:1

I understand it is just an estimate but we can strive to make it as accurate as possible. The more information we gain the more accurate we can get. Maybe someday someone will crack AACS and we can pull individual streams to get the exact stats.

It's earlier in the thread; look around where Cj first starts posting to see the insights he gives into the methods already in play here. And yes - absolutely - thank you someone else for seeing the potential of this thread so far. This is the most information gained about VC-1 in a long time, and some of the most significant to boot because it's real world numbers. This thread is what the 'science' in AVS is all about.

EDIT: Namechamps, check out posts #78 and #84 and I think that will give you the insights you're looking for here.

bobgpsr

12-22-06, 08:26 PM

Where did you get your # for THD, IME and U-Control?
...
If for BB the THD was 1.8mbps and the IME was 2.6 then you would end up with right around 12.5.
Yes the numbers from Cjplay: http://www.avsforum.com/avs-vb/showthread.php?p=9230016&&#post9230016

But I'd be happy to use 2.6 Mbps for IME and 1.8 Mbps for 5.1 16bit/48kHz Dolby TrueHD.

benwaggoner

12-22-06, 11:17 PM

Interesting you say that after your boss worked hard to try to keep people from understanding this after using one of those things along the lines of, "I'm the expert here and you're not." Should be considered one of the lower moments for Microsoft on this site IMO. Glad to see you are willing to admit the above (even though in this case it was helping your argument here).
I have no idea what you're talking about.

My mind goes to some of the battle scenes with lots of characters in them, but not sure if they would be the toughest. I think you are right about you guys being able to do these LOTR titles very well though, largely for the reasons you've stated.
The big battle in the rain in Two Towers would be the hardest thing I can think of in the trilogy. Lots of characters, dark, rain (rain!), and lots of hair on the warriors.

You've mentioned 12.5Mbps average for LOTR. What kind of peak bitrate would you be comfortable with for it?
Hard to say without having spent much time with the HD master. Low twenties, if they started today? Lower as time goes on. LOTR is probably easier than average content overall, since it's recent and presumably with a really high quality master.

So, how about something like "Saving Private Ryan"? There's that 1.85:1 vs 2.35:1 thing where somebody trying to keep people from the truth could try to keep them from understanding why that matters. What kind of bitrates would you be comfortable with for SPR?
I'm going to be less inclined to answer your questions if you continue that tone, FYI.

I've never seen that master, so I wouldn't want to speculate much. The opening scene would want a high ABR, since it used a high shutter speed, so not much motion blur and a lot of grain. But I don't remember the bulk of the picture being particularly complex.

benwaggoner

12-22-06, 11:23 PM

Well right, I mean those are obvious cases. But I guess what I meant by 'spread' is, does VC-1 need to spike higher as a ratio of its baseline bitrate in order to handle these scenes relative to say, MPEG-2, or does it scale similarly up to handle them? I ask because as I mentioned, when watching Superman Returns the bitrate could fluctuate wildly, whereas it seems more level throughout in the other codec encodes I have. (Though that's something I'll do some further testing on.)
Ah, gotcha.

If anything, VC-1 will need less spread, given a number of the new techniques we've implemented as previously discussed. One thing a "leading compressionist" told us about VC-1 v. MPEG-2 was that he had to get used to having 3rd pass not be mainly about adding more bits, but finding techniques that improved quality while potentially even lowering bitrate.

And it's that third pass tweaking that caught my attention when Amir alluded to it (as I mentioned in that thread as well); the sub-scene optimization as it were. So basically what I'm understanding is that in scenes that require a higher bitrate, or with some issue in the base encode, you are able to either a) increase the general bitrate but keep it low in segments of the scene that don't need the breathing room, or b) keep the bitrate steady but increase the bandwidth alloted to certain problem areas in said scenes if that'll solve the issue.
Either of the above, or sometimes just changing encoding mode based on the kind of content. For a trivial example, disabling B-frames in fades. Segment reencoding is typcially applied per-shot, not per scene. As the tool and codec have evolved, the average duration of a segment that gets adjusted has gone way, way down, as well as the number needed per title.

Is that more or less the deal? From a software perspective... or rather a functionality perspective... it's an impressive feature. But I imagine this is a very labor intensive process as well. I'll give you your due though on maximizing efficiency however. :)
It's a lot less labor intensive than it was, and is going to be even less in the future. Some of that is in training, and a lot of it is in workflow (making the decision making and management process simpler).

