Warner dropping VC1 for AVC? [Text View]

hdkhang

03-06-08, 09:18 PM

Unless authoring houses were getting incentives to use a particular codec, I can't see why both would stick around. It makes more sense for the industry to pick a codec and invest the time and money to improve it then to waste resources continuing to develope a redundant codec.

Why stop at AVC vs VC1? Why not just do as Sony initially planned and stuck with MPEG2? You fail to realise that competition does wonders for improvements in tech. Especially given the history that AVC was not up to snuff until the threat of VC1 came about.

But the R&B films person seemed to have different opinions

He'll also have you believe that if you copied a WAV file from CD to hard drive that jitter would be introduced. Choose to believe who you want to believe, but think about the science from time to time.

There are differences in looks between codecs, there are differences in look between implementations of the same codec. So to draw a conclusion is futile. Just focus on good looking "encodes" and disregard the codec used to achieve it. Now that it's all BD from here on in, lets try and exert our effort on stuff like removing the stupid region coding so that BD might be able to take off beyond those markets in the same region as the US... Australia for instance.

hdkhang

03-06-08, 09:20 PM

Compared using what test analyzing equipment ?

Your calibrated eyes ?

You never seem to hold those that claim h264 > VC1 to the same level of scrutiny... so you are guilty of exactly the same bias.

30XS955 User

03-06-08, 09:53 PM

PaulGo

03-06-08, 10:02 PM

In response to a comment I made in another thread:

That is correct.

BD has 48mbps compared to HD which only has 30mbps of bandwidth to work with.

The more bandwidth the less one has to molly coddle the encode. It is actually more efficient to use higher bandwidth rates. The higher bit rates are also an advantage dependent upon which CODEC you are using.

AVC tends to perform better given more bandwidth and is the preferred choice of many authoring houses that we work with given the opportunity to use higher bit rates.

VC-1 is probably the better choice for limited bandwidth applications. This may be due to the fact that is was designed for maximum performance in lower bit rate environments.

Again, this is not just my opinion ... but the opinion of several highly respected authoring labs, compressionists, and engineers we work with.

Everyone has an opinion, especially in the AVS Forum. However, I have spent 34 years in the industry working with the very best ... the elite so to speak ... and while we do not agree on everything ... we all agree on most things.

Having produced over 1,000 intellectual properties and worked on the development, distribution, sales, and marketing of over 5,000 programs during my career. I can honestly say I have seen, heard, witnessed, and dealt with enough to make a valuable contribution that is as good as any other opinion on this board.

The big difference between others and myself is that I don't assume I know any professional's experience level, work history, or background nor do I speak disrespectfully about them or their work experience. This is especially true if I have not seen their resume or know their entire body of work and book of business.

It seems some people here like to go off without have a clue about the facts and spew misinformation in an attempt to spread wholesale propaganda .... for reasons they believe will help them gain in the end.

All I can say is this. I am more motivated now than ever before to assure people know the truth and have what they need to decide for themselves what is best.

In the end, the truth almost always wins out ... just like it has in the past and is playing out now with technology that is being accepted by the majority of industry professionals and consumers alike.

Again, both VC-1 and AVC Codecs may serve a purpose in the end. Both do a good job. Neither of them are bad. However, I personally see advantages in using AVC for specific content and for certain when I have over 20mbps to work with.

However, that is just my opinion and the opinion of at several dozen other industry professionals. You do not have to believe any of us. Just look at our Encoder "Crash Test" Shootout on one of our upcoming releases and decide for yourself. At the very least, I think we will all agree on one thing in the end....that there are differences between the codecs.

_Avarice_

03-06-08, 10:05 PM

Get a hold of those straws yet, guys?

Jackietreehorn

03-06-08, 10:12 PM

To quote Maxpower:

"Look, VC-1 as an underlying technology is incredibly similar to AVC, there is no point in questioning the fact (as some would do), the worst part is that VC-1 was actually designed for HD content ground up, while AVC wasn't."

Even Sony insider concedes VC-1 is superior.

Unfortunately for you, Max is NOT a Sony insider, and his comment in no way constitutes proof that VC-1 in any way is superior to AVC.

kamspy

03-06-08, 10:17 PM

As long as it looks good, who cares? We've got 50GB to work with now. Use Mpeg-2 if it gets the rocks off. Just make it look good.

Joe Bloggs

03-06-08, 10:26 PM

He'll also have you believe that if you copied a WAV file from CD to hard drive that jitter would be introduced. Choose to believe who you want to believe, but think about the science from time to time.

There are differences in looks between codecs, there are differences in look between implementations of the same codec. So to draw a conclusion is futile. Just focus on good looking "encodes" and disregard the codec used to achieve it. Now that it's all BD from here on in, lets try and exert our effort on stuff like removing the stupid region coding so that BD might be able to take off beyond those markets in the same region as the US... Australia for instance.
Yes, I'm not sure I quite believe the WAV copying thing ;) (but then again I think he maybe just wasn't saying quite what he meant or something?)

As for his "AVC is better at higher bitrates" - who knows? maybe, I don't know? But both should be better/more similar to the original at very high bitrates. And as someone else said, it would also depend on the implementation (the particular VC1 or AVC encoder). He said he's going to produce a test disc as someone above has mentioned (but I'm sure that will only be released in the US).

I agree it would be good if region coding could be got rid of, or at least that all titles could be released everywhere at the same time (or all within 1 month at the most) at the same/similar price (none of the massive price increases for UK and other regions) :(

wakashizuma

03-06-08, 10:50 PM

With format war over, seems like folks are looking for new wars.
Now it's VC1 vs. AVC. Who cares as long as a title looks good?

Joe Bloggs

03-06-08, 10:54 PM

With format war over, seems like folks are looking for new wars.
Now it's VC1 vs. AVC. Who cares as long as a title looks good?
How will I know if the title looks good if I can't watch it because of region coding/double pricing/titles not being released in the UK?

Also even if they do eventually release a title in the UK, it may not look as good as the US version because they'll probably decide to stick a zillion different unwanted languages onto the UK version and leave us with a reduced video bitrate. And the first thing on the menu for Universal titles will be a language option asking us we speak Japanese :)

lgans316

03-06-08, 11:06 PM

It makes perfect sense for Studios to squeeze as many spoken languages tracks as possible that are concentrated on a geography. But including Japanese track on U.K / AUS editions doesn't make sense.

restart

03-06-08, 11:10 PM

Unfortunately, VC-1 is already out there, and this piece of relatively dead real estate occupies all chips that have to decode old VC-1 BD content forever.

This is Microsoft's gift to the chip companies of the world, a few mm^2 of dead real estate on the silicon for all their chips that have to be used whenever some old BD disks have to be run. (Not to mention being late by a year to the market).

I think what your are describing is an 'electronic trilobite.' :o

Like the segment registers (CS-GS) in the x86 cpus.

Joe Bloggs

03-06-08, 11:14 PM

It makes perfect sense for Studios to squeeze as many spoken languages tracks as possible that are concentrated on a geography. But including Japanese track on U.K / AUS editions doesn't make sense.
Well the UK HD-DVD titles by Universal Studios (or at least some of them) ask you what your language is at the start (obviously they can't be bothered to check what the settings are that we set up on our player), and they default to Japanese :rolleyes: (I think that's just for the multiple copyright pages that you are forced to sit through, maybe the menu etc.. I'm not sure if it affects the soundtrack).

But sticking many other unwanted languages (soundtracks) on the UK version, just because the UK is in Europe is of no use to the vast majority of the UK (just as it would be of no use to the vast majority of US citizens). It just means we get a worse picture/sound/extras/less or no interactive stuff :mad:

paintit77

03-06-08, 11:25 PM

It seems like people care about politicking VC1 vs. AVC more than actual real world results.

In my experience, VC1 titles have consistently been among my favorite HD releases. Scientifically, how is AVC actually better?

Its not and never has been. It will be better in about 12/18 months. It has some advantages long term due to it being essentially open source so it will evolve. One thing it will never be better than VC-1 at is system resource for encoding. Also, its memory management is larger do to its foot print. AVC was designed to run on a Mac cluster or a server. VC-1 was designed to work on a PC or server so it automatically uses a much more efficient (or some would say antiquated) in how it uses memory.

Many of the custom design and engineering software houses for the video/film industry are living in a world still ruled by Apple.

I have used both and with out a doubt, VC-1 at any bit depth is simple amazing.
The results speak for themselves.

Random Digital

03-06-08, 11:25 PM

With format war over, seems like folks are looking for new wars.
Now it's VC1 vs. AVC. Who cares as long as a title looks good?

Yes. Luckily one format is dead and one more is likely to follow (based on upcoming releases).

Joe Bloggs

03-06-08, 11:28 PM

Yes. Luckily one format is dead and one more is likely to follow (based on upcoming releases).
Do you mean HD-VMD in the second bit of the above statement?

lgans316

03-06-08, 11:31 PM

Well the UK HD-DVD titles by Universal Studios (or at least some of them) ask you what your language is at the start (obviously they can't be bothered to check what the settings are that we set up on our player), and they default to Japanese :rolleyes: (I think that's just for the multiple copyright pages that you are forced to sit through, maybe the menu etc.. I'm not sure if it affects the soundtrack).

But sticking many other unwanted languages (soundtracks) on the UK version, just because the UK is in Europe is of no use to the vast majority of the UK (just as it would be of no use to the vast majority of US citizens). It just means we get a worse picture/sound/extras/less or no interactive stuff :mad:

Yeah. But you also have to consider these goodies.

1) Bourne Identity - U.K / AUS version has 1 additional scene which is missing in the U.S version.
2) Serenity has a 20 min special feature that is missing in the U.S version.

Dolby True HD is however stripped off to accommodate more spoken language tracks.

May I know which interactive feature was missing from the U.K version ? I found everything to be intact.

Joe Bloggs

03-06-08, 11:36 PM

Yeah. But you also have to consider these goodies.

1) Bourne Identity - U.K / AUS version has 1 additional scene which is missing in the U.S version.
2) Serenity has a 20 min special feature that is missing in the U.S version.

Dolby True HD is however stripped off to accommodate more spoken language tracks.

May I know which interactive feature was missing from the U.K version ? I found everything to be intact.
Exactly I hate them reducing the picture/sound quality in order to give us unwanted languages.

As for the extra features - check King Kong (2005) for one. UK=no IME. US=lots

Hopefully with Blu-ray and more bandwidth and bigger discs (I hope 100GB/200GB come one day :)) things will be a few bits better :) - well except for things like region coding, price, PiP, player speed, compatibility etc.

MidnightWatcher

03-06-08, 11:41 PM

VC-1, to my eyes, looks slightly better than AVC. They both look very good though.

Joe Bloggs

03-06-08, 11:44 PM

VC-1, to my eyes, looks slightly better than AVC. They both look very good though.
Have you compared the same content at low and high bitrates?

Random Digital

03-06-08, 11:48 PM

Do you mean HD-VMD in the second bit of the above statement?

It would seem that many Blu-ray studios are going AVC. Of course now that all studios are Blu-ray.......

Blu-ray format seems to favor AVC more and offers better quality over all.

The lack of space and bandwidth on HD DVD was more favorable to VC-1

zBuff

03-06-08, 11:48 PM

All the guys at Doom9 rate AVC over VC1, so I welcome the change.

PRO-630HD

03-07-08, 12:56 AM

VC1 is the king to me. The model of efficiency and an image that looks silky smooth not edgy or harsh which some will mistake as looking sharper. Example bluray.com gives Troy a 4.5 for PQ and and the video bit rate is an honest 11 mbps average. Yes VC1 is that good.

lgans316

03-07-08, 01:16 AM

Titles with low bit rate VC-1 encode (less than 16 Mbps) that looks excellent or utmost good.

1) Troy D.C
2) Aviator
3) 2001: A Space Odyssey
4) The Departed
5) The Shining
6) Happy Feet
7) TMNT
8) Unforgiven
9) Seabiscuit
10) Harry Potter
11) Out of Sight

MidnightWatcher

03-07-08, 01:16 AM

Have you compared the same content at low and high bitrates?
No. I am referring to the overall feel when watching VC-1 vs AVC content. They're both capable of exceptional quality, but if I had to choose I pick VC-1.

Kram Sacul

03-07-08, 01:24 AM

VC-1 has more placebos than AVC. ;)

IMO they're both great at high bitrates with any deblocking/filtering left off.

MovieSwede

03-07-08, 01:46 AM

VC-1 has more placebos than AVC. ;)

IMO they're both great at high bitrates with any deblocking/filtering left off.

Well I dont think either AVC or VC1 should leave deblocking etc off. Because the worst thing on HDM is image breakup (if you discount stuff like bad story etc)

It also creates artifacts on its own.

Also I would like to add, as long as the studios uses BD25 they will have bitrate considerations of their own.

If we take as an example a BD25 that have 2 hour movie. That leaves 27,7 mbs as ABR.

Lets say you want a 24bit PCM track and a DD640.

Then we have just slight over 20mbs left for video as an ABR. Not counting if they want to include extras. Of course they can use TrueHD or DTS HD, but not all studios does that. So all in all you will have some great need of the best encoding tools in the market.

lgans316

03-07-08, 01:58 AM

What about Extras my dear MovieSwede ? Why pay $27.95 on a movie with minimal to 0 extras encoded using MPEG-2 on BD-25.

MovieSwede

03-07-08, 02:04 AM

What about Extras my dear MovieSwede ? Why pay $27.95 on a movie with minimal to 0 extras encoded using MPEG-2 on BD-25.

Well I try to avoid purchasing overpriced movies. ;)

But yes if you include extras you will have even greater use for an efficient codec.

phansson

03-07-08, 02:05 AM

I have seen some incredible image quality with MPEG2, AVC and VC-1 and codecs do not matter.

You can have 5 star video with any one of them.......

MovieSwede

03-07-08, 02:07 AM

I have seen some incredible image quality with MPEG2, AVC and VC-1 and codecs do not matter.

You can have 5 star video with any one of them.......

True, but in some cases they dropped the extras. But thats more of a studio thing then a codec thing I would guess. ;)

lgans316

03-07-08, 02:09 AM

Then fine. To me it's doesn't matter what video codec the Studios use. But as many videophiles claim that VC-1/AVC are twice efficient than MPEG-2 I expect the Studios to offer higher bit rates while choosing MPEG-2. Whether it will boost the PQ or not is a different story.

Btw I spent $35 on Rounin (Ronin) and realized I was gloriously phucked up by the raging FOX.

But one thing with Fox, Sony and BVHE is they don't starve the bit rates like Warner do especially during dark scenes.

phansson

03-07-08, 02:26 AM

True, but in some cases they dropped the extras. But thats more of a studio thing then a codec thing I would guess. ;)

Let me state for the record, I would prefer AVC/VC-1 over MPEG 2 because they take up less space.

Extras are nice but the best possible video and lossless audio are my main concerns.

webdev511

03-07-08, 02:42 AM

Titles with low bit rate VC-1 encode (less than 16 Mbps) that looks excellent or utmost good.

3) 2001: A Space Odyssey

+1 Hi-Def Digest gave it 5 out of 5 stars for PQ and I'm pretty sure there are a few scenes with a lot of black in them.

