No VC-1 link.


link update ...


Also, you really need to share the full details of what encoder and parameters you used - the differences between implementations of each standard are bigger than between the two standards.

pep 1.06 with this command line.

vc1_enc.exe -i 1280x720.yuv -o Temp\720p_1.vc1 -w 1280 -h 720 -framerate 50 -frametype progressive -rate 12200 -maxrate 40000 -buffer 1843000 -gopperiod 100 -mvrange 1 -bframes 2 -bframeposopt 1 -bdeltaqp AdaptiveWeak -loopfilter 1 -complexity 2 -chromasearch 4 -motionmatch 2 -perceptual 2 -dquantstrength 0 -favorinterlevel 3 -singlethread 1 -log Temp\stat_1.txt

vc1_enc.exe -i 1280x720.yuv -o Temp\720p_2.vc1 -w 1280 -h 720 -framerate 50 -frametype progressive -rate 12200 -maxrate 40000 -buffer 1843000 -gopperiod 100 -mvrange 1 -bframes 2 -bframeposopt 1 -bdeltaqp AdaptiveWeak -loopfilter 1 -complexity 4 -chromasearch 2 -motionmatch 1 -perceptual 2 -dquantstrength 0 -favorinterlevel 3 -singlethread 1 -2pass Temp\stat_1.txt


I make test with the last VC1 encoder SDK and the result is not better.

VC-1 has its place in the world as does AVC. However, if it keeps getting shoved down people's throats as the one and only greatest thing ever invented superior to anything and everything mankind has ever known ...

You keep saying this but nobody has ever claimed this.

Thanks for your comments, I hope you reconsider and post here more often, your opinions are valued by many people who participate here or read this forum. I was thinking along the lines of you statement below about AVC and VC-1. After reading some threads on Doom9 and some other AVS threads I developed this theory about VC-1. It was created for lower bit rate applications, so it does not scale up to a higher bit rates as well as the latest version of AVC. Therefore it cannot take advantage of the increased bit rate that Blu-ray has to offer. I think your test disc will help prove or disprove this theory.

In this same thread Golgot13 (who's active on Doom9) posted the following........

http://www.avsforum.com/avs-vb/showthread.php?p=11209688#post11209688

If you talked about me, I recommand you to see my old post (first on AVS):
I was one of first supporter of HD DVD, but after some help (exactly: nothing, only wind) from some company
I change my opinion (after lot of H264 development from many companies) about VC1 and HD DVD.

H264, to my mind, it's really a codec for HD DVD. And VC1, because it optimize for high bitrate, it's a codec adapted for BluRay (it is not a joke).

Regards,

Golgot13

Maxpower1987 posted the following recently in a different thread......

http://www.avsforum.com/avs-vb/showthread.php?p=12953879#post12953879

In regard to the new HP profile of AVC, is that included in BD players? or did they go with AVC without the HP changes?

High had to be included in the BD spec, AVC is no good at 1080p HD using simple or main profile.

I think when Amir was talking about incompatibilities he was referring to stuff that used H.264 before HP was incorporated into the spec for Blu-ray.

Amir commenting on the history/origins of some of these codecs.

http://www.avsforum.com/avs-vb/showthread.php?p=12947084#post12947084

Maxpower1987 commenting on high bit rate AVC vs. VC1........

http://www.avsforum.com/avs-vb/showthread.php?p=12943159#post12943159

I'm interested in seeing more high bitrate vc-1 titles. I want a PQ war going on between these codecs. VC-1 vs AVC. Fight!

If both are used at >30Mbps you won't be able to tell the difference. Seriously at that level, they are both the same.


I believe that Maxpower's comment about the performance of these two codecs at high bit rates is correct (and I suspect visual parity is reached before 30Mbps), but I really have no way of knowing, as I don't do encoding. So a few questions for those with experience in this arena.

1. Is Maxpower's statement accurate? At >30Mbps, is the performance of the two codecs visualy the same?

2. If #1 is true, is the point of judging these codecs an excercise in finding which one does better at lower bit rates?

3. If #2 is true does this really have any relevance to Blu-ray? Doesn't a studio's choice then come down to which tool is easier to work with or offers some other form of incentive for its use?

In this same thread Golgot13 (who's active on Doom9) posted the following........

http://www.avsforum.com/avs-vb/showthread.php?p=11209688#post11209688



Maxpower1987 posted the following recently in a different thread......

http://www.avsforum.com/avs-vb/showthread.php?p=12953879#post12953879



Amir commenting on the history/origins of some of these codecs.

http://www.avsforum.com/avs-vb/showthread.php?p=12947084#post12947084

Maxpower1987 commenting on high bit rate AVC vs. VC1........

http://www.avsforum.com/avs-vb/showthread.php?p=12943159#post12943159




I believe that Maxpower's comment about the performance of these two codecs at high bit rates is correct (and I suspect visual parity is reached before 30Mbps), but I really have no way of knowing, as I don't do encoding. So a few questions for those with experience in this arena.

1. Is Maxpower's statement accurate? At >30Mbps, is the performance of the two codecs visualy the same?

2. If #1 is true, is the point of judging these codecs an excercise in finding which one does better at lower bit rates?

3. If #2 is true does this really have any relevance to Blu-ray? Doesn't a studio's choice then come down to which tool is easier to work with or offers some other form of incentive for its use?

Mostly true. But, IMHO, there are probably some hard to encode sections where a codec has to gracefully degrade even at 50 mbps. It's unclear whether the ability to do this well outweighs issues like ease of use in multi-pass encoding.

- Tom

Hi all,


Hi Ben, you have time now to finish the codec's challenge :)

I hope to prouve you that H264 is really better than VC1 and can not be compare with VC1.

I understand you want to make a comparison in High Bitrate. But in High Bitrate all codec will give same result
(the difference between two codec will be small). First, Zodiaque (Sagitaire on Doom9) and me want to be sure
that we will use same software to encode and compare the result (PSNR, SSIM, ?).
And we will use too MPEG2 because with high bitrate I'm sure VC1 and MPEG2 (best one) will give same quality
(may be MPEG2 will be better ?)

The last core of PeP is used on CinevisionPSE (because today PeP = CinevisionPSE). We will use it to encode in VC1,
Zodiaque will provide you the preset which will use, I hope you will give a preset and no piece of advice (like last time)
to show that VC1 is better than AVC.

About Sonic and CinevisionPSE, they will stop to sell it because they sell BAE-700 AVC Encoder from Sony.

For Amirm, you said that VC1 is good because lot of authoring studio use it.
But you can ask today your (2 first) partners in Europe which bought BD authoring software
and AVC encoder to use it on new HD title....
They used it because you gave it for free....


Last, I prefer to use the codec's challenge thread in Doom9 forum (because there are not lot of troll like on this forum)



Regards,



Golgot13

(posting from my phone in the airport)

PEP 1.0.6's codec implementation was from 2006 - the studios doing HDM have had several substantial new versions since then, with another on its way.

I'm not sure how relevant comparing some private experimental build of H.264 is with a commercial product is. It's no like that's what studios would use. I could use VC-1 experimental builds no one else has just as easily :).

Also, shouldn't we be picking encoding scenarios compatible with the HDM formats. We an talk about VOD codecs if desired, but I would have though that off-topic here.

A 720/50P videosource isnt the best way to judge how well something works with 1080/23,976P filmbased source.

(posting from my phone in the airport)


PEP 1.0.6's codec implementation was from 2006 - the studios doing HDM have had several substantial new versions since then, with another on its way.


Nice we will use CinevisionPSE encoder, 1.07a or b core: the last VC1 core encoder from MS :)



I'm not sure how relevant comparing some private experimental build of H.264 is with a commercial product is. It's no like that's what studios would use. I could use VC-1 experimental builds no one else has just as easily :).

I don't like to see advice. I prefer only the evidence but you don't like it (no answer from you
for a video that you promised on Doom9), we prouve that H264 is better than VC1 on Doom9 forum
and you nothing...

Yes I want that you use the best VC1 encoder compliant with HDDVD/BD specification (vbv, GoP, ... limitations).
We can use Mainconcept encoder core use on Cinevision software (Sony use Mainconcept encoder
on Bluray authoring software to encode in MPEG2 HD the still menu or slideshow).



Also, shouldn't we be picking encoding scenarios compatible with the HDM formats. We an talk about VOD codecs if desired, but I would have though that off-topic here.


For VOD, H264 is the best codec because STP, set up box, have a H264 chipset.
VC1 is good for old CPU with High Bitrate in HD or on Embeded device in SD format (with low CPU).
For VoD on XBox360, the video will be better in H264 than in VC1 but you lock the market with your box...
XBox360 support H264 because it's on HDDVD specification (and in BD specification, may be BD on XBox360... )

I really think today with all H264 decoder chipset that VC1 is not necessary.

A 720/50P videosource isnt the best way to judge how well something works with 1080/23,976P filmbased source.


No way for 25fps ans 50P, why -> it is not supported by HDDVD player (Toshiba player in Europe).
I asked MS and Toshiba last year (see my old post), my answer was "HDDVD is not for european" :mad: ...
We wait the firmware update to support it since November after announce (for european HDDVD).
The first HDDVD in Europe (05/2006) can not be played on HDDVD hardware player (Toshiba) because it is in 25fps....

The challenge must use 23.976fps because it can be play on all HD players: HDDVD player with 3:2 pulldown flag and BD player without flag
We put a patcher software to remove 3:2 pulldown flag for H264 stream in Open Source community like MS with "TelecineRemover" tool.
So my every people can use H264 23.976P video from HDDVD on BD project without recoding.

No advantage for VC1.... :rolleyes:

At the risk of offering trollish or controversial 'advice' I'd suggest not bothering now to honor HD DVD max bit rates since (IMHO) they have become somewhat less relevant.

Though I have not been following the Doom9 threads on all this recently and don't know if the various BD requirements are all even discovered yet. They were not previously during the older shoot-out.

- Tom

At the risk of offering trollish or controversial 'advice' I'd suggest not bothering now to honor HD DVD max bit rates since (IMHO) they have become somewhat less relevant.

Though I have not been following the Doom9 threads on all this recently and don't know if the various BD requirements are all even discovered yet. They were not previously during the older shoot-out.

- Tom

I agree I would like to see the tests done at the typical Blu-ray bit rates, to see what these codecs can do at its full potential.

At the risk of offering trollish or controversial 'advice' I'd suggest not bothering now to honor HD DVD max bit rates since (IMHO) they have become somewhat less relevant.

Though I have not been following the Doom9 threads on all this recently and don't know if the various BD requirements are all even discovered yet. They were not previously during the older shoot-out.

- Tom
Depending on how much work is invovled in preparing these, I would be interested in creating a bit-rate vs. performance curve. I assume that there will be some kind of "score" assigned and if that is plotted vs. various bit-rate levels we might be able to create graphs that show the relative performance gain and the "leveling off" point in terms of performance for the codecs.

Depending on how much work is invovled in preparing these, I would be interested in creating a bit-rate vs. performance curve. I assume that there will be some kind of "score" assigned and if that is plotted vs. various bit-rate levels we might be able to create graphs that show the relative performance gain and the "leveling off" point in terms of performance for the codecs.
Yes, that's typically called a "rate-distortion curve" which measures "distortion" (measured in a variety of ways) compared to bitrate. PSNR and SSIM are common metrics.

We use them all the time internally in codec developement.

I don't like to see advice. I prefer only the evidence but you don't like it (no answer from you for a video that you promised on Doom9), we prouve that H264 is better than VC1 on Doom9 forum and you nothing...
You're equally welcome to provide a 1080p24 film source to test against. I'm working on mine when I have time, but I keep getting randomized off to other projects.

I expect whatever sort of clip I make will be argued to be somehow unrepresentative or cherrypicked or something anyway, so I'm not sure how helpful it'll be to have in the end :).

For VOD, H264 is the best codec because STP, set up box, have a H264 chipset.
Most of those chipsets also support VC-1. But that's a different set of test content anyway, outside the scope here.

I agree I would like to see the tests done at the typical Blu-ray bit rates, to see what these codecs can do at its full potential.If the performance improvement curve knee hits at 20 Mbps transfer rate for the new AVC and VC-1 codecs -- then why is the 40 Mbps video Blu-ray rate the über answer? Why not 100 Mbps? When does "full potential" finally happen?

Was not the 40 Mbps Blu-ray video transfer rate derived from the need to get 2½ hours or so of quality recorded MPEG-2 hi def video on a 50 GB disc with 8 Mbps left for for 5.1 PCM audio? Why is it now the magic answer for more advanced video codecs?

If the performance improvement curve knee hits at 20 Mbps transfer rate for the new AVC and VC-1 codecs -- then why is the 40 Mbps video Blu-ray rate the über answer? Why not 100 Mbps? When does "full potential" finally happen?

Was not the Blu-ray transfer rate derived from the need to get 2½ hours or so of recorded MPEG-2 hi def video on a 50 GB disc with room for 5.1 PCM audio? Why is it now the magic answer for more advanced video codecs?

For the at least near future Blu-ray spec will be the maximum bit rate available to the mass market end user. Even now at the higher bit rates some tweaking still needs to be done for certain movie segments. To me this indicates that the AVC and VC-1 codecs either have not topped out or have already reached their limitations. So testing at these bit rates to me would give a real world comparison of how these codecs perform using conditions where they will most likely be used to their maximum potential.

For the at least near future Blu-ray spec will be the maximum bit rate available to the mass market end user. Even now at the higher bit rates some tweaking still needs to be done for certain movie segments. To me this indicates that the AVC and VC-1 codecs either have not topped out or have already reached their limitations. So testing at these bit rates to me would give a real world comparison of how these codecs perform using conditions where they will most likely be used to their maximum potential.
So is the compressionist still allowed to interact and fix problems? Who is to say that given a 29.4 Mbps peak video rate on HD DVD versus 40 Mbps on Blu-ray that the compressionist's work results will not allow both to be virtually "transparent" to the "mass market end user"?

Seems that for video coding problem areas, multiple factors come into play. Smartness of the multi-pass video encode software, time the compressionist works on problem areas, video encoding toolset/efficiency provided to the compressionist, quality assurance inspectors of the resultant video encode, and finally max video transfer rate available.

So is the compressionist still allowed to interact and fix problems? Who is to say that given a 29.4 Mbps peak video rate on HD DVD versus 40 Mbps on Blu-ray that the compressionist's work results will not allow both to be virtually "transparent" to the "mass market end user"?

Seems that for video coding problem areas, multiple factors come into play. Smartness of the multi-pass video encode software, time the compressionist works on problem areas, video encoding toolset/efficiency provided to the compressionist, quality assurance inspectors of the resultant video encode, and finally max video transfer rate available.

To me (and I am not by any means an expert) it would seem the higher the bit rate the less work a compressionist would have to do since in theory we would be getting closer to the actual product. The test I would want to see is how well the codecs could do without any special tweaking and then perhaps a set amount of tweaking - to be determined by the experts as to what is prudent. You do bring up some very good points which need to be worked out, since it is how closely the final product matches the original.

If the performance improvement curve knee hits at 20 Mbps transfer rate for the new AVC and VC-1 codecs -- then why is the 40 Mbps video Blu-ray rate the über answer? Why not 100 Mbps? When does "full potential" finally happen?

Was not the 40 Mbps Blu-ray video transfer rate derived from the need to get 2½ hours or so of quality recorded MPEG-2 hi def video on a 50 GB disc with 8 Mbps left for for 5.1 PCM audio? Why is it now the magic answer for more advanced video codecs?

It's very simple. If setting the max bit rate higher does not change anything then it is harmless and of course shoot out competitors would be allowed to set it lower if desired. But if it does change things then it shows it is sometimes necessary.

My own personal opinion is that the max bandwidth of HD DVD was mis-calculated a bit and they later would have rather had more. It was not the show stopper some said it was but I always did believe it was at least a mild problem and a design oversight.

But remember, I was talking about peak bandwidths here, not the averages. The 2 1/2 hour calculation you mention is about average bit rate.

- Tom

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.


To me (and I am not by any means an expert) it would seem the higher the bit rate the less work a compressionist would have to do since in theory we would be getting closer to the actual product. The test I would want to see is how well the codecs could do without any special tweaking and then perhaps a set amount of tweaking - to be determined by the experts as to what is prudent. You do bring up some very good points which need to be worked out, since it is how closely the final product matches the original.

This is exactly why we are creating custom content for our encoder crash test designed specifically to try and make the codec's puke .... or perhaps even projectile vomit if we are lucky...:)

At least we are being honest about it instead of denying that is our goal ... and I would not care to have MS or anyone else with a vested interest pick the content for our testing. MA and others can participate to assure best performance forn thier encoder...we have no issue with that.

However, we are a 3rd party with no vested interest in the outcome .... we are only interested in the truth. Our goal is to throw the toughest possible content at these encoders and see how they perform. We will use a variety of sources from graphics and animation to film and video footage.

I am not here to cause trouble ... but to learn and grow from my experiences. It is important that we alwasy use the best tools at our disposal. To that end, this test is important to us. Besides, we like pushing things to their breaking point ... and when they break .... it makes us smile...:)

The day we allow any machine, any technology, any equipment, any tool, or even a codec for that matter become the final end all be all in our lives ... that is the day that all forward momentum and progress shall cease.


You're equally welcome to provide a 1080p24 film source to test against. I'm working on mine when I have time, but I keep getting randomized off to other projects.

I expect whatever sort of clip I make will be argued to be somehow unrepresentative or cherrypicked or something anyway, so I'm not sure how helpful it'll be to have in the end :).


Most of those chipsets also support VC-1. But that's a different set of test content anyway, outside the scope here.

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.
Interesting. A few points of clarification:

Are you speaking of VC-1 v. H.264 as standards, or comparing implementions of them? If comparing implementations, which of each?

What specific features of VC-1 and H.264 do you think would cause them to diverge in performance at different data rates? Can you describe visually what the differences you'd expect to see at those low and high bitrates?

Lastly, what data rates do you believe H.264 was designed for that VC-1 was not?

So it seems some folks want to see VC1 vs. AVC/H.264 at Blu-ray bit rates and others want to see the results at VOD bit rates. Personally, I'm more interested in the high bit rate results as they relate to HDM.

If anyone has 8GB webbspace i could provide some footage that is really hard to compress. :)

I love the simplicity of fact and genuine honesty of your posts. Thank you for taking the time to post here and I hope you continue to participate. Your input is welcome and greatly appreciated.

Thank you.

