Blu-ray, HD-DVD & HD Broadcasts(H.264 & MPEG-2) Screenshots*BIG FILES* - Page 7 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #181 of 2128 Old 04-04-2007, 11:41 AM
AVS Special Member
 
WayneL's Avatar
 
Join Date: May 2006
Posts: 2,612
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Icemage View Post

VC-1 has a strong loop filter. AVC also has a strong loop filter (if we are to believe what Amir and Ben have to say about it, it's weaker than VC-1's, but the effects are plainly visible in the X-Men 3 comparison).

If you look at the MI3 explosion shot encoded with VC-1, you'll see the same sort of effect: blended colors, less grain, less noise, less obvious macroblocking. Compare against the MI3 MEPG2, which shows stronger color contrast, perhaps at the price of compression artifacting (bearing in mind that MPEG2 doesn't use a loop filter).

I think this is why the question was asked about VC-1 with the loop filter off as compared to MPEG-2 (am I correct, Mr. Hanky?). The idea is to see just how much more efficient VC-1 is at similar bitrates as MPEG-2 without a loop filter to hide its flaws.

I can agree with your observed facts, except, is it VC-1 has less grain because MPEG2 has more noise? Can anyone upload a fragment of scanned 35mm film that shows us what real grain looks like. I'll bet it doesn't look like anything we see.

But I still have doubts that an objective comparison can be made with MI3:

1. are the differences in these scenes (loop filter notwithstanding) largely a result of choices by the compressionist, which could be quite different with different choices?
2. should we care about similar bit rates since VC-1 and MPEG2 have different bit rate objectives?
3. are these similar bitrates?

I think for my max viewing options - a 60" screen at 10', both are likely very comparable, except I can't decide if those who call VC-1 smooth, are doing so because there is less noise from the macroblocking

I admit to having neither, so Xylon's uploads are gold to me.
WayneL is offline  
Sponsored Links
Advertisement
 
post #182 of 2128 Old 04-04-2007, 11:46 AM
AVS Special Member
 
Mr. Hanky's Avatar
 
Join Date: Mar 2006
Posts: 6,064
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 31
Grain can have many kinds of appearances based on the grade and treatment of the film stock. I don't think there is only one characteristic look for film grain.

I need your sweet love, Rosetta Stone girl!
Mr. Hanky is offline  
post #183 of 2128 Old 04-04-2007, 12:09 PM
Senior Member
 
beagle five's Avatar
 
Join Date: Jan 2006
Posts: 234
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Steeb View Post

Yeah, that's great. It still means that most who know this about you will automatically take your opinions about a particular disc's PQ with a grain of salt, since you're altering the disc's look drastically by pumping up the artificial sharpening (and introducing artifacts all over the place, no doubt.)

To each his own - by all means, crank that sharpness as high as you can. Just realize that the vast majority of us don't want that - we want accurate representations of what's on the disc (which should be an accurate - within reason - representation of the master.)

so you are representing the opinion of the "vast majority"? thats nice to know.
well I know what I am doing and what I like, and you make it sound as if there is so much artefacts in the image that its impossible to look at it! well I dont know how terrible your sharpening is but mine isnt very bad at all, not on HD-DVD anyway, hardly any noticeble artefacts compared to the positive effects!
so maybe that means that I actually know what I am doing and is actually founding my opinion in the real world by trial and error!?
I would say that I spend at least a couple of hours EVERY DAY calibrating and playing with the settings on my projector or dvd or hd-dvd just to find the perfect setting. and it sure isnt 0 sharpening on hd-dvds, well in a perfect world where hd-dvds actually are sharp it would be but now they are not!

so MY opinion is, sharpness where sharpness is needed and as much as is needed
YOUR opinion ( not the vast majority I am afraid ) is that sharpness is terrible.

and you are talking about that you want it to look like the master, and this thread is about the fact that VC-1 seems less sharp, well shouldnt YOU also be turning up the sharpness if VC-1 has a less sharp picture then the master?

