How Will High Bit Rate VC-1 Compare to AVC

[Teazle]

Originally Posted by reiella

[ Add / Edit ]
And quite the point, do you look for shortcomings on the film when you notice that your PS3 says "VC-1 14mbps", or do you look for the "VC-1 14mbps" after seeing shortcomings?

Ok, but that isn't quite fair, since some shortcomings due to low bitrate might not be apparent unless you already knew what to look for having seen the source. This doesn't really hold for motion artefacts; more for situations where detail is obliterated or obscured due to low bitrate.

Addendum:
I've seen this myself in a very amateurish setting in ripping my DVDs to mp4s for playback on the PSP. Sometimes if the bitrate is set too low, especially for live action the result can look indistinct and hard to distinguish from poor focus. (For me the solution was to max out the bitrate setting for live action content and buy a bigger memory stick.)

[patrick99]

Originally Posted by reiella

But you'll trust the people who are putting out the disc to actually utilize the supposed bandwidth?

[ Add / Edit ]
And quite the point, do you look for shortcomings on the film when you notice that your PS3 says "VC-1 14mbps", or do you look for the "VC-1 14mbps" after seeing shortcomings?

As I said in response to an earlier comment, these are the types of responses I am used to seeing at AVS, from people who refuse to admit that the bitrate difference between BD and HD DVD has any importance.

And no, I don't trust the people putting out the disc to actually utilize the available bandwidth, since I have seen far too many cases where they have not.

[reiella]

Originally Posted by Teazle

Ok, but that isn't quite fair, since some shortcomings due to low bitrate might not be apparent unless you already knew what to look for having seen the source. This doesn't really hold for motion artefacts; more for situations where detail is obliterated or obscured due to low bitrate.

Addendum:
I've seen this myself in a very amateurish setting in ripping my DVDs to mp4s for playback on the PSP. Sometimes if the bitrate is set too low, especially for live action the result can look indistinct and hard to distinguish from poor focus. (For me the solution was to max out the bitrate setting for live action content and buy a bigger memory stick.)

Quite, but here's the thing, you can have wasted bandwidth as well. Most likly not the case with the frame we're refering to, but it's still there.

Are high-action scenes higher picture quality than slow action scenes? Because of where the compression falls?

Fun tidbit though with your rip comparison though, you initially do have a better idea of how it should look than say from a blind-test. Now consider the person who didn't see the title in a theatre... It's a hard matter to judge. Compounded with say those who couldn't see it at a theatre with a good projectionist. And again compared against those who saw the scene with their own eyes. It's a hard matter to know 'what the fault' is, which is partly why there are technical critics for movie releases, but *shrug*.

Consider: The upper-end is 40 mbps for video, does that mean every title that doesn't utilize 40 mbps for their video 'shortcoming'?

Originally Posted by patrick99

As I said in response to an earlier comment, these are the types of responses I am used to seeing at AVS, from people who refuse to admit that the bitrate difference between BD and HD DVD has any importance.

And no, I don't trust the people putting out the disc to actually utilize the available bandwidth, since I have seen far too many cases where they have not.

I did not mean available bandwidth. I meant supposedly utilized bandwidth. I'm sorry if my statement wasn't clear in that regards.

[Teazle]

In my amateurish AVC encodes for the PSP, where I start with an mp2 and wind up with an mp4 of the same resolution but perhaps half the file size, a nice easy way for me to waste bandwidth would be to use too many key frames. Doing that of course defeats the purpose of video compression. (I'm just trying to think of an uncontroversial example of a dumb use of high bandwidth which doesn't improve PQ.)

On the question of how often wasted bandwidth occurs in professional Blu-ray encodes. Fantastic 4 Rise of the Silver Surfer is a letterbox movie of only 92 minutes with very high bitrate throughout (it happens to be AVC btw). The film looks perfect and in the fast-action scenes a high bitrate is surely needed. But maybe here's a case where if they spent more time tweaking they could have gotten away with less in some places. I mean, I would hope that the 40mbps peak would be adequate for fast-action scenes which use the full 1920x1080 screen size, and they often use it for 1920x817 or whatever it is.

But I think they didn't bother much because the film is only 92 min. so there was less worry about final file size.