Richard J. Casey
President / Executive Producer
R&B Films, Ltd.

Hi all,


Hi Ben, you have time now to finish the codec's challenge :)

I hope to prouve you that H264 is really better than VC1 and can not be compare with VC1.

I understand you want to make a comparison in High Bitrate. But in High Bitrate all codec will give same result
(the difference between two codec will be small). First, Zodiaque (Sagitaire on Doom9) and me want to be sure
that we will use same software to encode and compare the result (PSNR, SSIM, ?).
And we will use too MPEG2 because with high bitrate I'm sure VC1 and MPEG2 (best one) will give same quality
(may be MPEG2 will be better ?)

The last core of PeP is used on CinevisionPSE (because today PeP = CinevisionPSE). We will use it to encode in VC1,
Zodiaque will provide you the preset which will use, I hope you will give a preset and no piece of advice (like last time)
to show that VC1 is better than AVC.

About Sonic and CinevisionPSE, they will stop to sell it because they sell BAE-700 AVC Encoder from Sony.

For Amirm, you said that VC1 is good because lot of authoring studio use it.
But you can ask today your (2 first) partners in Europe which bought BD authoring software
and AVC encoder to use it on new HD title....
They used it because you gave it for free....


Last, I prefer to use the codec's challenge thread in Doom9 forum (because there are not lot of troll like on this forum)



Regards,



Golgot13

Hi Movie Swede,

DO you have rights to this footage? I would be interested in speaking with you. We are looking for more content for our encoder tests which will be distributed worldwide.

Please send em a PM if you wish to discuss.

Thank you.

Rich

If anyone has 8GB webbspace i could provide some footage that is really hard to compress. :)

Personally speaking, some softness of image and shadow detail and how grain and noise is dealt with is where I notice differences.

I am not knocking your Codec at all. It is very god at lower bit rtaes ... beter than AVC. I am only just saying that there are differences as well as advantages and benefits to each one depending on content and application.

They are both excellent codecs in my pinion...but they do yield different results. If that were not true ... this thread / argument wold nbot even exist.

Interesting. A few points of clarification:

Are you speaking of VC-1 v. H.264 as standards, or comparing implementions of them? If comparing implementations, which of each?

What specific features of VC-1 and H.264 do you think would cause them to diverge in performance at different data rates? Can you describe visually what the differences you'd expect to see at those low and high bitrates?

Lastly, what data rates do you believe H.264 was designed for that VC-1 was not?

Most of those chipsets also support VC-1. But that's a different set of test content anyway, outside the scope here.

Yes, but H264 is better why we will need to use VC1.
If I think like you I can use MPEG2 because it support by chipset...




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.


Hi RBFilms,

I give you why the authoring house prefer AVC with high bitrate:
Because they use a H264 encoder solution which work near of real time but it need
a high bitrate (more than 18-20Mbps) to give good quality (CC-HDe, Toshiba encoder, Sony encoder,...).
It's only to encode quickly the video...

Today there is many H264 encoder which work well at 10-12Mbps...
It's possible to use it at 8-10Mbps but the encoding time to give best result is long.

About VC1 it has not better performance than H264 at low bitrate...
You can go to see on the codec's challenge thread on Doom9 forum to understand this...
(If it's really true we should have HD camcorder in VC1...)

Personally speaking, some softness of image and shadow detail and how grain and noise is dealt with is where I notice differences.
You never answered if you were comparing standards or implementations, and if the latter, which implementations.

Softness in shadow detail is a... unique complaint at least when it comes to our implementation.

They are both excellent codecs in my pinion...but they do yield different results. If that were not true ... this thread / argument wold nbot even exist.
This is AVS. We're not lacking for threads arguing about all sorts of differences that don't exist :).

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.


Bulls***s and FUD is thrown by both sides. Its up to rational and intelligent posters to determine which one is closer to the truth. AVS has a lot of long time members,pros and insiders with established reputation that can see thru all this crap and to which I personally value their opinion.

Please don't accuse some people of "propaganda . . . for reasons . . . help them gain in the end" since you are an INSIDER for the blu-ray forum too.

Ben, are those people from Doom9 (of which I am a long time forum member) doing the codec shootout using the same encoding tools from the studios?

Does it matter if they don't?

Thanks.

"propaganda . . . for reasons . . . help them gain in the end"

About me.

I never say that BluRay is better or bad: you can se my old post
for HDDVD interactivity, I always said it really a good job from MS.
I don't make any propaganda for a format, BD or HDDVD (I prefer HDi).

But I want that all people say the truth about VC1 quality.

What I can say: it's hard to open eyes at some people about VC1.

H264 is better than VC1 (MS never prouve it's wrong), and it will be much better
(see developpment of MPEG2 encoder since last ten years).

But I want that all people say the truth about VC1 quality.

What I can say: it's hard to open eyes at some people about VC1.

H264 is better than VC1 (MS never prouve it's wrong), and it will be much better
(see developpment of MPEG2 encoder since last ten years).

The truth about VC-1 and AVC are they are very good codecs.

But concluding one codec is superior than the other without taking into account the process necessary to achieve the optimal results is not fair. I really hope you are not basing your conclusion from the "codec shootout" from Doom9 using those files.

There is only ONE source of material I look for PQ evaluation and that is the movie that came from those shiny discs.

I am not dismissing the work done on Doom9 (since I participate there also) but you must know what matters most to real HD movie enthusiasts is the final image presented to the screen.


Anyone proclaiming this codec (AVC or VC-1) superior in an all encompassing manner is just . . . excuse my American, full of it.

matters most to real HD movie enthusiasts is the final image presented to the screen.

Agree.


Anyone proclaiming this codec (AVC or VC-1) superior in an all encompassing manner is just . . . excuse my American, full of it.


We must to use MPEG2...
Why we must change codec?
Why we make new codec ?
.....

-> AVC and VC1 have many advantage for HD video.
But we make it to have the best quality in HD with reasonable bitrate.
We make it to have a much effient codec than MPEG2.
AVC is more efficient so AVC is better/superior than VC1
(see industry's developpment: HD Camcorder, Camera, HD TV channel, IPTV,...)

Ben, are those people from Doom9 (of which I am a long time forum member) doing the codec shootout using the same encoding tools from the studios?
Not the current version. And they aren't doing any segment reencoding, so the workflow doesn't match. Nor are they using 1080p24 film source content.

Not that they aren't doing interesting work, but Doom9 is much more focused on the "media backup" space.

Does it matter if they don't?
It means they're testing different scenarios than we're talking about here.

Not the current version. And they aren't doing any segment reencoding, so the workflow doesn't match. Nor are they using 1080p24 film source content.


Thank you. Thats all I (we) need to know here at AVS.



Not that they aren't doing interesting work, but Doom9 is much more focused on the "media backup" space.


The main purpose of the video encoding section of Doom9 is compress as much movie file size as possible with the least compromise on PQ. Original source file and to which also has the best PQ to begin with is not welcome in that section mainly of course the big file sizes. Over the years we (they) tried every method but in the end any re-encoding using commercial tools will always introduced noticeably inferior in-all-aspects-PQ.

Preserving original PQ defintely not priority but cramming as much movie in that hardrive or blank DVD-R is.



It means they're testing different scenarios than we're talking about here.

I hope it stays that way. The only thing I want discussed here is what the studio's final product delivers not from someone's PC with thousands of variables using a combination of freeware and shareware with some commercial tools.

I've received about a half dozen complaints today alone on the bickering within this thread. If I get anymore, I'm going to close the thread.

No way for 25fps ans 50P, why -> it is not supported by HDDVD player (Toshiba player in Europe).
I asked MS and Toshiba last year (see my old post), my answer was "HDDVD is not for european" :mad: ...
We wait the firmware update to support it since November after announce (for european HDDVD).
The first HDDVD in Europe (05/2006) can not be played on HDDVD hardware player (Toshiba) because it is in 25fps....

Can Blu-ray players play back 25p/50i/p HD content?

What's unusual is that that HD-DVD was created by the DVD forum, and when DVD first came out in the UK they could do 50i at their full resolution. Yet when HD-DVD comes out, they can't, over a year after being out.

If anyone has 8GB webbspace i could provide some footage that is really hard to compress. :)

Why not ...

I have many beta AVC encoder for make test ...

Here's my question about video codecs (like VC1).
Microsoft made the VC1 codec. They also make the Xbox console. The xbox console can use 3d internally for the geometry etc. '3D' CGI films (or such sequences in live action films) have 3d geometry/a simulated camera in 3d space internally before being rendered to 2D (or maybe stereoscopic in future).

My question is - why don't codecs such as VC1 use some of this data to help more efficiently compress video sequences. ie. VC1 will currently try to track stuff by looking at the final render but couldn't it have all/some of that movement data if the rendering program told it some of this data? Also, could you use some of this 3d data to help create a sort of stereoscopic image without having to have two encodes (1 for each eye) on the disc?

Also, say there's a credit sequence created entirely digitally (perhaps using a layer of text over another layer) couldn't the VC1 encode it in 2 layers as well for efficiency, and also, often, each of the letters that make up the words will probably be an exact duplicate of a previous occurrence of that letter. Do the codecs take that into account to improve efficiency - or would it not really be worth the added complexity?

All I'm interested in is if a the movie is visually indistinguishable in its entirety from the original movie when viewed by someone that knows what to look for based on the optimum viewing angle and distance at HDM’s maximum resolution.

I want the above and I think it is going to take a higher bit rate than what we see typically now, but I have no way to prove it other than the conversations here and other sites that point out the flaws in certain encodes.

I would then be interested in what the minimum peak transfer rate on the various codec’s were to deliver on the above to satisfy my own curiosity, not that it would really matter. If we can get close the ideal as stated in the first paragraph, then everyone from the guy loving his 42 inch LCD to the guy with the 120+ inch projection system should be happy.

Here's my question about video codecs (like VC1).
Microsoft made the VC1 codec. They also make the Xbox console. The xbox console can use 3d internally for the geometry etc. '3D' CGI films (or such sequences in live action films) have 3d geometry/a simulated camera in 3d space internally before being rendered to 2D (or maybe stereoscopic in future).

My question is - why don't codecs such as VC1 use some of this data to help more efficiently compress video sequences. ie. VC1 will currently try to track stuff by looking at the final render but couldn't it have all/some of that movement data if the rendering program told it some of this data? Also, could you use some of this 3d data to help create a sort of stereoscopic image without having to have two encodes (1 for each eye) on the disc?

Also, say there's a credit sequence created entirely digitally (perhaps using a layer of text over another layer) couldn't the VC1 encode it in 2 layers as well for efficiency, and also, often, each of the letters that make up the words will probably be an exact duplicate of a previous occurrence of that letter. Do the codecs take that into account to improve efficiency - or would it not really be worth the added complexity?
There has been experimentation around this idea for a while - there was a company called Object Video that tried to do automatic extraction of objects to be able to compress them individually.

The biggest problem is that 3D sources aren't available for most movies. Even if 3D was used, that gets compositied into frames with natural images, gets film grain added, etcetera. Plus, it can take 10 minutes to do a 3D render of a frame of animation that has to play back in 1/24th of a second, so it wouldn't be feasible.

That said, what you're talking about does happen after a fashion in many games, as in-engine cut scenes. Those are a lot more efficient for storage than HD video, as you suggest. But it's not a feasible solution for typical movie content.

All I'm interested in is if a the movie is visually indistinguishable in its entirety from the original movie when viewed by someone that knows what to look for based on the optimum viewing angle and distance at HDM’s maximum resolution.

I want the above and I think it is going to take a higher bit rate than what we see typically now, but I have no way to prove it other than the conversations here and other sites that point out the flaws in certain encodes.
Oh, you'd be surprised how easy it would be to start doing some empircal research yourself in this area. There are free VC-1 and H.264 encoders out there. As long as you have some video sources, you can jump in and try some encodes yourself to get a sense for how changing bitrate changes video quality.

While on the topic of HDM similarities to originals/masters etc., has anyone seen--or actually measured--updated resolution data similar to what sspears reported (http://archive.avsforum.com/avs-vb/showthread.php?postid=2498224#post2498224) here back in '03? AIUI, 1080/24p movie masters (HD-D5 tapes @ ~270 Mbps) were analyzed with a HP spectrum analyzer. Maximum equivalent effective horizontal resolution (some scenes) was 800--1300 lines. That's equivalent lines/picture width, not height, and not D-Theater tapes either, as I suggested in this post (http://www.avsforum.com/avs-vb/showthread.php?p=9314235&&#post9314235). Also, as quotes/sublinks in that link imply, typical maximum effective resolution was about 800 lines with 1100 lines being exceptional and the high end (1300 lines) limited to computer-generated images--not sampled and subject to Nyquist limiting.

Have read here that HDMs are using rescanned masters, with some even downconverted to 1080/24p from 4k digital-intermediate recordings used for production. If older 1080/24p masters are/were typically 800 lines maximum (equivalent), would a resolution boost to, say, 1000 or 1100 lines maximum result in the 'wow' look reported for some HDMs? If not, what is a typical maximum? -- John

Is it true that Blu-ray supports 1920x1080 at 50i now (and 25p?), if so, when will HD-DVD also support this (and what is the reason for the delay) or have they decided not to?

Is it true that Blu-ray supports 1920x1080 at 50i now (and 25p?), if so, when will HD-DVD also support this (and what is the reason for the delay) or have they decided not to?

Well there isnt any pro 1080/25P content released on HD DVD.

Of course for consumers its another story.

Well there isnt any pro 1080/25P content released on HD DVD.

Of course for consumers its another story.

That's because HD-DVD doesn't support 1080 at 25p :D
It would be a bit silly if Planet Earth was released on HD-DVD with thousands of copies and nobody could play it :)

Also, aren't most UK/European HD broadcasts at 1920x1080@50i or 25p?
So it looks like only Blu-ray supports this?

That's because HD-DVD doesn't support 1080 at 25p :D
It would be a bit silly if Planet Earth was released on HD-DVD with thousands of copies and nobody could play it :)

The point would be if anybody planned to release Planeth Earth in 25P they sure would give us that update.

Its a chicken and egg problem.

The point would be if anybody planned to release Planeth Earth in 25P they sure would give us that update.

Its a chicken and egg problem.
My question is - as Blu-ray seems to support it (as the new "Death Sentence" Blu-ray seems to which is 50i- as indicated by that thread on the forum) why would they frame rate convert it from 25p/50i to 24p? Was it just to use the same encode on HD-DVD and Blu-ray (to allow for the fact that HD-DVD doesn't support it) and so that they could save on doing a 2nd encode for Blu-ray which does support it? Or was it because they wanted to sell the same encode to both UK and US customers, and US HDTVs don't support 25p/50i?

My question is - as Blu-ray seems to support it (as the new "Death Sentence" Blu-ray seems to which is 50i- as indicated by that thread on the forum) why would they frame rate it from 25p to 24p? Was it just to use the same encode on HD-DVD and Blu-ray (to allow for the fact that HD-DVD doesn't support it) and so that they could save on doing a 2nd encode for Blu-ray which does support it?

Even if the player supports it, does the displays support it?

European sets do support it, but I dont know if it supported as standard in american displays.

Even if the player supports it, does the displays support it?

European sets do support it, but I dont know if it supported as standard in american displays.

Then why not give the UK the original and the best frame rate and encode - instead of lowering the frame to lower, less realistic frame rates and reducing the picture quality?

Then why not give the UK the original and the best frame rate and encode - instead of lowering the frame to lower, less realistic frame rates and reducing the picture quality?

Because the UK market isnt big enough to justify a seperate encode.

Beside going from 25P-23,976P isnt that much of a problem. Its no degradation in PQ, and the framerate difference is hard to notice (4%).

And since DVD got away all these years in Europe with the same speeddifference, I doubt the studios care that much.

Because the UK market isnt big enough to justify a seperate encode.

Beside going from 25P-23,976P isnt that much of a problem. Its no degradation in PQ, and the framerate difference is hard to notice (4%).



This was done is the VC1 encoder at the time and results in some motion judder and a little mosquitoing it says in my notes. They would have liked to have done a 25p/50i version to get the full fidelity but the players were not all supporting this so they couldn't. She is waiting for Toshiba among others to update its firmware I believe!


And since DVD got away all these years in Europe with the same speeddifference, I doubt the studios care that much
Pretty much all UK SD-DVDs have been at 50hz (with film sources that were not 25p sped up - so we never got the 3:2 pulldown motion judder - not that speeding up is the right thing either). If the studio was British/European - I think they should care.

Hi all,
we discuss about VC1 versus AVC benchmark?

About 25fps, I said on my old post about it that
"HDDVD is not for european"; "The format is for US and Japan"
Because Toshiba HDDVD player doesn't support 25fps HD
(HDDVD 25fps work on Xbox, PC and LG player...)

Hi all,
we discuss about VC1 versus AVC benchmark?

About 25fps, I said on my old post about it that
"HDDVD is not for european"; "The format is for US and Japan"
Because Toshiba HDDVD player doesn't support 25fps
(HDDVD 25fps work on Xbox, PC and LG player...)

Well as its region free its very European ;)

But hey lets move back to the orginal topic.

The problem is that there are so many different sources to consider if you gonna judge codec quality. Consumer video differs extremly much from Pro film.

You never answered if you were comparing standards or implementations, and if the latter, which implementations.

Softness in shadow detail is a... unique complaint at least when it comes to our implementation.


This is AVS. We're not lacking for threads arguing about all sorts of differences that don't exist :).
Is there any difference to the way VC1 and Mpeg4/AVC handle shadow details/ black level details or details at any of the brightness ranges, with the implementations that the major studios use? (do any of the codecs 'compress' or limit the levels used in any way eg. in a way that would make them appear brighter or darker than in the master?

In Microsoft's view, is VC1 about equal to Mpeg4/AVC when used at high bitrates (say the maximum allowed on HD-DVD or Blu-ray), assuming no tweaking of the encode is done by the person doing the encoding?

In Microsoft's view, is VC1 about equal to Mpeg4/AVC when used at high bitrates (say the maximum allowed on HD-DVD or Blu-ray), assuming no tweaking of the encode is done by the person doing the encoding?

Yes it's true ... but at high bitrate MPEG2 will produce equal result to H264/VC1. For example with BD50, H264 and VC1 are completely useless codec. 1080p MPEG2 at 20 Mbps will produce the same quality by pixel than 480p MPEG2 at 6 Mbps. And BD can use very higher bitrate than 20 Mbps for the main video stream.