Me? I don't care which codec they use so long as they don't screw up the transfer (FOX). Granted I'd prefer if they leave enough room to include loss less audio, but that shouldn't be a problem unless the studios are going to cut corners (FOX) and just use a BD25 for catalog releases.

sharkcohen

03-07-08, 02:54 AM

Just give me good looking movies. I really don't care otherwise.

Stevie76

03-07-08, 03:11 AM

Wake me up when all these pointless nerd threads are gone ;)

Just turn off the bitrate meter and watch the damn movie guys ;)

Faceless Rebel

03-07-08, 03:43 AM

Titles with low bit rate VC-1 encode (less than 16 Mbps) that looks excellent or utmost good.

10) Harry Potter

I don't know why you keep listing this one when I've shown repeatedly there is noticeable and ugly blocking in HP OotP.

SyHD

03-07-08, 04:15 AM

Paramount did a couple of titles both for HD DVD and Blu-ray back when they were neutral using VC-1 for HD DVD and AVC for Blu-ray(I believe one was Flags of Our Fathers). I remember reviewers giving a slight edge to AVC due to sharpness("it does seem that the Blu-ray/AVC encode is the tiniest bit sharper, while the HD DVD/VC-1 is the slightest bit softer."). Of course the VC-1 was encoded on a 30 GB disc and the AVC on a BD-50. I guess we will never know now unless a foreign studio does a Blu-ray title using one codec and a domestic studio using the other ...both using the same master and optimizing/authoring on a BD-50 disc.

It will probably show VC-1 softer but with less grain and AVC sharper with more grain. Grain haters probably prefers VC-1 just like film novices prefers full screens ...without the "dreaded" black bars.

lgans316

03-07-08, 04:25 AM

You are wrong. Paramount used MPEG-2 on Blu-ray and VC-1 on HD DVD. None of the MPEG-2 encodes got better reviews for PQ than their VC-1 counterparts.

The only title that Paramount really delivered on Blu-ray is Flags.

SyHD

03-07-08, 04:46 AM

Paramount did a couple of titles both for HD DVD and Blu-ray ...

You are wrong. Paramount used MPEG-2 on Blu-ray and VC-1 on HD DVD. None of the MPEG-2 encodes got better reviews for PQ than their VC-1 counterparts.

The only title that Paramount really delivered on Blu-ray is Flags.

HD DVD Flags of Our Fathers - VC1
BD Flags of Our Fathers - AVC/BD50

HD DVD Coming To America - VC1
BD Coming To America - AVC/BD25

HD DVD Trading Places - VC1
BD Trading Places - AVC/BD50

This happened at the tail end of Paramount's neutrality. No, you are wrong. And why bring MPEG-2 into this when the subject matter is VC1/AVC?

Grubert

03-07-08, 06:27 AM

Beowulf HD DVD is AVC and it is Tier 0000.

K.L.

03-07-08, 06:47 AM

While many companies are developing H.264 tools, VC-1 is managed only by Microsoft, it's pretty clear which can emerge as a better production environment.

PeterTHX

03-07-08, 06:53 AM

Paramount also used VC-1 for Face/Off where Disney used AVC for the Japanese version, which is much cleaner. The US version uses the same HD master but has been filtered, probably to compensate for the lower bitrate & capacity.

Plus the audio is DD+ where the Disney is PCM.

webphilosopher

03-07-08, 08:24 AM

Here's hoping Warner doesn't rely too much on BD25 with AVC.

Corellianrogue

03-07-08, 08:36 AM

Exactly I hate them reducing the picture/sound quality in order to give us unwanted languages.

As for the extra features - check King Kong (2005) for one. UK=no IME. US=lots

Hopefully with Blu-ray and more bandwidth and bigger discs (I hope 100GB/200GB come one day :)) things will be a few bits better :) - well except for things like region coding, price, PiP, player speed, compatibility etc.

Is there somewhere on here, or anywhere on the internet, that compares different countries' versions of HD DVDs? (Or Blu-Rays for that matter.) I know there's DVDcompare.net but they have hardly any HD DVD comparisons on there. I don't mean a screenshot comparison, I mean listing the extras and if any cuts have been made or if some versions are the director's cut versions etc. (Although also mentioning PQ would be good too.)

beagle five

03-07-08, 08:37 AM

VC-1 is good at certain things but I think AVC is better.

JTYoung

03-07-08, 08:48 AM

These pissing contests really get old.
Who cares what anyone thinks is "better?" It is an opinion and even the "experts" have opinions on what they prefer. Quoting Richard from RB Films or Maxpower aren't definitive answers on what is better, nor is quoting any other "expert"

What everyone should be concerned with is getting a quality product no matter what encode is used.

webphilosopher

03-07-08, 09:03 AM

These pissing contests really get old.

And pissing contests get more difficult as we get older. :o

Rich Peterson

03-07-08, 09:16 AM

I have seen some incredible image quality with MPEG2, AVC and VC-1 and codecs do not matter.

You can have 5 star video with any one of them.......
I think what we enthusiasts should be most concerned with is how much effort it takes to get good results across all the various scenes in a Master with one set of parameters.

Both codecs are capable of producing great quality, but if one or the other required more tuning/tweaking or special attention in some scenes to get there, then I think the risk of poorer quality encodes using that codec will increase as the volume of disks being authored increases.

Remember, characteristics within a master can vary considerably given some indoor shots, outdoor, special effects, etc. and it's really more of an art to get the best possible results when trying to minimize bitrate.

I don't know for sure if one or the other codec requires more attention, but I am guessing one of them does and I think I know which one that is. I'll keep it to myself, though.

Joe Bloggs

03-07-08, 09:22 AM

Is there somewhere on here, or anywhere on the internet, that compares different countries' versions of HD DVDs? (Or Blu-Rays for that matter.) I know there's DVDcompare.net but they have hardly any HD DVD comparisons on there. I don't mean a screenshot comparison, I mean listing the extras and if any cuts have been made or if some versions are the director's cut versions etc. (Although also mentioning PQ would be good too.)
I don't know of any. Hopefully someone else knows. I would also like it if they list the video/audio bitrate of each one to see if we are/were getting a worse deal because of unwanted extra language soundtracks, etc.

Hopefully with Blu-ray the bitrate should be less of a problem (depending on how many unwanted languages they add too. Hopefully the unwanted language soundtracks will be as few and as low bitrate as possible). It's not good when they give us less (video bitrate/extras etc) but make us pay more.

iceperson

03-07-08, 09:35 AM

Beowulf HD DVD is AVC and it is Tier 0000.

^^^

From what I've read here at AVS Beowulf is supposed to be the best HDM title ever made.

ToddUGA

03-07-08, 09:36 AM

I swear fanboys will find anything to argue about.

BTW, the Mighty Thor can beat the Hulk any day of the week! :)

Nosferax

03-07-08, 09:39 AM

eci

03-07-08, 10:19 AM

^^^

From what I've read here at AVS Beowulf is supposed to be the best HDM title ever made.

Not possible. Beowulf is on HD DVD and 30GB isn't enough. Only Blu-ray can have the best HDM title ever made!

Joe Bloggs

03-07-08, 10:38 AM

dvd-basen.dk does that
It seems to have a link to different review sites.
But is there any that just lists the differences in a list?
something like:

USA,King Kong (2005) HD-DVD,x average video bitrate,y average audio english language bitrate,IME,these extras...
UK,King Kong (2005) HD-DVD,x average video bitrate,y average audio english language bitrate,no IME,these extras...

Corellianrogue

03-07-08, 10:50 AM

It seems to have a link to different review sites.
But is there any that just lists the differences in a list?
something like:

USA,King Kong (2005) HD-DVD,x average video bitrate,y average audio english language bitrate,IME,these extras...
UK,King Kong (2005) HD-DVD,x average video bitrate,y average audio english language bitrate,no IME,these extras...

Yeah, that's what's I'm looking for.^^^ Thanks for the dvd-basen.dk link though Nosferax. :) If you go to DVDcompare.net and search for any DVD (and they do have some HD DVDs and Blu-Rays) you'll see what I mean. It's a great site for DVDs if you still buy them. Ever since I found them years ago I've never bought a DVD without them. In fact I've been out shopping and seen DVDs in sales and unless I knew the UK version was the same or better than other versions I'd think "Damn! I wish I could check DVD Compare right now." and not buy them, lol!

phansson

03-07-08, 10:56 AM

I don't know why you keep listing this one when I've shown repeatedly there is noticeable and ugly blocking in HP OotP.

Yes there is some blocking in a couple scenes, but overall I think that the Order of the phoenix is a very nice looking title. Also, the black level tends to hold up better than most titles I have seen on either format.

Here's hoping Warner doesn't rely too much on BD25 with AVC.

For a shorter movie, I wouldn't have a problem with it. But I do agree, I hope they don't "rely" on it either.

Warner better get lossless more prevelent also....

Joe Bloggs

03-07-08, 11:04 AM

Yeah, that's what's I'm looking for.^^^ Thanks for the dvd-basen.dk link though Nosferax. :) If you go to DVDcompare.net and search for any DVD (and they do have some HD DVDs and Blu-Rays) you'll see what I mean.
Thanks for that. I checked dvdcompare.net and it doesn't have the list saying bitrate differences etc. but it does show some of the differences between the versions - it doesn't list the UK King Kong HD-DVD but just compare the US and Australia ones and see what a better deal the US gets (even assuming that Aus didn't get a reduced video/audio bitrate!) - I bet the UK got the same/similar version as Aus! :(

Corellianrogue

03-07-08, 11:12 AM

Thanks for that. I checked dvdcompare.net and it doesn't have the list saying bitrate differences etc. but it does show some of the differences between the versions - it doesn't list the UK King Kong HD-DVD but just compare the US and Australia ones and see what a better deal the US gets (even assuming that Aus didn't get a reduced video/audio bitrate!) - I bet the UK got the same/similar version as Aus! :(

Damn, I ordered King Kong from Australia and didn't check it at DVD Compare, doh! :( I think I can still cancel it since it was a sale for stocks of HD DVDs coming in on the 28th. I originally wasn't going to buy it at all since it wasn't the extended edition but for less than £5 I think it's worth it! :D (Hopefully I can find a US version for around £5 too.)

tvted

03-07-08, 11:23 AM

Kroenen

03-07-08, 11:24 AM

Not possible. Beowulf is on HD DVD and 30GB isn't enough. Only Blu-ray can have the best HDM title ever made!

:rolleyes:

Stop trolling and grow up.

angelo913

03-07-08, 11:26 AM

Not possible. Beowulf is on HD DVD and 30GB isn't enough. Only Blu-ray can have the best HDM title ever made!

To date BDM does not surpass PQ with it's BD50 to HDM's 30GB discs. So BD50 didn't show any advantage.

The only time optical discs offered the maximum bit rate to fill a disc was Super-Bit DVDs. So until Sony releases Super-Bit BD50 BDM I would agree.

HDM was using VC1 since day one and BDM was using MPEG2 back then too. Even today about 30% of BDM is MPEG2. HDM caused BDM to step up the PQ and started using AVC/VC1. I hope with the loss of competition and no comparison to VC1 or HDM hasn't caused PQ to go down hill, but we'll see.

I find it so funny when someone starts to question BDs motives and decision making, the BD fanboys get so defensive. :rolleyes: HD is dead. I want BDM to be as good and better than HDM in AV Quality and feature set.

...Angelo

tvted

03-07-08, 11:27 AM

And pissing contests get more difficult as we get older. :o

diaper capacity and fill rates. :o

ted

khwiggins2

03-07-08, 11:31 AM

:rolleyes:

Stop trolling and grow up.

I think he meant it jokingly.

angelo913

03-07-08, 11:35 AM

I think he meant it jokingly.

Re-read his message plus the use of an exclamation mark.

...Angelo

khwiggins2

03-07-08, 12:22 PM

Re-read his message plus the use of an exclamation mark.

...Angelo

It was a ridiculous statement, so I took it as a joke. Like saying the sky is falling. :)

jocktheglide

03-07-08, 01:02 PM

I swear fanboys will find anything to argue about.

BTW, the Mighty Thor can beat the Hulk any day of the week! :)

yeah, but if HULK gets more mad as they fight longer he ges stronger and faster unless THOR can calm him down and go for the kill then thats the only way I can see HULK getting his ass whomped on by weak THOR............

iceperson

03-07-08, 01:11 PM

I actually prefer VC1 over AVC. My tests have confirmed this.

So, what VC1 titles does your test confirm are better than the AVC beowulf?

30XS955 User

03-07-08, 02:06 PM

So, what VC1 titles does your test confirm are better than the AVC beowulf?

You will need to pick a better AVC title. Animation does not benefit much going HD from SD, especially for films like Beowolf which are already relatively soft compared to MPEG 2 titles like Crank or VC-1 titles like PotC:AWE.

iceperson

03-07-08, 02:10 PM

You will need to pick a better AVC title. Animation does not benefit much going HD from SD, especially for films like Beowolf which are already relatively soft compared to MPEG 2 titles like Crank or VC-1 titles like PotC:AWE.

PotC:AWE is AVC

Doh!

30XS955 User

03-07-08, 02:15 PM

PotC:AWE is AVC

Doh!

Which goes to show that AVC can be either sharp like AWE or soft like Beowolf.

Do you think Beowolf looks better than Crank?

iceperson

03-07-08, 02:19 PM

Which goes to show that AVC can be either sharp like AWE or soft like Beowolf.

Do you think Beowolf looks better than Crank?

I'm not the one who thinks 1 codec is better than another. I believe that all three can produce images that are indistinguishable from each other given enough bandwidth and attention during the transfer.

I'm curious, when you decided that VC1 was better than AVC was AWE one of the titles you used?

SyHD

03-07-08, 02:26 PM

Beowulf is a CGI title ...of course PQ will be none other than perfect. Having said that, the film does look soft in some spots. It has nothing to do with the transfer(digital to digital) but the technology and rendering of the film. They need to work out the depth of view issues. It wasn't very consistent in Beowulf.

I rather take Ratatouille over Beowulf ANY DAY OF THE WEEK.

30XS955 User

03-07-08, 02:29 PM

[QUOTE=iceperson;13320381
I'm curious, when you decided that VC1 was better than AVC was AWE one of the titles you used?[/QUOTE]

I have not seen AWE on Blu ray and was just going by the popular consensus.

DougPr

03-07-08, 02:41 PM

Icemage

03-07-08, 02:52 PM

As long as they avoid Mpeg-2 like the plague, I really don't care. Both codecs from Apple and Microsoft are extremely good in comparison. You have to have really good eyes to even tell, and even that is questionable since there are very few things that are encoded in VC-1 and AVC in a comparable bitrate to even do a side-by-side comparison. Full "bluray bandwidth" bitrate of either compression should reveal results that would require a magnifying glass on a 100" movie screen to tell a difference IMO.
+1

Both codecs are so close in performance at the bitrates used on Blu-ray that I'm surprised there's as much dissension as there is. It's not like there's any hardware or software that can't support either codec; it's required by the spec, just as it was for HD DVD.

I think most of the "flaws" people are pointing at for either codec on specific titles have a lot more to do with the quality of the encoding tools and the skill of the people doing the encoding, than any supposed deficiency in the codecs themselves.