This too low thing again... too high I understand, but how is too low possible? I mean that *was* the real-world readout, y'know?
Yeah, I don't get it either.

Like I said, I just compared that list to the list of titles I knew, and noticed some too high, others too low.

bobgpsr

12-22-06, 11:40 PM

Like I said, I just compared that list to the list of titles I knew, and noticed some too high, others too low.
How about mine? Should I use 1.8 Mbps for BB's TrueHD and 2.6 Mbps for IME?
Does the mux used at the time of BB's authoring require allocating those amounts? And that might change with some type of "dynamux"?

scaesare

12-23-06, 12:43 AM

I bought Superman Returns on Blu-Ray and wonder myself how much better the title could have been had they re-encoded it to better use the extra space BD has available.

...

A 4 hour movie cannot compare in quality to an hour movie on the same amount of space unless the people doing the encoding purposely make the shorter film not look as good as it should or can.

...

Are you saying that because of particular issues you saw with SR?

Because as an absolute statement, it's not necessarily true.

If a movie only needs 12/19 to get to the "no perceptible encoding artifacts" point, and you have the space to to allow 4 hours of that, then giving a 1 hour movie of similar content 20/30 is only going to buy you wasted bits.

(I'm ignoring audio for the sake of the illustration)

benwaggoner

12-23-06, 12:43 AM

How about mine? Should I use 1.8 Mbps for BB's TrueHD and 2.6 Mbps for IME?
Sorry, I don't have numbers for audio to even hint about.

Does the mux used at the time of BB's authoring require allocating those amounts? And that might change with some type of "dynamux"?
Again, the data I've got just lists the video ABR. The mux won't change that, but can change the total file size.

scaesare

12-23-06, 12:57 AM

xbdestroya

12-23-06, 01:09 AM

Of course, the other perspective is that the spikes are similar to MPEG2's, and that the efficiency of VC1 allows the bit rate in the "valleys" to drop much more rapidly than MPEG2 would.

;)

Although I appreciate the logic you're using in the contra, I really was just wondering what Ben's insights into bitrate spreads were. Remember I've actually watched these dips and spikes in front of my eyes on a bitrate readout for both codecs, and the question stems from genuine curiosity. Superman Returns was taking me from 6Mbps to 30Mbps... so it can be pronounced.

There's no question though that VC-1 has the greater ability to achieve a certain PQ standard relative to MPEG-2 within a certan bitrate constraint. If there's nothing else it seems I'm readily granting in this whole thing, believe me when I say I give that aspect it's due immediately and 100%. :)

Cjplay

12-23-06, 01:11 AM

I have a hard time arguing for a new format that has hit its limits at launch. If either format is successful, it will be around for at least 10 years. Cost of discs today is irrelevant over the life of either format. Physical limits of these formats do matter for their life span. When DVD was released, dual layer discs were too expensive, and it took several years before they became the norm for shipping movies. Now it adds a few cents to a disc and is common.

Well it almost doubles the glass master cost which matters to smaller producers... DVD was maxed out each time a possible config was made available. MPEG-2 got better as did AC3, so more fit on the disc. DVD added dual layer and again it is easily be maxed out. BD can easily be maxed out just as easily with the higher BR's. "Future-proofing" to me (later sentences in this post are others concern, not mine) isn't about space and bandwidth, but use of the current technology which I feel both are doing to a degree.

In the end, even WHV realizes the benefits of BD50 when packaging concerns are factored in. Alien Quadrilogy on 8 DVD discs or 4 for the HD format? I'm guessing HD DVD could do it in 5, but BD50 would be 3 and Fox would have a smaller package in that regard costing less to manufacture and ship. Sony knows this and the longer this war drags out, the better chance they have that packaging concerns will get higher and higher.

And oh I've gone cross-eyed... ;)

Add Andy Warhol's Empire to your list (485 minutes).

/carmi

Okay.

Cjplay

12-23-06, 01:15 AM

I would have thought IME would take more than 1.5mbps as it contains video and audio.

OMG I am so embarrassed. :o Add 192Kbps for an IME stereo track and any subs that include the IME add the same 32kbps.