I have seen many BD-50 MPEG-2 encodes and have not been impressed with any of them. MPEG-2 at 18-20 mbps isn't going to impress anyone. It is what got bluray such a bad reputation during it's 1st year. Lets put this into perspective HDTV can run MPEG-2 at 19.3 mbps. DVHS was running about 26+ mbps. 28.2 mbps minus the audio track of 576 kbps DD. Put MPEG-2 to bed and give me great looking AVC or VC1 encodes at 20 mbps which are easily twice as efficient, have less artifacting, and both smooth the individual pixels. So basically 20 mbps AVC or VC1 = 40 mbps MPEG-2 and still provides a better picture.

Yes it's true ... but at high bitrate MPEG2 will produce equal result to H264/VC1. For example with BD50, H264 and VC1 are completely useless codec. 1080p MPEG2 at 20 Mbps will produce the same quality by pixel than 480p MPEG2 at 6 Mbps. And BD can use very higher bitrate than 20 Mbps for the main video stream.

That 1080p at 20 mbps quality would probably equal 720x480p at 6 mbps only if the 1080p master did not really contain full detail, or at least an equivalent amount of detail per pixel. This may be sometimes true under current telecine practices but is not theoretically true or even true under the best masters, computer generated masters, or test patterns.

I suspect you can easily verify this by taking a detailed 720x480 clip and laying multiple copies out side by side in a 3x2 pattern, then encoding it as 2160x960p. This would have the same number of pixels as 1080p but would likely have visibly lower quality if encoded at 20mbps if the original 720x480 clip was highly detailed, moving, and reasonably hard to encode.

Generally I think the bits per pixel ratio is fairly constant for any given resolution and material except when there is no high frequency detail to encode anyway, say in upconverts or other very soft material.

- Tom

Is there any difference to the way VC1 and Mpeg4/AVC handle shadow details/ black level details or details at any of the brightness ranges, with the implementations that the major studios use? (do any of the codecs 'compress' or limit the levels used in any way eg. in a way that would make them appear brighter or darker than in the master?
In our implementation, we've done a lot of work with Differential Quantization and other techniques to avoid slight compression artifacts that would be invisible on a calibrated CRT, but can be seen on a really bright LCD. For example an 8x8 block of Y'=16 next to a 8x8 block of Y-=16.

But nothing in there would change the overall luma range. We do have a "remap" feature, so for example you could tell the encoder to treat Y'=15 as Y'=16 (mainly useful for problems in how credits get rendered), but nothing that can change overall contrast or brightness.

In Microsoft's view, is VC1 about equal to Mpeg4/AVC when used at high bitrates (say the maximum allowed on HD-DVD or Blu-ray), assuming no tweaking of the encode is done by the person doing the encoding?
Sure. All codecs converge on performance as bitrates go up, and the differences become more interesting as bitrates go down. At HD DVD rates, you'd still see a difference between MPEG-2 and VC-1. Not sure about Blu-ray; I've honestly not done much MPEG-2 at those rates.

Ben any comment about this?

I don't think we'll be seeing long term support of VC-1 going forward. MS delivered the last tools for VC-1 already and has moved away from the codec in terms of future support from what I've heard.

Contributing Writer
Home Theater Magazine
"If it isn't in HD, I'm not watching it"

Has anybody ever tried doing a 2-pass mpeg-2 encode of material that has already passed through a vc-1 encoding stage (carrying forward same rez, same bitrate)? This could yield some intriguing results, especially for more typical film-soft content. ;)

Ben any comment about this?
There's nothing about Blu-ray that makes VC-1 look worse :). I'd expect that the companies happy with VC-1 for HD DVD would be just as happy with it on Blu-ray.

Also, companies wishing to avoid VC-1 due to political issues during the format wars will now be free to adopt it on the technical merit of our implemetation.

There's nothing about Blu-ray that makes VC-1 look worse :). I'd expect that the companies happy with VC-1 for HD DVD would be just as happy with it on Blu-ray.

Also, companies wishing to avoid VC-1 due to political issues during the format wars will now be free to adopt it on the technical merit of our implemetation.

I understand your comment but the comment by Kris Deering was "I don't think we'll be seeing long term support of VC-1 going forward. MS delivered the last tools for VC-1 already and has moved away from the codec in terms of future support from what I've heard." Will Microsoft be providing the long term support for VC-1? Has Microsoft halted development of tools for VC-1? Kris Deering seems to be stating this.

I understand your comment but the comment by Kris Deering was "I don't think we'll be seeing long term support of VC-1 going forward. MS delivered the last tools for VC-1 already and has moved away from the codec in terms of future support from what I've heard." Will Microsoft be providing the long term support for VC-1? Has Microsoft halted development of tools for VC-1? Kris Deering seems to be stating this.
It depends by what you mean by tools. Now that we have the VC-1 Encoder SDK launched, we've finally enabled 3rd parties build robust VC-1 support into their own products, so it's less neccessary for us to provide our own 1st party encoding products for every market. But we've got substantial ongoing work improving our VC-1 implementation. Lots of those improvements are directly applicable towards the use of VC-1 in Blu-ray.

Also, may I also mention I picked a lousy week to go on vacation :).

It depends by what you mean by tools. Now that we have the VC-1 Encoder SDK launched, we've finally enabled 3rd parties build robust VC-1 support into their own products, so it's less neccessary for us to provide our own 1st party encoding products for every market. But we've got substantial ongoing work improving our VC-1 implementation. Lots of those improvements are directly applicable towards the use of VC-1 in Blu-ray.

Any comments on the use of VC-1 on recent Disney titles Becoming Jane and Gone Baby Gone? Becoming Jane to me looks just gorgeous.

Also, may I also mention I picked a lousy week to go on vacation :).

I suppose that depends upon whether you had planned on spending it here. ;)

- Tom

It depends by what you mean by tools. Now that we have the VC-1 Encoder SDK launched, we've finally enabled 3rd parties build robust VC-1 support into their own products, so it's less neccessary for us to provide our own 1st party encoding products for every market. But we've got substantial ongoing work improving our VC-1 implementation. Lots of those improvements are directly applicable towards the use of VC-1 in Blu-ray.


Can you elaborate on the improvements of using VC-1 in Blu-ray?

One additional question, since VC-1 was the dominant format for HD-DVD why did Toshiba develop an AVC encoder which they insisted Paramount use for its Star Trek HD-DVD releases?

Any comments on the use of VC-1 on recent Disney titles Becoming Jane and Gone Baby Gone? Becoming Jane to me looks just gorgeous.
I haven't seen or wasn't involved with either. But here's good evidence that the end of the format war will be good for VC-1 on Blu-ray.

Can you elaborate on the improvements of using VC-1 in Blu-ray?
Well, I can't say TOO much about the stuff we're working on now. But a given is always improved performance - faster encoding makes everyone happier.

And of course, we're always working on reducing the amount of manual reencoding required. We've just come up with another big wave of improvements in that regard, and there's always more we're working on. We've talked about DQuant - differential quantization - in the past, which figured out how much to compress different regions of the frame in order to achieve optimal quality at the lowest bitrate. That continues to be a big area of research and tuning.

One additional question, since VC-1 was the dominant format for HD-DVD why did Toshiba develop an AVC encoder which they insisted Paramount use for its Star Trek HD-DVD releases?
What is it that you see wrong with this? I mean Sony is a direct competitor to the other studios using BD format yet they collaborated in making discs for them. Just because companies agree on one thing, doesn't mean they agree on everything.

Toshiba is strongly behind AVC as are a number of other CE companies. They bid AVC for DVD forum codec shoot out and have built one of the best AVC encoders out there after years of R&D, well before Microsoft ran into them and decided to support HD DVD. The fact that AVC didn't get more design wins given Toshiba's strong support should be puzzling to you given your stance regarding VC-1, not what you are saying here....

One additional question, since VC-1 was the dominant format for HD-DVD why did Toshiba develop an AVC encoder which they insisted Paramount use for its Star Trek HD-DVD releases?
VC-1 was dominant for HD DVD since it was the best available codec implementation. Both formats support both codecs, so if Toshiba made a better-yet H.264 implementation, it wouldn't have hurt the format or anything.

Anyway, it's the tools and the implementaitons that matter more than the underlying codec in the end (a great MPEG-2 encoder could beat Apple's first H.264 encoder by a healthy margin!).

Well, I can't say TOO much about the stuff we're working on now. But a given is always improved performance - faster encoding makes everyone happier.

And of course, we're always working on reducing the amount of manual reencoding required. We've just come up with another big wave of improvements in that regard, and there's always more we're working on. We've talked about DQuant - differential quantization - in the past, which figured out how much to compress different regions of the frame in order to achieve optimal quality at the lowest bitrate. That continues to be a big area of research and tuning.

Thanks for the information! I hope Microsoft continues to invest and improve VC-1. The competition between both codecs will only be a win-win situation for the studios and the viewers since it will provide better tools and hopefully better results.

Where does one go to find information on Microsoft's VC-1. I can find plenty of information regarding MPEG-4/AVC available on the internet but for VC-1, everything dies at the SMPTE pay me site.

Thanks in advance.

Where does one go to find information on Microsoft's VC-1. I can find plenty of information regarding MPEG-4/AVC available on the internet but for VC-1, everything dies at the SMPTE pay me site.

Thanks in advance.

Why

http://www.microsoft.com/resources/mediaandentertainment/vc-1.mspx

Of course!

Well thanks for that reply, that link is readily available. I was looking for strictly technical information though.

The IEEE Overview of the H.264/AVC Video Coding Standard (.pdf) that is readily available?

Well thanks for that reply, that link is readily available. I was looking for strictly technical information though.
What kind of technical details are you looking for? Basically the SMPTE spec, without getting it from SMPTE?

What kind of technical details are you looking for? Basically the SMPTE spec, without getting it from SMPTE?

lol we must have hit the button at the same time! See the post above your last^

The IEEE Overview of the H.264/AVC Video Coding Standard (.pdf) that is readily available?
There is some information here: http://www.microsoft.com/windows/windowsmedia/howto/articles/vc1techoverview.aspx

There was also a more technical paper presented at a signal processing conference but I can't seem to find it....

Just what I was looking for, primer type info. Thanks!

Ben - any update on providing the samples for testing in the codec evaluation?

From what I have been able to tell from my testing (full disclosure I was at Microsoft in IPTV until recently which is a primarily h.264 division).....

h.264 is better implimented in hardware than VC-1. The best looking realtime h.264 encoders looked better than the best looking realtime vc-1 encoder.

For Video On demand you had to set the rules for a fair fight.... I could do VC-1 encodes at 3 megabit 1920x1080p 23.976 Fps that would take 100+ hours for every hour of content, that would blow away what a realtime h.264 encoder would do at 9 megabits.

If I tweaked the settings so that h.264 and VC-1 took the same amount of time on the same cpu the H.264 would win on "faster" encodes and VC-1 would win on slower encodes, and they would tie some where in the middle.

The VC-1 files would then play back on less PC than the H.264 encodes, but if you were playing back on something like a SIGMA set top box then it didn't matter they both played back the same.

If you are an operator optimizing for bandwidth costs this makes VC-1 very appealing for VOD, and H.264 very appealing for live. You can spend more time on a VOD encode and recoup your extra encode time in saved bandwidth and storage pretty quickly, and get the best quality per bit out of a realtime encoder for live.

The problem with this story is that many vod's are created from live. Literally just hit record on the Showtime feed and capture the h.264 to disk and push it to the file server to be a Vod later.

In a Shiny Disc Scenario VC-1 should win big. If you want a disc to look as breath taking as possible encode it on the slowest settings in VC-1 and you will beat H.264 every time especially with the short key frame distances required in BD and HD-DVD.

I can try and get some content to encode if you are really interested. To do a "fair" comparison you really need an H.264 god to go up against the VC-1 gods... And I like Ben and Amir too much to crush their VC-1 encodes with my H.264 encodes... And if you put my VC-1 up against any h.264 it will lose so badly that it would hardly be fair... :-)

The amount of time needed to do a real test is extensive. I spent 3 weeks doing tests internally to come up with what I shared above, and they were not conclusive. If you do encodes for BD, you might get different results than if you do them for live streaming, and each codec has sweet spots, so you'd have to do a best of 50 or so with some rules set for each of the test scenarios.

-Brandon Wirtz

Ben - any update on providing the samples for testing in the codec evaluation?
Oh, yeah, forgot to post it back here.

I did a quick sample up on a different forum, with a quarter-HD sample of it:

http://forum.doom9.org/showthread.php?t=135938

I hope to finish it up proper after NAB.

A couple of other recent blog posts involving VC-1 encoding:

http://on10.net/blogs/benwagg/21750/
http://on10.net/blogs/benwagg/21587/

Speaking of VC-1, there's also this fun Silverlight presentation about it:

http://www.microsoft.com/resources/mediaandentertainment/presentation/default.html

I can try and get some content to encode if you are really interested. To do a "fair" comparison you really need an H.264 god to go up against the VC-1 gods... And I like Ben and Amir too much to crush their VC-1 encodes with my H.264 encodes... And if you put my VC-1 up against any h.264 it will lose so badly that it would hardly be fair... :-)

Oh, don't be nice on my account; challenge accepted :).

Challenge accepted :).

Hi Ben,

you want to make a challenge ?
When we asked (Sagitaire and me) you, you after dispeared of forum...

Today, with your last versoin of PeP core, it's impossible to do better quality that H264 from x264.
And H264 codec is better than VC1 codec (MSU is agree with me...).
Your partners (first one) in Europe which have VC1 encoder (for free) start to use H264 on BD...


For Amirm, there is many H264 solution for authoring studio:
- Cinevision (Mainconcept SDK inside) from Sonic
- CinemaCraft HD encoder (CC-HDe) from CTC
- BAE-VA700 from Sony (Sonic sell it too ?)
- Toshiba H264 encoder (software in command line)
- MEI H264 Encoder (never see...)
- Nexcode encoder (last development of Tiger AVC) from Thomson
- H264 encoder from Inlet

And I don't talk about Mainconcept encoder (simple version) or Elecard encoder,...
All companies have H264 encoder on list of product:
- Ateme developed a version (which work...)
- NTT have a hardware module in real time for H264 HD( SC50KE )
- Vitec Multimedia annouce real time H264 HD encoder
.....

If you want you can make a challenge between your encoder and a H264 solution from Thomson
(I listen, but I didn't see, from my best source it is a famous encoder).


Last, I will try to say you, Ben, "hello" at NAB but it will be hard because I must to see many (much) H264 product...



Regards,



Golgot13

Thanks for the information! I hope Microsoft continues to invest and improve VC-1. The competition between both codecs will only be a win-win situation for the studios and the viewers since it will provide better tools and hopefully better results.


The war is close for optical disc support....
H264 is better than VC1 (Ben or Amirm did not prouve it's false).

There is only the VoD market for VC1 but it will restrict at XBox Live use
because Mainconcept was bought by Divx Network....
The market in europe for IPTV and VoD in HD picture size is only in H264.

Oh, yeah, forgot to post it back here.
http://on10.net/blogs/benwagg/21587/

Ben,

All I can say is Wow! The Vegas clip is very impressive.. the guys over at Adobe better get cracking on Flash updates..

Golgot13, it seems you're even more emotionally invested in VC-1 versus H.264 than I am :).

H.264 is a fine codec for lots of stuff, as is VC-1. I don't think it's useful to talk about which is "best" in the abstract, only for particular scenarios and implementations.

FWIW, it looks like the end of the format war is increasing the use of VC-1 on Blu-ray, as the political issues of using a Microsoft-originated technology go away.

Golgot13, it seems you're even more emotionally invested in VC-1 versus H.264 than I am :).

Yes, because I don't like the lie, you know :p ...


H.264 is a fine codec for lots of stuff, as is VC-1.

Yes and no. Last version of PeP core increase the quality in low bitrate but WE (and you too) can not compare H264 and VC1.



I don't think it's useful to talk about which is "best" in the abstract, only for particular scenarios and implementations.


This thread talk about the benchmark between H264 and VC1, so we want and must talk about the quality to see which is the best.
I think I test all scenarios (Vod in SD or HD, streaming in SD or HD, handheld use,...) H264 give the best result.



FWIW, it looks like the end of the format war is increasing the use of VC-1 on Blu-ray, as the political issues of using a Microsoft-originated technology go away.

About BluRay, VC1 is a good codec for this format (not like for HDDVD) because the bitrate is high (more than 18Mbps average)....
VC1 give a better quality (similar but not same at MPEG2 for same high bitrate)
because you have a good video preprocessing solution (eg Xscaler,...), without people will afraid...
But today, studio company want to put more lossless audio track so we will have less bitrate for video so H264 will be use.

In 2009, all BD title will use exclusively H264 codec (like AC3 on DVD, I don't say that VC1 is like DTS, it's like MP3) and Dolby TrueHD.
All your authoring partner companies use (or search) H264 solution too...


I wait your video (all scenario will welcome) to show at all people here or in Doom9 that H264 is better/efficient than VC1.


About authoring companies in US, I surprise by the "bad" quality of video:
I surprise by the "low", to my mind, quality with the bitrate use on BD.
Because Sony recommand to have a average bitrate higher than 20Mbps !!!??
Yes this bitrate is necessary with some encoder like encoder from Sony, Toshiba, MEI,...
But some best solution (eg special version x264 encoder which work with BD or HDDVD) we can have the same quality at 10-13Mbps average....

If some studio want to make a challenge between all their encoder and codec, it will be very nice.
But sure they do not want to show that some people (from doom9 or avsforum) can do better work than themselves and their expensive solution...

Is it true that AVC doesn't produce good results at lower bit rates than VC-1 ? Some of the members in doom9 mentioned that AVC needs to be work extremely hard to produce good results at low bit rates than VC-1. Please throw in some light on this statement.

Is it true that AVC doesn't produce good results at lower bit rates than VC-1 ?

No, H264 produce better quality than VC1 at lower bitrate. it depend too of Profile and level: restriction of software or chipset player; like VC1.


Some of the members in doom9 mentioned that AVC needs to be work extremely hard to produce good results at low bit rates than VC-1. Please throw in some light on this statement.

Can you give me the link to correct the thread on Doom9?

Thanks Golgot13 but I am afraid I don't have the link to the correct thread because I read it long time back. Why I put forth this question is based upon my viewing experience I found out VC-1 encoded discs to produce slightly artefact free picture than AVC at lower bit rates though the contents were different.

Why I put forth this question is based upon my viewing experience I found out VC-1 encoded discs to produce slightly artefact free picture than AVC at lower bit rates though the contents were different.