Corellianrogue

03-07-08, 03:01 PM

Which goes to show that AVC can be either sharp like AWE or soft like Beowolf.

Do you think Beowolf looks better than Crank?

I don't care too much whether VC-1 is better than AVC or not but Beowulf is incredible, it's flawless and easily the best PQ I've ever seen. And yes, I own Crank and I think Beowulf blows it out of the water. That's nothing against Crank, it's got great PQ (and it's a cool movie) but Beowulf is just flat out better, no question. And it may not be in 3D but at times it looks like it is! :eek:

olivaw

03-07-08, 03:17 PM

...Beowulf is incredible, it's flawless and easily the best PQ I've ever seen....

Too bad beowulf's characters seem as lifeless as a dead fish eye.

30XS955 User

03-07-08, 03:54 PM

I don't care too much whether VC-1 is better than AVC or not but Beowulf is incredible, it's flawless and easily the best PQ I've ever seen. And yes, I own Crank and I think Beowulf blows it out of the water. That's nothing against Crank, it's got great PQ (and it's a cool movie) but Beowulf is just flat out better, no question. And it may not be in 3D but at times it looks like it is! :eek:

May I ask what you mean by Beowolf looking great? Is it the fx? I know you probably don't mean it looks great because the animation looks life-like, because if it were that than obviously real-life movies would always look better, you know?

SyHD

03-07-08, 03:59 PM

Have you even seen the movie or are you just trolling?

The characters are lifeless. Don't get me wrong ...the technology behind the film is fascinating but the facial expressions were unnatural, mechanical, and bordering on creepy. The technology is still a long ways from the Uncanny Valley. I think thats what he means.

PaulGo

03-07-08, 04:34 PM

Any discussion about which codec is superior will not accomplish anything. The studios will use the codec that they feel most comfortable with as long as it produces "good enough" results.:) If one codec can give them better results for a specific movie (Michael Bay preferred AVC for Transformers because he thought it produced superior results) then they will use that.

Their has been some talk that Microsoft is getting very little renumeration out of the VC-1 patent pool because of all the payemts they need to make to other patent holders who hold underlying patents. Perhaps because of this Microsoft may not find it economically feasible to provide as much support for VC-1 and that may also enter into the decision to use AVC.

briankmonkey

03-07-08, 04:49 PM

The characters are lifeless. Don't get me wrong ...the technology behind the film is fascinating but the facial expressions were unnatural, mechanical, and bordering on creepy. The technology is still a long ways from the Uncanny Valley. I think thats what he means.

I've only watched trailers but I agree from what I saw. Didn't really leave me with any desire to rush to the theater. I'm sure I'll rent it one day, maybe while I still have a HD DVD player.

whippersnapper

03-07-08, 05:08 PM

And pissing contests get more difficult as we get older. :oBut certainly gives one something to look forward to!:)

Corellianrogue

03-07-08, 05:18 PM

May I ask what you mean by Beowolf looking great? Is it the fx? I know you probably don't mean it looks great because the animation looks life-like, because if it were that than obviously real-life movies would always look better, you know?

I'm talking about the quality of the transfer, it's even better than Ratatouille and Shrek 3. The depth is incredible and becomes almost 3D in places, especially near the end. I'm not talking about how good the CGI is (although it's incredible and scenes of a particular character at the end are pretty much indistinguishable from real life, plus Angelina Jolie's scenes look real too) I'm talking about the image quality that the HD DVD is displaying. Ratatouille on Blu-Ray and Shrek 3 come close (yes I know these are all CGI movies but not only am I comparing like with like but CGI movies look better in HD anyway) but Beowulf is clearly the best HD PQ so far. If you don't believe it because it's an HD DVD or you think I'm just saying it because I support HD DVD you don't have to worry because I've seen ads online for the Blu-Ray version here in the UK coming soon so both HD DVD and Blu-Ray will share the honour of having the greatest looking HD movie ever seen! :D

Joe Bloggs

03-07-08, 05:31 PM

I'm talking about the quality of the transfer, it's even better than Ratatouille and Shrek 3.
For CGI films like Beowulf, Ratatoille and Shrek do they bypass film altogether? I thought for the transfer of these films they would/do just copy it digitally so there wouldn't be a difference in the quality of transfer? Or do they actually scan film?

Kram Sacul

03-07-08, 05:46 PM

They're 100% digital. I think the "transfer" means the look of the movie (detail, color, etc) and/or compression quality.

Corellianrogue

03-07-08, 06:00 PM

They're 100% digital. I think the "transfer" means the look of the movie (detail, color, etc) and/or compression quality.

Yeah, that's what I meant, I probably should have said encode not transfer. Although the movie's still "transfered" from somewhere even if it's not film, lol!

SlickVik

03-07-08, 10:54 PM

AVC is better than VC1 imho.

eci

03-08-08, 01:13 AM

AVC is better than VC1 imho.

Please explain.

nakedeye

03-08-08, 01:46 AM

I find that I can always seem to pick out AVC movies vs. VC-1 movies. It could perhaps be how the studios mastered them (wienstien for AVC is a good example), but even Transformers has a bit of a soft AVC edge that I have posted about in the past.

To me VC-1 is more "film like" Mpeg2 is more video ish .

That being said, they can all look fantastic. I hope they go the VC-1 route however.

Kram Sacul

03-08-08, 03:15 AM

Please explain.

Because AVC has a great taste? :D

I don't really like vc-1 partly because in a lot of comparisons it seems to smooth over things. Maybe it just wasn't fed enough bitrate though or the loop filter or whatever was too aggressive. In any case this should not have happened (filtered number):

http://img516.imageshack.us/img516/2827/wwws254da0.png

and this too (blurry UFOs):

http://img408.imageshack.us/img408/8512/wtcartifactsqe2.png

AVC or VC-1, IMO it doesn't matter. Just no more low bitrate crap.

MaynardJames

03-08-08, 03:49 AM

Please explain.

VC-1 > AVC by a long shot.

Please explain.

benwaggoner

03-08-08, 04:16 AM

x264 > pep :D
Comparing the hundreds of great looking PEP-encoded titles to which x264 encoded discs?

All the guys at Doom9 rate AVC over VC1, so I welcome the change.
...when they're trying to test how low a bitrate you can get away with while backing up already compressed media without getting distracting artifacts. Different market, and one that rests on postprocessing filters as much as anything.

While many companies are developing H.264 tools, VC-1 is managed only by Microsoft, it's pretty clear which can emerge as a better production environment.
Untrue. Main Concept (primary vendor of H.264 implementations for professional video apps like Sorenson Squeeze, Sonic Cinevision (non-PSE), Canopus ProCoder, Rhozet Carbon, and many others) has their own in-house implementation of VC-1 they're now selling. Other companies like Telestream have built their own implementations based on source code licensed from Microsoft.

http://www.mainconcept.com/site/developer-products-6/vc-1-pro-sdk-20559/information-20575.html

Lots of vendors are also incorporating our VC-1 Encoder SDK into their own products.

Kram Sacul

03-08-08, 04:39 AM

I think Microsoft is seriously underestimating the potential of VC-1. Think of all the money they could get by selling hats and coffee mugs with a VC-1 logo on them. AVC wouldn't know what hit them . :D

Faceless Rebel

03-08-08, 04:39 AM

Animation does not benefit much going HD from SD

WRONG BEYOND ALL IMAGINATION

Icemage

03-08-08, 06:54 AM

WRONG BEYOND ALL IMAGINATION
It's not by accident that many of the titles listed at the top of the tier threads here at AVS (on both formats) are animated CGI.

Richard Paul

03-08-08, 07:26 AM

Main Concept has their own in-house implementation of VC-1 they're now selling.Just curious but how many Blu-ray/HD DVD movies were encoded with that encoder?

Lots of vendors are also incorporating our VC-1 Encoder SDK into their own products.Well I think that has a lot to do with the fact that Microsoft included VC-1 support in WMP but did not include MPEG-4 AVC support. In fact even today Microsoft doesn't support MPEG-4 AVC in WMP and I have never heard of a good explanation for that.

MovieSwede

03-08-08, 08:10 AM

Because AVC has a great taste? :D

I don't really like vc-1 partly because in a lot of comparisons it seems to smooth over things. Maybe it just wasn't fed enough bitrate though or the loop filter or whatever was too aggressive. In any case this should not have happened (filtered number):

Thats why you shouldnt watch movies upzoomed...

Your eyes cant see all thoose issues during playback.

If you go in 400% of a still frame, of course you can find issues, but they dont aim that you wiew it that way. And neither does the compressionist.

How the movie looks during normal playback is really what counts.

Joe Bloggs

03-08-08, 08:21 AM

Thats why you shouldnt watch movies upzoomed...

Your eyes cant see all thoose issues during playback.

If you go in 400% of a still frame, of course you can find issues, but they dont aim that you wiew it that way. And neither does the compressionist.

How the movie looks during normal playback is really what counts.
Does that mean we should watch on HDTVs that are as small as possible because the compressionist doesn't want us to watch on a bigger one because we will see more issues because the compressionist didn't use a high enough video bitrate or the right codec or didn't tweak the encode enough?

And what will happen when we have Quad HD HDTVs at about 3840x2160 resolution. Won't we want to upconvert our existing discs, in the best quality possible (which won't be possible if someone has chosen the wrong video bitrate/codec/done a bad transfer/encode?

iceperson

03-08-08, 09:44 AM

Tom McMahon

03-08-08, 10:19 AM

...

I think most of the "flaws" people are pointing at for either codec on specific titles have a lot more to do with the quality of the encoding tools and the skill of the people doing the encoding, than any supposed deficiency in the codecs themselves.

Agree.

Lee Stewart

03-08-08, 10:48 AM

Just curious but how many Blu-ray/HD DVD movies were encoded with that encoder?

140 BD's

http://www.blu-raystats.com/stats.php

316 HD DVD's

http://www.blu-raystats.com/stats.php

30XS955 User

03-08-08, 11:10 AM

http://www.avsforum.com/avs-vb/showthread.php?t=998844

You could not be more wrong.

In my opinion, Beowolf does not look half as good as Transformers, either in detail or sharpness.

TheLoveone

03-08-08, 11:30 AM

VC-1 > AVC by a long shot.

:rolleyes:

I've compared, using ONLY BD. Have you?

:rolleyes:

We are all still waiting for an explanation of your methods, because everything I've read says the opposite (such as their comparison at Doom9). In fact I think you are the only one in this thread saying this.

What BD title did you compare that had AVC and VC-1 encodes at the same bitrate?

Steeb

03-08-08, 11:37 AM

140 BD's

http://www.blu-raystats.com/stats.php

316 HD DVD's

http://www.blu-raystats.com/stats.php

The question was: how many Blu-ray/HD DVD movies were encoded with the Main Concept encoder that Ben mentioned. He did not ask how many titles used the VC-1 codec.

TheLoveone

03-08-08, 11:37 AM

To date BDM does not surpass PQ with it's BD50 to HDM's 30GB discs. So BD50 didn't show any advantage.

The only time optical discs offered the maximum bit rate to fill a disc was Super-Bit DVDs. So until Sony releases Super-Bit BD50 BDM I would agree.

HDM was using VC1 since day one and BDM was using MPEG2 back then too. Even today about 30% of BDM is MPEG2. HDM caused BDM to step up the PQ and started using AVC/VC1. I hope with the loss of competition and no comparison to VC1 or HDM hasn't caused PQ to go down hill, but we'll see.

I find it so funny when someone starts to question BDs motives and decision making, the BD fanboys get so defensive. :rolleyes: HD is dead. I want BDM to be as good and better than HDM in AV Quality and feature set.

...Angelo

I'm not sure what you are talking about, I own several movies on BD that are high bitrate AVC/VC-1 encodes that go beyond 40+Mbps. In fact high bitrate encodes seem to be the norm now for BD exclusive releases. In fact I only own about 6 movies and 3 of them do this. 3:10 Yuma (VC-1), Ratatouille (AVC), and The Prestige (AVC) all go above 40Mbps just for video on my bitrate monitor.

mhafner

03-08-08, 12:05 PM

Yeah, that's what I meant, I probably should have said encode not transfer. Although the movie's still "transfered" from somewhere even if it's not film, lol!
Yes, you can not just compress the Beowulf digital master element as it is not in HD format. You need to switch color gamut, likely resample from > 1080p to 1080p, go from > 8 bit samples to 8 bit and do color subsampling to 4:2:0 before you can compress.

Corellianrogue

03-08-08, 03:48 PM

I'm not sure what you are talking about, I own several movies on BD that are high bitrate AVC/VC-1 encodes that go beyond 40+Mbps. In fact high bitrate encodes seem to be the norm now for BD exclusive releases. In fact I only own about 6 movies and 3 of them do this. 3:10 Yuma (VC-1), Ratatouille (AVC), and The Prestige (AVC) all go above 40Mbps just for video on my bitrate monitor.

And yet the reviews say there's no PQ difference between the HD DVD and Blu-Ray version of The Prestige. Hmmm. (Yes, I do have The Prestige on HD DVD and yes it does look great. :D)

TheLoveone

03-08-08, 04:27 PM

And yet the reviews say there's no PQ difference between the HD DVD and Blu-Ray version of The Prestige. Hmmm. (Yes, I do have The Prestige on HD DVD and yes it does look great. :D)

What does that have to do with my post, I was just correcting some misinformation that no BD's take advantage of maximum bitrate yet.

Corellianrogue

03-08-08, 05:32 PM

What does that have to do with my post, I was just correcting some misinformation that no BD's take advantage of maximum bitrate yet.

I was just pointing out that if The Prestige really does use 40Mbps+ on Blu-Ray then obviously there are diminishing returns since the HD DVD version looks the same yet must have a lower bitrate as HD DVD's limit is in the 30s. (I've forgotten the exact bitrate.) So maxing out the bitrate (for video), or at least going above a certain bitrate doesn't seem to have a "Superbit" effect on Blu-Ray like people are asking for.

Richard Paul

03-08-08, 06:28 PM

140 BD's

http://www.blu-raystats.com/stats.php

316 HD DVD's

http://www.blu-raystats.com/stats.phpWell Lee its good to know how many VC-1 encodings there are but that doesn't answer my question (http://www.avsforum.com/avs-vb/showpost.php?p=13325804&postcount=141). I was asking Ben how many Blu-ray/HD DVD movies were encoded with the Main Concept VC-1 encoder.

Lee Stewart

03-08-08, 06:30 PM

Well Lee its good to know how many VC-1 encodings there are but that doesn't answer my question (http://www.avsforum.com/avs-vb/showpost.php?p=13325804&postcount=141). I was asking Ben how many Blu-ray/HD DVD movies were encoded with the Main Concept VC-1 encoder.

My bad:o

Good to know though:D

sspears

03-08-08, 06:54 PM

Just curious but how many Blu-ray/HD DVD movies were encoded with that encoder?

Hopefully not too many!

There is a lot of talk of VC-1 vs. AVC, but I believe the implemenation of each is more important. Threads on Thompson vs. Sony vs. MEI vs. Cinemacraft vs. Toshiba vs. Sonic, w/r/t AVC, would be more valuable. I say this because not all AVC encoders are created equal.