I am just so tired of these selfproclaimed forum police people that just does what the "gurus" does and other things is tabu and terrible and shall be silenced since its wrong wrong wrong, reminds my of religon... has it gone that far...

lighten up people!

there are different tastes and things we all want out of our homecinema, and after we have put a lot of money on the stuff why couldnt we watch it the way that gives us the most pleasure without getting some "gotta look like the master or else you are out of here"- crap slapped on your face!?!?

here is a screenshot from my setup, please show me all those terrible artefacts that is going to ruin my reputation here!
beagle five is offline  
post #184 of 2128 Old 04-04-2007, 12:20 PM
AVS Special Member
 
Icemage's Avatar
 
Join Date: Feb 2007
Posts: 1,978
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by WayneL View Post

I can agree with your observed facts, except, is it VC-1 has less grain because MPEG2 has more noise? Can anyone upload a fragment of scanned 35mm film that shows us what real grain looks like. I'll bet it doesn't look like anything we see.

I don't know, but I do know the high-bitrate AVC and MPEG-2 titles that I own on Blu-ray have plenty of grain when expected. Do we have any reason not to suspect that VC-1 at high bitrates would be the same? I think the only title we could truly see this with would be Flight Plan on Blu-ray, but I don't own that title. Is there anyone out there that has that disc and can comment on the nature of the grain in the encode?

Quote:


But I still have doubts that an objective comparison can be made with MI3:

1. are the differences in these scenes (loop filter notwithstanding) largely a result of choices by the compressionist, which could be quite different with different choices?

Bit starvation is undoubtedly the cause of at least some of the macroblocking we're seeing. Optimization is all well and good, as is loop filtering, but a complex scene is a complex scene, period.

Quote:


2. should we care about similar bit rates since VC-1 and MPEG2 have different bit rate objectives?

I would say no, except insofar as answering the question of whether the typically low-grain look of VC-1 is directly due to the application of low bitrate plus heavy loop filtering. I'd really like to see what VC-1 can do with more bitrate.

Quote:


3. are these similar bitrates?

I think for my max viewing options - a 60" screen at 10', both are likely very comparable, except I can't decide if those who call VC-1 smooth, are doing so because there is less noise from the macroblocking

I admit to having neither, so Xylon's uploads are gold to me.

It's probably a safe bet to say that VC-1 will exhibit lower amounts of macroblocking than MPEG2 (if it doesn't then Microsoft has everyone fooled, including the experts) at any given bitrate.

The issue really at hand is, in typical scenes and high-stress scenes, how much bitrate does each codec need to not exhibit macroblocking without loop filtering, and how much damage control can the loop filtering perform without negatively impacting image quality?

One interesting case is what I watched last night: CHRONOS on Blu-ray. It's an MPEG-2 encode at a whopping 24.0Mbps+ consistent bitrate. Not once did I see the bitrate meter on my PS3 drop below 23Mbps. Macroblocking was virtually non-existent as far as I could tell on my 1080p RP display, and this is with some very challenging material; there are many scenes with shifting lines and streaks of light from long exposure shots that would make macroblocking very evident, but I saw no traces of it.
Icemage is offline  
post #185 of 2128 Old 04-04-2007, 12:36 PM
AVS Special Member
 
Mr. Hanky's Avatar
 
Join Date: Mar 2006
Posts: 6,064
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 31
Quote:
Originally Posted by Icemage View Post

It's probably a safe bet to say that VC-1 will exhibit lower amounts of macroblocking than MPEG2 (if it doesn't then Microsoft has everyone fooled, including the experts) at any given bitrate.

You may be right, but it is my own suspicion, that under duress, the key distinction is that vc-1 will exhibit less visible macroblocking (as opposed to "less macroblocking"), but still the same macroblocking instances of mpeg2 under similar duress. I think this avenue of thought could be an interesting investigation.

This makes for an interesting case of semantics, where you can, indeed, make claims that one codec can breakthrough to new bitrate thresholds w/o macroblocking like the other codec, but the reality is that the 2 codecs are macroblocking at a similar frequency of incidence, but the macroblocking is far less visible in the one codec. The claim is still true to the extent that if you can't see the macroblocks (or if they can at least slip the eye), did they really occur? Well, it really depends on how "technical" you want to be about it, right?