What software did you use for H264 and VC1?
You can send me a little video I will show you the best result of H264 development with it (PM me).

In Japan, there is nice development of H264 encoder (CinemaCraft HD from CTC, or KDDi or H264 X pass encoder from NTT,...).

Some people say with H264 codec, the bitrate can start (with best H264 encoder) at 2bit by pixel...

In 2009, all BD title will use exclusively H264 codec (like AC3 on DVD, I don't say that VC1 is like DTS, it's like MP3) and Dolby TrueHD.
Clearly wrong. At least I don't see Fox stopping to use DTS-HD Master Audio instead of TrueHD.

I think we should consider that we also have different type of sources, when we judge quality.

DV, HDCAM, AVCHD, HDV, Super35, Imax, S16 etc

There is a big difference between a bit grainy film source and homevideo.

Colors, grain, Depth of field, etc makes a huge impact.

Clearly wrong. At least I don't see Fox stopping to use DTS-HD Master Audio instead of TrueHD.

Because some BD player (first version of Samsung, LG,...) don't support Dolby TrueHD. They don't have HDMI 1.3a so it use only DTS core...
Now, all new BD player will support Dolby TrueHD (see announce from manufactor) ....

I think we should consider that we also have different type of sources, when we judge quality.

DV, HDCAM, AVCHD, HDV, Super35, Imax, S16 etc

There is a big difference between a bit grainy film source and homevideo.


Yes sure, but H264 give better result than VC1 in all scenario.

Lot of compressionist (more in US than in Europe) put default setting of their encoder !!!
And PeP give nice (not best) picture with default setting (MS save the work of US compressionist).

Different scenario need a specific setting (it depend of colour, grain,...).

To my mind I recommand to use video preprocessing of Microsoft and "special version" x264 encoder (don't need to spend lot of money...).

Yes sure, but H264 give better result than VC1 in all scenario.



Do you have any clip with filmsources to prove it?

Do you have any clip with filmsources to prove it?

Yes, but it's copyrighted and it's from client of company where I work.

There is Elephant Dream (no problem of copyright with it) but it's CG animation it is not filmsource.

You know I (Zodiaque too) wait the video promised by Ben since the end of August 2007 to prouve that VC1 can do better than H264...

You know I (Zodiaque too) wait the video promised by Ben since the end of August 2007 to prouve that VC1 can do better than H264...
And I posted a link to a rough cut a while ago!

http://forum.doom9.org/showthread.php?t=135938 (http://forum.doom9.org/showthread.php?t=135938)

I welcome your feedback!

And I posted a link to a rough cut a while ago!

http://forum.doom9.org/showthread.php?t=135938 (http://forum.doom9.org/showthread.php?t=135938)

I welcome your feedback!

Nice one month ago and on another thread...
PM me next time.

QHD is not compliant with BD format (I don't know for HDDVD),
so we could not check if your encoded file use the restriction of BD format or HDDVD format....

I really want to show before NAB that H264 is the best, but all companies at NAB know it
because they propose H264 product (very few VC1 product, one sure, your product).
it propose H264 product

There is Elephant Dream (no problem of copyright with it) but it CG animation it is not filmsource.

Perhaps, ED (Elephant Dream, not Erectile Dysfunction ;) ), can be used as an example, if someone could re-author a version of it that has "film grain" noise artificially added to it? It would still not be an "exact" representation of performance on film material, but I cannot imagine getting any closer to that than using the actual thing. It would also have the extra benefit of the film grain effect being present in a metered/controlled amount. What do you say?

QHD is not compliant with BD format (I don't know for HDDVD),
so we could not check if your encoded file use the restriction of BD format or HDDVD format...
Yes, I just posted it to show what the content looks like, so we can get some validation that it's a reasonable test.

I don't want to do all the work on the clip and have people argue it's somehow not a valid test after the fact. I want to get buy in about what needs to be in there to make it a valid test.

Perhaps, ED (Elephant Dream, not Erectile Dysfunction ;) ), can be used as an example, if someone could re-author a version of it that has "film grain" noise artificially added to it? It would still not be an "exact" representation of performance on film material, but I cannot imagine getting any closer to that than using the actual thing. It would also have the extra benefit of the film grain effect being present in a metered/controlled amount. What do you say?
That might work, but there are other ways that film is different than animation, and it'd be preferable to do something using actual film sources, like in my (yes, unbearably late) project. I changed jobs earlier in the year, so it's something I work on as I have time.

If someone has their own film source content they'd like to contribute, I'd be happy to test with someone else's :).

I actually prefer VC1 when I rip to my HTPC. It is alot easier to play on machines. Some of my machines need a little tweaking when playing back HDAVC.

VC1 seems like a fine codec. to me. I don't notice anything quality wise. I think it has to do more with the encoding than the codec itself.

That might work, but there are other ways that film is different than animation,...

This is true, and if those properties can be synthesized, then they should be introduced as "controlled" factors (similar to "x% of film grain" technique) to additional variants of the ED sample, imo. That way you have a very implicit data set of how each "film" property affects the codec's efficacy, independently. Once you have defined the primary "film property" effects that can impact the encoding, I imagine you can have a test sample variant of ED that incorporates all of the "film" properties together, in a known and controlled relation. It would greatly accelerate the understanding of which "film" properties matter and maybe some that have little/no effect (despite preconceptions to the contrary) wrt stuff that can offset the ideal encoding, using codec xyz. For all we know, maybe noise is the primary factor relative to all other known "film" properties, or vice versa, or maybe they all have a somewhat equal impact? It is really the ideal way to distill such an investigation down to discrete and controllable parameters.

Otherwise, you could use a real film sample which would certainly be as representative of the "film" effect as is possible, but then you have no control over the "film" properties- you get what you get, which may not be the same as a different film scenario. You'll have the noise effect at x% and various other "film" properties at various % and various % relative to each other...all unknown. Hence, how could you make any conclusion about the impact of "film" properties that would be relevant to any film scenario? You only have the information to say how that particular film sample interacts with the codec, and can only generalize that it should be a similar effect for all other film scenarios that are reasonably similar to that sample. It's a lot of unknowns there, and maybe too much to be of much use, altogether. Any kind of projection you develop out of that test sample will really be a "shot from the hip", because we have skipped a step in the investigation that gives us the understanding of what are the primary factors, in the first place, and at what %.

/\ while grain can by simulated on cartoons, and it would be good for some tests, I think actual real film & HD video content (ie. non-cartoon/CGI) should be used as well. This is because, although CGI can be more controlled, it won't look/be the same as filming stuff with real cameras. One codec could be better at encoding CGI content (even with added noise/grain) but the other one could be better/more accurate at "real" camera recorded content.

Yes, I'm agree about CGI video but we need a full HD movie.
Ben can you share the FHD version of your video.
We will see the encoded file on BD or HDDVD player box.

Yes, I'm agree about CGI video but we need a full HD movie.
Ben can you share the FHD version of your video.
We will see the encoded file on BD or HDDVD player box.
If I only had an analog HD video capture card in my hands, then I'd capture a scene from one of my countless HiVision LDs. I'd guess there's enough random noise in them to be a good analog source for encoding tests.

Can anyone recommend a (not too pricey) HD YUV capturing solution (PCI card, or external box)?

Hi Ben,

Yes, I just posted it to show what the content looks like, so we can get some validation that it's a reasonable test.

I wait full HD format version of your video...


I don't want to do all the work on the clip and have people argue it's somehow not a valid test after the fact. I want to get buy in about what needs to be in there to make it a valid test.

No problem, we will use "for you" the last core of PeP/CInevisionPSE (1.07a or 1.07b).
We will post all data (options, preset, metrics: PSNR, vbv,...) not like your test on your old website page about VC1 and H264...

I really want to make the comparison before NAB but I think you don't want...


Today I'm sure that x264 can produce better result than most HD studio encoder solution for BD (or HDDVD).

Can anyone recommend a (not too pricey) HD YUV capturing solution (PCI card, or external box)?

BlackMagic card, it use by professional but the price is low (start at 1000$):

http://www.blackmagic-design.com/products/hd/


Your video will be nice to test interlaced source but not progressive like video from Ben (filmsource).

BlackMagic card, it use by professional but the price is low (start at 1000$):

http://www.blackmagic-design.com/products/hd/


Your video will be nice to test interlaced source but not progressive like video from Ben (filmsource).
Thanks Golgot13. Will take a look at this card.

By the way: do you have the x264 encoder settings handy, which allows to produce a HDDVD compliant elementary stream?

Thanks Golgot13. Will take a look at this card.
By the way: do you have the x264 encoder settings handy, which allows to produce a HDDVD compliant elementary stream?

See the setting in thread "Codec Challenge" (for HDDVD and BD) on Doom9:
http://forum.doom9.org/showthread.php?t=114928

But you need a special version of x264 (for vbv compliant) and you will need to patch the stream with H264info
(Trahald from Doom9 make a good work, I showed him for PPS and 3:2 pulldown flag to convert H264 stream: HDDVD <-> BD ).

See the setting in thread "Codec Challenge" (for HDDVD and BD) on Doom9:
http://forum.doom9.org/showthread.php?t=114928

But you need a special version of x264 (for vbv compliant) and you will need to patch the stream with H264info
(Trahald from Doom9 make a good work, I showed him for PPS and 3:2 pulldown flag to convert H264 stream: HDDVD <-> BD ).
Thanks for pointing me to the right thread, that's exactly what I'm looking for!

...But you need a special version of x264 (for vbv compliant) and you will need to patch the stream with H264info
(Trahald from Doom9 make a good work, I showed him for PPS and 3:2 pulldown flag to convert H264 stream: HDDVD <-> BD ).
Which is why no one is using it for actual professional disc production.

There's a big difference between a technology and a product. We have to do a lot of beta testing and validation of a particular build before we certify it as being compatible with particular players and specs.

Which is why no one is using it for actual professional disc production.

This version is not public, it's a special version compiled by specialist of x264.
I used it on some HDDVD project (more than 2), it work well on Toshiba players.
And same video stream work well on BD players (after patch NAL/header of H264 stream).


There's a big difference between a technology and a product. We have to do a lot of beta testing and validation of a particular build before we certify it as being compatible with particular players and specs.


About your quality of your test, I remember that one guy from Doom9 show you
there was some "vbv" problem (and quality of it) with your PeP 1.06 (MS release a new version 1.07...).

On via and doom9 communities, they do too many test: you will surprise
if you know how many professionel test their plateform with x264 stream...
Today, x264 can generate a fully compliant stream with BD or HDDVD ( and DVB, IPTV, 3G, ISDB, DMB,...).


Put your video, Ben, and we will show you !

...
Put your video, Ben, and we will show you !

That does pretty well sum it up. I would like to see this.

- Tom

That does pretty well sum it up. I would like to see this.

- Tom


Same, but it will not be possible this week:
Ben participate at DVDA seminar before NAB this saturday. But I don't understand why he will talk about H264...
"Encoding for the Next Generation Discs:: MPEG-2/H.264/VC1
Presenter: Ben Waggoner, Senior Program Manager, Video Encoding, Microsoft"

I understand why, in US market, authoring studio prefer VC1. I'm sure people on this seminar will not see and know
the truth about H264 or MPEG2 on next generation disc...


I will really happy if some people of DVDA could see the video from Ben encoded in H264 before the seminar.
And I will be happy if some people can talk about this thread and the Doom9's thread (Codec's challenge) with Ben on DVDA seminar...

Last, I think today MS will not participate at next MSU (Moscow State University) video codec comparison to show the VC1 potential.
Last year they said me I talked late about it (I said it in May, MSU close the participation in August)...

Thank you for doing a very effective job of informing people and getting out the truth! :)

Same, but it will not be possible this week:
Ben participate at DVDA seminar before NAB this saturday. But I don't understand why he will talk about H264...
"Encoding for the Next Generation Discs:: MPEG-2/H.264/VC1
Presenter: Ben Waggoner, Senior Program Manager, Video Encoding, Microsoft"

I understand why, in US market, authoring studio prefer VC1. I'm sure people on this seminar will not see and know
the truth about H264 or MPEG2 on next generation disc...


I will really happy if some people of DVDA could see the video from Ben encoded in H264 before the seminar.
And I will be happy if some people can talk about this thread and the Doom9's thread (Codec's challenge) with Ben on DVDA seminar...

Last, I think today MS will not participate at next MSU (Moscow State University) video codec comparison to show the VC1 potential.
Last year they said me I talked late about it (I said it in May, MSU close the participation in August)...

That does pretty well sum it up. I would like to see this.

- Tom
Well, what I was expecting to happen is that I'd post a rough cut and people would give me some feedback on whether or not it was appropriate content for a test, and if not, what sorts of modifications would make it a better test. As it is, I've gotten about a dozen posts complaining it isn't done for every post actually trying to help the process along.

What I don't want to happen is to spend the 40-50 hours it'll take to really finish it up right and then have people tell me after the fact why it's not an appropriate test. I really want to do this as a useful community resource, not something to get turned into AVS gotcha sniping.

Yes, it's taken me a long time to get it done (and note that I'm not even on the codec team anymore), but I want to finish it up. So, for those of you anxious to see it, please, give me feedback about the rough cut on the Doom9 thread!

http://forum.doom9.org/showthread.php?t=135938

NAB is finish, so we can restart the thread to finish the challenge.
I wait the HD video....
Sure. But do you want to weigh in on modifications to the test clip? I don't think I've see any detailed response from you on that.

I surprise by all news from you at NAB and DVDA.
You said on this thread we can not compare VC1 and H264,
because many people (me first or second) say H264 is better than VC1.
But my feedback from DVDA/NAB is: you said VC1 give better result than H264
at same bitrate !!!!!
Can you show your videos and explain what software and preset did you use?
I think what you heard reported is my quoting a number of Hollywood compressionists and directors who have been comparing the latest PEP and competing H.264 encodres, who said that VC-1 is doing a much better job of preserving film grain texture and avoding blocking in blacks.

This isn't just from my testing, but a variety of studios doing A/B compares with their own in-progress Blu-ray titles. These conversations were under NDA, of course, but I'll see if I can get some attributable quotes.

Sure. But do you want to weigh in on modifications to the test clip? I don't think I've see any detailed response from you on that.

I posted a answer:
http://forum.doom9.org/showthread.php?p=1124639#post1124639

We test the codec efficiency so put what you want because we don't judge the video quality
(colour, noise, movie grain, effect, ...).
Yes the video quality will influence the result but I'm sure of H264 codec.



I think what you heard reported is my quoting a number of Hollywood compressionists and directors who have been comparing the latest PEP and competing H.264 encodres, who said that VC-1 is doing a much better job of preserving film grain texture and avoding blocking in blacks.

Yes, but I'm not sure that you use the last encoder or the best preset for H264 encoding process...
I'm sure for VC1 encoding process you take much time than H264 encoding (not same pre-processing,...)
You can put the name of H264 encoder and his preset that you use (you don't like sometime to answer a specific ask ...)


This isn't just from my testing, but a variety of studios doing A/B compares with their own in-progress Blu-ray titles. These conversations were under NDA, of course, but I'll see if I can get some attributable quotes.

I don't listen this...
You show the difference between your VC1 encoding (with pre-process) and a H264 encoding.
Strange you don't invite any company which make a H264 encoder like Mainconcept (Sonic use their SDK),
CTC, Thomson, Sony (two japaneses present at NAB and not the guy who make a demo at DVDA), ...

VC1 will die soon because H264 HD hardware encoder appear which encode in real time (I find one which can make a good quality,
because 20Mbps is enough...) and no company developp a good VC1 (except Inlet which is not compliant with BD or HDDVD).

VC1 will die soon because H264 HD hardware encoder appear which encode in real time (I find one which can make a good quality,
because 20Mbps is enough...) and no company developp a good VC1 (except Inlet which is not compliant with BD or HDDVD).
It woud be a sad, sad day if real-time encoders are used for Blu-ray. That would put speed over quality. Such devices are very inflexible and difficult to improve. With speed on their side, people will just use them as is and we would be locked in time. Such was the case with MPEG-2 encoders....

It woud be a sad, sad day if real-time encoders are used for Blu-ray. That would put speed over quality. Such devices are very inflexible and difficult to improve. With speed on their side, people will just use them as is and we would be locked in time. Such was the case with MPEG-2 encoders....

???
On last version PeP, there is more optimization for speed encoding than quality encoding (support more core, ...).
And you correct some bug, some bug find by Doom9 guy (vbv...) to increase the PeP VC1 quality.


Amirm, may be you can answer at my ask: what H264 encoder Ben use for DVDA demo and why no H264 company is invited
to discuss about video quality (it's easy to show what you want if there is nobody to say and show it is not true).


I hope to see one day your HD source to show.
It's very simple to finish and know who say true: put the video source...

On last version PeP, there is more optimization for speed encoding than quality encoding (support more core, ...).
Making things faster is good. Hardwiring them in some real-time box, using specialized DSPs and hard to program FPGAs, not.

And you correct some bug, some bug find by Doom9 guy (vbv...) to increase the PeP VC1 quality.
I am sure Microsoft is in the process of sending him a free cookie for finding a bug ;) :).


Amirm, may be you can answer at my ask: what H264 encoder Ben use for DVDA demo and why no H264 company is invited
to discuss about video quality (it's easy to show what you want if there is nobody to say and show it is not true).
I have no idea. I left Microsoft in January and don't keep up with stuff like this.

I hope to see one day your HD source to show.
It's very simple to finish and know who say true: put the video source...
The message that HD optical encoding invovles manual intervention always seems lost in arguments in this thread. You don't use a compact car to race around the track, nor use the latter to go to work everyday. Just because ever car has 4 wheels, doesn't mean they are all the same. Automated encoding using random tools here and there is of no interest or value to people creating titles people care about here. For personal archiving, sure. But not for HD optical.

I am sure Microsoft is in the process of sending him a free cookie for finding a bug ;) :).

No, because the guy has not officially the soft...:rolleyes: ;)
But it's only to prouve MS make sometime mistakes...:p



The message that HD optical encoding invovles manual intervention always seems lost in arguments in this thread. You don't use a compact car to race around the track, nor use the latter to go to work everyday. Just because ever car has 4 wheels, doesn't mean they are all the same. Automated encoding using random tools here and there is of no interest or value to people creating titles people care about here. For personal archiving, sure. But not for HD optical.

