HiFi Writer Blog

Regarding my earlier post (two down) on possible differences between US and Australian Blu-ray discs for quality, Sandra has written to me:

I was also present the night John talks of. There is an alternate hypothesis to encoding variations or imagination to explain these perceived differences. Is it possible that physical variations in the disks arising from differences blu ray disk production quality control standards or practices could lead to slight differences which are material for players i.e. which lead to visible differences for the user? I know the “digital is digital” argument regards 1’s and 0’s but it seems clear that some brand new disks sometimes have playability issues on some players. Could it also be possible that the tiniest variations in physical disks could result in slightly visible differences at the player level?Is this possibility being suggested in the following white paper from quality control specialists in the blueray production process? See:

http://www.media-tech.net/fileadmin/templates/BD_News/OTO222_p30-33_whitepaper.pdf

On p2 under the heading “Coating thickness uniformity: local defects” they state (2nd para):

“Bumps in the coating layer do not directly affect the signal quality. However, they may cause optical aberrations that lead to focus and tracking errors.”

One does not have to understand all the details of this highly technical paper, to get the point that bluray production is FAR more technically demanding than is standard DVD. This is after all new technology. I do hope I am wrong, as the implications are gruesome. I cannot help but wonder whether occasionaly the pressures on increasing yields could cause production houses to drop standards, or whether just natural random variation in the physical disk, even within agreed tolerances, can cause some players to playback ever so slightly differently even the same encode?In any case I agree with you that we need to do further comparisons, on the same gear, blind, to see if any variances can be repeatedly detected.

In the mean time, from your knowledge of bluray production and players, do you think this alternative hypothesis i.e. we just happened to strike a “dud” Australian disk, at all possible?

This is an interesting theory, but I have strong doubts about it.

There are two ways that data reading difficulties could, arguably, feed into performance. One is if the data itself is corrupted as a result, and clearly this could have major implications. This way is not in dispute by anyone, and I will return to it in a moment.

A more subtle way that is sometimes raised is that the data ends up not being corrupted, but only because the equipment is working harder at correcting the data. That in itself is said to feed back into the reproduction chain in some way, reducing quality.

I’ve always thought that this latter argument is not very strong at all, since the effects (for example, greater power draw by the processing chips) would be of a magnitude similar to other pseudo-scientific nostrums peddled in the hifi and home theatre arenas (eg. the £500. P.W.B. Quantum Clip).

So let us return to corrupted data. There is a major disconnect between the claimed corruption of data drawn from a Blu-ray disc, and the alleged effects on the screen. Intuitively it would seem that, say, the loss of some proportion of the data would translate to the loss of some proportion of the picture, perhaps fine detail or colour fidelity.

That’s how things work in analogue systems. But modern digital stuff does not work like that.

The data on a Blu-ray disc is not an analogue of the original signal. Through the compression systems it has been dismantled and repackaged in a completely different way that lacks all proportion to the original signal. Ultimately those proportions will be restored (more or less, because the compression is lossy), but while the data is being laid down on the disc, while it resides on the disc, and while it is being read from the disc, it has no apparent relationship to the original data, except via advanced mathematical transformations. Misplacing bits beyond the capabilities of the system to correct them will result in quite unpredictable errors.

When it comes to reading the disc, a significant number of read errors will cause the signal reconstruction to simply fail. What you will get is not subtle loss of detail, but gross errors such as macro blocking, dropped frames and even total signal loss.

That is how Blu-ray (and DVD) signal problems manifest. Not in loss of sharpness, detail, contrast and ‘life’ in the picture.

Yesterday I received a very interesting email from a reader, headed as shown above. The matters raised will be, I think, of interest to other readers on this Blog, so I’m answering it here. First, here’s the email (lighted edited for grammar):

Hi, while comparing Blu-ray players and HDMI cables last night we (4 of us) discovered while using The Fifth Element Blu-ray as a reference (which we had 2 copies of) that there was a significant difference between them. One of the disks was a American import, the other an Aussie, both were the remastered versions.The American version was clearly sharper, more detailed, better contrast and had just more life to the picture, discussing this further one of the group said this was not the first time he had witnessed this, and went on to say he saw similar improvements on other Blu-rays such as The Dark Knight Region A over the B version .