I need your sweet love, Rosetta Stone girl!
Mr. Hanky is offline  
post #186 of 2128 Old 04-04-2007, 01:39 PM - Thread Starter
AVS Special Member
 
Xylon's Avatar
 
Join Date: Dec 2005
Location: Liecheinstein
Posts: 7,380
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Fantastic Four

7.79 GB*9.75 Mbps

7.48 GB*9.90 Mbps.

17.60 GB
Xylon is offline  
post #187 of 2128 Old 04-04-2007, 01:42 PM - Thread Starter
AVS Special Member
 
Xylon's Avatar
 
Join Date: Dec 2005
Location: Liecheinstein
Posts: 7,380
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
The FF H.264 is captured as is. It has a higher level of brightness and contrast. Is it psssible to lower it down without affecting the image quality? I'm not sure so I left it alone.

Mpeg-2 version I will post later.
Xylon is offline  
post #188 of 2128 Old 04-04-2007, 01:44 PM - Thread Starter
AVS Special Member
 
Xylon's Avatar
 
Join Date: Dec 2005
Location: Liecheinstein
Posts: 7,380
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Mr. Hanky View Post

On a more serious note, do you anticipate getting any screenshots up for the Departed and that Lady in the Water flick, Xylon?

Yes, eventually.
Xylon is offline  
post #189 of 2128 Old 04-04-2007, 01:53 PM
AVS Special Member
 
cyberbri's Avatar
 
Join Date: May 2004
Location: San Jose, CA
Posts: 8,095
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 52
Quote:
Originally Posted by Xylon View Post

The FF H.264 is captured as is. It has a higher level of brightness and contrast. Is it psssible to lower it down without affecting the image quality? I'm not sure so I left it alone.

Yes, the brightness of the black in the black bars is different for each capture. Just try adjusting the level and see how it looks. Nice choice of scenes, btw.
cyberbri is offline  
post #190 of 2128 Old 04-04-2007, 02:04 PM
AVS Special Member
 
Mr. Hanky's Avatar
 
Join Date: Mar 2006
Posts: 6,064
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 31
Is she supposed to be completely naked, or is that a strap to something (that they forgot to digitally remove or keep obscured) that appears below her forearm?

I need your sweet love, Rosetta Stone girl!
Mr. Hanky is offline  
post #191 of 2128 Old 04-04-2007, 02:42 PM
Senior Member
 
tindizzle's Avatar
 
Join Date: Dec 2006
Posts: 255
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Mr. Hanky View Post

Is she supposed to be completely naked, or is that a strap to something (that they forgot to digitally remove or keep obscured) that appears below her forearm?

Unfortunately, she was wearing a bra IIRC.
tindizzle is offline  
post #192 of 2128 Old 04-04-2007, 03:07 PM
AVS Special Member
 
Kram Sacul's Avatar
 
Join Date: Sep 2003
Location: Los Angeles, CA
Posts: 5,257
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Xylon View Post

The FF H.264 is captured as is. It has a higher level of brightness and contrast. Is it psssible to lower it down without affecting the image quality? I'm not sure so I left it alone.

If you mean the black level you can correct it in Photoshop with the levels filter. Just put the input sliders to 16 and 235. This expands it to pc levels.

How are you capturing the h.264 broadcasts? With CoreAVC? If so, there are various options in there that we should know that are enabled or not, like the deblocking feature and color space output.
Kram Sacul is offline  
post #193 of 2128 Old 04-04-2007, 03:07 PM
Advanced Member
 
HD-DVDwonder's Avatar
 
Join Date: Sep 2006
Posts: 679
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by beagle five View Post

so you are representing the opinion of the "vast majority"? thats nice to know.
well I know what I am doing and what I like, and you make it sound as if there is so much artefacts in the image that its impossible to look at it! well I dont know how terrible your sharpening is but mine isnt very bad at all, not on HD-DVD anyway, hardly any noticeble artefacts compared to the positive effects!
so maybe that means that I actually know what I am doing and is actually founding my opinion in the real world by trial and error!?
I would say that I spend at least a couple of hours EVERY DAY calibrating and playing with the settings on my projector or dvd or hd-dvd just to find the perfect setting. and it sure isnt 0 sharpening on hd-dvds, well in a perfect world where hd-dvds actually are sharp it would be but now they are not!

so MY opinion is, sharpness where sharpness is needed and as much as is needed
YOUR opinion ( not the vast majority I am afraid ) is that sharpness is terrible.

and you are talking about that you want it to look like the master, and this thread is about the fact that VC-1 seems less sharp, well shouldnt YOU also be turning up the sharpness if VC-1 has a less sharp picture then the master?