Cinemacraft does well on Animation, assuming you turn off all of the stupid NR/LPF filters that are on by default for the unsuspecting compressionist. Sony's does not do as well on animation. On the flip, Cinemacraft does not do nearly as well on live action while Sony's does. MEI's does poorly in areas of black where low amplitude noise, such as film grain, gets quantized out and you are left with alternating blocks of pure DC, which results in chroma blocking. Its pretty severe in POTC: Curse of the Black Pearl. Main Concept's (Sonic) has the same issue.

Thompson's is probably only used by Technicolor and MEI's by PHL. You can get a Cinemacraft box for about $60k. Its a single box solution that runs on multi-core. The Sony, like MEI, Thompson and PEP (VC-1) are all distributed over several HW clients.

Richard Paul

03-08-08, 07:38 PM

Hopefully not too many!Do you think the Main Concept VC-1 encoder isn't that good of an encoder?

There is a lot of talk of VC-1 vs. AVC, but I believe the implemenation of each is more important.I would agree with that though I have often heard conflicting reports about the same encoder. For instance I have heard some people say that the Toshiba MPEG-4 AVC encoder was somewhat good while others called it pretty bad. I haven't heard whether Toshiba will market it for Blu-ray but what did you think of it?

Faceless Rebel

03-09-08, 01:05 AM

It's not by accident that many of the titles listed at the top of the tier threads here at AVS (on both formats) are animated CGI.

Well, I was referring more to traditional style cel animation. I realize that no one actually uses cels anymore, everything is digitally painted, but the traditional type of animation has a distinct look. Both traditional and CGI animation are stunning in HD. Even something relatively simple like The Simpsons Movie looks stunning in 1080p.

I defy anyone here to watch this short sequence (http://students.washington.edu/lzone/[FZ] AIR - Extended Opening A (FEB94D03).mp4) (Riight-click/Save As, 155MB, 3:21) and then claim that animation does not benefit much from HD. If you can't see the improvement, then I'm sorry, you're blind and you should apply for government disability benefits right away.

For those curious, that is the complete text-less opening sequence for the Japanese TV anime Air. It's ripped from the Japanese Blu-ray box set which currently goes for $260 at YesAsia.com. It's properly encoded in 1080p AVC and contained in .mp4 format and is ready for transfer to a PS3 if you want to view it from there, otherwise you'll need a powerful computer which has FFDShow or CoreAVC installed. I can tell you that I run Core 2 Duo 2.8ghz and using FFDShow I have to turn all filters off for it not to skip. Not that you need any filters for this as it is already as perfect as it will ever get. The point is, you will not be watching this on a puny computer. Also note that there is no show here whatsoever, it's just the opening. I want to make sure I am not violating any copyrights here even though Air is not yet licensed for distribution outside of Japan in HD. However even with no show it at least gives you an idea of animation in HD looks like. Also you can listen to one of my favorite anime OPs ever.

Random Digital

03-09-08, 01:23 AM

It seems that AVC is fast becoming the standard on Blu-ray.

stumlad

03-09-08, 03:56 AM

I'm sure it was answered already, but when will the switch occur.. right after they stop making HD DVDs? If so, this will not help in proving that the extra bit-rate matters or doesn't matter because it will likely be higher bit rate with different codec, and possible different compressionists.

Just a note to those who have HD DVDs from Warner and thinking about selling to buy the BD version. Since there's no advantage as far as the transfer (since it's the same), you may as well hold onto your HD DVD until they re-release the titles with new transfers along with maximizing bandwidth, etc... Otherwise you will have bought the DVD, then HD DVD, then blu-ray, then blu-ray again...

Tim Glover

03-10-08, 01:04 AM

Interesting thread. :) In reality, all of the codecs used thus far have shown to be fantastic when the effort is there. For those demanding no more MPEG-2, I say pop in the Kingdom of Heaven Blu-ray. It remains one of the best images I've ever seen. It's one that perhaps looks the most filmlike of any HDM title.

Most of us here are kind of over-informed and let our bias get in the way sometimes. With HD-DVD & Blu-ray films I find myself getting more absorbed in the film, not the codec.

Faceless Rebel

03-10-08, 05:40 AM

Well, if Warner wants to keep using VC-1, they probably could. But they need to stop releasing HD DVD optimized encodes because when Warner does it the inevitable result is soft, detail-lacking, and littered with compression artifacts. The latest victim of the Warner low-bitrate VC-1 method is Appleseed Ex Machina.

http://bluray.highdefdigest.com/1259/appleseedexmachina.html

In my estimation, the 1080p/VC-1 transfer featured on this Warner Blu-ray release is merely a descendant of its immaculate digital source. The picture lacks the refined detailing and crisp resolution I've come to expect from high definition CG -- the picture is consistently hazy, random nighttime scenes are a bit blurry, and several shots are completely out of focus. As it stands, linework and edges are dull, clothing textures lack definition, and smaller objects look only slightly better than they do on the standard DVD. While daytime exterior scenes are vastly improved, they still exhibit a a mild softness that robs the transfer of the three-dimensional pop found in top tier releases like 'Ratatouille' and 'TMNT.' The transfer's flaws smack of overindulgent DNR and compression inadequacies.

Less problematic (but equally distracting) are a series of light artifacts, minor aliasing issues, and heavy color bands that litter the image from beginning to end. Look no further than the opening assault on the cathedral to catch a dozen examples of each (as well as the rampant softness I mentioned earlier). The banding becomes so brazen at times that it stretches across the entire width of the screen -- some shots are plagued by rows of bands that shoot through the sky, lay across the character models, and drape along every element in the foreground. The effect is bizarre to say the least and occasionally makes the image look as if it's being projected on a set of closed window blinds.

MovieSwede

03-10-08, 07:31 AM

Well, if Warner wants to keep using VC-1, they probably could. But they need to stop releasing HD DVD optimized encodes because when Warner does it the inevitable result is soft, detail-lacking, and littered with compression artifacts. [/url]

Strange because 300, Blade Runner, TMNT, Matrix were all optimized for HD DVD. (Not in all cases since they also optimize for BD25/HD30)

If your theory was true, you would have experienced it in every HD DVD title released.

zBuff

03-10-08, 08:39 AM

It seems that AVC is fast becoming the standard on Blu-ray.

Only because it has quickly become the standard for digital recording and transmission worldwide.

amirm

03-10-08, 01:23 PM

Well, if Warner wants to keep using VC-1, they probably could. But they need to stop releasing HD DVD optimized encodes because when Warner does it the inevitable result is soft, detail-lacking, and littered with compression artifacts. The latest victim of the Warner low-bitrate VC-1 method is Appleseed Ex Machina.

http://bluray.highdefdigest.com/1259/appleseedexmachina.html

You didn't quote this part:

"Despite its issues, the Blu-ray transfer still has the clear stability and swagger of a high definition release, boasting a fresh image and a healthy bit rate. Even the squishiest details are more crisply rendered here than they are on the underwhelming standard DVD. "

:).

The issue you have with quoting such things is that you have a+b = c and with "c" in your hand (final product), you are trying to figure out "a" and "b" and you just can't do that with one equation :).

Banding can indeed be created with the codec. But the top reason by far it is there is due to poor effects processing (while creating the original movie) and conversion from 10 bits to 8 bits. Set your desktop to fewer colors on your computer and what do you see? Banding. Take the same image and downconvert it 640x480 and chances are, you won't see it anymore as that filtering removes it. But in its original version, it is easily visible without proper dither. To wit, the logo for one of the studios has banding in it in every one of their HD releases. Needless to say, it is not bitstarved. It is there in the source and folks didn't notice it. But these new formats are so transparent that such artifacts are visible, especially if one pays attention to it.

So per above, the artifact can be in the codec or in the source/conversion to 8-bits giving you "a" and "b." Without more data, you simply cannot figure out its source.

As to softness, again, that could be caused by strong loop filters and such, but once more, it is also easily found in the source. When we were working on Bandai's Anime for HD DVD, I was surprised how soft it was in many scenes. So the post house played the master in another monitor next to the encode and the same softness was there too. Without access to the master ("a"), you can't put the blame on the codec ("b"). My impression from talking to folks is that true 1080p Anime source is very hard to come by.

So let's not put such things forward as proof of things. Instead, with the format war over, let's focus on understanding the causes for these things. Go and ask your insiders to contact the studio and find out what they say about the review. I know I used to do this all the time and it is eye opening what you hear (and learn!).

amirm

03-10-08, 01:28 PM

Only because it has quickly become the standard for digital recording and transmission worldwide.
If by that you mean folks "follow the crowd" that is pretty true :). But otherwise, from technical point of view, the tools are very different. For broadcast, you have to be real-time and everything in hardware. This means some things can be better, but often they are worse when you are trying to spit things out on a frame by frame basis. Best offline encoders have radically different user interface and feature set than real-time encoders.

By the way, the other reason AVC is taking off there is that a number of MPEG-2 patents are about to expire and those companies are banking on fresh new revenues from AVC (patent life is about 17-20 years depending on which country you have registered it). Given how many major companies are in this boat, and substantial revenue MPEG-2 brings them, you can get a feel how strong the business motivation is to move folks this way.

Joe Bloggs

03-10-08, 02:51 PM

Banding can indeed be created with the codec. But the top reason by far it is there is due to poor effects processing (while creating the original movie) and conversion from 10 bits to 8 bits. Set your desktop to fewer colors on your computer and what do you see? Banding. Take the same image and downconvert it 640x480 and chances are, you won't see it anymore as that filtering removes it. But in its original version, it is easily visible without proper dither.
Would you agree that we should be encoding on HDM in something like 10 bit colour instead of 8 bit colour, because of this, even if some of the current HDTVs can't properly output all 8 bits correctly?

Someone pointed out that 10 bit colour should sometimes take less bandwidth to encode than 8 bit dithered (due to it being more efficient than dithering to 8 bit and storing the 8 bit dithered version on the disc - ie. more efficient for the codec in detecting differences between frames etc.).

Also, when we get higher res screens we can have an option to dither in the HDTV or Blu-ray player or with screens that display many more colours correctly (say those with many LED backlights/OLED/Laser) they may not have to dither at all and everything will be better.

amirm

03-10-08, 03:25 PM

Would you agree that we should be encoding on HDM in something like 10 bit colour instead of 8 bit colour, because of this, even if some of the current HDTVs can't properly output all 8 bits correctly?
I am not sure I understand :). Are you asking if the current HD formats should have allowed for 10-bits? If so, yes, I wish they did have that as a standard feature. Fortunately, there is a way to add 10-bit support in a backward compatible way using optional payload that gets ignored by current decoders. Once silicon becomes availble with 10-bits, then the encoder can be updated to put the other 2 bits in the optional payload and a new player with the updated silicon used to extract the full 10-bits (it is a bit more complicated than this but you get the picture :) ).

Without support for 10-bits, then we wind up adding dither and that is not something that is good when you want to compress things as that reduces codec efficiency.

As to displays, they have their own flaws that cause banding (and if your set is not bought in the last couple of years, most of the banding you see is caused by your display and not the source!).

Someone pointed out that 10 bit colour should sometimes take less bandwidth to encode than 8 bit dithered (due to it being more efficient than dithering to 8 bit and storing the 8 bit dithered version on the disc - ie. more efficient for the codec in detecting differences between frames etc.).
Oops, you just made the same point as I mentioned above :). Not sure it really does take less than 8-bits. But in balance, one does lose efficiency when you add noise to the source in the process of (proper) conversion from 10 to 8 bits. Whether that evens out the difference in data rate, would require lots of experimentation on different types of sources.

Also, when we get higher res screens we can have an option to dither in the HDTV or Blu-ray player or with screens that display many more colours correctly (say those with many LED backlights/OLED/Laser) they may not have to dither at all and everything will be better.
This is generally true. However, there may be a need for dither inside the display if it processes the signal at higher resolution than it can display (e.g. in the course of YUV to RGB conversion).

Joe Bloggs

03-10-08, 03:47 PM

I am not sure I understand :). Are you asking if the current HD formats should have allowed for 10-bits? If so, yes, I wish they did have that as a standard feature. Fortunately, there is a way to add 10-bit support in a backward compatible way using optional payload that gets ignored by current decoders. Once silicon becomes availble with 10-bits, then the encoder can be updated to put the other 2 bits in the optional payload and a new player with the updated silicon used to extract the full 10-bits (it is a bit more complicated than this but you get the picture :) ).

By new silicon, new mean new silicon in the players (or players and TV)?
Assuming you weren't doing any processing of the picture (say just outputting a 1080p24 video via HDMI and nothing else, with no scaling or PiP or subtitles or anything) you could output with no extra silicon (as it's just extra data)? Though I assumed they had already made chips for this (and 1080p60 :)) but maybe they're just not ready to be put into players yet?

Also, if you encode with some bits that can be ignored by players that don't support 10 bit but can be used by players that do support 10 bit, wouldn't they still put some dither in the 8 bit one (in order to prevent banding in their version) and so it wouldn't look the best?

Wouldn't it be better to put a second disc in the package labelled in big letters "High Colour Version - for High Colour enabled players only" with a bit of (Java?) code at the start that checks your machine model and version number to check whether it is high colour (10 bit) enabled, and if not, it will say so and ask you to put in the other disc instead (eg. it has a list of all current model and serial numbers that it knows are 8 bit machines only - and for any other machine it will assume it is 10 bit capable - or for the first view discs produced in 10 bit it will warn you first - or have another list of model & minimum version numbers that can do 10 bit)?

amirm

03-10-08, 04:00 PM

By new silicon, new mean new silicon in the players (or players and TV)?
Yes. Without it, no one is going to encode 10-bit content.

Assuming you weren't doing any processing of the picture (say just outputting a 1080p24 video via HDMI and nothing else, with no scaling or PiP or subtitles or anything) you could output with no extra silicon (as it's just extra data)? Though I assumed they had already made chips for this (and 1080p60 :)) but maybe they're just not ready to be put into players yet?
No, no. There is fair amount of changes to the decoder to decode the extra 2-bits and then merge them back in with the 8-bit core. This has nothing to do with frame rate or HDMI. The core decoding engine in the SoC needs to be updated to handle the optional stream and merging it. This will make the chip somewhat more expensive, and increase its power consumption plus memory usage. So it doesn't come for free but relatively speaking, it might be a good way for a silicon provider to differentiate its product.

Also, if you encode with some bits that can be ignored by players that don't support 10 bit but can be used by players that do support 10 bit, wouldn't they still put some dither in the 8 bit one (in order to prevent banding in their version) and so it wouldn't look the best?
Potentially yes and that is why I said the process was more complicated than I described :). I say "potentially" because you could choose to disadvantage 8-bit users by not dithering and hence, reduce the burden on the optional layer.

To expand, if dither is added to arrive at the 8-bit stream, then you subtract that from the 10-bit source. What is left, is what you have to encode and send to the decoder. The decoder performs the inverse math and you get to the original 10-bit fidelity. Needless to say, the payload gets larger for the optional layer due to (inverse) dither to be carried.

Wouldn't it be better to put a second disc in the package labelled in big letters "High Colour Version - for High Colour enabled players only" with a bit of (Java?) code at the start that checks your machine model and version number to check whether it is high colour (10 bit) enabled, and if not, it will say so and ask you to put in the other disc instead?
From a business model point of view, there are two issues with this:

1. Folks will sell the 8-bit version in eBay and as a result, reduce the potential sales of title.

2. There is higher manufacturing cost.

These two factors combined, may not make up for increased sales to enthusiasts of a "10-bit" title.