We were using some pretty good gear, and whilst we could see improvements in players and then HDMI cables, the improvement in software was also significant.

This is to me terribly frustrating as I do not want to start replacing my Aussie Blu-rays for American ones. Would you be able to find out any info on this if possible please I would really appreciate it.

Here’s a list of the visual equipment :-

JVC HD1 1080p projector
Oz Theatre screens Majestic 110 inch
Sony BDP-S1E
Fully light controlled room (dark walls and ceiling etc)

Compared equipment:-
OPPO BDP 83 Blu-ray
Sony 5000 ES Blu-ray
Kordz master reference 7m HDMI cable

Regards John

There are several issues here, but I’ll start with the last concern: whether John should replace his Australian Blu-ray discs for American ones. The answer is: sometimes. But only if there really is a significant difference, only if it is the US version that is significantly better than the Australian version, and only if he has a player which will work with the US version.

If there are significant differences, I would generally expect that to be because the actual encoding of the data is different (different codec, for example, or different video bitrate). That would generally be because different companies have produced the discs for the different parts of the world, so these discs are more likely to be region coded. Obviously, an Oppo BDP-83 Blu-ray player purchased in Australia is likely to be fitted out to play discs from other regions, but few other players can do this.

In the specific cases mentioned — The Fifth Element and The Dark Knight — the US versions are not region coded. They will play happily in any player in the world.

Now let us assume for a moment that the US versions of these two movies do look better than the Australian versions, that does not mean that US Blu-rays in general look better than Australian ones. If the Blu-ray versions of all movies look different in their US and Australian versions, then the default assumption would be that of the ones for which a ‘this is better than that’ judgement could be made, in roughly half the cases the Australian version would look better than the US version.

Consider a gross example: the Australian version of Watchmen comes from Paramount with a much higher bitrate MPEG4 AVC transfer in Australia, than the US Warner Bros VC1 version. However the US version is the longer Director’s Cut, so there’s a tradeoff.

Now we come to the specifics of the claims for these two movies. I have written extensively about The Fifth Element in my review of the Australian version here. I have written extensively about the differences and similarities between this and the two US versions here, finishing it off quite conclusively, I think, here. I believe that in these posts I establish that the video in the Australian version released by Sony and the remastered US version is identical. Having said that, I have been advised by Sony that it has disposed of distributorship, at least in Australia, for this movie. I just checked and this movie does not appear to be available in either DVD or Blu-ray format from any of the major on-line Australian disc retailers (DVD Orchard, EzyDVD, DVD Crave, Atlantic DVD, DVD Downunder, Devoted DVD). So if you want this movie, for the moment you’re going to have to buy from overseas.

As for The Dark Knight (my review of which is here), I invite readers to compare scans of both the Australian and US versions of this disc which were conducted using the tool BDInfo. These are on the Unofficial Blu-ray Audio and Video Specifications Threads, hosted at the AVS Forum. Those two posts show that both versions of this disc have exactly the same average video bitrate. From my experience with Warner Bros titles, to me that means that these are bit for bit identical copies. I note that the average bitrate for the English language Dolby TrueHD tracks is also identical. So the only difference seems to be the additional streams in the Australian version, adding a number of audio and subtitle languages.

So, are there actual visible differences between the US and Australian versions of these two movies? People like me are inclined to say ‘No’. If the video streams are identical, then they should be decoded identically on any given player. So I’m inclined to think that the differences John reports are imaginary. It’s the same phenomenon that can lead people to think that putting a plain piece of cardboard under one leg of a couch in your listening room can make your system sound better.

I suppose that the additional audio and subtitles streams included in the Australian version of The Dark Knight could, in some way, interfere with a player’s performance in extracting data from the disc, since there is a bit more of it, but it doesn’t seem likely. In any case, it is the US version of The Fifth Element that has the extra audio data, so such an effect should be around the other way with this movie.

I have previously received a report of alleged inferior performance of the Australian version of The Fifth Element, so perhaps there is some other version floating around out there. Unless proven, though, I’ll stick with the opinion that the differences apparent between the two versions (and, the differences between competent HDMI cables) are imaginary. And I’ll remain of that opinion until a well-conducted, statistically significant, double blind trial is conducted showing otherwise, and is successfully repeated a couple of times.