On the thread we talk about codec with HD optical restriction, so we can use any tool to my mind.
But you're right I can discuss about H264 studio professional encoder from some companies (I know 4 different package) :p :p
Some of this encoder can do better quality than VC1 from PeP, and have many pre-processing and preset for BD use....:D

Automated encoding using random tools here and there is of no interest or value to people creating titles people care about here. For personal archiving, sure. But not for HD optical.

You're right on DSPs not being the way to go, DSPs are too limited in their capabilities. FPGAs are useful only for verification as part of the answer, they are too slow to run real time.

Once you have a strong enough algorithm, you can go to verilog and solve your problem there. That and a mill or two should about do it to beat any DSP or sw solution. As long as you find people willing to take these chips off your hands, even the mill or two can go away quickly.
---

I wouldn't call the tools used in automated encoding "random". They're likely field tested and often do have tricks not known to the outside world. Besides the wisdom of not sharing your best secrets to competitors, the people making these real time encoders used in broadcast don't often feel a need to get into a pissing contest with the riff-raff ;) because the only opinions they need to worry about are how their customers who build broadcast equipment to process video think about the video quality.

FWIW, the x264 generated and even the hand tuned 'optimal' VC-1 encodes of Elephant Dreams are not quite excellent, and they're already breaking many of the limits on max bitrate imposed by the realities of broadcast to achieve their current quality, which can still be improved.

You're right on DSPs not being the way to go, DSPs are too limited in their capabilities. FPGAs are useful only for verification as part of the answer, they are too slow to run real time.
Compared to software, FPGAs do provide significant speed up. I think you are comparing them to ASICs in which case, that is true. But often, FPGAs are used to implement a fixed algorithm faster while the rest is powered by a general pupose CPU (the one in a PC or otherwise).

Once you have a strong enough algorithm, you can go to verilog and solve your problem there.
Given the low number of units sold for encoder, it is not usually economical to pay the NRE for ASICs. So the solution usually stays in the domain of DSP+FPGA or CPU+FPGA or some other similar permuation.

On last version PeP, there is more optimization for speed encoding than quality encoding (support more core, ...).

This is not true. PEP 2.1 is slower overall. Don't confuse PEP with the SDK. PEP is superior and has features the SDK does not support in terms of quality.

PEP supports the same # of cores as it always has, 4. 1.8x real-time is about that fastest PEP has ever encoded HD content. This was very expensive HP HW with a SAN that was well over $100k all by itself. Add in the cost of the HP HW and its more than my hosue. :)

BTW - VC-1 adoption is actually picking up on BD. A big shootout was done recently between Cinemcraft AVC, Sony AVC and PEP. PEP was the winner. At least this is the information that was given to us by those that did the shootout.

I posted a answer:
http://forum.doom9.org/showthread.php?p=1124639#post1124639

We test the codec efficiency so put what you want because we don't judge the video quality
(colour, noise, movie grain, effect, ...).
Yes the video quality will influence the result but I'm sure of H264 codec.
The reason why I'm interested in doing that project is to create a test clip that's relevant to the AVS-centric goal of being able to encode Hollywood movies. So I want to make a clip that matches the mix of different kinds of feature content in a realistic way. I'm looking for feedback on what aspects of features we need to capture; edits, transitions, textures, lighting, etcetera.

You seem really excited to use this clip to prove a point, but I'd appreciate your and everyone's help and feedback in making it something that's actually useful and relevant for this community.

Ben

Would a strobing clip like the high shutter scenes in Saving Private Ryan be harder to encode since every frame is much more on its own. Or does the abscense of motionblur help?

???
On last version PeP, there is more optimization for speed encoding than quality encoding (support more core, ...).
And you correct some bug, some bug find by Doom9 guy (vbv...) to increase the PeP VC1 quality.
You seem to talk a lot about PEP, but it doesn't sound like you've been using recent versions.

The "multi-core" enhancement in the recent versions is a workflow tweak to allow it to run multiple instances on the same hardware, so two workers can be running together on an 8-core blade. There's been other work around supporting newer Intel/AMD instructions. But that's all about making existing algorithms faster, not figuring out how to reduce encoding time at the expense of speed, which is what real-time encoding is all about.

Amirm, may be you can answer at my ask: what H264 encoder Ben use for DVDA demo and why no H264 company is invited
to discuss about video quality (it's easy to show what you want if there is nobody to say and show it is not true).
You can ask me, I'm right here :).

The venue only had 1024x768 projectors, so we didn't do a lot of quality demos. I did show a few SD VOD clips to demonstrate how different encoders result in different quality, comparing Apple's and Main Concept's H.264 to VC-1, and how quality improvements are limited by display in the end.

Sony talked some about their new encoder, although I had to run off to a meeting and didn't hear it all. I'm not sure who else all presented; the schedule was pretty dynamic.

Would a strobing clip like the high shutter scenes in Saving Private Ryan be harder to encode since every frame is much more on its own. Or does the abscense of motionblur help?
In general short-shutter content is harder to encode due to the lack of motion blur, and also because it's typically grainy. Fast moving sharp edges are quite a challenge. If there are repeat frames, those definitely help, but flash frames can be challenging depending on the encoder (ours does really well). Partial flash frames can be the worst, where part of the frame is flashing and the rest isn't.

Speaking of my NAB presentation, the actual deck I used is now up here:

http://www.on10.net/blogs/benwagg/22040/

I haven't seen any of the footage you've made available, but wouldn't some material similar to what was available on that Nine Inch Nails BD/HD DVD be useful as a torture test for encoders? As I recall, you (Ben) had indicated this material was particularly difficult. It's not representative of typical motion picture material, but it might be interesting to include.

I haven't seen any of the footage you've made available, but wouldn't some material similar to what was available on that Nine Inch Nails BD/HD DVD be useful as a torture test for encoders? As I recall, you (Ben) had indicated this material was particularly difficult. It's not representative of typical motion picture material, but it might be interesting to include.
I actually have some 1080i source I can throw into the mix, and it's even actually complete! Interesting stuff, if there's interest in it. I've mainly focused on the 24p stuff since that's the AVS bread-and-butter.

Here's a SD downconvert of it I did earlier (I could do a lot better):

http://silverlight.services.live.com/invoke/31260/NABTutorial1/iframe.html

I actually have some 1080i source I can throw into the mix, and it's even actually complete! Interesting stuff, if there's interest in it. I've mainly focused on the 24p stuff since that's the AVS bread-and-butter.

Here's a SD downconvert of it I did earlier (I could do a lot better):

http://silverlight.services.live.com/invoke/31260/NABTutorial1/iframe.html
That would be fine for 1080i testing (is it 1080i material that stresses encoders though? I was under the impression, at least with the NIN concert disc, that it was the frenetic pace of the show/editing, not just that it was 1080i material). Agreed that 24p content is likely more important overall for movie enthusiasts, but as a straight codec comparison, 1080i content is inevitable (and may show strengths/weaknesses with the encoder).

And here's a somewhat better QHD (960x540p30) version of it.

http://www.on10.net/blogs/benwagg/Sample-Encoder-Test-Clips/

Not sure if I can get that big a file posted on Microsoft.com, but I can set up it up on the Peer to Peer Service That AVS Doth Not Speak Of when I have a free moment.

I actually have some 1080i source I can throw into the mix, and it's even actually complete! Interesting stuff, if there's interest in it. I've mainly focused on the 24p stuff since that's the AVS bread-and-butter.

Here's a SD downconvert of it I did earlier (I could do a lot better):

http://silverlight.services.live.com/invoke/31260/NABTutorial1/iframe.html

My own opinion is 1080i going forward will be mostly real time broadcast encoders, and some camcorders. I think it's less important for a shoot out.

- Tom

My own opinion is 1080i going forward will be mostly real time broadcast encoders, and some camcorders. I think it's less important for a shoot out.

- Tom

My new camcorder improved on the old one. The old one is a panasonic HDC-SD5 stuck at 1920x1080i60 (or 1920x540p60 since they are encoded field picture...). The new one's a Canon Vixia HF10 with support for 1920x1080P30/24 ontop of the 1080i60 standard. So if this model does as well as I think it will then camcorders may very well just have 30p/24p as standard supported modes.

If anyone's interested in the 1080i source, here's a test torrent file I just did to show what 1080i ATSC should look like (but rarely does). Still some blocking artifacts in the harder scenes, but nothing too horrible, and reasonably handled by a decent deblocking decoder. Something else all ATSC decoders should have but often don't.

http://216.99.212.233:6969/torrents/TallShip_1080i_ATSC.zip.torrent?9BA0395FC0E7B5D2203F9BEF4CB3 AB54B0B8D367

Hi ben,

I read your presentation of DVDA, nice, but there is some mistake and error:

- Page 14, you said "best codec for preserving texture detail at lower bitrate"
I hope you will prouve it....

- Page 15, you said "MPEG4 Part" but it's "MPEG4 Part10"
About H264, in High Profile we can put a specific quantization matrix
to have best picture quality (and preserve texture detail a very low bitrate).

- About VOD, H264 on TS stream you know "well" ?
You know it's possible to put a very good quality of video with multiple 5.1
(or 7.1) audio stream in HE-AAC audio stream and multiple subtitle stream
(in teletext format).
I know it's "allow" to mux VC1 on TS but I don't know any professional software
which can multiplex VC1 + multi HE-AAC audio stream + one or subtitle stream....


But the rest of presentation is good. About the speed of PSE 2.1, to test
I use only the "vc1 core" not the GUI (with lot of good video preprocessing)

My new camcorder improved on the old one. The old one is a panasonic HDC-SD5 stuck at 1920x1080i60 (or 1920x540p60 since they are encoded field picture...). The new one's a Canon Vixia HF10 with support for 1920x1080P30/24 ontop of the 1080i60 standard. So if this model does as well as I think it will then camcorders may very well just have 30p/24p as standard supported modes.

I have to correct this somewhat in case people think I'm promoting the canon HF10. After examining at the mts streams between 24P, 30P & 60I, it looks like while the sensor may (or may not) capture 1080P, the avc encode from the camcorder is still field picture at 60I, --- and the 24P encode still looks like a 60I encoding with the 3:2 pulldown flags marked correctly with repeated fields in a 60I encode. Somewhat disappointing since I was hoping for a true 24P encode of 1920x1080P, but it still does get played as 24P on a TV that takes 24P input.

Neo, the point with your cam is to import the stream into the editor and then output your movie as 24P.

I think panny did this apporch so it would be compatible with all 60i sets.

Neo, the point with your cam is to import the stream into the editor and then output your movie as 24P.

I think panny did this apporch so it would be compatible with all 60i sets.

The problem I'm trying to figure out with the new AVCHD camcorders is if the CCD/CMOS sensors themselves are capturing 1080P30/24 or 1080I60. Canon offers the choice of 1080i60/1080P24/1080P30 to the user. The display will match what the TV can handle. At this point all new TVs are going to be 1080P within a year and if the camcorder handles the progressive pictures, then there's advantages to have them match --- which is why I think the true 1080 progressive modes in camcorders will be the next marketing battleground in avchd camcorders.

The difference is that in the progressive case, we get 24 or 30 still pictures every second, and each line in that picture is from the same moment in time.

With the field/interlaced modes, the ccd/cmos sensor is likely (I'm almost 100% certain on this one), only taking 540 lines of still pictures 60 times every second.

Although I can see advantages of more 'pictures'/s less resolution vs more resolution less 'pictures'/s, my preference right now is still for progressive capture.

I have a Canon HV20 which shoots both 60i and 24p embedded in the 1440x1080 HDV format. If you capture directly off the sensor using the HDMI out, the word is you can get the full 1920x1080. Anyway, I've had great success removing pulldown to get true 24p and encoding it with x264 AVC, VC-1 (WMV) or even high-bitrate mpeg2.

That said, I've had a fair amount of trouble encoding 1080i content. I'd like to reduce the bandwidth to fit an hour of video on a dual-layer DVD...averaging around 16 M bps. I've tried MPEG2, WMV and XVID. All show considerable artifacts compared to the original source. (Granted, I haven't tweaked the WMV as Ben has suggested in other forums). x264 produces excellent results but I couldn't find a way to specify field order so my PS3 sometimes played back the file with interlacing artifacts (it was possible to keep restarting the video until the field order was correct!). Today I found that embedding the x264 AVC stream in a TS file with TS-Remux seems to correct the field order problem. Great results and I can mux it together with wav, ac3 or mpa. FWIW.

With the field/interlaced modes, the ccd/cmos sensor is likely (I'm almost 100% certain on this one), only taking 540 lines of still pictures 60 times every second.

I dont think its the best solution since cmos uses a bayer pattern.

That makes it hard just to drop half the lines, since the cam need the colorvalue for every sensor to be able to calculate the final image.

It could use a zoomed in area but then they loose horisontal resolution aswell.

The easiest way would be to process the image in the cam as 1920*1080/60P and then convert it to 1920*1080/60i.

Panasonics HVX200 does process the image as 1920*1080/60P and then makes conversions to 1280*1080/60i, 960*720/60P 720*480/60i etc.

x264 produces excellent results but I couldn't find a way to specify field order so my PS3 sometimes played back the file with interlacing artifacts

Thats because HDV goes the opposite of DV in terms of field order.

Best way is to deal with it in post, before it hits the encoder.

Thats because HDV goes the opposite of DV in terms of field order.

Best way is to deal with it in post, before it hits the encoder.

Yes, HDV is top field first. x264 has an interlaced parm but no way (that I can tell) to set the field order. That seems strange to me.

In 2009, all BD title will use exclusively H264 codec (like AC3 on DVD, I don't say that VC1 is like DTS, it's like MP3) and Dolby TrueHD.
http://dvd.themanroom.com/news/Exclusive_Universal_Talks_Blu-ray_Audio_Plans/2651

:D

Oh, and I got around to sticking up the 1080i30 source with 5.1 audio:

http://216.99.212.233:6969/torrents/TallShip_lag_YUY2_5.1.avi.torrent?96548FD06299672814021EB0D5 BD9FA826024E68

I'm out of town for most of the week, so here's something for my bandwidth to do.

http://dvd.themanroom.com/news/Exclusive_Universal_Talks_Blu-ray_Audio_Plans/2651

:D


Yes I know, there is on DTS company some people like on MS company
who make lobbing (M. Lou****).
And they work well with companies which start HD with HDDVD format.
I remember that this DTS european guy was very friendly with MS
(good HDDVD partisans)

Yes I know, there is on DTS company some people like on MS company who make lobbing (M. Lou****).
So there is no person in Dolby who tries to get their codec to be used? That business just falls on their lap?

Two codecs were approved to be part of the spec so that content owners would have a choice of which one to use. As such, it is natural to see competition between the companies to get design wins. If one of them builds a more positive relationship with a studio and promises better support, etc., and as a result gets the business, more power to them.

And they work well with companies which start HD with HDDVD format.
I remember that this DTS european guy was very friendly with MS
(good HDDVD partisans)
I am sure that is the reason. Universal who has been using Dolby technologies up until now, woke up one morning and said, "wait, there is this guy who liked MS in Europe, let's use DTS codec from here on!!!"

Say, do you have an explanation of why Fox uses DTS lossless also? Did they not know that DTS was in bed with MS and HD DVD?

http://dvd.themanroom.com/news/Exclusive_Universal_Talks_Blu-ray_Audio_Plans/2651

:D

Per Universal, all three Mummy films will include DTS-HD Master Lossless Audio. That's right, folks. Instead of lazily porting over what was already completed for HD DVD, Universal has heard your cries for lossless audio and is taking action.
What are they doing for the video? Lazily porting over what was already completed for the HD DVD?

Universal has heard your cries for lossless audio and is taking action
Universal, can we have 4000x2000 lossless video too please (with the best quality film scans)?

What are they doing for the video? Lazily porting over what was already completed for the HD DVD?

All of the currently announced titles have been re-encoded. I can't imagine they would do this for every catalog title, but who knows.

The facilites in Europe were unable to get the Dolby encoder at the time. DTS and Sonic were the only two willing to help them. They tried to use TrueHD but were unable to get access to the encoder. At that time it was a PC app. The official product was Mac based and shipped much later. By this time they had already purchased DTS, so no point in spending more money.

All of the currently announced titles have been re-encoded. I can't imagine they would do this for every catalog title, but who knows.
FWIW, many European Universal HD DVDs have new encodings. E.g. the VC-1 encodings of Waterworld, Scorpion King, Bourne Identity and Out of Sight are all different compared to the USA discs. So it seems to me that Universal generally isn't shy of reencoding.

I bet they skimped on the b-frames! :p

FWIW, many European Universal HD DVDs have new encodings. E.g. the VC-1 encodings of Waterworld, Scorpion King, Bourne Identity and Out of Sight are all different compared to the USA discs. So it seems to me that Universal generally isn't shy of reencoding.
No they re-encode the European HD-DVDs at lower video bitrates probably because they decide to add loads of unwanted languages and use less extras for the European versions. So the higher quality and more feature packed versions won't be the European ones.

I bet they skimped on the b-frames! :p

You say that kiddingly but the silly B-frame restriction in BD is probably the reason some encode cannot be used as is.

You say that kiddingly but the silly B-frame restriction in BD is probably the reason some encode cannot be used as is.
Does that mean they have to use more I or P frames? Shouldn't that mean that trick play functions (fast forward/rewind) will work faster/smoother with less B frames?

You say that kiddingly but the silly B-frame restriction in BD is probably the reason some encode cannot be used as is.

Naturally, another example of "standards drift" employed by certain companies to trip-up would-be competition, I'm sure. :rolleyes:

On a more serious note, I think the prevailing notion is that we DON'T want recycled hdvd encodes, anyway, which were likely targeted for a lower bitrate/lower capacity medium, in the first place. So your odd inclusion referencing b-frame with bd as a negative association is the usual red herring.

Does that mean they have to use more I or P frames? Shouldn't that mean that trick play functions (fast forward/rewind) will work faster/smoother with less B frames?

Fewer B frames means more P frames. I frames should not really change. In general we recommend 1 B frame (vs. 2 B frames) for the best quality. B frames are quantized more, which reduces the bit usage. (ie quality)

One way to implement fast forward is to discard B frames. A P frame only encode makes this more difficult. All of the original WMV HD discs were P frame only (ie no B frames, still had I frames). I believe the HD DVD of Cinderella Man is P frame only.

One way to implement fast forward is to discard B frames. A P frame only encode makes this more difficult. All of the original WMV HD discs were P frame only (ie no B frames, still had I frames). I believe the HD DVD of Cinderella Man is P frame only.
But discarding the B frames would make the fast forward more choppy than increasing the frame rate so that the 24p frames went by at a higher frame rate, like 60 fps (assuming the player could handle that data rate - if it couldn't, I suppose they'd have to drop frames like B frames till they could handle the data rate).