But sure, from technical point of view, it would work. And I suspect folks making eye candy stuff do this as the selling advantage of 10-bit would be higher for them and they are less concerned about piracy.

Also, you mention high-color. That is yet another thing that needs its own optional payload. While I have heard of some interest in the content community around 10-bit, I have not heard the same for high color. Perhaps the reason for this is the fact that current HD masters are already 10-bits, but do not have higher gamut than what you have at home.

Joe Bloggs

03-10-08, 04:14 PM

Also, you mention high-color. That is yet another thing that needs its own optional payload. While I have heard of some interest in the content community around 10-bit, I have not heard the same for high color
I was probably confusing the two things :) - I was just meaning one version has a lot more colours than the other, and so wouldn't have banding and wouldn't need dithering in the encode and hopefully would be a lot more true to life (if wanted - or more cinema colour range if wanted) etc.

EDIT:
If people selling one of the disc's on ebay would be a problem for the studios, why not make the first 10 bit encoded movies the movies that were shorter in length (say ones that are under 2 hours) ones that would normally easily fit on a 25GB disc, even with the bitrate set to max), but encode them on a 50GB disc instead. 8 bit players would play the version on the first layer, and 10 bit capable players would play the version on the second layer (with the Java code at the start to determine which player plays what)?

amirm

03-10-08, 04:39 PM

I was probably confusing the two things :) - I was just meaning one version has a lot more colours than the other, and so wouldn't have banding and wouldn't need dithering in the encode and hopefully would be a lot more true to life (if wanted - or more cinema colour range if wanted) etc.

They are two different things :). Put in full context, consumer HD video is less than 100% in the following areas:

1. 8-bit vs 10-bit which we have been talking about.

2. 4:2:0 sampling where the color resolution is reduced relative to 4:2:2 used in professional space, and 4:4:4 which you use on your computer and compositing/effects work. You have half and a quarter resolution respectively.

3. Color gamut which for HD, is only slightly better than SD standard established decades back. This determines which hues can be reproduced, not their resolution. Think of different shades of red for example.

Your eye can see #1 and #3 easier than #2.

Faceless Rebel

03-10-08, 09:12 PM

You didn't quote this part:

"Despite its issues, the Blu-ray transfer still has the clear stability and swagger of a high definition release, boasting a fresh image and a healthy bit rate. Even the squishiest details are more crisply rendered here than they are on the underwhelming standard DVD. "

:).

Great, it's better than the standard DVD. Considering what we are talking about here, the Blu-ray release, that's damning with faint praise if I ever saw it.
My impression from talking to folks is that true 1080p Anime source is very hard to come by.

Currently available in Japan on Blu-ray:
Ghost in the Shell (movie)
Ghost in the Shell 2: Innocence
Steamboy
Air (TV series) - opening sequence linked above, without banding or softness
Patlabor and Patlabor 2

Coming soon to Japan Blu-ray:
Ghost in the Shell: SAC Trilogy
Akira

Available in Japan and USA so you can see it yourself:
Freedom parts 1, 2, 3, 4 (on HD DVD)
Honneamise
Tekkon Kinkreet
Jin-Roh: The Wolf Brigade (Japan pricing for this one)

I admit that of these, I've only actually seen GitS and Innocence, and Steamboy. I've also watched Freedom part 1 but I wasn't paying that much attention to it. It had some people eating instant noodles or something. ;) Obviously I've seen the Air OP sequence, since it's linked above.

Of the few I've actually watched, I didn't notice any distracting banding and I certainly didn't see any problems with softness. GitS is not in a good condition but with the source elements they used, it was perfect. Obviously Innocence cost $20 million to make so it looks pristine, but no banding or softness there either. Steamboy was perfect as well. The Air intro you can go look at yourself, I have no need to comment on it.

It seems clear to me that there are many potential 1080p sources of anime.

amirm

03-10-08, 09:38 PM

Great, it's better than the standard DVD. Considering what we are talking about here, the Blu-ray release, that's damning with faint praise if I ever saw it.
I am not trying to bestow praise on it. I just thought it was funny that you said you don’t like lower bit rate encodes and the very reviewer you quoted, says he thought the bit rate was high :). For all I know, some of the cuts in that movie were in SD and not HD. Or that blurring was added on purpose (or by mistake).

It seems clear to me that there are many potential 1080p sources of anime.
Sorry but I see no information in what you posted that indicates they are 1080p sources. Just because they are not “soft,” it doesn’t mean they are 1080p. A 720p source can still look very sharp relative to 480p SD DVD….

jd213

03-10-08, 10:08 PM

My impression from talking to folks is that true 1080p Anime source is very hard to come by.

The "folks" you talked to don't know much about anime, then.

Most anything from the 1990s and before was done using handpainted cels captured on film, and can be scanned at 1080p or higher, just like any live action film. There's no technical reason that could limit a digitized photo of a painting to 1280x720, is there?

Theatrical and HD broadcast shows from the digital era of anime can be animated at 720p or higher. I've seen no indication that there's a vast conspiracy to really animate most of it at 720p and then advertise it as 1080p when it's released on Blu-ray, so you're going to have to provide a bit more proof if you want your above statement to have any credence.

Faceless Rebel

03-10-08, 10:13 PM

In fact, several animes which were broadcast or shown in Japan in 720p actually were made in 1080p by forward-thinking studios with suitably large budgets. GitS: SAC 1st and 2nd GIG were both digitally painted in 1080p and shown in 720p on Japanese satellite TV. The film Solid State Society is also a true 1080p source. Even recent theatrical animes with sufficient budgets are made for 1080p, Innocence was one, but so were the two Makoto Shinkai films The Place Promised in Our Early Days and 5 Centimeters Per Second, to name just two.

There are rumors that Ergo Proxy was also animated in 1080p. It was only shown in 720p, like all the others.

amirm

03-10-08, 10:45 PM

The "folks" you talked to don't know much about anime, then.
These "folks" live in Japan and run professional post production shops that produce DVDs for a living. They may not know everything, but they knew more than I did :).

Most anything from the 1990s and before was done using handpainted cels captured on film, and can be scanned at 1080p or higher, just like any live action film. There's no technical reason that could limit a digitized photo of a painting to 1280x720, is there?
I took their comment to be relevant to scans that exist today. And CGI generated in the modern age. The impression I was left with was that unlike US studios that routinely captured in 1080p for the last few years, the same was not done for Anime. This is a broad generalization and I only offer it as such relative to library of content out there.

I've seen no indication that there's a vast conspiracy to really animate most of it at 720p and then advertise it as 1080p when it's released on Blu-ray, so you're going to have to provide a bit more proof if you want your above statement to have any credence.
I did some tests a while back that showed a BD/HD DVD title which claimed to be "96Khz" audio, to be in reality 48Khz audio. And this was from someone waxing poetic about benfits of high sample rate audio. So while I am not saying anything specific about any title here, just because it shows up on BD, it is not sufficient proof for me that they are made from 1080p sources. They may well be. But the proof point put forth carries as little credence in my book :).

jd213

03-10-08, 10:56 PM

These "folks" live in Japan and run professional post production shops that produce DVDs for a living. They may not know everything, but they knew more than I did :).

If they live in Tokyo I could go talk to them and see how much they really know.

I took their comment to be relevant to scans that exist today.

There has been hundreds of hours of content released on DVD with newly scanned HD remasters, particularly from Bandai Visual. There is no shortage of anime video masters in 1080p.

sperron

03-11-08, 12:04 AM

I did some tests a while back that showed a BD/HD DVD title which claimed to be "96Khz" audio, to be in reality 48Khz audio. And this was from someone waxing poetic about benfits of high sample rate audio. So while I am not saying anything specific about any title here, just because it shows up on BD, it is not sufficient proof for me that they are made from 1080p sources. They may well be. But the proof point put forth carries as little credence in my book .

No, you showed that the electronic instruments in the soundtrack section you checked used 48Khz samples. Richard investigated and found that some of the music in fact used 96Khz samples (I believe he said the drum samples) while most of the sound track was generated with 48Khz samples. It certainly didn't hurt anything to use 96Khz.

B Leisle

03-11-08, 12:41 AM

So this thread started as Warner dropping VC-1 then we had the typical AVC versus VC-1 battle, now we're discussing, of all forsaken things, anime? Yeeeeesh.

On topic, I don't care which Warner uses, they both look good.

amirm

03-11-08, 12:46 AM

No, you showed that the electronic instruments in the soundtrack section you checked used 48Khz samples. Richard investigated and found that some of the music in fact used 96Khz samples (I believe he said the drum samples) while most of the sound track was generated with 48Khz samples.
Not at all. First of all, 90% of the title is synthesized tracks which we all agree were 48khz. Assuming for the moment that the other 10% is 96Khz, the video equiv. of this would be a title that had 90% of it encoded at 540p, yet having big bold letters advertizing "1080p." I hope we both agree that would not be right or defensible.

Turns out, the other 10% doesn't appear to be 96Khz either. I tested the drum sequence and found it to also have no high frequency content: http://www.avsforum.com/avs-vb/showpost.php?p=12881140&postcount=278

It certainly didn't hurt anything to use 96Khz.
That is like saying you are cool if a BD-50 has half of it filled with white noise :).

If higher video rates are better, why waste half the bandwidth that was used for audio instead of using it toward video?

PaulGo

03-11-08, 01:22 AM

sperron

03-11-08, 02:41 AM

If higher video rates are better, why waste half the bandwidth that was used for audio instead of using it toward video?

Blu-Ray has 8mb/s reserved for audio, so assuming the place authoring the disc chose to, they could max out the video bitrate and still use a 96/24 DTS-MA 5.1 audio track. Unlinke DVD, audio on Blu-Ray has no effect on video bitrate as long as they stay under the 8mb/s limit for all the audio tracks included on the disc. Of course there has to be room enough on the disc, so there still might be a trade off for very long movies. At any rate it's a non issue since none of the majors have been using anything better then 48/24.

On topic: sspears pretty much put this one to bed. The implementation of the codec matters as much as which codec happens to be used. Unless some professional compressionists show up and want to start disecting all the various implementations, us laymen probably just have to let this one go for now.

Grubert

03-11-08, 08:51 AM

These "folks" live in Japan and run professional post production shops that produce DVDs for a living. They may not know everything, but they knew more than I did :).

You should introduce them to the replicators who used to tell you that BD50 was science fiction. They'd get on like a house on fire.

sspears

03-11-08, 01:27 PM

Blu-Ray has 8mb/s reserved for audio

You are going to start seeing a lot more than 8 Mbps devoted to audio. I think you will see three TrueHD tracks used a lot in the near future.

One of the early BD title tests had a PCM, TrueHD and DTS-HD MA track. They were all identical. In the end, the audio used up more disc space than the video. I don't believe this ever shipped due to the pushback.

The implementation of the codec matters as much as which codec happens to be used.

Like Microsoft, the other HD optical encoders focused on progressive content first. Some can encode interlaced, but not all. The really, really sad part is some native interlaced content is being converted into 24p because that is what customers want, 24p! I have personally tried to convince them to encode interlaced as interlaced, but they must have 24p.

Ruined

03-11-08, 01:41 PM

To be fair to Richard here is his response,

http://www.avsforum.com/avs-vb/showpost.php?p=12875030&postcount=238

but why the heck are we drudging up the past? The war is over let's not start old battles again!:)

I don't know, but it's pretty clear Richard's marketing of Nature's Journey was false advertising. 48khz soundtrack (even the drum parts as amir demonstrated above) encoded at 96khz. That's like encoding a DVD's 480p transfer in 1080p and calling it "1080p Full HD Resolution".

Although, it looks like 96khz in general is "audiophile" snake oil, as this AES article proves that 96khz offers no audible improvement over 44.1khz:
http://www.aes.org/e-lib/browse.cfm?elib=14195

Clearly Richard is attempting to set apart his product from the rest by using high resolution encoding as a marketing bullet point. However, it is important to keep in mind that this does not necessarily make his product better as shown by the AES article. And some advice to Richard, if you are going to make a big stand on 96khz, next time don't use a 48khz master for your "96khz" product :)

P.S. - If Richard wants to continue to argue that there is some hidden 96khz joy in his 48khz Nature's Journey product, let's see some actual timecodes of where this 96khz magic exists on the disc.

bommai

03-11-08, 01:44 PM

Blu-Ray has 8mb/s reserved for audio, so assuming the place authoring the disc chose to, they could max out the video bitrate and still use a 96/24 DTS-MA 5.1 audio track. Unlinke DVD, audio on Blu-Ray has no effect on video bitrate as long as they stay under the 8mb/s limit for all the audio tracks included on the disc. Of course there has to be room enough on the disc, so there still might be a trade off for very long movies. At any rate it's a non issue since none of the majors have been using anything better then 48/24.

On topic: sspears pretty much put this one to bed. The implementation of the codec matters as much as which codec happens to be used. Unless some professional compressionists show up and want to start disecting all the various implementations, us laymen probably just have to let this one go for now.

Where did the 8Mbps limit for audio come from. I have a Bluray disk "Chris Botti and friends" that has the 96/24 5.1 PCM track and my PS3 displays the audio as 13.8 Mbps. Sounds great.

Joe Bloggs

03-11-08, 02:19 PM

Like Microsoft, the other HD optical encoders focused on progressive content first. Some can encode interlaced, but not all. The really, really sad part is some native interlaced content is being converted into 24p because that is what customers want, 24p! I have personally tried to convince them to encode interlaced as interlaced, but they must have 24p.
Tell them they are crazy, and that we want whatever it was originally recorded at.

nomunk

03-11-08, 05:09 PM

All I have to say is that almost all of the best looking Titles on HD-DVD and Blu-Ray are encoded in VC-1, compare that number with the titles encoded using AVC/MPEG 4, the results speak for themselves

VC-1 all the way

amirm

03-11-08, 06:34 PM

Where did the 8Mbps limit for audio come from. I have a Bluray disk "Chris Botti and friends" that has the 96/24 5.1 PCM track and my PS3 displays the audio as 13.8 Mbps. Sounds great.
Indeed, I tested Chris Botti BD and it has an honest 96Khz track. Here is the report and spectrum display of that title: http://www.avsforum.com/avs-vb/showpost.php?p=12661528&postcount=136

I think what OP meant was that BD uses a limit for video of 40mbit/sec and the last 8 mbit cannot be used for that. It is a silly limitation really. The only thing it helps with is making the decoder a bit easier to build. Then again I think most hardware decoders can go past 40 mbit/sec anyway. So whatever it was meant to help with, doesn't need it :p.

darinp2

03-11-08, 06:42 PM

Tell them they are crazy, and that we want whatever it was originally recorded at.Or at least use the original framerate or a multiple of it if they are going to convert from interlaced to progressive. For instance 1080i30 (also written as 1080i60 sometimes) being converted to 1080p30 before compression. The worst would be 1080i30 content being converted to 1080p24 because they think enthusiasts want 1080p24 more than 1080p30, when true enthusiasts want what is best for the particular content (basically like you said).

