Id love to see a VC1 title encoded to fit blu-ray specs @ 50GB and the same title encoded for a 30GB HD-DVD encoded with its specs.
For the time being though Im highly skeptical on VC1, It seems more often than not to smear the picture with the loop filter and still introduce macroblocks. Sharp picture is one of the main highlights of HD video, and they are essentially taking that away in order to hide the codecs deficiencies at low bitrates.
We have to remember that this release is not up to par in PQ compared to Aeon Flux or MIII. This release is just too grainy. I personally think that this needs remastering. Don't get me wrong the PQ details are there but the "cranked up" grain nullifies it.
A lot of people were complaining why I only put up BD25 Blu-ray against HD DVD. So to placate them this time I used BD50. Hopefully that ends that issue. BTW anyone can help list more BD50 dual format releases? We need more titles to check up on.
Its much appreciated Xylon, In the future Id love to see you do more of these kinds of comparisons, with the best of the best on each format pitted against each other. Its a shame that we rarely get the same movie on both formats with encodes that cater to both formats maximum specs.
Anxiously awaiting that Flags of our Fathers test though!
Transparent when you watch the movie. They never claimed lossless. You cant se this issues at 24fps. Your eyers isnt that good.
The encoder actually watch the master and compare. And if a scene doesnt look right, they do a segment reencode. They dont do frame by frame analysing. Because we dont watch movies that way.
As for Mpeg2 preserving more grain and detail. It shows the number on the helicopter better but at the same time shows the pilot head worse. As for grain. At lot of thoose grain are blockartifacts, not filmgrain.
But the funny thing is, when you watch the movie you cant se either of these things. Because it happens so fast. Thats how we they can fit the movie at less then 5% space of what it would be if it werent compressed.
and secondly i'd like to ask you to include (if possible) the actual bitrate at the timestamp of the screencaptures together with the average you already provide.
Bitrate applies to a scene, not a single frame as being done by Xylon.
And I fail to see why the bitrate matters, other than more popcorn for people to throw at one side or the other to justify their feeling that brand X is better than brand Y.
All that matters in the end is how these films look in our living rooms and for those of us with screens less than 100' I am having a very hard time believing that there is going to be any noticeable presentation difference with the examples that we have seen so far.
If anything this trumps the "mpeg2 sucks" and the "hd dvd doesn't have enough space" arguments.
The single frame represents the instantaneous state of the scene, that comes as a consequence of the ongoing bitrate. Granted, they don't rigidly/exactly correspond to each other down to the millisecond, but they do share a general correlation.
Margot Robbie is THE most exquisite human creature on the face of the planet, in this moment in time.
I did not asked for the instantaneous bits consumed, this would need a stream analyser. And we could fall on a "I frame" vs "b frame" situation, something like 80Mbps vs 2Mbps. This indeed would be ridiculus.
I think that mpc with haali shows a "running 1 sec average" as actual bitrate and this is what i would like to know.
As about the conclusions that can be extracted .. let those who care extract them.
I see plenty of blocking on the vc-1 encodes even with the loop filter. I cannot believe this blocking would be present on the master. A movie with lots of grain and motion would stress an encoder since grain is temporally and spatially random. High motion can reveal block edges. So the combination of both are a good encoder test, trying to preserve grain structure while avoiding blocking. Too much filtering to reduce blocking can cut off high frequency detail (aka grain structure). But filtering may be applied by an vc-1 encoder in order to adhere to the space & bandwidth constraints of the format.
Awesome can't wait. TD was heralded as the first "perfect" disc as I recall. But I always thought it was a bit under par in terms of fine object detail. it'll be interesting to compare BD w/ the HD DVD.
Forget the macroblocking. How about that filtering on the vc-1? That surely isn't on the master.
Looking forward to more vc-1 vs mpeg-2 comparisons. Training Day should be interesting. The vc-1 version should be better.
I think we can all agree that that image is pretty poor in both versions.
Certainly nowhere near what we have seen either VC1 or even mpeg2 do in a best case scenario.
I pretty much feel a lot of the problem is in the master, as th macroblocking in the VC1 sample is definitely "atypical". Sure, there is additional macroblocking in the Mpeg sample, but this looks more sharply defined, leading me to believe that it was added in the mpeg encode after the others from the master.
I think we are seeing a lot of macroblocking in both versions. This is very much *not* typical behaviour for VC1.
I am with those who suspect that it was a bad master that had macroblocking present. A master is not necessarily free of any artifacts.
Rubbish. HDCAM SR and D5 tapes have no blocking like this. And neither does the DI all elements were made from.
And let's assume the master had blocking which it did not. Why would VC1 erase the number and MPEG2 less/not? VC1 filtering more than MPEG2 from a blocky master with a number?
I'd love to see a comparison of Underworld Evolution bluray, to the European VC1 HD DVD release.
I have both discs and can help, if needed.
Hmmmm it looks like I'm going to need help with Underworld. The brightness and constrast is a bit higher on HD DVD even after using 16-235 TV levels on both. Its the same filters used with previous pictures. I checked and re-checked my settings and this only happens on Underworld. I don't see the same results with other movies.
The only time this happens (sometimes) and not usually caused by improper settings or filters used is when comparing Mpeg-2 broadcasts and HD discs.
Seems like the gamma was boosted on the HD-DVD, or the BRD is darker. You can replicate this in PS with levels, 1.1/0.9 on middle slider.
The HD-DVD has better details (eyelashes, lips, hair). Of course it would, it's vc-1, with the same file size. It better be better. There is a lot more noise visible on the HD-DVD though. Not sure if it's grain or what.
And oh yeah, the mpeg-2 version of Training Day looks terrible, at least on that shot. Maybe the rest fares better but you can already see ringing, blocking, and noise all over the car, like it came from a HD broadcast. It even has a broadcast file size: 13.9gigs. Really?
WTF? Training Day is only 13G on blu-ray? Is this for real? I though Warner blu-ray are all the same VC1 as HD-DVD? Someone need to make a list of these ****** transfer!
That HD-DVD UnderWorld is interesting. Where does it come from? Compare to MPEG2 with the same bitrate and size, VC1 sure must look quite a bit better if the source is similiar. I may want to get my first HD-DVD for my unopen XBOX360 HD-DVD drive!