I am just so tired of these selfproclaimed forum police people that just does what the "gurus" does and other things is tabu and terrible and shall be silenced since its wrong wrong wrong, reminds my of religon... has it gone that far...

lighten up people!

there are different tastes and things we all want out of our homecinema, and after we have put a lot of money on the stuff why couldnt we watch it the way that gives us the most pleasure without getting some "gotta look like the master or else you are out of here"- crap slapped on your face!?!?

here is a screenshot from my setup, please show me all those terrible artefacts that is going to ruin my reputation here!

Yes to each his own, but for most displays, optimal sharpness setting is set to as low as possible. There are number of HD DVDs from Universal such as Waterworld and Fear and Loathing in which the EE is bothersome even w/ the sharpness at 0.

Secondly, you may not notice from viewing distance but compare a 0 sharpness image to a 50 or 100 sharpened one and you will see not only how cruddy the halos look but also how harsh it looks due to artifacts that appear when sharpening film grain or digital noise. And for some BD titles that exhibit blocking, increasing sharpness would just make them much more apparent.

third, you will have to post a much bigger picture if you want to evaluate, especially from a camera. From your post it seems you are confusing sharpness w/ detail and definition.

For some ppl do actually prefer an oversharpened picture to a smoother one - but I can't stand EE and the harsh look - and when you put your nose up to the display, its swimming w/ artifacts
HD-DVDwonder is offline  
post #194 of 2128 Old 04-04-2007, 03:29 PM
Member
 
dethis's Avatar
 
Join Date: Mar 2007
Posts: 23
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by WayneL View Post

I can agree with your observed facts, except, is it VC-1 has less grain because MPEG2 has more noise? Can anyone upload a fragment of scanned 35mm film that shows us what real grain looks like. I'll bet it doesn't look like anything we see........

.

Take a look at "ftp://vqeg.its.bldrdoc.gov/HDTV/SVT_MultiFormat/"...
dethis is offline  
post #195 of 2128 Old 04-04-2007, 03:41 PM
 
Steeb's Avatar
 
Join Date: Jul 2006
Location: Vegas
Posts: 3,268
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 54
Quote:
Originally Posted by beagle five View Post

so you are representing the opinion of the "vast majority"? thats nice to know.

No, but from what I've read, you're definitely in the minority. If I'm wrong, where are all of the proponents of artificial sharpening backing you up? You're on your own, bud. Again, do what you like - if you want to crank your sharpness setting to the Max, go for it, Champ. Have at it. Enjoy.

Quote:
Originally Posted by beagle five View Post

well I know what I am doing and what I like, and you make it sound as if there is so much artefacts in the image that its impossible to look at it! well I dont know how terrible your sharpening is but mine isnt very bad at all, not on HD-DVD anyway, hardly any noticeble artefacts compared to the positive effects!

"Hardly any noticeable artifacts," huh? So you're admitting that what you're doing is causing artifacts - it's just that you like how it looks, right? That's fine and dandy, but don't expect to be taken seriously when you give your opinions of HD DVD releases.
Quote:
Originally Posted by beagle five View Post

so maybe that means that I actually know what I am doing and is actually founding my opinion in the real world by trial and error!?

No, it means you like things artificially sharpened to the point of artifacting.

Quote:
Originally Posted by beagle five View Post

I would say that I spend at least a couple of hours EVERY DAY calibrating and playing with the settings on my projector or dvd or hd-dvd just to find the perfect setting. and it sure isnt 0 sharpening on hd-dvds, well in a perfect world where hd-dvds actually are sharp it would be but now they are not!

Perhaps that's your problem. Calibrate it once - properly - and leave it alone. Better yet - have it done professionally. While he's there, ask him his opinion on cranking up the sharpness setting.

Your settings shouldn't have to be changed on a day-to-day or title-by-title basis. But again, to each his own. If you want to futz with your setup every day, go for it.
Quote:
Originally Posted by beagle five View Post

so MY opinion is, sharpness where sharpness is needed and as much as is needed
YOUR opinion ( not the vast majority I am afraid ) is that sharpness is terrible.

Wow - you just don't get it, do you? I'm all for detail and (higher) definition. What I'm against is artificial sharpness. I want to see what's on the disc, not an altered version.