Stacey,

Is there anything people here can do to convince these guys that if they really want to deliver interlaced content as progressive they should deliver 1080i30 content as 1080p30 and should never convert 1080i30 native content to 1080p24? If there are any issues with the PS3 with 1080p30 off disc (like delivering it as interlaced) then Sony could address those with a firmware update. Maybe we should let them know that if they convert 1080i30 to 1080p24 we will try to get the word out that they made things worse. But 1080i30 to 1080p30 and they can put 1080p30 on the back of the disc case instead of 1080p24. And everybody knows that 30 is greater than 24. :) Although I wouldn't want the masses to start demanding that 1080p24 content get converted to 1080p30 because 30 is bigger.

--Darin

Lee Stewart

03-11-08, 06:44 PM

amirm

03-11-08, 06:58 PM

Amirm:

According to Wiki - VC-1 can go all the way up to 135 Mbits/sec and handle 2048x1536 resolution.

Have you had any "hands on" experience working at these levels?

http://en.wikipedia.org/wiki/VC-1
Not sure where that data has come from but it is not surprising. I once asked what the resolution limit in our free encoder was and I was told 10,000 by 10,000 pixels. But I have not tried that myself.

On peak rates, I am sure there are videos with very short peaks ranging up to those levels but I have not disected one to know for sure. Ben may have more data.

In general, the codec is resolution and bitrate independent. Unlike hardware encoders, one can input any resolution and data rate and let the codec rip. Now, decoding the thing in real-time will be an issue but not encoding.

darinp2

03-11-08, 07:03 PM

I think what OP meant was that BD uses a limit for video of 40mbit/sec and the last 8 mbit cannot be used for that. It is a silly limitation really. The only thing it helps with is making the decoder a bit easier to build. Then again I think most hardware decoders can go past 40 mbit/sec anyway. So whatever it was meant to help with, doesn't need it :p.Could help with the XBOX360 if Microsoft wanted to do a Blu-ray add-on for it. 40Mbps of AVC/MPEG4 is already quite a bit to do along with other things (like audio decoding).

You used to push that enthusiasts here should have been happy about HD DVD's overall limit of about 30Mbps compared to Blu-ray's higher limit because it made making a decoder easier, but now you are turning around and claiming it is silly for Blu-ray to have their 40Mbps limit for video. Interesting, especially as the higher you get the more there are diminishing returns, so it seems they were more justified in stopping at 40Mbps for just video than HD DVD was in stopping at ~30Mbps for everything.

--Darin

Joe Bloggs

03-11-08, 07:20 PM

Or at least use the original framerate or a multiple of it if they are going to convert from interlaced to progressive. For instance 1080i30 (also written as 1080i60 sometimes) being converted to 1080p30 before compression. The worst would be 1080i30 content being converted to 1080p24 because they think enthusiasts want 1080p24 more than 1080p30, when true enthusiasts want what is best for the particular content (basically like you said).

Stacey,

Is there anything people here can do to convince these guys that if they really want to deliver interlaced content as progressive they should deliver 1080i30 content as 1080p30 and should never convert 1080i30 native content to 1080p24? If there are any issues with the PS3 with 1080p30 off disc (like delivering it as interlaced) then Sony could address those with a firmware update. Maybe we should let them know that if they convert 1080i30 to 1080p24 we will try to get the word out that they made things worse. But 1080i30 to 1080p30 and they can put 1080p30 on the back of the disc case instead of 1080p24. And everybody knows that 30 is greater than 24. :) Although I wouldn't want the masses to start demanding that 1080p24 content get converted to 1080p30 because 30 is bigger.

--Darin

But if it was recorded at 1080i60 (or 1080i30 as some people call it :)) I think it would be better to encode at that, because if you converted it to 1080p30 you would have things that were at 2 different points in time now at the same point in time. With a 1080i60 encode, the player or the High def tv could do the de-interlacing and end up with 1080p60 - not quite as good as true 1080p60 but getting there ;)

PS: Is 1080p30 an actual supported blu-ray rate & resolution anyway - I thought 1080i60 was but not really 1080p30???

amirm

03-11-08, 07:26 PM

Could help with the XBOX360 if Microsoft wanted to do a Blu-ray add-on for it. 40Mbps of AVC/MPEG4 is already quite a bit to do along with other things (like audio decoding).
Yeh, that is one pain in the neck codec to optimize and lower bit rates help some there. Since 360 was not a consideration in design of BD spec, perhaps it was a limitation in one of the systems they had in mind (PS3? Performance of the original ASIC they had in their first gen BD recorders in Japan?).

You used to push that enthusiasts here should have been happy about HD DVD's overall limit of about 30Mbps compared to Blu-ray's higher limit because it made making a decoder easier, but now you are turning around and claiming it is silly for Blu-ray to have their 40Mbps limit for video.
You are reading too much into my post :). I am making two points. First, if you are a fan of "more is better" then you should have an issue with a hard limit at 40 when the system can do 48. That doesn't mean I am in that camp however :). The second reason has to do with the fact that I just don't like systems with hardwired partitions for system resources at a principal level. Every modern system should use dynamic resource scheduling, period. Let me give you an example of how we have moved on in this area.

Every computer system has what is called a disk cache. When you read a block from a file on disc, the operating system holds on to it ("caches" it) even after the program that read it goes away. Then if some other program wants to read the same data, it gets it from memory, rather than reading it again from disk, saving a lot of time waiting for disk. Copy a file once and then do it again. The second time it goes faster! That is the disk cache in play. Indeed, if you see the disk light blinking too much the second time, it means you don’t have enough memory in your machine.

In the old days, the disc cache was a pre-configured thing whether you talked about big-boy operating systems like Unix or DOS. You had to decide how much to set aside for it. Allocate too little and your file system performance would suffer. Make it too large, and programs would run out space, causing paging/swapping which had even a bigger performance problem.

Then came the age of dynamic disk cache. Here, the system decides the ratios on the fly. If it is short of memory for programs, it shrinks the disk cache. If it is not, it lets it grow without bounds. That way, if you have 2 gigabytes of RAM, you can allocate almost all of it to make your little laptop run faster (and use less power to boot) if all you are doing is email and browsing. Ditto for a server.

Interesting, especially as the higher you get the more there are diminishing returns, so it seems they were more justified in stopping at 40Mbps for just video than HD DVD was in stopping at ~30Mbps for everything.
--Darin
I am glad to see you and I agree on "diminishing returns" part. But if you are from the "bit meter" camp, then surely 48 is even better than 40! With no good reason given for 40 limitation, then it is an arbitrary way of disappointing those fans :D.

amirm

03-11-08, 07:31 PM

And to be complete, let me say as I have said in the past that it sucks that every stream encoded is allocated its peak/worst case bandwidth, not what it uses at that moment. It is like your gas/electric company taking the highest amount you use in a single day in a month, and using that to charge for every day of the month! It sucked in HD DVD that tools did that, and it sucks nearly as much in BD format.

It is like variable rate encoding is treated like a kludge and afterthought in these systems, rather a key principal for the design of the entire system.

It sucks even more that if there are three audio tracks, they all take up bandwidth even though you never listen to more than one. But that is a rant for another thread :D.

amirm

03-11-08, 07:35 PM

...careful, you are wandering down the "path of the troll" (again) with a remark such as this.
It was a genuine obeservation trigged by comments a few posts above that lower bit rate is the cause of many things evil. However, may apologies to anyone seeing it your way.

sspears

03-11-08, 07:37 PM

Is there anything people here can do to convince these guys that if they really want to deliver interlaced content as progressive they should deliver 1080i30 content as 1080p30 and should never convert 1080i30 native content to 1080p24?

I don't like 1080i60 to 1080p30 either. I personally think the best approach, if you insist on converting to progressive, is 1080i60 to 720p60. You are scaling down horizontally, but up vertically. ie 1920x540p (a single field) to 1280x720p (single frame). You don't introduce any odd motion problems this way, like you do going to 30p.

In the case of PiP, they have to convert to 24p because BD requires the PiP rate to be the same as the primary video rate. I consider this a good requirement, its too bad the PiP stuff is usually a mix of interlaced and progressive with telecine patterns.

As someone pointed out, this thread has gone in many directions. It may be more valuable to break them into separate threads, that way those who care about Anime can read the Anime thread and those who care about 96 kHz audio can, well, read about that too. :)

sspears

03-11-08, 07:43 PM

On BD, the video cannot go above 40, but the audio can go beyond 8. Most BDs don't have a peak of 40 Mbps. Generally they are set between 30-35 Mbps. You see peaks above 40 on the PS3 meter because of the VBV buffer.

At the end of the day, the average bitrate is more important than the peak. What do you want to look better, the talking head you can see or the fast action thats so fast its virtually a blurr?

Season six, part one of the Sorpanos was originally encoded with an average bitrate for a 30 GB disc. This was done because they had planned to use BD50. Then HBO could not get BD50, so they re-encoded the video to fit on a BD25 and there was a hit to video quality.

Joe Bloggs

03-11-08, 07:49 PM

In the case of PiP, they have to convert to 24p because BD requires the PiP rate to be the same as the primary video rate. I consider this a good requirement, its too bad the PiP stuff is usually a mix of interlaced and progressive with telecine patterns.

I think that BD limitation is not good either. It means you can't have a 1080p24 film with a 60hz PiP video comentary :(

Mr. Hanky

03-11-08, 07:56 PM

...one more reason for Hollywood to get out of the 10,000 BC age of 24 fps production. :p

darinp2

03-11-08, 08:01 PM

I am making two points. First, if you are a fan of "more is better" then you should have an issue with a hard limit at 40 when the system can do 48.As I have said elsewhere, just because I'm not going to believe somebody who claims that 60 degrees is the most comfortable room temperature for most people doesn't mean I am going to believe somebody who claims it is 85 degrees. Anybody who believes that higher bitrate is better no matter what is ignorant and so is anybody who believes that lower bitrate is better no matter what.

I wondered if I would get the, "If you think a little more is better then you must think that more is always better" kind of argument and I still wonder how rational people come up with that stuff when there are so many examples in life of where more is better when low, but not when high (where lower often becomes better). Just as simple examples that people have to deal with, the number of calories to eat per day, the number of hours to sleep per day, etc. Anybody who can make it without somebody else dressing and feeding them should understand that just because more calories per day than 100 is good for humans doesn't mean that everybody should eat more than they do, because they will be healthier. I realize this is somewhat of a tangent, but this whole class of argument that if you think things should go one direction from where they are that you must believe that continuing in that direction forever is a good thing seems ridiculous to me. Does anybody rational really fall for that kind of argument? If so, they must not really grasp how the real world works, where many ideal points are between the endpoints.
I am glad to see you and I agree on "diminishing returns" part.I've never once disagreed that there are points of diminishing returns.
But if you are from the "bit meter" camp, then surely 48 is even better than 40!A person who understands that how a format is designed as far as bitrate limitations matters doesn't have to believe that allowing 48 for video with the overall video and audio budget of 48 would be better than having the video limit at 40, as I think I pretty much covered above as far as how that is faulty logic. I could play the game of claiming that anybody who thinks that designing to 30Mbps overall was better than designing to 48Mbps overall because it makes decoding easier must also believe that designing to 10Mbps overall would be even better because somehow they must be in the "lower is better" camp no matter what, but that would be ridiculous and a waste of people's time.

--Darin

amirm

03-11-08, 08:04 PM

I don't follow most of what you are saying Darin :). So let me ask this very simply. Are you defending static limit for video as having some value?

Joe Bloggs

03-11-08, 08:17 PM

What I think is "more is usually better" in video bitrate, in that, all things being equal, it should get you closer to the original (uncompressed/less compressed original).
(except in Nature's Journey where they said more stars were visible in the lower bitrate version :)).

There may sometimes be times when more compression might look better visually to some people who dislike film grain or video noise and would prefer it more smooth, but generally a lower bitrate will look less like the original HD master you had, all other things being equal, probably :)

Mr. Hanky

03-11-08, 08:20 PM

It's a moot point, amir, as you already brought up the point that the limit isn't entirely "static", anyway, due to the operation of the video buffer (that enables higher than media-bound peak bitrates).

eci

03-11-08, 08:20 PM

...one more reason for Hollywood to get out of the 10,000 BC age of 24 fps production. :p

I sure hope they don't! I don't want movies to look like home videos. Movies need to retain the sense of unreality they have. Movies are an escape. I don't want to feel like I am watching the news.

Mr. Hanky

03-11-08, 08:28 PM

eci

03-11-08, 08:33 PM

It only "feels" like watching the news, because that is the perspective you have been born into. The reality is that it simply is what it is, and a whole new generation of moviegoers could just as easily see it as the "norm" and older movies as flickery, slideshow antiques of ages past.

I feel there is room for both "eras" to be appreciated. Certainly, dismissing one over the other simply out of the innate difference is not a particularly enlightened view.

Actually my view is *very* enlightened. You not agreeing does not make it otherwise.

I don't want to go to the movies and feel like I am watching "The Making Of" special feature. Has nothing to do with perspective whatsoever. A 60fps movie will be more "real" since that is how you see things in real life. No unreality at all.

The good thing is, no one agrees with you that matters. Movies will be 24fps for a long time to come, and certainly not because of technology deficiencies.

Joe Bloggs

03-11-08, 08:37 PM

Movies will be 24fps for a long time to come, and certainly not because of technology deficiencies.
And that will force the PiP 60hz video to be converted to 24fps too.
And as has been said, they're converting proper 60i titles (I'd imagine documentaries/concerts etc.) to 24fps because they 'think' everyone prefers it (when that's not the case).

It would be easier for a player to reduce the frame rate to 24fps than to increase it with fake motion interpolation.

eci

03-11-08, 08:39 PM

And that will force the PiP 60hz video to be converted to 24fps too.
And as has been said, they're converting proper 60i titles (say I'd imagine documentaries/concerts etc.) to 24fps because they 'think' everyone prefers it.

PiP? I thought we only cared about the movie?

What do you 'think' everyone prefers?

I know that everyone I know who sees 120HZ motionflow in action on a display thinks it looks horrible for 24fps movies. Animation, fine.

darinp2

03-11-08, 08:51 PM

I don't follow most of what you are saying Darin :). So let me ask this very simply. Are you defending static limit for video as having some value?From a design standpoint I think the 40Mbps limit for the video when the overall is 48Mbps makes a fair amount of sense. If HD DVD had gone with a 1.5x spin rate then I think limiting the video to somewhere in the 40Mbps ballpark would have made sense in their design too. At one point I proposed that HD DVD up their overall bandwidth by spinning at 1.5x, but still limit the video to 30Mbps peaks and I could have understood them going to that as a modification to try to keep their audio and secondary things from eating into the video peaks.

As far as the other stuff, which I understand may have been confusing, let me try to address it by asking a question? If somebody argued that people should sleep for only 1 hour per night to be healthiest, would you disagree? I assume you would argue that more than 1 hour of sleep per day would be better for most people. If you did that and they said that you must believe that sleeping 24 hours per day is the healthiest since you are in the "more is better" camp, would you take them seriously? Basically, the argument style of acting like somebody must believe the extreme is the best if they argue that point x isn't ideal gets old. It isn't rational.