On a more serious note, I think the prevailing notion is that we DON'T want recycled hdvd encodes, anyway, which were likely targeted for a lower bitrate/lower capacity medium, in the first place. So your odd inclusion referencing b-frame with bd as a negative association is the usual red herring.
Lack of portability of content means much slower rate of availability of 150+ Uni titles in BD. While some people here may wish for a new encode, there is a much bigger hit to the BD market in general, due to reduced title count in the interim. Indeed, some titles may not get moved or be delayed for a few years if the cost doesn't justify the revenues.

Red herring? Be sure to read the definition in a proper business dictionary ;).

Lack of portability of content means much slower rate of availability of 150+ Uni titles in BD. While some people here may wish for a new encode, there is a much bigger hit to the BD market in general, due to reduced title count in the interim. Indeed, some titles may not get moved or be delayed for a few years if the cost doesn't justify the revenues.

Red herring? Be sure to read the definition in a proper business dictionary ;).
If there's no new video encode at higher bitrate (and no new film scans where necessary), for people who already have the HD-DVD there would be less incentive for them to buy the Blu-ray.

If there's no new video encode at higher bitrate (and no new film scans where necessary), for people who already have the HD-DVD there would be less incentive for them to buy the Blu-ray.
The market for most titles is quite small. I suspect the market for HD DVD owners wanting to re-buy a title in BD due to a re-encode, is a fraction of that :).

My comment above was for BD market moving forward, i.e. for new users. For them, having 150 titles now, is better than a handful rencoded and rest, well, resting on a shelf :).

Yes I know, there is on DTS company some people like on MS company
who make lobbing (M. Lou****).
And they work well with companies which start HD with HDDVD format.
I remember that this DTS european guy was very friendly with MS
(good HDDVD partisans)

What!?!

No they re-encode the European HD-DVDs at lower video bitrates probably because they decide to add loads of unwanted languages and use less extras for the European versions. So the higher quality and more feature packed versions won't be the European ones.
You are mistaken in several ways. For one "unwanted languages" is a matter of taste. Here in Germany nobody (i.e. 99.9% of the population) would ever buy a HD DVD or Blu-Ray disc without a german dub track on it. So "unwanted languages" is flat out wrong. Next: Actually out of the 4 European HD DVD titles I mentioned 3 had a *higher* video bitrate compared to the USA disc.

You are mistaken in several ways. For one "unwanted languages" is a matter of taste. Here in Germany nobody (i.e. 99.9% of the population) would ever buy a HD DVD or Blu-Ray disc without a german dub track on it. So "unwanted languages" is flat out wrong. Next: Actually out of the 4 European HD DVD titles I mentioned 3 had a *higher* video bitrate compared to the USA disc.
I meant the additional languages were unwanted for me. I bought the version of the disc sold in the UK, yet Universal's discs first asked me to select my language, and defaulted to Japanese! And those extra audio languages on Universal HD-DVDs sold in the UK (ie. French, Italian, German, Spanish, Japense Dolby Digital Plus 5.1) used up bitrate that could have been used for extra picture quality and/or more extra features in higher quality (and the special features on HD-DVDs were usually standard definition and not generally anamorphic for SD clips from the film)

Next: Actually out of the 4 European HD DVD titles I mentioned 3 had a *higher* video bitrate compared to the USA disc.
Where can I find a site that shows the UK and USA video bitrates for those 4 HD-DVDs as I can't find any. I do know that the USA often gets more extra features as they don't put as many extra languages on the disc so they have more bitrate/disc space spare.

EDIT: I found one place where you say the USA version of scorpion king is 17.0 avg mb/s and the European is 18.44 but on the video picture specs page of the forum it says the USA one is 17.7 avg mb/s., but that still gives the European one a little bit better bitrate.

Black Levels:
Do the studios reduce black levels in a film so that it uses up less video bitrate? Or do some films not look quite so good on LCD because of some LCDs having poor black levels? On some Universal HD-DVD titles, the darker scenes can look too dark but if you turn the brightness up, the brighter scenes can look too bright.

I meant the additional languages were unwanted for me. I bought the version of the disc sold in the UK, yet Universal's discs first asked me to select my language, and defaulted to Japanese! And those extra audio languages on Universal HD-DVDs sold in the UK (ie. French, Italian, German, Spanish, Japense Dolby Digital Plus 5.1) used up bitrate that could have been used for extra picture quality and/or more extra features in higher quality (and the special features on HD-DVDs were usually standard definition and not generally anamorphic for SD clips from the film)
Ok, I can agree with that... :)

Where can I find a site that shows the UK and USA video bitrates for those 4 HD-DVDs as I can't find any. I do know that the USA often gets more extra features as they don't put as many extra languages on the disc so they have more bitrate/disc space spare.

EDIT: I found one place where you say the USA version of scorpion king is 17.0 avg mb/s and the European is 18.44 but on the video picture specs page of the forum it says the USA one is 17.7 avg mb/s., but that still gives the European one a little bit better bitrate.
I don't know any site for video bitrates of the European/UK discs. I've did the comparison myself since I originally had the USA discs and later replaced them with the German HD DVDs. Not sure why the video picture specs page shows a different number than I calculated. Probably I'm using a different method to calculate the bitrate. I guess that the video specs page includes overhead caused by the container. Personally, I've demuxed the raw video bitstream of the USA and German disc and compared the size. There can't be a more exact result than this. So I think my number is more exact than the one from the video specs page.

Do the studios reduce black levels in a film so that it uses up less video bitrate?

Adjusting lift (control used to set black level), while color grading, is done for creative reasons. Sometimes black level can get messed up if incorrectly converted to and from RGB somewhere in the chain later in life. (pre-encoding)

Or do some films not look quite so good on LCD because of some LCDs having poor black levels?

Not limited to LCD technology, but sure. As black level goes up, color saturation goes down.

In some instances, the image is purposely crushed into the extreme black and extreme white range. The mathematical result in those areas are nice regions of blocks with zero detail, that provide a little "data tax holiday" for the rest of the image (that contains detail). The visual result is utterly masked (unless you image process to focus in on what is going on in those ranges), since they are in a region that is, by definition, utterly black or utterly white.

Examples of this can be found in the hdm encodes for 300 and Transformers. ;)

300 and Transformers.

Transformers actually had a lot of grain, it was removed by the encoder. The BD version should have much more grain, and detail.

The bits are spent per block. If large parts of the image are flat, those bits will go to parts that are not. Until you hit the lowest QP, there is always something to spend bits on. :)

Transformers actually had a lot of grain, it was removed by the encoder. The BD version should have much more grain, and detail.

The bits are spent per block. If large parts of the image are flat, those bits will go to parts that are not. Until you hit the lowest QP, there is always something to spend bits on. :)

This would be a good test for the casuals. I'd be willing to bet that most (before some you want to argue CASUALS! not ENTHUSIASTS!) would pick the "cleaner" look with removed grain.

Of potential interest to this thread, I did this demo showing how VC-1 has evolved over the past five years:

http://on10.net/blogs/benwagg/What-a-difference-a-half-decade-makes-Live-VC-1-today-and-at-launch/

Hi Ben,

Could you please tell us in layman terms on what aspects in VC-1 have been improved ? :o

Hi Ben,

Could you please tell us in layman terms on what aspects in VC-1 have been improved ? :o
For live encoding?

Faster performance means it can do more work per pixel. This means it can find better matches between frames for changes between frames, meaning fewer bits are needed to hit the same quality level.

Also, lookahead rate control means it can peek into the future to see what changes are coming in the video, to make sure it's leaving enough bits, but not too many, for future frames.

And B-frames improve efficiency, and also make it possible to correct for flash frames.

And lots of other stuff that there aren't layman's terms for, like Adaptive Deadzone :).

Thanks Ben. Finally understood a portion of your comments. I hope that macroblocking would be completely eliminated even with low bit rates.

Thanks Ben. Finally understood a portion of your comments. I hope that macroblocking would be completely eliminated even with low bit rates.
Completely eliminated is probably too high a bar for 600 Kbps :).

@ Ben and Amirm:
A little comparison between some codec (VC1 and H264 are on list):
http://conferences.infotoday.com/documents/45/SM2_Ozer.pdf

So today, I 'm not along (I'm remember last year...) to say VC1 is not better than H264.

And Ben, you work on integration of H264 on SilverLight, Whoaaahhhh.....
MS understand that the war is off :
http://www.streamingmedia.com/article.asp?id=10431


@ Amirm:
If I use your reasoning, H264 is used by lot of people (electronic devices, IPTV, TV Broadcast, BD title,...)
so it's the best...
I'm not agree with the reasoning but the result is true: H264 is better than VC1.

Now, all authoring studio use and will use H264 codec.
I saw and tested a hardware H264 HD encoder with HDSDI:
- Real time
- HDSDI 3Gbit too
- And give better result than VC1 from PeP/CinePSE (without preprocessing) at same bitrate :) :)
The file generated is compliant with BD specification (and HDDVD), price will less than 10K$ (version light less than 6K$ will available too).


I need a little help:
I have a HD source uncompressed (YUY2) to share to finish the Codec's challenge
but I need a website or ftp with good data transfer rate to upload on it.


Last, H264 and TrueHD are the winner (but with datarate of BD, it's possible to use a many 5.1 PCM), see some BlockBuster (MiB,...)

Now, all authoring studio use and will use H264 codec.
I saw and tested a hardware H264 HD encoder with HDSDI:


So the studios will use AVC because U think its better?

What we seen so far both codec have been used by different studios.

And Universal has gone from DD to DTS.

I personally like VC-1. I am a home theater PC enthusiast. While H264 is a great codec, it is very taxing on a system and much more so on a lower end system. Mpeg2 is too inefficient and results in file sizes that suck up my hard drive.

VC-1 seems to produce the smallest file size and still looks just fine to me. Plus every one of my PCs can play it. I only have 3 that can properly play AVC.

At first, I would think Mpeg2 is the way to go as it is the least lossy, but then again you might also have more problems staying under the bitrate cap with it.

Transformers actually had a lot of grain, it was removed by the encoder. The BD version should have much more grain, and detail.

The bits are spent per block. If large parts of the image are flat, those bits will go to parts that are not. Until you hit the lowest QP, there is always something to spend bits on. :)

Quite a bit still came through on both the HD DVD and the HD-lite Dish Network PPV. I do recall a good bit of grain in the theater as well.

So the studios will use AVC because U think its better?

No, because H264 is better (much than VC1).
Many people think VC1 is better than H264, but it's not true (see my old post on this forum and Doom9).

No, because H264 is better (much than VC1).
Many people think VC1 is better than H264, but it's not true (see my old post on this forum and Doom9).

I have seen the post, and im not convinced that its better (It could be). And it really doesnt matter since many studios seem to be satisfied with VC1. Just as some seem to like AVC.

And frankly both codecs works really good. So its not that i will avoid a purchase because of the codec.

If everything you say is right, then the studios will switch to AVC within a year. If you wrong and VC1 has some advantages over AVC, we maybe gonna se studios switch to VC1.

But so far when i have seen movies encoded with both codecs from the same master. VC1 has managed to compete with the AVC using lower peak and avarage bitrate.

If AVC is even better, well thats great, more competition and more reason for the VC1 to improve itself. And the winners are we. :)

They look the same to me. I don't think you can judge by individual frames as these are moving images.

When I look at my 720p PJ screen, I can see SDE as I am sitting about 8-9 feet away from a 94" screen. It doesn't bother me, and it lets me know it is in focus. I specifically noticed the other day when testing it that it all disappears when I actually play the movie.

I imagine it is the same type of thing. You can use all the still frames you want, but motion truly does change what you see. Funny thing is on my HD-A3, I actually think AVC is a little weaker as my box cannot handle it as easily as the other. A good example of this was Stardust from what I remember. It didn't look as good in my A3 as it did on HTPC. On the other hand Top Gun in AVC looked fine on both. HDAVC is notorious for its hardware requirements. On the other hand, my HD HTPC has no problems with it whatsoever (of course that system is a little overkill - AMD x2 BBE OCd to 6000+ (3.0) / 128MB 6600GT / 2GB Ram / XP Pro). I also have no problems on my Dell 4700 with an HD2600XT card installed.

I like VC-1 alot. It is easier to work with in my opinion on the computer. You would think AVC would be, but I always have problems with it. I have been able to successfully run VC1s on my Pentium M 2.1 laptop with good success. AVC is just not doable for me.

But again. I think in a real world comparison where there is motion, the difference would be indinstinguishable. Captured images don't really tell the full story.

Just my own personal opinion. I will tell you the one that I like the least is actually Mpeg2. It takes up too much space.

I'm not a pro but have encoded a few hundred movies using different codecs
and different packages over the last 4 years. This comparison looks redicilous (p.41-42)
and I think the presenter should explain a few more things before I believe this garbage.

And what does this line mean: "Tests ridiculously aggressive; at 1500kbps they all look good" (p.40)
Was this something in the sub-1Mbps range? For hidef? WTF does this have to do with real life hidef encoding?

Diogen.

You're right. I don't know how people can even watch material in VC-1. It shouldn't even be considered HD. cough cough. :rolleyes:

http://www.projectorreviews.com/Images-projectors-Q1-07/HomeCinema1080_PhantomStarLarge.jpg

Phantom of the Opera in VC-1 is just atrocious and renowned for its poor PQ. :rolleyes:

Hidefdigest.com (http://hddvd.highdefdigest.com/phantomoftheopera2004.html)

My surprise enjoyment of 'Phantom of the Opera' continued with the film's impressive picture. I would even venture to say that of all the initial HD DVD titles I have reviewed thus far, 'Phantom' has produced some of the most striking images. I don't know if I've ever described a video transfer as "delicate" before, but that is exactly the trick 'Phantom' pulls off here, perfectly straddling the line between technical razzle-dazzle and a palpable sense of reality.

I have seen the post, and im not convinced that its better (It could be). And it really doesnt matter since many studios seem to be satisfied with VC1. Just as some seem to like AVC.

?
You saw "Codec challenge" on Doom9 forum, and you're not convinced...
You can read too:
http://conferences.infotoday.com/documents/45/SM2_Ozer.pdf



And frankly both codecs works really good. So its not that i will avoid a purchase because of the codec.


You may be not purchase because a codec is used, but people will.
eg DVD and Dolby Digital, I remember in 1999 my boss who said me put Dolby stereo audio track
and not PCM because people like to see Dolby logo on DVD box...


If everything you say is right, then the studios will switch to AVC within a year. If you wrong and VC1 has some advantages over AVC, we maybe gonna se studios switch to VC1.


Yes, MS said we will see many new BD title in VC1, but I see today more H264 BD title
and many BD blockbuster announced are in H264.


If AVC is even better, well thats great, more competition and more reason for the VC1 to improve itself. And the winners are we. :)

"...VC1 to improve itself."
I wait to see a better implementation of VC1 which give better result than H264,
but today I'm sure it's not possible.

MS don't want to participate at Codec comparison by laboratory or expert (eg MSU
http://compression.ru/video/ ). I know why, but many people (and you) need more to understand ...

?


You may be not purchase because a codec is used, but people will.
eg DVD and Dolby Digital, I remember in 1999 my boss who said me put Dolby stereo audio track
and not PCM because people like to see Dolby logo on DVD box...
...

Please. The people will not buy a movie because it is VC-1 are probably 1% of 1% of the population. Most bluray buyers probably don't know the difference between DD and TrueHD... because they have it hooked up a 2 channel TV speak system. Furthermore, The only way a person is really going to know is if their SA displays (which most don't) or if they put it in a computer and have the diagnostics enable. The idea that any significant number of people are going to not buy a BD because it is in VC-1 is ridiculous. I am legend is in VC-1 and it has not seemed to stop people from picking it up.

I am not saying that there are not difference in how the PQ looks, but you will only see any if you intentionally look and even then I think it would be placebo most of the time. I remember watching one and thinking it must be AVC only to find out it was actually VC-1.

?
You saw "Codec challenge" on Doom9 forum, and you're not convinced...
You can read too:
http://conferences.infotoday.com/documents/45/SM2_Ozer.pdf


Yes I saw some handcamcorder footage someone had shot wich was a crappy source to begin with. Every HD DVD and BD I have is procontent and most of it is filmbased.

Whats important would be how the codecs now work inside the specifictions of BD. Like peak of 40mbs with an ABR of 20.

As for the article, it was very biased. If they spent as much time with VC1 as they did with AVC im sure it would result in a better encode.

Everyone is aware of why Golgot13 is hell bent on VC-1, right? He once had a trial version of PEP and then when it expired, and was expected to purchase, he was insulted.

Two codec shootouts were done recently by 3rd parties. MS was not involved in the shootouts. We were informed after the fact. One was a studio and the other a compression facility. The same three encoders were used. PEP VC-1, Sony AVC and Cinemacraft AVC. PEP won both shootouts. Cinemacraft came in a distant 3rd.

VC-1 share will grow on BD. I expect it to surpass AVC later this year or early next in terms of % of BD titles in VC-1.

MS don't want to participate at Codec comparison by laboratory or expert... I know why, but many people (and you) need more to understand ...Good for you.
Now, can you keep it to yourself until asked to share?

Your anti-VC1 crusade over the last year or so went from mildly entertaining to really annoying.

Diogen.

Everyone is aware of why Golgot13 is hell bent on VC-1, right? He once had a trial version of PEP and then when it expired, and was expected to purchase, he was insulted.

Two codec shootouts were done recently by 3rd parties. MS was not involved in the shootouts. We were informed after the fact. One was a studio and the other a compression facility. The same three encoders were used. PEP VC-1, Sony AVC and Cinemacraft AVC. PEP won both shootouts. Cinemacraft came in a distant 3rd.

VC-1 share will grow on BD. I expect it to surpass AVC later this year or early next in terms of % of BD titles in VC-1.

It finally comes out. First I thought, English not being his native language, it was just coming across as rude and confrontational. Turns out, someone has very little feelings.

Why is everyone circling the wagons all of a sudden? Is AVS suddenly becoming a VC1 forum?

- Tom

Why is everyone circling the wagons all of a sudden? Is AVS suddenly becoming a VC1 forum?

- Tom

I didn't really see it that way.