[patrick99]

Originally Posted by reiella

Consider: The upper-end is 40 mbps for video, does that mean every title that doesn't utilize 40 mbps for their video 'shortcoming'?



I did not mean available bandwidth. I meant supposedly utilized bandwidth. I'm sorry if my statement wasn't clear in that regards.

I am afraid I am still not clear on the meaning of the supposedly utilized bandwidth concept.

When I see bitrates in the high twenties, they often go along with PQ that looks suspect. For example, in Silver Surfer, when the bitrate is well into the mid 30's, the PQ generally looks excellent, but when it is in the twenties, it doesn't look as good. Whoever did the compression seems to have felt that Jessica Alba warranted mid 30's bitrates, but that mid to high twenties were enough for Chris Evans.

[patrick99]

Originally Posted by Teazle

On the question of how often wasted bandwidth occurs in professional Blu-ray encodes. Fantastic 4 Rise of the Silver Surfer is a letterbox movie of only 92 minutes with very high bitrate throughout (it happens to be AVC btw). The film looks perfect and in the fast-action scenes a high bitrate is surely needed. But maybe here's a case where if they spent more time tweaking they could have gotten away with less in some places. I mean, I would hope that the 40mbps peak would be adequate for fast-action scenes which use the full 1920x1080 screen size, and they often use it for 1920x817 or whatever it is.

But I think they didn't bother much because the film is only 92 min. so there was less worry about final file size.

I don't agree that it looks perfect; it often looks perfect, but sometimes not. And likewise it doesn't have very high bitrate throughout; the bitrate level seems to fall off somewhat as the movie progresses, and even earlier in the movie it is somewhat inconsistent.

[Teazle]

Originally Posted by patrick99

I don't agree that it looks perfect; it often looks perfect, but sometimes not. And likewise it doesn't have very high bitrate throughout; the bitrate level seems to fall off somewhat as the movie progresses, and even earlier in the movie it is somewhat inconsistent.

I take your word for it as this was a renter for me and I only spot-checked with the bit meter when I saw it ... maybe I didn't pick a good example.

About reiella on "supposedly utilized bandwidth", I assume s/he is just pointing to the possibility that bandwidth can be "wasted" in that if the encoding is done stupidly or incorrectly, higher bitrate need not redound to better PQ.

[patrick99]

Originally Posted by Teazle

I take your word for it as this was a renter for me and I only spot-checked with the bit meter when I saw it ... maybe I didn't pick a good example.

About reiella on "supposedly utilized bandwidth", I assume s/he is just pointing to the possibility that bandwidth can be "wasted" in that if the encoding is done stupidly or incorrectly, higher bitrate need not redound to better PQ.

One example where the relatively high bitrate did not seem to result in appropriately good PQ was Disturbia.

[reiella]

Originally Posted by Teazle

I take your word for it as this was a renter for me and I only spot-checked with the bit meter when I saw it ... maybe I didn't pick a good example.

About reiella on "supposedly utilized bandwidth", I assume s/he is just pointing to the possibility that bandwidth can be "wasted" in that if the encoding is done stupidly or incorrectly, higher bitrate need not redound to better PQ.

Quite, the bandwidth is used, just not to good effect. Your example is better than my stupid one following so .

Say you have a solid color patch [such as say, the ending credits for a moronic example], instead of saying 'black'/'black'/'black' for each frame, say 'black for' x frames. Same net effect, but the former would utilize more data [and thus a higher bandwidth].

Low Bandwidth can be the cause of problems, obviously, however, it isn't the problem in itself. The problem, of course, being the symptom of the poor picture quality.

The hyperbolic example of why it's ok for studios not to utilize a constant 40 mbps bitrate [that is an available option, neh?].

My personal view, and take on it, should be look for the problem first, don't go looking for the problem after seeing what may be the cause of a potential problem. Unless, of course, you're a technical critic, in which case, hats off to you .

There is a very strong point that you may not know 'what you're missing' without knowing the possible sources of technical flaws [which is what I was rambling about with the original frame of reference remarks].