Cjplay.

darinp2

12-23-06, 04:43 AM

In the end, even WHV realizes the benefits of BD50 when packaging concerns are factored in. Alien Quadrilogy on 8 DVD discs or 4 for the HD format? I'm guessing HD DVD could do it in 5, but BD50 would be 3 and Fox would have a smaller package in that regard costing less to manufacture and ship. Sony knows this and the longer this war drags out, the better chance they have that packaging concerns will get higher and higher.Most of us have been hearing about how more discs help sell products and so I found this interesting. From what you are saying it sounds like the lower packaging costs might be more important than being able to tell people they are getting more discs. How do you think this relates to 1 disc vs 2 discs for single movie releases? I've been thinking that as extras go to HD (which they should) instead of the SD that most of them seem to be now even on these HD formats, that studios might make 2 HD30s common for HD DVD. I'm sure 2 disc sets can help sell movies, but wonder about oversaturation if a high percentage of releases become dual disc sets. Do you have any thoughts on that part of this? Do you think studios would want to fit the extras onto a single HD30 for most releases in a year or two for HD DVD (assuming it is still supported of course) and retain 2 disc sets for special things, or make 2 disc sets very common?

--Darin

Vern Dias

12-23-06, 09:18 AM

Just a thought:

Most bitrate indicators available for public consumption average/update over a finite time period. Typically 1 second or so. While these might be useful for indicating a rough value or a reasonable trend, they certainly do not give the viewer an accurate value for the instantaneous minimum or maximum of the bitrate on the media.

So I would take what any bit rate indicator tells me with at least a teaspoonful of salt..... and I would give a lot more credibility to those who actually master the titles or build the tools used to master the titles.

Vern

xbdestroya

12-23-06, 09:40 AM

Most bitrate indicators available for public consumption average/update over a finite time period. Typically 1 second or so. While these might be useful for indicating a rough value or a reasonable trend, they certainly do not give the viewer an accurate value for the instantaneous minimum or maximum of the bitrate on the media.

Right. But those bitrate meters aren't going to indicate a 30Mbps bitrate unless there was a stretch at 30Mbps, know what I mean?

bobgpsr

12-23-06, 10:24 AM

The mux won't change that, but can change the total file size.But since I start with the total file size it does affect the derived ABR calculation. Do we need to use peak bit rate for IME (say 3 Mbps) and TrueHD (another 3 Mbps) for our calculations with titles that did not use some type of dynamic mux'ing? Since that is what must be allocated? Hey Cjplay, fell free to chime in. :)

benwaggoner

12-23-06, 02:18 PM

But since I start with the total file size it does affect the derived ABR calculation. Do we need to use peak bit rate for IME (say 3 Mbps) and TrueHD (another 3 Mbps) for our calculations with titles that did not use some type of dynamic mux'ing? Since that is what must be allocated? Hey Cjplay, fell free to chime in. :)
Yeah, given the other VBR elements in the stream, trying to figure out video ABR but just subtracting everything else from the file size isn't going to get you very precise.

Someone's just going to need to make a EVOB parser and measure individual streams.

bobgpsr

12-23-06, 02:47 PM

Someone's just going to need to make a EVOB parser and measure individual streams.Yes, that sounds like an interesting programming project. I wonder just when the AACS encryption kicks in. :p

Thanks Ben for answering as much as you can. :)

benwaggoner

12-23-06, 03:26 PM

Yes, that sounds like an interesting programming project. I wonder just when the AACS encryption kicks in. :p

Thanks Ben for answering as much as you can. :)
My understanding is that you wouldn't need to decrypt AACS in order to count packets for each type of track...

namechamps

12-23-06, 04:30 PM

Yes the numbers from Cjplay: http://www.avsforum.com/avs-vb/showthread.php?p=9230016&&#post9230016

But I'd be happy to use 2.6 Mbps for IME and 1.8 Mbps for 5.1 16bit/48kHz Dolby TrueHD.

Yeah oops. I hadn't seen where cjplay mention those #. Anyways my point wasn't that my # were correct rather that just a small change in the estimated bitrate of the other streams could affect the video stream. Right now trying to get an exact # is rather difficult since there are numerous other streams and at best we have a good guess as to their bitrate.

namechamps

12-23-06, 04:36 PM

OMG I am so embarrassed. :o Add 192Kbps for an IME stereo track and any subs that include the IME add the same 32kbps.

Cjplay.