Another example would be speed limits. If I asked you what you thought the ideal speed limit for your neighborhood is you could probably come up with a reasonable number. But somebody who wanted higher could argue that you must believe that .1 MPH would be better (since you want lower than they do) and somebody who wanted lower could argue that you must believe there shouldn't be any speed limit in your neighborhood at all (since you want higher than they do).

--Darin

Joe Bloggs

03-11-08, 08:52 PM

PiP? I thought we only cared about the movie?

I don't only care about just the movie. I'm also interested in how it was made, etc., and yes I probably would watch the PiP commentaries after I'd watched it without (I just hope it doesn't decrease the video quality too much).

They're putting PiP commentaries on the discs, and the limitations of the BD spec mean they will have to make those PiP commentaries 24fps if the film is 24fps, even though the commentaries are not fiction, and even though they may have been recorded at 60p or 60i.

What do you 'think' everyone prefers?
I think some people like some things whereas other people like other things.
And in my opinion, for live events (and to a lesser extent slow pans or fast action movies) etc. 24fps sucks.

I know that everyone I know who sees 120HZ motionflow in action on a display thinks it looks horrible for 24fps movies. Animation, fine.
I've read that some people like it for some movies, but not for others (as it can make some sets look too fake etc.). I think motion flow 'as an option' is good.

And motionflow isn't even proper 120hz or even proper 60hz. It's guessing, which causes problems. For waggon wheels etc. motionflow won't help - it just smooths the existing motion - it doesn't give the the real motion. If they increased the frame rate from 24 to something like 72 you could more easily convert it back to 24fps than convert 24fps properly to 120fps.

Joe Bloggs

03-11-08, 09:02 PM

Another example would be speed limits...
--Darin

But bitrate is not like speed limits, if you want the video to look as close to the master as possible no matter what, you are better off compressing it less I would think (as long as the player can handle that bitrate and it doesn't introduce any other problems).

darinp2

03-11-08, 09:29 PM

But bitrate is not like speed limits, if you want the video to look as close to the master as possible no matter what, you are better off compressing it less I would think (as long as the player can handle that bitrate and it doesn't introduce any other problems).I was referring more to design of a format (like the 40Mbps for video instead of 48Mbps we were discussing). Whatever limits are chosen, players need to be able to handle those. It also applies to where they should split movies across multiple discs. There are points at which not even experts with the best displays anybody could buy in the next 5 years could tell the difference between that bitrate and a higher bitrate. Just to be clear, this doesn't mean I am going to believe any deception about lower bitrates than x being better or false information about some stuff that was encoded. There are of course points of diminishing returns and points were no improvement could be perceived. Some will use that fact to try to sell unrealistically low bitrates and claim they make no difference, and then play the game that anybody who thinks higher than that would be better must be in the "higher is better" camp and so therefore must believe that no matter what, going higher would be better.

--Darin

amirm

03-11-08, 10:03 PM

From a design standpoint I think the 40Mbps limit for the video when the overall is 48Mbps makes a fair amount of sense.
--Darin
That's not the question I asked Darin. :) Once more, since the format allows 48 mbit/sec, is the restriction of the video to 40 mbit/sec in your opinion is a good thing? I am not asking whether 40 mbit/sec provides good picture or not. But rather, the merits of a decision to put a hard limit below the maximum for the format.

And please, no analogies. I can follow technical reasoning just fine :).

Mr. Hanky

03-11-08, 10:19 PM

That's not the question I asked Darin. :) Once more, since the format allows 48 mbit/sec, is the restriction of the video to 40 mbit/sec in your opinion is a good thing? I am not asking whether 40 mbit/sec provides good picture or not. But rather, the merits of a decision to put a hard limit below the maximum for the format.

You've already answered your own question with the advent of video buffering/disc caching.

Aside from that, your point becomes increasingly irrelevant when accounting for the presence of a pcm or lossless audio track and/or multiple lossy track for other languages. The "full package" is using the gross available bandwidth of the format quite fine.

What you are arguing really comes into play in the scenario where you only choose to put one lone lossy audio track to accompany the video. The only time there could have been any benefits to such an approach would have been on the limited bandwidth scenario typified by your pet format, hdvd. Given that br can match that performance threshold with extra bandwidth for video AND audio to spare on top of that, your point seems largely trivial.

amirm

03-11-08, 10:40 PM

You've already answered your own question with the advent of video buffering/disc caching.
I am not seeking knowledge. I am seeking consensus ;) :). So let’s see what Darin says.

What you are arguing really comes into play in the scenario where you only choose to put one lone lossy audio track to accompany the video.
I don’t follow what math you are using for that. A losslessDTS track at 24-bit/48Khz/5.1 could go as low as 3-4 mbit/sec . Heck, even uncompressed PCM is at 5.8 mbit/sec. So clearly we are still below 8 mbit/sec if there is only one lossless track.
The only time there could have been any benefits to such an approach would have been on the limited bandwidth scenario typified by your pet format, hdvd.
Well, you are answering the same question I did not ask Darin :). I am the last guy who needs convincing that 40 mbit/sec for video is enough ;). That is not the question. What I am asking if we all agree that there is no good reason, for you as a consumer, to like having such a restriction.

As to HD DVD, it doesn’t matter what it did or did not do. We are talking about the only HD optical format surviving. If it has a restriction like this, let’s discuss why it is there and whether it has a good reason for existing. Maybe the spec can be relaxed to let the video go up to 48 mbit/sec. If so, folks may want to push the powers to be, to make this change.
Given that br can match that performance threshold with extra bandwidth to spare on top of that, your point seems largely trivial.
To whom? To me. Sure. Per above, I am not the one to convince on this thing. But there are folks in this thread and elsewhere who have expressed the desire to be able to use the last bit of data rate available for video. I am asking, it seems for a number of times :), whether those of you arguing back, see a benefit. Not that it is “good enough” (:D).

sspears

03-11-08, 11:45 PM

and even though they may have been recorded at 60p or 60i.

Not sure if it is still the case, but I have heard some say they like to record commentaries at 60i because they like the look.

You can make 24p look more like video by not using a 180 degree shutter. I think that is currently happening on this season of Jericho. Its driving me crazy! It looks like a soap.

Joe Bloggs

03-11-08, 11:52 PM

Not sure if it is still the case, but I have heard some say they like to record commentaries at 60i because they like the look.

Maybe but it won't look exactly the same once they convert the 60i commentary to 24p if it's for PiP (as a PiP commentary for a 24p film), even though the commentary doesn't have film's 180 degree shutter.

But it's not as bad as converting 60i titles to 24p :rolleyes:

darinp2

03-12-08, 12:45 AM

That's not the question I asked Darin. :) Once more, since the format allows 48 mbit/sec, is the restriction of the video to 40 mbit/sec in your opinion is a good thing?I believe I already answered your question:
I don't follow most of what you are saying Darin :). So let me ask this very simply. Are you defending static limit for video as having some value?I most definitely am defending static limit for video as having some value for formats with rules like these have as far as things not selected currently also counting against the total bitrate. Whether 40Mbps is the best value or not could be debated, but I believe a lower static limit for the video (which both Blu-ray and HD DVD put into their designs) definitely has some value. Since you seem to believe otherwise, how about addressing whether you think 48:48 would have been better as a decision than 46:48 (we can address the real 40:48 separately). I could see how somebody who didn't understand that all tracks coming off the disc count against the 48 total would think that having the limits be 48:48 would be smarter than 46:48, but those who actually understood how things work would probably have realized that the cases where the video would need more than 46Mbps and other things came to less than 2Mbps would be almost nil. We generally aren't talking about silent movies here.

So, how about it, does your position that there this is no value in having a lower static limit for video when the overall limit is 48 (correct me if that isn't your position) apply to 46:48 or 47:48 instead of 48:48 also? If somebody proposed a limit of 46 for the video when the overall was 48, would you have said that it was silly to put anything less than 48 for the video into the design? The person proposing a lower limit like 46 for just the video would probably have understood that going with 48:48 would have meant making all the player manufacturers design to and validate something that might not ever happen in the real world where discs have audio and other things. As I said, we can address whether 40 was going too far or not, but I think those who understand how DVD and these formats work as far as all tracks coming off the disc counting against the total should understand the merits of not forcing the player manufacturers to design to and validate something that won't happen or has almost no value.

If you still want to claim there is no value in having a lower static limit for video than the total, would you argue against 47.4:48? And did you argue against HD DVD putting a lower static limit for video (although with a small gap that isn't far from the range of a single DD+ track at 640k)? If you didn't argue against that for HD DVD back when the design still could have been changed to allow the video to go up to ~30Mbps instead of topping out at 29.4Mbps, why not?

You can of course argue that things not being played currently not counting should have been changed instead of sticking with that from DVD and propose a different design for that, but as long as that is the rule, then it is part of the decision making process for designing other parts of the system.

--Darin

Mr. Hanky

03-12-08, 01:07 AM

I don’t follow what math you are using for that. A losslessDTS track at 24-bit/48Khz/5.1 could go as low as 3-4 mbit/sec . Heck, even uncompressed PCM is at 5.8 mbit/sec. So clearly we are still below 8 mbit/sec if there is only one lossless track.

Yes, so what exactly do you plan to achieve in pq improvement with the remaining audio bandwidth, if it wasn't "restricted" to only audio? Is +2.2 Mb/s going to make any difference, whatsoever, assuming you are already maxing out the 40 Mb/s video bandwidth? Is the balance any different even with 1 lossless and 2-3 lossy alt language tracks?

Any way you slice it, you would be using the bulk (or more) of that bandwidth for audio functions in a typical real product, anyway. So the "restriction" is a moot point. Whatever audio track bandwidth that is left will be too small to do video any good. At best, it could maybe be used to squeak out an extra sd pip video track, and even that would be pushing it.

So if the net impact enables a more simple approach/implementation for bandwidth legal streams of muxed video and audio, that would seem to be a design win. :eek: ...no need to scramble to come up with a fancy statistical multiplexer thingamajig to scavenge that last Mb/s of resource, right? :rolleyes:

What I am asking if we all agree that there is no good reason, for you as a consumer, to like having such a restriction.

It has no impact on the consumer, either way, so it's hardly even a "restriction" beyond a silly game of semantics. It's like asking "the consumer" if they are worried about the number of logical sectors on disc format A vs. disc format B. It's a scenario that would never come into play unless you are proposing a hyper-detail silent movie or something with a lone DD audio track.

But there are folks in this thread and elsewhere who have expressed the desire to be able to use the last bit of data rate available for video.

This sounds more like a fictional interest that you have conjured based on oversimplifying or outright taking someone out of context. Are you sure these "persons" are here and asking that, specifically, or are you just taking liberties with cherry-picked posts? Maybe you should check, first, rather than sidetrack more relevant discussions?

iceperson

03-12-08, 09:26 AM

I love how the discussion morphed from AVC and VC1 to "Hey look, a shiny penny. Do you think it was a good idea to pick up the shiny penny? Let's discuss the merits of the shiny penny vs. unshiny pennies."

Oh well, it seems some people will never tire of rehashing arguments about the format they didn't choose...

PaulGo

03-12-08, 10:36 AM

Xylon

03-12-08, 10:39 AM

I want to see screenshots :D

amirm

03-12-08, 11:14 AM

I most definitely am defending static limit for video as having some value for formats with rules like these have as far as things not selected currently also counting against the total bitrate.
I am trying to find out why you are defending it. All I have heard so far is what I volunteered. Namely, it makes it a little easier to implement. But assuming that SoCs already handle higher peaks, and PS3 has headroom to do the same, then we are down to no benefit. Right?
Since you seem to believe otherwise, how about addressing whether you think 48:48 would have been better as a decision than 46:48 (we can address the real 40:48 separately).
I have already said as a matter of sound computer science design, having 48:48 is best. You let the system/user decide how to allocate audio+video+interactivity ratios like every other modern system rather than whoever thought creating a static partition was a good idea.

I could see how somebody who didn't understand that all tracks coming off the disc count against the 48 total would think that having the limits be 48:48 would be smarter than 46:48, but those who actually understood how things work would probably have realized that the cases where the video would need more than 46Mbps and other things came to less than 2Mbps would be almost nil. We generally aren't talking about silent movies here.
You know, what they say is right. Us country folks, are slow to follow arguments :). So without using big words that we can’t understand (:D), can you please explain what can you do in 46:48 that you can't do in 48:48? That’s what I am trying to find out.

So, how about it, does your position that there this is no value in having a lower static limit for video when the overall limit is 48 (correct me if that isn't your position) apply to 46:48 or 47:48 instead of 48:48 also?
Yes, no extra functionality comes from having anything other than 48:48. Because if I wanted to, I can create any partitioning I want after the fact with 48:48. You want 40.99 and 7.01, I can give it to you if I have 48:48 :D. But you can't do the inverse. Maybe I just have a moving test signal and no audio. Why not have 48 mbit/sec for the whole thing? Why are you foreclosing that option on folks by arbitrarily taking away the last 20% of the video bandwidth? Folks paid for the drive and media which can handle it. Why not let people do with it as they please?

If somebody proposed a limit of 46 for the video when the overall was 48, would you have said that it was silly to put anything less than 48 for the video into the design?
Yes, it is even sillier to have 46:48. You would not ease implementation details yet you have to explain to someone, why you thought there had to be such a limit.

As I said, we can address whether 40 was going too far or not, but I think those who understand how DVD and these formats work as far as all tracks coming off the disc counting against the total should understand the merits of not forcing the player manufacturers to design to and validate something that won't happen or has almost no value.
So that is it then? Cheaper player trumps better specs? How much cheaper is PS3 as a result of it anyway? How about stand-alone players? And would you likewise oppose extending the format to support 4:2:2 instead of 4:2:0? I think that increases the cost too and folks could argue it has almost no value. Ditto for 10-bit vs 8-bit. If 20% of video rate is OK to hold back to save money, what is your position on these things? Do you think they can fit in 40 mbit/sec too? If not, how about getting ahead of the game by pushing to have this limit relaxed so that you can be more ready for these features? ;)

iceperson

03-12-08, 11:42 AM

I am trying to find out why you are defending it. All I have heard so far is what I volunteered. Namely, it makes it a little easier to implement. But assuming that SoCs already handle higher peaks, and PS3 has headroom to do the same, then we are down to no benefit. Right?

I have already said as a matter of sound computer science design, having 48:48 is best. You let the system/user decide how to allocate audio+video+interactivity ratios like every other modern system rather than whoever thought creating a static partition was a good idea.

You know, what they say is right. Us country folks, are slow to follow arguments :). So without using big words that we can’t understand (:D), can you please explain what can you do in 46:48 that you can't do in 48:48? That’s what I am trying to find out.

Yes, no extra functionality comes from having anything other than 48:48. Because if I wanted to, I can create any partitioning I want after the fact with 48:48. You want 40.99 and 7.01, I can give it to you if I have 48:48 :D. But you can't do the inverse. Maybe I just have a moving test signal and no audio. Why not have 48 mbit/sec for the whole thing? Why are you foreclosing that option on folks by arbitrarily taking away the last 20% of the video bandwidth? Folks paid for the drive and media which can handle it. Why not let people do with it as they please?

Yes, it is even sillier to have 46:48. You would not ease implementation details yet you have to explain to someone, why you thought there had to be such a limit.