Quote:
Originally Posted by beagle five View Post

and you are talking about that you want it to look like the master, and this thread is about the fact that VC-1 seems less sharp, well shouldnt YOU also be turning up the sharpness if VC-1 has a less sharp picture then the master?

Wow. Just wow. That certainly wouldn't work. You can't just create detail that's not there. You can't make a movie more detailed by bumping up the sharpness, you can only hurt the PQ by introducing all sorts of artifacts. Don't you understand that? (Of course not, hence your maxed out sharpness setting.)

Not to mention the fact that I'm nowhere near sold on the whole "VC-1 softens the picture" theory.

Steeb is offline  
post #196 of 2128 Old 04-04-2007, 06:20 PM
AVS Addicted Member
 
trbarry's Avatar
 
Join Date: Jan 2000
Location: Gainesville FL USA
Posts: 10,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Amirm started a discussion of how AVC tends to soften images. It is my own possibly unjustified speculation that in retaliation there is now a campaign to state VC1 softens pictures and that statement will be endlessly repeated until folks either believe or ridicule it.

It is much like one side or the other declaring the (format) war over and stating all objectives have been met.

- Tom

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

Tom Barry - Find my video filters at www.trbarry.com
trbarry is offline  
post #197 of 2128 Old 04-04-2007, 06:39 PM
AVS Special Member
 
Mr. Hanky's Avatar
 
Join Date: Mar 2006
Posts: 6,064
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 31
In part, I believe this falls somewhat into the "just desserts" category, since all of this "softening' talk centers around the deblocking loop filter which is used by both vc-1 and avc. Since the respective deblocking filters are implemented slightly different to achieve the same end, that gives a bit of wiggle room for the inevitable "we do it better than the other guy" sort of statements. One implements in the frequency domain and blends across 1 pixel while the other implements in the spatial domain and blends across 3 pixels. So depending on the macroblock size, one can claim the other is blending more pixels, and hence will observe a greater softening effect. However, one can also argue that if the degree of macroblocking is considerable, the softening effect can be just as dominating whether you are blending 1, 3, or whatever pixels. So the hypocracy is exposed of accusing the other guy of softening, when the first guy is just as vulnerable to softening if the macroblocking is severe enough.

Ironically, the workaround to this is simply to run more bitrate (which seems to be the philosophical antithesis of vc-1), so you aren't macroblocking and blending block edges in the first place.

Moral to the story- it is a silly strategy to point fingers at the competition for some downside, if you are similarly vulnerable to the same downside.

That's my 2 cts on that matter.

I need your sweet love, Rosetta Stone girl!
Mr. Hanky is offline  
post #198 of 2128 Old 04-04-2007, 10:16 PM
Senior Member
 
restart's Avatar
 
Join Date: Jul 2006
Posts: 421
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Mr. Hanky View Post

...

Ironically, the workaround to this is simply to run more bitrate (which seems to be the philosophical antithesis of vc-1), so you aren't macroblocking and blending block edges in the first place.

...

Except that in order to gain a higher compression ratio and to adhere to the lower peak video mux rates a HD-DVD VC-1/AVC encode must rely on the loop filter more. This is especially true for films of long duration which have grainy film stock. Loop filters do not add detail but rather forgive the sins of not having enough space/bandwidth to encode the large amount of detail and grain present in many films. That is why I have said no loop filter design is perfect. Playing around with loop filters at low bitrates can never add detail to a picture but rather acts as an attempt to fix the underlying space/bandwidth weakness of the physical disk format.

That is why the loop filter argument is a red herring.

I hope this explains the benefit of having higher physical disk specs.

PS. This is not a VC-1/AVC argument but a bandwidth argument.
restart is offline  
post #199 of 2128 Old 04-04-2007, 10:31 PM
AVS Special Member
 
Mr. Hanky's Avatar
 
Join Date: Mar 2006
Posts: 6,064
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 31
Hey, you're preaching to the choir, unless I misunderstand something in your reply.