Sorry cjplay i didn't realize the 1.5mbps rate came from you. I just thought it was a number someone pulled out of the air. I still am suprised that IME can be done with so little bits. So less than 2mbps for audio, video, and subs? Is that an average encode. For some reason I thought i recalled a 4mbps being tossed around in regards to IME. If IME can be done w/ so little bits well that is good news for HD DVD because it isn't as bit starved as I thought.

darinp2

12-23-06, 05:01 PM

But since I start with the total file size it does affect the derived ABR calculation. Do we need to use peak bit rate for IME (say 3 Mbps) and TrueHD (another 3 Mbps) for our calculations with titles that did not use some type of dynamic mux'ing?It depends on whether you are trying to calculate ABR or PBR for the video. For ABR you shouldn't have to use the peaks of the IME and TrueHD.

--Darin

scaesare

12-24-06, 01:14 PM

Sorry cjplay i didn't realize the 1.5mbps rate came from you. I just thought it was a number someone pulled out of the air. I still am suprised that IME can be done with so little bits. So less than 2mbps for audio, video, and subs? Is that an average encode. For some reason I thought i recalled a 4mbps being tossed around in regards to IME. If IME can be done w/ so little bits well that is good news for HD DVD because it isn't as bit starved as I thought.

IIRC the cap on PiP rates is 2Mb for VC1 & AVC and 4Mb for MPEG2.

-sc

benwaggoner

12-24-06, 01:34 PM

IIRC the cap on PiP rates is 2Mb for VC1 & AVC and 4Mb for MPEG2.

And that's the maximum allowed PiP rate - one can use a lower peak than that. For lots of PiP content, that'd work just fine (e.g. talking heads). Bear in mind VC-1 is an enhanced implementation of what is also a web codec - we do great in < 1 Mbps.

namechamps

12-24-06, 02:22 PM

IIRC the cap on PiP rates is 2Mb for VC1 & AVC and 4Mb for MPEG2.

-sc

Sounds good to me. I learned something today. Makes my belief that HD DVD has enough bandwidth and cap that much stronger.

Cjplay

12-24-06, 10:30 PM

Most of us have been hearing about how more discs help sell products and so I found this interesting. From what you are saying it sounds like the lower packaging costs might be more important than being able to tell people they are getting more discs. How do you think this relates to 1 disc vs 2 discs for single movie releases? I've been thinking that as extras go to HD (which they should) instead of the SD that most of them seem to be now even on these HD formats, that studios might make 2 HD30s common for HD DVD. I'm sure 2 disc sets can help sell movies, but wonder about oversaturation if a high percentage of releases become dual disc sets. Do you have any thoughts on that part of this? Do you think studios would want to fit the extras onto a single HD30 for most releases in a year or two for HD DVD (assuming it is still supported of course) and retain 2 disc sets for special things, or make 2 disc sets very common?

--Darin

Even today, 2-disc sets are held back except for those distributors that want to make it feel like an extra value. Ask yourself why POTC2 had 2 versions and charged 10 dollars more for the SE? However, several other Disney and other studio releases even for the brand new material have only 1 disc. Pixar titles tend to have 2 as do the special animated releases, but I've seen countless others that are singles. Warner's own Space Jam (great soundtrack, film did well for the studio) was a single-disc. Space Jam went 2-disc recently, though, and consumers bought it because of the perceived value of more extras from the film. Consumers still see it as perceived extra value. Of course they see movies that way as being over and above CD's and multi-track sets from downloading vendors. Packaging costs are not the primary concern, but a valid one and constantly taken into account.

I feel 2-disc editions will not saturate the market, but if the studio produced extras for a title while it was shooting and didn't pursue them for the DVD if it went badly in the box office, with enough space on the HD format the extra could still make it on the disc.

TV shows fitting onto less discs makes a better thing for the consumer as well as the distributor. Watch one season of Friends on one disc or six? Think about it. Also, multi-feature sets are the biggest perceived value and they're getting those onto one disc, not just those weird multi-packs from Best Buy. Go to Wal-Mart and see what I mean. You don't see 2 discs in those, do you? Like I said, packaging can be a conern, but not the primary one.

Cjplay.

Cjplay

12-24-06, 10:33 PM

But since I start with the total file size it does affect the derived ABR calculation. Do we need to use peak bit rate for IME (say 3 Mbps) and TrueHD (another 3 Mbps) for our calculations with titles that did not use some type of dynamic mux'ing? Since that is what must be allocated? Hey Cjplay, fell free to chime in. :)

You want ABR to prove a trend. PBR's are only the concern when you're worried about why something isn't in the mux and the disc isn't full.