And of course Jessica Alba warrants mid-30s bitrate, why do you think most people buy Fantastic Four? :P

[PaulGo]

Getting a great movie quality is more than just bit rate. Even with a very high bit rate the type of encoder plays a significant role. Also, for difficult passages you need a specialist to tweak the encoder for the best results. When you do this tweaking I assume you are making compromises to get the best visual quality. The higher bit rate that Blu-ray allows should make it easier to get a higher quality encode. I am excited that with HD-DVD out of the way all the movie studios will try to optimize their product for this higher bandwidth. The question in my mind is if VC-1 is as capable of producing (at a high bit rate) the quality of AVC. I believe Richard Casey thinks for most purposes AVC is superior since he stated he has a preference for using it in the future. The test disc that he is developing is designed to rigorously test these codecs and show what flaws they have.

[PaulGo]

Mr. Hanky, I reactivated the old thread on AVS that I was referring to:
"Codec Wars" : The attempt of an objective AVC/h.264 versus VC-1 benchmark

Ben Waggoner from Microsoft stated he would have a 24p clip.

[WickyWoo]

I don't trust Ben Wagggoner to provide something not geared toward VC-1's strengths as far as I can throw the entire MS campus

[PaulGo]

Originally Posted by WickyWoo

I don't trust Ben Wagggoner to provide something not geared toward VC-1's strengths as far as I can throw the entire MS campus

It will be interesting to see what happens.

The more I think about it and the more research I do I think my comment above is very valid:

"It seem Microsoft is reluctant to participate in these comparisons. I was trying to think as to why this might be. The most obvious answer is they have nothing to gain. VC-1 has good tools, but I was thinking about how and why it was created. It was created for lower bit rate applications, so perhaps it does not scale up to a higher bit rate as well as the latest version of AVC. So perhaps it cannot take advantage of the increased bitrate that Blu-ray has to offer."

[bkbluray]

If I'm not mistaken, 3:10 to Yuma is a high bit-rate VC-1 codec and looks just as good as any Disney AVC encode.

[Clark Kent]

No one should forget MPEG2. Shorter movies on a 50 GB BD can look phenomenal with MPEG2 at high bitrates. For anything longer than 90 minutes though I would use AVC due to space constraints. While Disney's high bitrate VC-1 encodes looked very good I have consistently preferred AVC encodes to VC-1 encodes. Most of the VC-1 encodes I've seen look artificially smooth while AVC seems to remain sharp.

[PaulGo]

Originally Posted by bkbluray

If I'm not mistaken, 3:10 to Yuma is a high bit-rate VC-1 codec and looks just as good as any Disney AVC encode.

I have looked at several reviews - this one seems representative of some minor faults - could the result have been better with AVC? Unless we had it encoded both ways we will not know. But when Richard (R&B Films) releases his demo disc we will have a better idea as to which codec is better.

However, on very high contrast transitions, there are inconsistently visible halos. Fortunately, they are of very low amplitude and are very thin; on properly adjusted displays of under 65-inches, they will probably not be noticed. And in more conventionally lit scenes, halos are not visible at all. Grain can be a bit apparent, but it’s not intrusive. Nitpicks aside, this is a very pleasing film-like transfer.

http://www.dvdfile.com/index.php?opt...=6484&Itemid=3

[patrick99]

Originally Posted by PaulGo

I have looked at several reviews - this one seems representative of some minor faults - could the result have been better with AVC? Unless we had it encoded both ways we will not know. But when Richard (R&B Films) releases his demo disc we will have a better idea as to which codec is better.

However, on very high contrast transitions, there are inconsistently visible halos. Fortunately, they are of very low amplitude and are very thin; on properly adjusted displays of under 65-inches, they will probably not be noticed. And in more conventionally lit scenes, halos are not visible at all. Grain can be a bit apparent, but it’s not intrusive. Nitpicks aside, this is a very pleasing film-like transfer.

http://www.dvdfile.com/index.php?opt...=6484&Itemid=3

3:10 to Yuma does have some minor EE, but nothing like that in Disney's High School Musical 2.

[Neo65]

Originally Posted by Teazle

In my amateurish AVC encodes for the PSP, where I start with an mp2 and wind up with an mp4 of the same resolution but perhaps half the file size, a nice easy way for me to waste bandwidth would be to use too many key frames. Doing that of course defeats the purpose of video compression. (I'm just trying to think of an uncontroversial example of a dumb use of high bandwidth which doesn't improve PQ.)