While I can acknowledge where you are coming from when you suggest no loop filter would be perfect, I don't eschew its use entirely, in practice. I think it is a clever part to the whole package, but it should only come into play when all else fails (as far as premium hd production goes), imo. It should be there when no amount of available bitrate can help the situation (which I'm sure there are situations like that), like a safety net. I can certainly agree that "difficult to see" macroblocking is preferable over "easy to see" macroblocking, but either of those should take a back seat to no macroblocking, altogether, when possible.

I need your sweet love, Rosetta Stone girl!
Mr. Hanky is offline  
post #200 of 2128 Old 04-05-2007, 12:30 AM
AVS Special Member
 
benwaggoner's Avatar
 
Join Date: Sep 2003
Location: Portland, OR
Posts: 2,616
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by patrick99 View Post

I thought that someone had posted a question in the insiders thread asking for comment from Microsoft insiders on the difference between the results under VC-1 and MPEG2 in the case of the star from MI3 that has been posted earlier in this thread. I didn't recall having seen any response from Microsoft insiders to that invitation to comment.

Right.

The MPEG-2 looked like artifacts to me. Of course, without access to the source, it's hard to verify.

Digital Media Technology Insider with Microsoft

My compression blog
benwaggoner is offline  
post #201 of 2128 Old 04-05-2007, 12:30 AM
AVS Special Member
 
Issac Hunt's Avatar
 
Join Date: Aug 2005
Posts: 1,644
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10


here's the additional detail that ben seems to think is artifacts.

Quote:
Originally Posted by trbarry View Post

Amirm started a discussion of how AVC tends to soften images. It is my own possibly unjustified speculation that in retaliation there is now a campaign to state VC1 softens pictures and that statement will be endlessly repeated until folks either believe or ridicule it.

are you really suggesting that this thread is part of some orchestrated campaign against a particular codec? i envy you your imagination!
Issac Hunt is offline  
post #202 of 2128 Old 04-05-2007, 12:34 AM
AVS Special Member
 
benwaggoner's Avatar
 
Join Date: Sep 2003
Location: Portland, OR
Posts: 2,616
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by trbarry View Post

Amirm started a discussion of how AVC tends to soften images. It is my own possibly unjustified speculation that in retaliation there is now a campaign to state VC1 softens pictures and that statement will be endlessly repeated until folks either believe or ridicule it.

It is much like one side or the other declaring the (format) war over and stating all objectives have been met.

I again reiterate that folks are welcome to use the free Windows Media Video 9 Advanced Profile codec to see what it can do as far as detail, blocking, etcetera. It doesn't have segment reencoding and it's an older version of the codec, but it'll give a decent baseline of what is and isn't possible with VC-1.

Anyone want to mess around, I can tell you the settings to use to match HD DVD/BD encoding as closely as possible.

Digital Media Technology Insider with Microsoft

My compression blog
benwaggoner is offline  
post #203 of 2128 Old 04-05-2007, 12:37 AM
AVS Special Member
 
benwaggoner's Avatar
 
Join Date: Sep 2003
Location: Portland, OR
Posts: 2,616
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by restart View Post

Except that in order to gain a higher compression ratio and to adhere to the lower peak video mux rates a HD-DVD VC-1/AVC encode must rely on the loop filter more. This is especially true for films of long duration which have grainy film stock. Loop filters do not add detail but rather forgive the sins of not having enough space/bandwidth to encode the large amount of detail and grain present in many films. That is why I have said no loop filter design is perfect. Playing around with loop filters at low bitrates can never add detail to a picture but rather acts as an attempt to fix the underlying space/bandwidth weakness of the physical disk format.

It's important to distinguish between a simple deblocking filter, and an in-loop deblocking filter. The in-loop filter is much more useful and subtle in application, since it's part of the decoding of reference frames, and so allows later frames to be more accurate. So, in practice it's more complex than you suggest, and in fact can add detail to the picture by saving bits that would otherwise have been spent on correcting (potentially subtle) errors propogated from previous frames.

Also, neither optical format really has enough bits to not use deblocking in all cases. It's really a question of how much is too much, and how much is not too much.

Digital Media Technology Insider with Microsoft

My compression blog
benwaggoner is offline  
post #204 of 2128 Old 04-05-2007, 02:44 AM
AVS Special Member
 
patrick99's Avatar
 
Join Date: Oct 2006
Posts: 3,912
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 11
Quote:
Originally Posted by benwaggoner View Post

Right.