Cjplay.

Cjplay

12-24-06, 10:35 PM

Yeah, given the other VBR elements in the stream, trying to figure out video ABR but just subtracting everything else from the file size isn't going to get you very precise.

Someone's just going to need to make a EVOB parser and measure individual streams.

It'll get you within 1Mbps. It's a big start.

Cjplay.

Cjplay

12-24-06, 10:36 PM

Sorry cjplay i didn't realize the 1.5mbps rate came from you. I just thought it was a number someone pulled out of the air. I still am suprised that IME can be done with so little bits. So less than 2mbps for audio, video, and subs? Is that an average encode. For some reason I thought i recalled a 4mbps being tossed around in regards to IME. If IME can be done w/ so little bits well that is good news for HD DVD because it isn't as bit starved as I thought.

4Mbps is the PBR for AVC and VC-1 encodes. MPEG-2's PBR is 6Mbps.

Cjplay.

scaesare

12-25-06, 01:20 AM

4Mbps is the PBR for AVC and VC-1 encodes. MPEG-2's PBR is 6Mbps.

Cjplay.

I stand corrected.

(Although Ben... you commented on my 2/4 asertion without correcting me... are you in agreement that CJ's 4/6 is the correct set of PBR's for PiP?)

darinp2

12-25-06, 01:37 AM

Even today, 2-disc sets are held back except for those distributors that want to make it feel like an extra value. Ask yourself why POTC2 had 2 versions and charged 10 dollars more for the SE? However, several other Disney and other studio releases even for the brand new material have only 1 disc. Pixar titles tend to have 2 as do the special animated releases, but I've seen countless others that are singles. Warner's own Space Jam (great soundtrack, film did well for the studio) was a single-disc. Space Jam went 2-disc recently, though, and consumers bought it because of the perceived value of more extras from the film. Consumers still see it as perceived extra value. Of course they see movies that way as being over and above CD's and multi-track sets from downloading vendors. Packaging costs are not the primary concern, but a valid one and constantly taken into account.
...
TV shows fitting onto less discs makes a better thing for the consumer as well as the distributor. Watch one season of Friends on one disc or six? Think about it.Thanks for your take. I was thinking about it today and for Netflix customers this is even more important since it is a pain to wait for the next discs. For Netflix themselves it seems like a smaller number of discs saves them postage and handling costs (people keep the discs longer), but too low and they are sending out discs that they paid a lot for and are only charging $6/month for.
Also, multi-feature sets are the biggest perceived value and they're getting those onto one disc, not just those weird multi-packs from Best Buy. Go to Wal-Mart and see what I mean. You don't see 2 discs in those, do you? Like I said, packaging can be a conern, but not the primary one.I hadn't really noticed them before, but was in Fry's yesterday and noticed they they have discs with 3 movies on them. Some of them are actor or actress themed (like a disc with 3 movies with Drew Barrymore) and others are things like all 3 versions of "Poison Ivy".

--Darin

benwaggoner

12-25-06, 04:33 AM

I stand corrected.

(Although Ben... you commented on my 2/4 asertion without correcting me... are you in agreement that CJ's 4/6 is the correct set of PBR's for PiP?)
Sorry, yes, he was correct.

Although with VC-1, you don't need to get near 4 Mbps PBR if it's optimally encoded, for most typical PIP content.

Talkstr8t

12-26-06, 02:38 PM

I would be a fervent Blu-ray support if:

The players were more full featured. Networking in only 2 BD products?? Lack of mandatory audio and video features means a wide dynamic range of possible hardware configs. In spite of "mandatory TrueHD support", HD-DVD situation is ever more diverse than Blu-ray. Yes, your player might decode TrueHD, but it may well be output in lossless 2.0 or as Dolby Digital 5.1. Meanwhile every Blu-ray player will output TrueHD as Dolby Digital 5.1.
Reliance on MPEG2. Sony has tried to flog crusty technology for too long while pricing it like it is the new hotness. MPEG2 come on.Blu-ray has no reliance on MPEG2, some studios do. Oh, and how do you explain that many recent titles in MPEG2 have equivalent PQ to the same title on HD-DVD in VC-1?
Java for interactivity?? That's so 90's ...high efficiency codecs and XML based interactivity are distinctly more in tune with what's going on todayHuh. The US broadcast market has standardized on Java (OCAP / ACAP) and completely shunned XML (HTML). XML has it's virtues, but as a content type, not a content platform.