I think someone posting a paper entitled "Comparing and Using
Online Video Codecs", with a stated goal of comparing sub-1.5Mbs encodes in the "Blu-ray & HD DVD Areas > HDTV Software Media Discussion" section of the forum as supporting evidence when denigrating one codec over another, it seems a bit disengenuous.

I'd think the same thing if somebody used a Flash video in a web page as evidence that the On2 codec wasn't useful for authoring HD video streamed from a hard drive.

When you add in past posting history, it's not so hard to notice an agenda.

Everyone is aware of why Golgot13 is hell bent on VC-1, right? He once had a trial version of PEP and then when it expired, and was expected to purchase, he was insulted.

Sorry, but it's not true and I'm not insulted VC1, I say VC1 is not better than H264 with my test.
And I wait that some people prouve us (guys on Doom9 and me) with measure and test VC1 is the best...
Don't forget that MS updated his VC1 website because they said lot of bad informations about H264 and tests
(they updated with old informations: I recommand all guy to read the "Codec Challenge" thread on Doom9 forum )

I had PeP at first time with trial version many month, near of one year, after my company want to purchase it
(MS prefer to "give" at only only company in Europe, MS don't know that it's prohibited by competition law...).
I never say that my company don't have CinevisionPSE... (and you can find it on Russian warez website)

I was the first who said that VC1 is adapted at BD format because it need more bitrate than H264
to give a good result with low CPU power use. But today, all is change: H264 is decoded by GPU
and there are big optimization to play H264 on many software player.

So I have access at it and say that VC1 (from best implementation) can not give better result than H264.
And I'm sure that VC1 codec is not the best codec, I really want that MS prouve it's not true...


Last I don't like to loose my time:

I only want that MS or guy prouve that VC1 is better than H264 with Open Community test (open at all people)
We can do it: I have a HD video (with no problem of copyright) I need a website to share it and I will wait the VC1 result of MS team.

We can do it: I have a HD video (with no problem of copyright) I need a website to share it and I will wait the VC1 result of MS team.

How can you prove that h.264 is better then VC1 with filmbased content using HD video? They are two very different animals to encode.

To judge you would need a huge collection of different clips to see the true performance of a codec.

While you maybe can prove that h.264 is the better codec on this type of content, you must agree that its hardly conclusive for the performance of a codec?

Well, I'm thinking of selling all of my VC-1 encoded discs since someone just convinced me it looks sooo bad than the other codecs regardless of the bitrate ;)

Shoot Em Up and King Kong is not looking good anymore ;)

Sorry, but it's not true and I'm not insulted VC1, I say VC1 is not better than H264 with my test.
And I wait that some people prouve us (guys on Doom9 and me) with measure and test VC1 is the best...
Don't forget that MS updated his VC1 website because they said lot of bad informations about H264 and tests
(they updated with old informations: I recommand all guy to read the "Codec Challenge" thread on Doom9 forum )

I had PeP at first time with trial version many month, near of one year, after my company want to purchase it
(MS prefer to "give" at only only company in Europe, MS don't know that it's prohibited by competition law...).
I never say that my company don't have CinevisionPSE... (and you can find it on Russian warez website)

I was the first who said that VC1 is adapted at BD format because it need more bitrate than H264
to give a good result with low CPU power use. But today, all is change: H264 is decoded by GPU
and there are big optimization to play H264 on many software player.

So I have access at it and say that VC1 (from best implementation) can not give better result than H264.
And I'm sure that VC1 codec is not the best codec, I really want that MS prouve it's not true...


Last I don't like to loose my time:

I only want that MS or guy prouve that VC1 is better than H264 with Open Community test (open at all people)
We can do it: I have a HD video (with no problem of copyright) I need a website to share it and I will wait the VC1 result of MS team.

You think MS (who is not the only developer of VC-1) is going to bother to kowtow to your wishes?

It is interesting how you phrased. I would flip it around. I don't think that H264 is any better than VC-1.

Well, I'm thinking of selling all of my VC-1 encoded discs since someone just convinced me it looks sooo bad than the other codecs regardless of the bitrate ;)

Shoot Em Up and King Kong is not looking good anymore ;)
Maybe this guy has some beef with VC-1 but if MS can't prove that VC-1 is NOT better in PQ than AVC in a open public test when all things are equal, then the only thing I take from this thread is:
AVC provides the BEST PQ/encode but VC-1 can generate pretty well PQ/encodes as well.

We are on an AV enthusist forum.
Aren't we here to get the best and push to get the best standardized and not settle for "good enough" when the best can be had ?


It seems if MS doesn't want to participate in an open public encode comparison, all we hear is propaganda that VC-1 is this and that and it's the optimum codec with better compression at lower bit rates and file size.
But if VC-1 could increase it's bit rate and file size equal to AVC, would the encode be just as good or "good enough" to fool the viewer?

You all know the saying, "Put up or shut up." :)

You all know the saying, "Put up or shut up." :)


Try reading this post: http://www.avsforum.com/avs-vb/showpost.php?p=14071238&postcount=446

They don't need to impress fanatics and biased on AVS when they're winning the battle where it actually matters for their business.

Try reading this post: http://www.avsforum.com/avs-vb/showpost.php?p=14071238&postcount=446

They don't need to impress fanatics and biased on AVS when they're winning the battle where it actually matters for their business.

As the maintainer of the officially dated titles release sheet, I really have to call bunk on that statement of VC-1 being used on a majority of titles this year.
Of the titles I have added to the list so far this year, and they are now populating september, Warner is the only major post house using VC-1, This is opposed by Sony, Paramount, Fox, Disney all leaning towards AVC, maybe this will change down the road, however I do not believe that this year will be a VC-1 coup year. That said, to me the codec makes no difference as long as the PQ is the best representation of the master. I just find all the spin an annoyance as it makes reading some threads difficult.

Warner is the only major post house using VC-1, This is opposed by Sony, Paramount, Fox, Disney all leaning towards AVC, maybe this will change down the road, however I do not believe that this year will be a VC-1 coup year.

Well right now AVC has huge advantage since Universal hasnt released BD yet. If they gonna continue to use VC1 on BD. VC1 shares will raise during the second half of the year.

But I wonder what studio did the codec test on its own. If it wasnt Warner or Universal we maybe gonna se anyone of the other to use VC1 aswell?

But hey thats just wait and see.

If the year old codec shootout at doom9 is of any indication (I'd understand if some people don't agree: animation, computer graphics, too clean source, etc.), when you go above 12Mbps bitrate you can forget about discerning the difference in the output quality with a naked eye (OK, the absolute majority won't). Only quantitative measures (PSRN, SSIM) can show differences between encodes. And those measures are debatable, too.

This is the reason why I think the reference to some really lame comparison (see attachment in post 441 above) done by a person claiming to be an industry insider looks really really lame...

Diogen.

Maybe this guy has some beef with VC-1 but if MS can't prove that VC-1 is NOT better in PQ than AVC in a open public test when all things are equal, then the only thing I take from this thread is:
AVC provides the BEST PQ/encode but VC-1 can generate pretty well PQ/encodes as well.
Really? You just ignored what Stacey said in studios doing their own evaluation with their own content and arriving at opposite result. Yet you choose to believe a guy you don't even know saying that ain’t so?

We are on an AV enthusist forum.
Aren't we here to get the best and push to get the best standardized and not settle for "good enough" when the best can be had ?
You should push for the best. And since this is AV Science, you need to understanding the science. There really is no shortcut to getting past the posturing. And to that end, I wrote an extensive tutorial on the differences: http://www.avsforum.com/avs-vb/showthread.php?p=9931723&&#post9931723.

Here is my net of the thread. Someone keeps asking Microsoft to go out for a fist fight on some synthetic test clip. Yet he doesn't seem to understand that performing codec testing is super complex, super complicated, and requires extensive resources from displays to viewers. And when it is all said and done, someone will still cal foul because as I said, the process is so complicated that there are always, always, flaws.

The guy quotes Jan Ozer but has no idea who he is. I have worked with Jan for a decade to do benchmark tests and I could say a lot. But smart people here have already figured out key issues with his presentation. Yet the guy who claims to know codecs are best, doesn’t catch the fact that no methodology for the tests was specified. Camcorder content was used. Strange criteria was used, etc. Oops, I started to say a lot :D. But you get where I am going. The guy confuses tests for streaming on internet for HD encoding.

To be fair, AVC is quite strong at very low bit rates (think typical youtube) where lack of distortion as opposed to strict fidelity is more important. Keep in mind that nobody has made any claims regarding VC-1 cleaning AVC’s clock at sub-mbit data rates. While VC-1 is still quite strong there, as I just mentioned, the analysis becomes quite subjective and using the codec properly for the application matters a lot and even then, someone will still like the other image. In that sense, you can expect VC-1 to lose some and win some. As will be the case with AVC. But this is all for a different domain as multiple people have already tried to say.

So where does it leave someone who is deciding to choose a codec? Let's be clear, that someone is not anyone here but the studios. And studios have their material, their own experts, and resources to do the tests to their satisfaction. Those tests will also have faults, but at least, they meet the requirements of the people running them. That puts them way, way ahead of any tests someone wants to run on the Internet especially since winning it gets you something (Hollywood titles using it).

If you are not satisfied with VC-1 being an excellent codec after 500+ titles encoded in it, then I don't know what to say. OK, I will say this :). Benchmarks are necessary when you don't have data otherwise. This is why DVD Forum conducted codec tests (where I might add, AVC lost to VC-1). But now that we have the real titles, we don't need synthetic tests. Go read the reviews. If you find VC-1s faulted for compression artifacts more than other codecs, come and talk about that. I think that would be a much more useful conversation than this one.

Also, keep in mind that one of the main appeals of VC-1 is the superb tool /team behind it. You can build the most powerful engine you want for a car. But if it has no instrumentation or control, the race car driver won't be able to figure out how to drive it. Being a software company does have its advantages! And having people like Stacy, well, priceless! :D

It seems if MS doesn't want to participate in an open public encode comparison, all we hear is propaganda that VC-1 is this and that and it's the optimum codec with better compression at lower bit rates and file size.
You don't just "hear it." You see it used. You see us explain the science. You see the company not caring what some unknown person on the Internet wants when its business in this area solely relies on studios making their own selection.

When I ran the codec team, I could easily ask the crew to play in this test. The team was used to me interrupting their work to please AVS folks :). But seeing the attitude presented here, made me do anything but. When the company has no motivation to do something from business point of view, you want to play along nicely to get their interest to participate. You don't keep sticking your finger in their eye. Because if you do, they have a nice defence mechanism: they ignore you :p.

Instead, we keep getting these demanding posts and claims that AVC has already won. If so, then he should go and be happy that his codec is best. Why the constant posts here? If he is convinced AVC is best, then he is done and the results should come his way. What is he looking for with all of these posts?

You all know the saying, "Put up or shut up." :)
Yup, Microsoft put up. It won two rounds of DVD Forum tests to earn its spot as one of the chosen codecs not just in HD DVD, but also BD. AVC did not win yet it was chosen anyway. Until I left the company, VC-1 had the largest share of HD content encoded and has won a number of perfect review scores. Major Hollywood studios standardized on it despite AVC companies begging them to use their product on daily basis and making all kinds of offers including free encodes, tools, etc. If that is not “put up,” I don’t know what is.

Well right now AVC has huge advantage since Universal hasnt released BD yet. If they gonna continue to use VC1 on BD. VC1 shares will raise during the second half of the year.

But I wonder what studio did the codec test on its own. If it wasnt Warner or Universal we maybe gonna se anyone of the other to use VC1 aswell?

But hey thats just wait and see.

I really don't expect anyone to really change the codec they are comfortable using. unlike the arguements in this thread the visual quality has been demonstrated to be nearly equal which is an adequate answer for the studio's. For them it is about cost, ease of use, and compressionist comfort, (and comfort comes from familiarty with a tool) the more you use something the better you are with it.
So I personally would expect Warner and Universal to continue to use VC-1, and the others to continue to use AVC.
Again with HDDVD out of the picture and warner's use of the higher available bandwith even they seem to think that they were bit-constrained before (mentioned before by some who doesnt post here anymore) so I hope to see nothing but great master reproduction on every release. be the codec VC-1, or AVC, or Mpeg-2 for that matter.

Really? You just ignored what Stacey said in studios doing their own evaluation with their own content and arriving at opposite result. Yet you choose to believe a guy you don't even know saying that ain’t so?


You should push for the best. And since this is AV Science, you need to understanding the science. There really is no shortcut to getting past the posturing. And to that end, I wrote an extensive tutorial on the differences: http://www.avsforum.com/avs-vb/showthread.php?p=9931723&&#post9931723.

Here is my net of the thread. Someone keeps asking Microsoft to go out for a fist fight on some synthetic test clip. Yet he doesn't seem to understand that performing codec testing is super complex, super complicated, and requires extensive resources from displays to viewers. And when it is all said and done, someone will still cal foul because as I said, the process is so complicated that there are always, always, flaws.
.

That is an excellent point and fantastic post overall. thank you. Even amateurs like myself who have encoded their videos to VCD, DVD, Mp2, Mp1, Divx, AVC, H264, H263, X264 and now VC-1 know that you cant just go by a few tests. You could encode a video with three different encoders using the same codecs and get three different results. All three of these codecs are great and each has their strengths and limitations. From what I have read, Mpeg2 will generally give you the best results, but it of course takes up huge amounts of data. AVC is a very mathmetical driven codec that is very demanding, gives very good quality at low bitrates and has good compression. VC-1 seems to be optimized for its current application in HD and offers the best compression in size but still results in a very good image. The they all work a little differently.

It is like when people try to compare bitrates between AAC and Mp3. One uses a mathmatical compression, the other a combination of mathmetical and eliminated inaudible sound. They both sound great to me.

Furthermore, we are at 1080p right now. I know we all like to think this is as good as it gets, but didn't we think the same thing 10 years ago when DVD came out? Or 30 years ago when LD came out? 10 years from now (or sooner) when UHD comes out we will claim we won't even be able to watch SD HD anymore after watching UHD. :)

As the maintainer of the officially dated titles release sheet, I really have to call bunk on that statement of VC-1 being used on a majority of titles this year.
Of the titles I have added to the list so far this year, and they are now populating september, Warner is the only major post house using VC-1, This is opposed by Sony, Paramount, Fox, Disney all leaning towards AVC, maybe this will change down the road, however I do not believe that this year will be a VC-1 coup year. That said, to me the codec makes no difference as long as the PQ is the best representation of the master. I just find all the spin an annoyance as it makes reading some threads difficult.

Um, last I checked Warner was the biggest. And doesn't Universal (another big one) seem to prefer VC-1?

Um, last I checked Warner was the biggest. And doesn't Universal (another big one) seem to prefer VC-1?

All I do is add up the numbers.
1 + 1 + 1 = 3

I add up the VC-1 releases i get X
I add up the AVC releases I get Y
currently Y > X and doesn't appear to be going the other way any time soon, but honestly I don't really care as long as the PQ is acceptable to me, I happen to have 20/10 vision, and visual quality is more important to me than audio quality.
:)

All I do is add up the numbers.
1 + 1 + 1 = 3

I add up the VC-1 releases i get X
I add up the AVC releases I get Y
currently Y > X and doesn't appear to be going the other way any time soon, but honestly I don't really care as long as the PQ is acceptable to me, I happen to have 20/10 vision, and visual quality is more important to me than audio quality.
:)

Statistictly is has pretty much been a 3 way split on the bluray side. HD DVD was overwhelmingly VC-1 at 86% last I checked. So you could make the argument that those studios that were HD DVD exclusive also tended to use VC-1 and it could tip the scales one they start releases BD. Problem is that Universal and Paramount were notorious for holding back releases so we will have to see if that will change. I think they are letting the other studios get the water warm first before they jump in.

So it could very well stay a three way split for some time. It may eventually go VC-1, but who knows what will happen in the future. I honestly am more worried about the transfer than the codec. bad transfer + perfect codec still equals bad PQ.

Statistictly is has pretty much been a 3 way split on the bluray side. HD DVD was overwhelmingly VC-1 at 86% last I checked. So you could make the argument that those studios that were HD DVD exclusive also tended to use VC-1 and it could tip the scales one they start releases BD. Problem is that Universal and Paramount were notorious for holding back releases so we will have to see if that will change. I think they are letting the other studios get the water warm first before they jump in.

So it could very well stay a three way split for some time. It may eventually go VC-1, but who knows what will happen in the future. I honestly am more worried about the transfer than the codec. bad transfer + perfect codec still equals bad PQ.

Or perfect transfer, and codec + DNR and EE = Craptastic, part of the reason why VC-1 is getting a bad rep with it being almost all that universal used and we know the history of releases there. :(

I believe the VC1 vs. AVC discussions on this forum started with this Tom McMahon post
the day after he joined this forum (as quoted by Ben, couldn't find the original).
http://archive2.avsforum.com/avs-vb/showthread.php?p=6575927&&#post6575927

I think since VC1 was associated with Microsoft: anti-MS meant anti-VC1 and since MS supported HD DVD - anti-HD DVD.

Since HD DVD lost the war, I think the only thing missing for the BD side to be completely happy is VC1 to die as well.
Since this doesn't seem to be happening, the anti-VC1 posts grow more desperate. Can't figure out why...

Diogen.

Well, I'm thinking of selling all of my VC-1 encoded discs since someone just convinced me it looks sooo bad than the other codecs regardless of the bitrate ;)

Shoot Em Up and King Kong is not looking good anymore ;)
Have you watched the UK release of King Kong?

I believe the VC1 vs. AVC discussions on this forum started with this Tom McMahon post
the day after he joined this forum (as quoted by Ben, couldn't find the original).
http://archive2.avsforum.com/avs-vb/showthread.php?p=6575927&&#post6575927

I think since VC1 was associated with Microsoft: anti-MS meant anti-VC1 and since MS supported HD DVD - anti-HD DVD.

Since HD DVD lost the war, I think the only thing missing for the BD side to be completely happy is VC1 to die as well.
Since this doesn't seem to be happening, the anti-VC1 posts grow more desperate. Can't figure out why...

Diogen.

you really should read some doom9 posts. The VC-1 guys post there too. Lets not start a format war here please.

The holy grail is transparency to the master, in a lossy codec at current an reasonable bandwidth. other than that who cares.

you really should read some doom9 posts. The VC-1 guys post there too. Lets not start a format war here please.

The holy grail is transparency to the master, in a lossy codec at current an reasonable bandwidth. other than that who cares.

What we really need is lossless video.... :) Now we are talking. funny thing is, we could probably do that with those Holographic Discs they talked about a few years ago. Now that would be sweet.