The MPEG-2 looked like artifacts to me. Of course, without access to the source, it's hard to verify.

Thanks for the comment.

Most who have commented in this thread on the difference in results in this star image from MI3 have said that the MPEG2 version looks sharper and more detailed than the VC-1 version. That is certainly my view. To me the VC-1 version looks "softened." But, as you say, without being able to make a comparison to the source it is not possible to know for sure.
patrick99 is offline  
post #205 of 2128 Old 04-05-2007, 03:52 AM
AVS Addicted Member
 
trbarry's Avatar
 
Join Date: Jan 2000
Location: Gainesville FL USA
Posts: 10,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by benwaggoner View Post

I again reiterate that folks are welcome to use the free Windows Media Video 9 Advanced Profile codec to see what it can do as far as detail, blocking, etcetera. It doesn't have segment reencoding and it's an older version of the codec, but it'll give a decent baseline of what is and isn't possible with VC-1.

Anyone want to mess around, I can tell you the settings to use to match HD DVD/BD encoding as closely as possible.

I'll reply since you quoted me in that post. I've used WMV9 Adv Profile and I think it does a very good job. I've also used X264 (great open source AVC codec) and it also does a very good job. I like them both MUCH better than MPEG-2.

And I think if you've seen one in-loop filter you've mostly seen them all.

But I was mainly commenting not on the quality of the codecs or filters but the direction and motivation of the various posts we've seen on this issue the last couple months.

- Tom

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

Tom Barry - Find my video filters at www.trbarry.com
trbarry is offline  
post #206 of 2128 Old 04-05-2007, 04:38 AM
Senior Member
 
Dot50Cal's Avatar
 
Join Date: Jul 2006
Posts: 291
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by rdjam View Post

All spin... all spin... The BD version looks dirty and clotted, but you seek to explain this as just a "higher detail"? rotfl

Reading through this thread I cant help but comment on how off base you are. You HAVE to know it as well, or else AVS is in a VERY sad state of affairs.

Xylon, Your doing an awesome job! Keep up the great work! I work with X264 myself, so seeing this is making me go back and test various settings in regards to the loop filter. Thanks!
Dot50Cal is offline  
post #207 of 2128 Old 04-05-2007, 04:57 AM
Member
 
totsugeki's Avatar
 
Join Date: May 2006
Posts: 34
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by WayneL View Post

Can anyone upload a fragment of scanned 35mm film that shows us what real grain looks like. I'll bet it doesn't look like anything we see.

512x512 crop (of a sky) from a 3840x2160 65mm master. Sorry it's not 35mm. You can download whole frames from ftp://vqeg.its.bldrdoc.gov/HDTV/SVT_MultiFormat/


For images of high resolution 35mm scans see http://www.clarkvision.com/imagedetail/index.html or http://www.xs4all.nl/~bvdwolf/main/f...e/example1.htm for example. Those are still film scans though, and I don't know how much film used for movies differs.
totsugeki is offline  
post #208 of 2128 Old 04-05-2007, 05:02 AM
Senior Member
 
Dot50Cal's Avatar
 
Join Date: Jul 2006
Posts: 291
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
totsugeki thanks for posting that. From viewing quite a lot of bluray movies the grain in that picture looks very much like what you see in Xmen 3 on Bluray. Seems to re-enforce the thought that the AVC is doing the encoding proper.
Dot50Cal is offline  
post #209 of 2128 Old 04-05-2007, 08:53 AM
AVS Special Member
 
MovieSwede's Avatar
 
Join Date: Feb 2007
Location: Gothenburg
Posts: 6,761
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 37
Grain comes in different flavours, but that doesnt mean that a grainy encoding shows proper grain.

Good movies are as rare as an on topic discussion.
MovieSwede is offline  
post #210 of 2128 Old 04-05-2007, 09:34 AM
AVS Special Member
 
nataraj's Avatar
 
Join Date: Mar 2000
Location: WA
Posts: 8,020
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by totsugeki View Post

512x512 crop (of a sky) from a 3840x2160 65mm master. Sorry it's not 35mm.

When it comes to that level of detail - don't we have to deal with scanning inaccuracies as well ? Or is that small compared to the details we are seeking here ... ?
nataraj is offline  
Reply HDTV Software Media Discussion

User Tag List

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


Forum Jump: 

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

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