So that is it then? Cheaper player trumps better specs? How much cheaper is PS3 as a result of it anyway? How about stand-alone players? And would you likewise oppose extending the format to support 4:2:2 instead of 4:2:0? I think that increases the cost too and folks could argue it has almost no value. Ditto for 10-bit vs 8-bit. If 20% of video rate is OK to hold back to save money, what is your position on these things? Do you think they can fit in 40 mbit/sec too? If not, how about getting ahead of the game by pushing to have this limit relaxed so that you can be more ready for these features? ;)

Great, now that that's settled let's talk about something that actually matters to the consumer. Like the color of blu ray cases. Or, maybe we can talk about deficiencies in the blu ray logo...

amirm

03-12-08, 12:02 PM

Great, now that that's settled let's talk about something that actually matters to the consumer. Like the color of blu ray cases. Or, maybe we can talk about deficiencies in the blu ray logo...

So limiting the video rate to 20% lower, has the same importance as the color of the box to you? By the way, it does to me :D. I just didn't think that was the case with other folks arguing for the best. And thinking data rate is a linear scale of quality....

Xylon, pictures are coming! But trust me, they won't put you out of the great job you do with your thread! :) Much appreciate the work you do there.

Mr. Hanky

03-12-08, 12:06 PM

If it is anything other than a silent movie, then yeah! ;) I don't know anyone here who has argued that audio tracks should be eliminated altogether from an hdm release, if it could make the video better. Doesn't seem like the consumer would really dig that, but that's just me.

iceperson

03-12-08, 12:11 PM

So limiting the video rate to 20% lower, has the same importance as the color of the box to you?.

Actually, I'm saying the color of the box is more important, because you told me that 30Mbit for audio and video was all I'd ever need and last I checked, 40 was greater than 30...:D

amirm

03-12-08, 12:21 PM

Actually, I'm saying the color of the box is more important, because you told me that 30Mbit for audio and video was all I'd ever need and last I checked, 40 was greater than 30 no matter how you slice it...:D

Once more, there is no HD DVD. So no sense is worrying what it did or did not do. I am trying to find out why the vocal supporters of "higher is better" are all of a sudden of the "good enough" mentality when it comes to this limit and keep defending something that is clearly a system limitation with no good benefit that I can think of in this day and age.

amirm

03-12-08, 12:26 PM

iceperson

03-12-08, 12:28 PM

Once more, there is no HD DVD. So no sense is worrying what it did or did not do. I am trying to find out why the vocal supporters of "higher is better" are all of a sudden of the "good enough" mentality when it comes to this limit and keep defending something that is clearly a system limitation with no good benefit that I can think of in this day and age.

The real question is this: "Is the limitation a detriment to the consumer?" If that's your argument (that you could get noticeably better video somewhere between 40.1 and 48 than you get at 40) then let's hear your case. Otherwise you're just stirring the pot.

bah, you posted while i was typing =)

Joe Bloggs

03-12-08, 12:42 PM

OK, OK, I fess up. I am having fun with you all :D. As I have said, the limit is of no real consequence. However, I would have appreciated some of you pushing the BD insiders to remove it. It adds nothing so why have it there....

By the way, the pictures ARE coming though ;) :D.
The thing is when there are multiple soundtracks (including different language tracks) on the discs they sell, it probably won't be much of an issue anyway, if at all.

Obviously there are cases when we see artefacts, but as it also needs audio, and if they're going to continue putting different language tracks on etc. there's not much that can be done (unless they could greatly increase the bitrate - by a lot more than 8 mbps, then we could get increased pixel resolution - say 4K, increased colour resolution, frame rates, etc.)

DJ79

03-12-08, 12:49 PM

I find it amusing how this system limitation (agreed by all by now?) was perverted into a selling point during the war -- dedicated 8 mbps for audio.

bjmarchini

03-12-08, 01:29 PM

I'd hate for this thread to be hijiacked into a VC1 vs AVC thread, but I hope Warner sticks with VC1 (high bit-rate of course, like New Line) as they've thus far had stellar PQ on the majority of their titles. If it ain't broke, don't fix it....

I agree to and it is alot easier to play on PCs.

Richard Paul

03-12-08, 02:12 PM

They're putting PiP commentaries on the discs, and the limitations of the BD spec mean they will have to make those PiP commentaries 24fps if the film is 24fps,I would point out that sspears considered it to be a good requirement (http://www.avsforum.com/avs-vb/showpost.php?p=13354893&postcount=200) which you didn't notice (http://www.avsforum.com/avs-vb/showpost.php?p=13354985&postcount=202) when you responded to his post. kjack, a Sigma Designs employee, has also said there are technical advantages for having that requirement. I believe that by requiring the primary and secondary video streams to have the same frame rate it allows for simpler processing compared to having to deal with video streams with two different frame rates.

I find it amusing how this system limitation (agreed by all by now?) was perverted into a selling point during the war -- dedicated 8 mbps for audio.From what I remember most Blu-ray supporters considered the 48 Mbps maximum AV bitrate to be an advantage because it allowed for a higher overall video/audio rate and not because the video rate was lower than it. Now there might have been a few poorly informed Blu-ray supporters who thought that was an advantage but there were also poorly informed HD DVD supporters. For instance I remember some HD DVD supporters who thought Blu-ray players weren't able to play back DVDs so poorly informed posters were not exclusive to any particular format.

Joe Bloggs

03-12-08, 02:48 PM

I would point out that sspears considered it to be a good requirement (http://www.avsforum.com/avs-vb/showpost.php?p=13354893&postcount=200) which you didn't notice (http://www.avsforum.com/avs-vb/showpost.php?p=13354985&postcount=202) when you responded to his post. kjack, a Sigma Designs employee, has also said there are technical advantages for having that requirement. I believe that by requiring the primary and secondary video streams to have the same frame rate it allows for simpler processing compared to having to deal with video streams with two different frame rates.

I respectfully disagree with that though. I don't think it is good for me or anyone who wants wants to watch a 60hz PiP in the corner of a 24p movie. Nor the content makers who may have 60p commentaries they want to include as a PiP extra.

It may make the job of people making the chips or the players easier, but a PiP commentary isn't fiction, what valid reason, apart from making the job of chip makers/player manufacturers job easier would there be in forcing it to be 24p (which may also mean a standards conversion has to be done too, with the artefacts that creates).

It would probably be easier for the player manufacturers to only output 720x576 video at a low frame rate instead of high definition. Would that be a good requirement? It would make their job easier.

PS: With 120hz HDTVs doing 5:5 ie. showing each frame of a 24fps 5 times, couldn't the chip makers have done something similar? eg. they could ask that the PiP video be a multiple of the main video frame rate. That should be simpler. eg. a 24p film with a 48p or 72p PiP window.

Also, didn't HD-DVD manage different frame rates for both (I didn't see any but I'm sure it did) so I don't think it can be that complicated.

darinp2

03-12-08, 02:51 PM

I am trying to find out why you are defending it. ... Wow. Is this the kind of "logic" you used when you were in design meetings? I'm assuming you've been in plenty of design meetings.
But assuming that SoCs already handle higher peaks, and PS3 has headroom to do the same, then we are down to no benefit. Right?"No benefit" is not true. There may be very small benefit, but I don't want the people who have to design and validate things to have to design and validate things that are not ever going to happen. And as another example, if a Blu-ray add-on could be made for the XBOX360 at the current limit but couldn't qualify at 48 and a Blu-ray add-on is released for the XBOX360, then that could be considered a benefit to consumers. I already mentioned the XBOX360 add-on thing and yet you conveniently ignored it for some reason (maybe because it didn't support your "no benefit" claim). And a lot of SOCs were and will be designed after the bitrate specs for the format were put in place. If somebody wants to assume the SOCs that came after the bitrate specs were put in place can actually do more than the specs require then that is fine, but it doesn't change the reason(s) for making those specs what they were.
You know, what they say is right. Us country folks, are slow to follow arguments :).Since you keep using the ridiculous argument that people must be in the "higher is better" camp from any bitrate if they think higher than some specific bitrate (like 30Mbps for everything) is good, you either don't comprehend some stuff or you want to waste people's time, or you don't care what is logical and just want to show that you are good at arguing. You can play the game like your "us country folks" above, but did you really not understand this:Another example would be speed limits. If I asked you what you thought the ideal speed limit for your neighborhood is you could probably come up with a reasonable number. But somebody who wanted higher could argue that you must believe that .1 MPH would be better (since you want lower than they do) and somebody who wanted lower could argue that you must believe there shouldn't be any speed limit in your neighborhood at all (since you want higher than they do).You are being like a person who claims that anybody who wants a higher speed limit than them must be in the "higher is better" camp, so must not want any speed limits. I would think you could understand how somebody playing that game wouldn't be logical even if you were country folk. Yet you keep doing pretty much the same thing. Why? Is it a game, or do you actually believe you are making a valid argument?
So without using big words that we can’t understand (:D), can you please explain what can you do in 46:48 that you can't do in 48:48? That’s what I am trying to find out.You are trying to find that out as if it means you must be right that any limit below the max is silly, when that just isn't the case. They don't have to be able to do something for the lower limit to have a good reason for it or for it to end up with some benefit. Especially the 29.4 instead of ~30 for HD DVD or a 47.4 instead of 48 for a format like Blu-ray (as I said, we can address whether 40:48 was going too far). There is close to zero benefit to allowing something without sound to go to higher than 47.4Mbps when it is likely that real content will never get there because the vast majority of real content has sound and as long as there is a 640k DD track backing TrueHD tracks, DTS cores of 768k and higher or PCM, the video couldn't get above 47.4Mbps. If I was in a design meeting and somebody said that the idea of making the limit for the video anything below the overall (48 in this case) was silly, I would wonder how seriously I should take their opinions. The people designing HD DVD seemed to be smart enough to understand how little value there would be in forcing designers and those validating to have to go above 29.4Mbps when the overall limit was ~30Mbps. Was that "silly"? And please don't say that HD DVD is dead so it doesn't matter if you want to have an intellectual discussion. That tactic is used when somebody doesn't have an answer and so wants to avoid the question.
And would you likewise oppose extending the format to support 4:2:2 instead of 4:2:0? I think that increases the cost too and folks could argue it has almost no value.People can argue all sorts of dumb things. Just because somebody can argue that something has almost no value doesn't mean that designers shouldn't look at what values things have (along with what they take to do). Your argument that "folks could argue it has almost no value" could be used to not even have these formats, but that doesn't mean we shouldn't. In other words, looking at things that have no value or almost no value and considering not supporting them is the smart thing to do, while ignoring anything of value just because some outlier can claim it has almost no value isn't. Extending the format to 4:2:2 instead of 4:2:0 or 10 bit instead of 8 bit are not even close as far as value propositions as limiting the video to at least 640k less than the overall, so the designs don't need be made for something that would likely never happen and validation doesn't need to be done for things that would likely never happen.
Ditto for 10-bit vs 8-bit. If 20% of video rate is OK to hold back to save money, what is your position on these things? Do you think they can fit in 40 mbit/sec too?They wouldn't necessarily need to because if they open the specs to add 4:2:2 or 4:2:0 they could address whether the bitrate limitations should stay the same or be increased for machines that support those other things.

--Darin

WTS

03-14-08, 03:09 PM

HI,

Considering this thread started out as a codec thread I'd just like to say I could care less what codec is used as long as the end product is as good as they can make. And STOP using all that video enhancement crap so people with dinky little 50" screens can say "Wow is that ever sharp looking". Yet when you really blow it up it looks like crap because of all the EE!

Blows me away why these studios/production houses think they need to do this still.

Xylon

03-16-08, 04:55 AM

Wake me up when all these pointless nerd threads are gone ;)

Just turn off the bitrate meter and watch the damn movie guys ;)

Believe it or not some members here watch their movies with the bitrate meter on all the time and judge the overall PQ of the movie and everything that comes with it. If the number doesn't reach a point to where they think excellent PQ doesn't even start ;) then a whole movie is soft and "if more bitrate were used the PQ would be better" :D Nevermind the fact that maybe the softness is attributed to the master print or :eek: style of the movie not the advance codec used.

It would be funny when the next firmware updates for BD players will show bitrates at VBR 40 mbps even tough there was an error on the software where its adding 10 mbps on the actual bitrates. And before the error is discovered because the bitrate is so high they will explain away the PQ this time because of the " . . . . . master print or style of the movie" :D

Xylon

03-16-08, 05:00 AM

Xylon

03-16-08, 05:03 AM

^^^

From what I've read here at AVS Beowulf is supposed to be the best HDM title ever made.

So far . . . . . . .

Xylon

03-16-08, 05:04 AM

Not possible. Beowulf is on HD DVD and 30GB isn't enough. Only Blu-ray can have the best HDM title ever made!

GO TO YOUR ROOM!!! AND NO PS3!!! ;)

Xylon

03-16-08, 05:08 AM

And yet the reviews say there's no PQ difference between the HD DVD and Blu-Ray version of The Prestige. Hmmm. (Yes, I do have The Prestige on HD DVD and yes it does look great. :D)

That movie is something I need to look at.

lgans316

03-16-08, 05:10 AM

Not possible. Beowulf is on HD DVD and 30GB isn't enough. Only Blu-ray can have the best HDM title ever made!

Hope that your comment was a sarcastic one.
Paramount is using modest bit rates >= 18 Mbps with preference to AVC over VC-1. Their recent releases have been top notch and they can be considered on par with Disney/Sony in terms of throwing extras on HD.

Xylon

03-16-08, 05:13 AM

Well, I was referring more to traditional style cel animation. I realize that no one actually uses cels anymore, everything is digitally painted, but the traditional type of animation has a distinct look. Both traditional and CGI animation are stunning in HD. Even something relatively simple like The Simpsons Movie looks stunning in 1080p.

Other than the obvious ringing throughout the movie. Animation released before don't have this problem.

Is it the master? transfer? :eek: codec? :)

Xylon

03-16-08, 05:22 AM

Icemage

03-16-08, 06:29 AM

I don't blame the *ahem* codec for the softness, jaggies and banding for this movie. That's the way it is. Style and look pursued by the movie.

I do blame the codec for not sufficiently given more bits. 15.60 GB VC-1 is NOT going to cut it for this kind of fast paced, lots of action movie.

Hey! Appleseed is actually optimized for BD25!!!!:p
Is that 15.6 GB video size only, or audio/video mux? If I do the math, I get something like 18.6 Mbps video average bitrate assuming that figure is for video only. Pretty low for an action film, though the fact that it's largely CGI probably helps it somewhat, as I remember the original Appleseed has no grain added (everything looked squeaky clean - clearly artistic intent).

I wouldn't blame jaggies on the codec or encoding. You don't normally get jaggies from the encoding process; that's more often coming from filtering (unlikely in this case), or in the case of CGI source like this, possibly just lazy rendering without at least one anti-aliasing pass. If the latter, then it's possible that the banding is also due to lazy rendering rather than being part of the 4:2:0 colorspace conversion or posterization due to overly low bitrates (though with the above math, this still remains a distinct possibility).

There's really no excuse for softness, though; as I mentioned above, the original presentation of the first movie was clean and crisp, and unless the director had a change of heart in the sequel (which I'll admit I have not yet seen), Ex Machina should have followed suit.