On the question of how often wasted bandwidth occurs in professional Blu-ray encodes. Fantastic 4 Rise of the Silver Surfer is a letterbox movie of only 92 minutes with very high bitrate throughout (it happens to be AVC btw). The film looks perfect and in the fast-action scenes a high bitrate is surely needed. But maybe here's a case where if they spent more time tweaking they could have gotten away with less in some places. I mean, I would hope that the 40mbps peak would be adequate for fast-action scenes which use the full 1920x1080 screen size, and they often use it for 1920x817 or whatever it is.

But I think they didn't bother much because the film is only 92 min. so there was less worry about final file size.

When the source is lossy compressed video like mpeg2, there is already significant loss of detail from the original, so you can't create more actual entropy than what is available in your mpeg2 source, you can at most create more noise/artifacts but by transcoding with a different codec, hence different DCT, different math, you're getting even further away from the original master.

That's not how higher bitrates are useful. If you have tools to analyze the encode itself, you'll see that each macroblock has a quantization number/scale/index --- this number determines the lossyness of the encode. IE: after the DCT, what number(s) to divide the coefficients by (hence discarding the remainder of that division). Note that there is also a quantization matrix as each term in the macroblock is divided by a different number.

After the stream is encoded, during decode, these numbers are multiplied back with the quantization factor(s) before going the reverse path of IDCT and motion compensation etc to build the macroblock to what the original is supposed to be. Except, because of the quantization step, that original value is no longer available. This is because the remainder is discarded. Eg : 2008/13=154 [with 6/13 discarded], 154*13=2002 !=2008.

The reason why you need the higher bitrate is that while some streams have easy segments, in just about any movie there's always some that are complex enough that if you let the quantization number to stay small, these complex segments will not fit within the limit of the spec --- ie: the ceiling of available bandwidth.

When a few frames together exceed the buffer limits and max bitrate allowed (28 [actually less with various audio] for HD DVD, 40+ [can be higher] for BD), the size of those compressed frames have to be reduced, meaning quantization has to be increased. This is what the HD DVD people talk about when they say tweak the encode --- select regions of interest and reencode the frames at lower bitrates/higher quantization while keeping the important regions quantization lower. Under ideal conditions, this is just tedious and more work, under less ideal conditions, the areas not within the ROI could become over compressed or the ROI also has to have their quantization factors increased.

Another side effect with higher quantization is that the in loop filters are adaptive, meaning they kick in more with higher quantization as they are designed to hide/smudge the artifacts that arise with higher quantization.

[PaulGo]

This was just posted by Richard Casey (R&B Films) on AVS, I bolded the portion that supports my thoughts about AVC vs VC-1:

Originally Posted by RBFilms;

I am popping in here albeit I swore that I swore off this forum a few weeks ago. AVC seems to use the higher bit rates more effectively than VC-1...while VC-1 does a great job at lower bandwidth rates. I will be running some encode tests fo my own to see and making them avalable to the public.

I believe that Ben from MS has agreed to participate. We welcome his support for our testing...as long as it is sincere and they actually call us back to us this time... We have heard this promise before and it never happened. We were told by Amir that MS is to busy with "Big Studios" to support someone like us ... even though a promise was made by MS ... they did not follow through in a timely manner and did not support us in the end.

I do not visit this forum often ... may check back in a month ... so don't take it personally if I do not respond to posts. I really don;t have the time these days. Also, I am spending more time in the Blu-ray Forum now. The group over there seems to be less biased and far more open minded to all possibilities. AVS seems has come across on the past as primarily focused on supporting VC-1, HD-DVD, MA, and Toshiba. I myself am open to all possibilities for superior performance and ultra fidelity for the ULTIMATE Home Theater presentation. I prefer to remain open minded about technology and innovation moving forward.

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 ... rather than letting water seeks its own level so to speak ... it may end up just like HD-DVD.

This has always been the MS powerhouse 800 Pound Gorilla Dictator marketing strategy ... telling people what they what rather than listening to what people want .... like Apple and other great companies have done in the past. Not that I like iPods or the crappy sound quality of iTunes ... not by any means. It is pure junk but it does serve today's lifestyle.

In the end, I respect Apple's "Affinity" marketing approach compared to Microsoft's Push Marketing. At least I feel good about making a choice with Apple. I don't mind being manipulkated ... I just respond better when it is done with some class, style and grace.

[PaulGo]

Neo65, thank you for the excellent post about compression technology!