I am actually fine with either. So is Mpeg2 better than both as it is less lossy? I would think it would unless it is a bottleneck bandwidth issue. You can fit more OJ in a can if it is concentrate.... but you need a really good prosecution lawyer. :) Sorry, I couldn't resist that one.

you really should read some doom9 posts.I do. Post there sometimes, too.
Lets not start a format war here please.Had no intention. Was trying to avoid the replay of all this again...
The holy grail is transparency to the master, in a lossy codec at current an reasonable bandwidth. other than that who cares.Good point.
So is Mpeg2 better than both as it is less lossy?No, but at 18+Mbps it is almost as good as AVC/VC1 when encoding animation (from doom9 shootout)...

Diogen.

What we really need is lossless video.... :) Now we are talking. funny thing is, we could probably do that with those Holographic Discs they talked about a few years ago. Now that would be sweet.

I am actually fine with either. So is Mpeg2 better than both as it is less lossy? I would think it would unless it is a bottleneck bandwidth issue. You can fit more OJ in a can if it is concentrate.... but you need a really good prosecution lawyer. :) Sorry, I couldn't resist that one.

The keyword is bandwidth. VC-1's popularity in HDDVD was definitely because of its remarkable efficiency in working with HDDVD's bandwidth, and why Mpeg-2 can sometimes give phenomenal results and occasionally s**k, bandwidth is finite and you choose the tool required to give you the best result. AND yes lossless video is what i dream for as well but can you ever see that as a download, i think not and so I am not holding my breath, as media is always going to be a compramise between solid and downloads.
:o

The keyword is bandwidth. VC-1's popularity in HDDVD was definitely because of its remarkable efficiency in working with HDDVD's bandwidth, and why Mpeg-2 can sometimes give phenomenal results and occasionally s**k, bandwidth is finite and you choose the tool required to give you the best result. AND yes lossless video is what i dream for as well but can you ever see that as a download, i think not and so I am not holding my breath, as media is always going to be a compramise between solid and downloads.
:o

So if VC-1 is more efficient in terms of bandwidth, wouldn't produce better results than AVC at the same bitrate?

It is like the comparison of 128 MP3 versus 128 WMA. WMA sounds alot better at the same bandwidth as it is squeezing more substance through it.

So if VC-1 is more efficient in terms of bandwidth, wouldn't produce better results than AVC at the same bitrate?

It is like the comparison of 128 MP3 versus 128 WMA. WMA sounds alot better at the same bandwidth as it is squeezing more substance through it.
exactly what this thread is about. WELCOME. within a specific bandwidth what creates a better picture. To state it again this is a moot point to me but I am an interested bystander who wants in the end a PQ appealing to my eyes. NO banding, NO DNR, NO EE, as close to the master as the available bandwidth allows, and I know this will raise hackles but if that means stealing bandwidth from lossless audio so be it. :D

From what I understand, the only real comparison between the three is The Elephants Dream in Europe on HD DVD as it included all three codecs?

So I guess with any type of motion video, Mpeg2 would actually be the least optimum as it is the least lossy and requires the highest bandwidth.

From what I understand, the only real comparison between the three is The Elephants Dream in Europe on HD DVD as it included all three codecs?

So I guess with any type of motion video, Mpeg2 would actually be the least optimum as it is the least lossy and requires the highest bandwidth.
Which would give the best performance when played back in fast forward reverse speed - mpeg2 or avc or vc1? Which would give the fastest response time when the user presses frame advance/reverse or fast forward? Which could go to any point in the movie in the fastest possible time? Also, have we actually seen high bitrate progressive footage comparing all 3 codecs? Which are more accurate and best representative of the actual frame - I or B frames? And which codec uses the most I frames per minute of video - mpeg2 or avc or vc1?

Inquiring minds want to know
:eek:

Inquiring minds want to know
:eek:
I think, in response times (like fast forward/reverse/going to any point in a movie in the fastest time, mpeg2 would be the fastest, simply because it uses more I frames per minute than avc or vc1. I-frames are more accurate (though they use more bandwidth than b frames. But b-frames are only approximations I think - ie. they look at previous and subsequent reference frames (I or P - and P frames are only approximations too, so as time goes on newer B frames can get progressively worse, until the next I frame). They try to encode the frame based on the differences between them, a little bit like motion interpolation in TVs isn't always accurate (it's not as bad as that) but as someone pointed out in this thread I think, it may not be fair to compare a B frame to an I frame in a screenshot, so that shows that their quality and accuracy won't be as good (even though they are more efficient in bandwidth terms to encode).

Though the other codecs (avc and vc1) could be told to use less b frames and act just like mpeg2 if they wanted too. Also, at very high bandwidth all 3 should in theory be very similar. The other codecs will probably look better than mpeg2 at lower bandwidths. But I'd like to see the direct comparisons too.

From what I understand, the only real comparison between the three is The Elephants Dream in Europe on HD DVD as it included all three codecs?
Remember, you can only ever compare implementations, not codecs in the abstract.

Elephant's Dream is a comparison between three older implementations with a particular source. It's very hard to generalize from any one example, even if it's done very well.

By example, they switched VC-1 encoders between the first and second versions of that HD DVD, with massively better results the second time around.

So I guess with any type of motion video, Mpeg2 would actually be the least optimum as it is the least lossy and requires the highest bandwidth.
Less lossy is probably not the right way to think about it. At a given bitrate, it'll be LESS accurate than the other codecs; VC-1 in particular can be thought of as a superset of MPEG-2, with more options to accurately represent the source.

Since HD DVD lost the war, I think the only thing missing for the BD side to be completely happy is VC1 to die as well.
Since this doesn't seem to be happening, the anti-VC1 posts grow more desperate. Can't figure out why...

Certainly the BD studios are now much more interested in using VC-1 than before with the politics of the format war gone. I'd expect the percentage of BD titles using VC-1 to be quite a bit higher holiday '08 than holiday '07.

Less lossy is probably not the right way to think about it. At a given bitrate, it'll be LESS accurate than the other codecs; VC-1 in particular can be thought of as a superset of MPEG-2, with more options to accurately represent the source.
What if there is a complex object moving in a very unpredictable way around the screen (ie. lots of apparently random motion (apparently random to the encoder anyway))? Will mpeg2 (more I frames, less B frames) really be less accurate?

?
You saw "Codec challenge" on Doom9 forum, and you're not convinced...
You can read too:
http://conferences.infotoday.com/documents/45/SM2_Ozer.pdf
Utterly irrelevant to HDM, given the bitrates being used.

Also, lots of the results are inexplicable. Baseline H.264 decodes faster than Main? And looking at the screen shots, they're not even close to luma normalized; looks like some screen shots use 0-255, other 16-235.

Lastly, the source is HDV, IIRC, certianly not 24p feature film content.

MS don't want to participate at Codec comparison by laboratory or expert (eg MSU
http://compression.ru/video/ ). I know why, but many people (and you) need more to understand ...
Have you read the MSU specs? They're much more a test of single-threaded encoding speed than anything else; no professional encoding would use less than 10x the MIPS/pixels as they allow.

Anyway, they're more than welcome to test our codec the SDK is a free downlad; it just wasn't relevant enoough to us for us to pay the significant participation fee. We had other stuff we wanted to spend our marketing budget on that quarter.

What if there is a complex object moving in a very unpredictable way around the screen (ie. lots of apparently random motion (apparently random to the encoder anyway))? Will mpeg2 (more I frames, less B frames) really be less accurate?
Well, there's nothing about the different codecs that sets a maximum number of I-frames. The issue is that MPEG-2 needs to have more frequent I-frames than VC-1 or H.264 since it using a floating point DCT instead of an reversable integer transform. This means you get more "drift" the farther away you get from an I-frame, making for reduced quality and bad keyframe popping when you too far between keyframes. This can often be seen in bitrate-starved cable/sat MPEG-2 broadcasts.

For content with so much random motion that an I-frame would be a more efficient way to encode that frame, any decent implementation of any codec will insert an I-frame at that point. There's a lot of on-the-fly adjustment to GOP patterns we do in VC-1, like I-frame insertions, turning off B-frames during fades to take advantage of the intensity control, using intra B-frame ("skip frames") during single frame flashes, etcetera.

Statistictly is has pretty much been a 3 way split on the bluray side. HD DVD was overwhelmingly VC-1 at 86% last I checked. So you could make the argument that those studios that were HD DVD exclusive also tended to use VC-1 and it could tip the scales one they start releases BD. Problem is that Universal and Paramount were notorious for holding back releases so we will have to see if that will change. I think they are letting the other studios get the water warm first before they jump in.
We're also seeing increasing interest in previously BD-only studios that didn't do much with VC-1. I don't think that woudl be reflected in the title announcements you would have recieved yet. The trend should be clearer by the end of the holiday season.

So it could very well stay a three way split for some time. It may eventually go VC-1, but who knows what will happen in the future. I honestly am more worried about the transfer than the codec. bad transfer + perfect codec still equals bad PQ.
Critical point - source quality differences overwhelm codec differences. It's only when you have good source that codec choice really makes that much of a difference.

This means you get more "drift" the farther away you get from an I-frame, making for reduced quality and bad keyframe popping when you too far between keyframes. This can often be seen in bitrate-starved cable/sat MPEG-2 broadcasts.

I think I've seen this in one or more of the HDScape titles (which I think were AVC). Do both AVC and VC1 use integers instead of floating points to do better keyframing than mpeg2? (I thought floating point was supposed to be accurate).

Will mpeg2 (more I frames, less B frames) really be less accurate?
Also, MPEG-2 typically has more B-frames than VC-1. VC-1 normally alternates B and P frames, while MPEG-2 typcially has two B-frames between P-frames.

IBP versus IBBP

Certainly the BD studios are now much more interested in using VC-1 than before with the politics of the format war gone. I'd expect the percentage of BD titles using VC-1 to be quite a bit higher holiday '08 than holiday '07.
This this fact or speculation ?

Also, MPEG-2 typically has more B-frames than VC-1. VC-1 normally alternates B and P frames, while MPEG-2 typcially has two B-frames between P-frames.

IBP versus IBBP
So mpeg2 has more B-frames than VC-1?
How is VC1 more efficient than Mpeg2 then? I thought it was due to greater use of B (and P) frames than mpeg2?

I know VC1 and AVC can both have more accurate motion vector precision (to 1/4 pixel) than Mpeg2 (1/2 pixel) so maybe that improves bandwidth efficiency too), and the fact that the newer codecs can go to smaller block sizes.

Also, mpeg2 doesn't have a deblocking filter, so while AVC and VC1 can smooth over blocks where those blocks could be more obvious in mpeg2, that won't be totally accurate either - they'll look smoother but still not accurate like the original uncompressed video, and smoothing it out (a bit like how DNR works) might make it harder for the compressionist to know that they really need to increase the bitrate at that point as it will be missing details in the picture - which would have been more obvious if encoding in mpeg2 as the blocks would be easier to see.

This this fact or speculation?
Prediction? This is what I hear is going on, but there's not thing to say studios wouldn't change their mind for some reason.

But if it turns out to be wrong, you certainly may enjoy the pleasure of telling me I was wrong :).

So mpeg2 has more B-frames than VC-1?
How is VC1 more efficient than Mpeg2 then? I thought it was due to greater use of B (and P) frames than mpeg2?
It's more that the improvements in VC-1 allow the use of fewer B-frames. Remember, the data that goes into a B-frame is thrown away after playback. B-frames are based on the previous/next I-frame or P-frame, but no frame is ever based on a B-frame. Only the data in I-frames and P-frames is used in future frames. However, the bidirectional prediction makes B-frames more efficient, so you get a sweet spot for a given codec as to the optimum pattern for typical content. This can vary; for example, I like to use IBBP for animation content, and IBBBBP for screen captures with VC-1.

I know VC1 and AVC can both have more accurate motion vector precision (to 1/4 pixel) than Mpeg2 (1/2 pixel) so maybe that improves bandwidth efficiency too), and the fact that the newer codecs can go to smaller block sizes.
Both help, definitely. There's also motion vector modes that can handle more complex motion, more efficient entropy coding, etcetera.

There's a variety of informational resources at

http://www.microsoft.com/vc-1

Including this groovy Silverlight animation:

http://www.microsoft.com/resources/mediaandentertainment/presentation/default.html

Also, mpeg2 doesn't have a deblocking filter, so while AVC and VC1 can smooth over blocks where those blocks could be more obvious in mpeg2, that won't be totally accurate either - they'll look smoother but still not accurate like the original uncompressed video, and smoothing it out (a bit like how DNR works) might make it harder for the compressionist to know that they really need to increase the bitrate at that point as it will be missing details in the picture - which would have been more obvious if encoding in mpeg2 as the blocks would be easier to see.
A post-processing deblocking filter doesn't really add any visual information. But an in-loop deblocking filter like in VC-1 and H.264 does actually increase the SNR of the video, since it means that future frames based on a highly compressed on don't have to spend bits fixing block errors. So that in-loop filter does improve the true quality of future frames.

One of the bigger practical differences between VC-1 and H.264 are the loop filter implementations, with VC-1 applying to a smaller number of pixels.

Remember, you can only ever compare implementations, not codecs in the abstract.

Elephant's Dream is a comparison between three older implementations with a particular source. It's very hard to generalize from any one example, even if it's done very well.

By example, they switched VC-1 encoders between the first and second versions of that HD DVD, with massively better results the second time around.


Less lossy is probably not the right way to think about it. At a given bitrate, it'll be LESS accurate than the other codecs; VC-1 in particular can be thought of as a superset of MPEG-2, with more options to accurately represent the source.

So would Mpeg 2 be best at slower motion films? I would think VC-1 would be the best then at high motion seens as it approaches the Bitrate cap.

So would Mpeg 2 be best at slower motion films? I would think VC-1 would be the best then at high motion seens as it approaches the Bitrate cap.
Higher motion (particularly complex motion, where a lot of things are going different directions at once), or higher detail, like complex textures. Fades to/from black. There's a whole lot of incremental improvements.

First, @ Ben:
No answer about this:
And Ben, you work on integration of H264 on SilverLight, Whoaaahhhh.....
MS understand that the war is off :
http://www.streamingmedia.com/article.asp?id=10431




Lastly, the source is HDV, IIRC, certianly not 24p feature film content.

And ?
You gave a 29.97i video for test (from 24P source...):rolleyes::p


Have you read the MSU specs? They're much more a test of single-threaded encoding speed than anything else; no professional encoding would use less than 10x the MIPS/pixels as they allow.


VC1, PeP/CinePSE, was the last codec which supported multicore (multi-thread) CPU....
They test the effiency of codec, there is no consideration of time.

Professional encoding need only efficiency codec and a sofware solution (GUI with access at all feature, mutlipart encoding, 2 pass...).
Today, big authoring studio say that Thomson Encoder (Nexcode encoder) is the best solution. It's a solution on Linux OS (no MS inside :D )
and they give 5/6 PC and the network hardware so the price is not nice.

Ben I recommand you to test yourself x264 (last with all optimization) and VC1 at 6Mbps or 9Mbps.
9Mbps with good parameter is enough for 1920x1080@24fps (with or without grain thank DarkShakira).
If you have the BD specification, the video will be compliant with BD (read page 255 and 364, VC1 at 383 of BD Rom2 Part3 book).

9Mbps with good parameter is enough for 1920x1080@24fps (with or without grain thank DarkShakira)..
No, this is not (good) enough. Absolutely not. That may work with some material, but not in general.

No, this is not (good) enough. Absolutely not. That may work with some material, but not in general.

I dont think 9mbs will hold up for the standard AVS pause and enlarge evaluation. ;)

I think Amir stated that VC1 starts to look good at about 10mbs and after
20mbs its really diminishing returns.

So it could be fun to see how AVC works at 6mbs, 10mbs, 20mbs, 30mbs. (1920*1080/24P with grain)

No, this is not (good) enough. Absolutely not. That may work with some material, but not in general.

So far 9 Mbps has been good enough for our brothers @ Warner.:D

So far 9 Mbps has been good enough for our brothers @ Warner.:D

You mix up peak and avarage bitrate. Even if you can find a Warner movie with that low avarage bitrate, Im very sure it has alot higher peaks.

First, @ Ben:
No answer about this:
And Ben, you work on integration of H264 on SilverLight, Whoaaahhhh.....
MS understand that the war is off :
http://www.streamingmedia.com/article.asp?id=10431
I think you're really the only one thinking about this as a "war" - we've always used a mix of standardized and Microsoft-created codecs appropriate to different platforms.

Honestly, you seem much more committed to there being a winner and loser here than any of the actual professionals in the field of codec design.


You gave a 29.97i video for test (from 24P source...):rolleyes::p
Once again I note that you agitate for 24p source, but every time I ask you to participate in the task of actually figuring out what the source should be like, you never make any actual contribution. I've got a number of open questions up on the Doom9 forum. If you actually want something to happen, try actually helping.

VC1, PeP/CinePSE, was the last codec which supported multicore (multi-thread) CPU....
They test the effiency of codec, there is no consideration of time.
VC-1 has always been multi-threaded.

The MSU test plan for 2007 absolutely had a hard time requirement on a single core, single-threaded processor.

Ben I recommand you to test yourself x264 (last with all optimization) and VC1 at 6Mbps or 9Mbps.
9Mbps with good parameter is enough for 1920x1080@24fps (with or without grain thank DarkShakira).

x264 is an impressive effort in many ways, and I've spent time with it myself, but it's been used for exactly 0 studio HDM titles that I'm aware of, so isn't really pertinent to this discussion.

...
x264 is an impressive effort in many ways, and I've spent time with it myself, but it's been used for exactly 0 studio HDM titles that I'm aware of, so isn't really pertinent to this discussion.

As the alternate codec in the Doom9 shootout that appeared to give better results (that time) than VC1 and noting the number of comments, including yours, referencing that shootout I think we have already pretty much established its relevance here.

And it is not out of the question that someone could put together a package to use it as the basis of a studio encoding platform for BD.

- Tom

As the alternate codec in the Doom9 shootout that appeared to give better results (that time) than VC1 and noting the number of comments, including yours, referencing that shootout I think we have already pretty much established its relevance here.


- Tom
The tools needed for production of physical media must have a strict set of features. Any old encoder is not suitable for their use or they would already have put it into production.

And it is not out of the question that someone could put together a package to use it as the basis of a studio encoding platform for BD.
At which time, studios will be in a position to do their own benchmarks and wouldn't need something put up by doom9.