- Talk

Rio

12-26-06, 03:45 PM

Because right now there is an American version of KK with three languages and bonus materials, and a Europe/Japan version with six languages and no bonus materials. Same with Miami Vice for PiP, according to announcements. Just for your information, total used capacity of the KK US version is 29,481,049,776 bytes and the file size of feature only is 28,912,744,448 bytes (including all audio tracks as well as U-Control stuff). As to audio bit rate, KK has English DD+ 1.5Mbps, French DD+ 768kbps, Spanish DD+ 768kbps and commentary DD+ 384kbps.

I feel that assuming main video's ABR by subtracting PiP's "ABR" is not a suitable way, since PiP is not staying through out the entire movie. It just shows up in some part, with short duration.

bobgpsr

12-28-06, 06:15 PM

Just for your information, total used capacity of the KK US version is 29,481,049,776 bytes and the file size of feature only is 28,912,744,448 bytes (including all audio tracks as well as U-Control stuff). As to audio bit rate, KK has English DD+ 1.5Mbps, French DD+ 768kbps, Spanish DD+ 768kbps and commentary DD+ 384kbps.No matter what I do I keep seeing two big EVO files for the main feature (13,884,074 KB + 14,352,028 KB = 28,235,102,000 bytes total). Where is your 28,912,744,448 bytes coming from? What files?

I re-did my calcs to use your reduced bitrates for the French and Spanish DD+ tracks. Where does your info on the audio bitrates come from?

My updated numbers for BB (13.29 Mbps) and KK (14.98 Mbps) (now only using 2 Mbps for IME/U-Control):
http://www.avsforum.com/avs-vb/showthread.php?p=9247400&&#post9247400

Rio

12-28-06, 09:21 PM

Numbers came from WinDVD HD DVD on Toshiba Qosmio laptop. It shows audio bitrate. As to file size, I selected 2 EVO files and showed properties for 2 files and read byte size (instead of KB size). Your number is measured in KB instead of bytes.

FEATURE_1.EVO : 13,884,074KB (14,217,291,776 bytes)
FEATURE_2.EVO : 14,351,028KB (14,695,452,672 bytes)
Total: 28,235,102KB (28,912,744,448 bytes)

bobgpsr

12-29-06, 12:59 PM

I selected 2 EVO files and showed properties for 2 files and read byte size (instead of KB size). Your number is measured in KB instead of bytes. Thank you Rio. You are correct. I edited my earlier post accordingly.

BuGsArEtAsTy

12-29-06, 01:08 PM

LOTR is easier than some, since it's 2.35:1, so you get nice big black bars that take essentially no bandwidth, letting you spend more bits per pixel on the image area.
The black bars for LOTR are hard-coded into the stream?

benwaggoner

12-29-06, 02:46 PM

The black bars for LOTR are hard-coded into the stream?
Yes, all HD streams for both formats are encoded as 16:9. So any letterboxing or other matting is added to the frame before the encode.

BuGsArEtAsTy

01-27-07, 01:45 AM

Yes, all HD streams for both formats are encoded as 16:9. So any letterboxing or other matting is added to the frame before the encode.
Any specific reason for this? It seems like a waste of space (although with pure black bars the wasted space would be low).

Anyone with an HTPC or something equivalent: How does it work for PC software players? Do they strip off the black bars, or do they always have a 16:9 aspect ratio, for both movies and extras? If the latter, that'd be annoying. It's great for my TV though, since my TV locks into widescreen mode regardless if the video is widescreen or not.

It seems fortunate that LOTR is 2.35:1. That would make your claimed 12 Mbps reference quality video (http://www.avsforum.com/avs-vb/showthread.php?p=9589524&&#post9589524) all the more achievable. Using that 12 Mbps figure:

12 Mbps VC-1
No TrueHD
3.072 Mbps for audio tracks (1.536 + 0.768 + 0.768 Mbps)
0.256 Mbps commentary
0.096 Mbps for subtitles (32 Kbps + 32 Kbps + 32 Kbps)
No IME
-----------------------
Total ~ 15.5 Mbps ~ 116 MB/min

Thus 30000 MB = 258 minutes, enough for the longest of the LOTR movies.

Now, this is cutting it close of course, but there is a little bit of cushion in that one doesn't necessarily have to have a 1536 Kbps audio track. eg:

12.5 Mbps VC-1
No TrueHD
2.048 Mbps for audio tracks (0.768 + 0.640 + 0.640 Mbps)
0.256 Mbps commentary
0.096 Mbps for subtitles (32 Kbps + 32 Kbps + 32 Kbps)
No IME
-----------------------
Total ~ 15 Mbps ~ 112 MB/min

Thus 30000 MB = 268 minutes

LOTR: ROTK is 250 minutes. Going backwards: 30000/250 minutes = 120 MB/min = 16 Mbps. Thus, even 13 Mbps for video is doable, since 3 audio tracks plus an additional commentary track and subtitles can all be encoded in less than 3 Mbps.

Of course, somebody will complain about the lack of lossless audio tracks, but I personally don't care, especially since movies this long are so uncommon. And if we exclude lossless audio tracks, peak bitrate is also pretty much irrelevant here.

Somebody else may complain about the lack of PiP, but I can live without it if need be (although I'd prefer to have it). Would Peter Jackson even do a PiP for this series? And for the time being it may be moot, since the format neutral studio New Line has this title, and Blu-ray currently cannot properly do PiP anyway.

I think this does illustrate (again) that HD45, HD51, or BD50 would be advantageous, but nonetheless even with the potential limitations (no lossless audio, and no IME), I think it's impressive that movies like this are still doable with top quality video on HD30.

TrevorS

01-27-07, 04:15 PM

Again I don't have a problem with VC-1 *or* HD DVD... I think VC-1 is great and efficient, and I think HD DVD is, well... it gets the job done. But at the same time, it's not as if they are forever extensible into infinity as something that can preserve at four hours what we've come to expect in the two hour range. I'm just saying, Batman Begins isn't at the top of the PQ pile, and until Ben or Amir or someone else starts giving up the identities of some of these other sub-13 titles, I frankly have to consider playing flexibly with wordings on the part of MS.

I have a concern with this statement. The definition of PQ being used here is based entirely on Fettastic's Tier thread, which has nothing inherently to do with the accuracy or difficulty of the encoding and transfer.

Over and over again, I see people on both sides of the format fence declaring the correctness of a transfer based on the definition described in the first post of his tier threads. Unfortunately, it appears as though very few are taking the time to even begin to comprehend what those criteria are actually saying.

Having a high rating in Fettastic's Tier threads has LITTLE TO NOTHING to do with the accuracy or difficulty of the transfer. The most obvious examples of that are animation titles, but it isn't limited to animation.

Having a high rating in Fettastic's Tier threads has EVERYTHING to do with the material and the artistic choices of the director. And material and choices intended to please the eye score highest.

These two fundamentally different criteria MUST not be confused, and yet MANY PEOPLE ON THESE FORUMS ARE CONFUSING THEM REGULARLY.

PS. I'm not suggesting there is anything inherently wrong with those threads, but the continual miscomprehension and misapplication of their content is an ongoing and severe disservice to BOTH formats!

TrevorS

01-27-07, 06:01 PM

God of Biscuits

02-14-07, 08:08 PM

I would be a fervent Blu-ray support if:

-snip-

2. The players were more full featured. Networking in only 2 BD products?? Lack of mandatory audio and video features means a wide dynamic range of possible hardware configs.

3. Reliance on MPEG2. Sony has tried to flog crusty technology for too long while pricing it like it is the new hotness. MPEG2 come on. Java for interactivity?? That's so 90's ...high efficiency codecs and XML based interactivity are distinctly more in tune with what's going on today

Addressing #2: Specious argument. Networking on "only 2" BD products? What percentage of the entire BD player market do you think that comprises?

Addressing #3: This is just plain WRONG. There is no reliance in blu-ray on MPEG-2.

From blu-ray com:
What video codecs will Blu-ray support?

MPEG-2 - enhanced for HD, also used for playback of DVDs and HDTV recordings.
MPEG-4 AVC - part of the MPEG-4 standard also known as H.264 (High Profile and Main Profile).
SMPTE VC-1 - standard based on Microsoft's Windows Media Video (WMV) technology.

Please note that this simply means that all Blu-ray players and recorders will have to support playback of these video codecs, it will still be up to the movie studios to decide which video codec(s) they use for their releases.

So *what* were you saying about VC-1 being a point in HD-DVD's favor?