Stanley Kubrick Collection

[movielocke]

Clearly you're right

[MichaelB]

Just out of interest, does anyone who has actually watched the Blu-ray (a group that I assume still excludes Nothing) regard it as a disappointment?

[knives]

No, I'm not a Lyndon expert, but I didn't notice anything egregious. If this is a hack job it's the complete opposite sort of hack job from say Pabst's Comrades.

[Nothing]

Clearly you're right

That Kubrick didn't intend for the film to be vertically pan and scanned is unequivocal. That EITHER this version OR every past home video and television version of the film up until the year 2001 is distorted is also unequivocal. That Kubrick did not intend for the film to be screened at 1.78:1 is unquivocal too, since no such theatrical ratio existed in 1975. That Kubrick intended for the film to be screened at 1.66:1 (or "no wider than 1.75:1") is strongly suggested by multiple sources. The combination of all of the above is enough to conclude that the telecine is flawed, but thanks anyway for your supercillious comment.

In answer to Michael's question, amongst Kubrick enthusiasts on the Glenn Kenny board and elsewhere there certainly is discontentment amongst those who own the BD.

[Roger Ryan]

Personally, I prefer the barrel of a gun to be circular which makes the accuracy a lot better than one that is oval
Therefore, I tend to think that the earlier DVD image is the one distorted, not the Blu-ray.

[Brian C]

... they should perhaps consider that if this ISN'T stretched it means that every other transfer of the film, from 1980 through to 2001 (incl. the 1990 lasterdisc transfer and the 2001 DVD transfer, which have a very similar framing) is squeezed.

You keep saying this like it's your trump card, but I honestly don't know why you think this possibility is so remarkable. If "every other transfer" was the same, then the only way to verify accuracy would have been to compare to a 35mm print. In other words, who would have noticed, or been able/motivated to make the case like you have here? I know you like to present Kubrick as an infallible god in these matters, but as I said earlier, it doesn't seem at all unlikely that he wouldn't have noticed (or cared, frankly).

For that matter, you would have never noticed the difference either had the aspect ratio issue not led you to it. Don't get me wrong, your detective work here is commendable. It's good to know the facts ... but your conclusions just don't logically follow.

Personally, I prefer the barrel of a gun to be circular which makes the accuracy a lot better than one that is oval
Therefore, I tend to think that the earlier DVD image is the one distorted, not the Blu-ray.

Heh.

[swo17]

The barrel does look more perfectly round in the BD grab to my eyes as well. Though with the gun turned at an angle, you wouldn't expect to see a perfect circle.

[jedgeco]

Just out of interest, does anyone who has actually watched the Blu-ray (a group that I assume still excludes Nothing) regard it as a disappointment?

Not one bit.

[Nothing]

Personally, I prefer the barrel of a gun to be circular which makes the accuracy a lot better than one that is oval Therefore, I tend to think that the earlier DVD image is the one distorted, not the Blu-ray.

Considered this briefly earlier, however the gun barrel is close the edge of the frame so could very likely be subject to lens distortion (and may also be at an angle, as swo17 notes).

Agreed that one cannot conclusively say whether it is the new transfer or all of the previous transfers that were distorted, but given the accompanying facts (the incorrect aspect ratio, the vertical p&s, that this is the only transfer that looks like this, that Kubrick approved the earlier versions) it doesn't seem an unreasonable conclusion to draw. Even without the distortion argument, the incorrect aspect ratio and p&s are enough to conclude that the transfer is flawed.

As for all the folks jumping on to say the BD looks 'great', the question is how well acquainted they were with the film before, how many times had they seen it, have they seen it in the cinema? Without this past familiarity such comments are fairly inconsequential.

[Brian C]

Without this past familiarity such comments are fairly inconsequential.

They are anyway, if we're being honest. Cinema conditions vary so widely and for so many reasons that even numerous theatrical screenings - even provided that people have uncommonly accurate memories of what they saw (I suspect that most people, though not all, are frankly full of pure shit when they talk about color schemes, framing, etc., at screenings a few days ago, much more so when we're dealing with several decades) - can hardly be considered to provide authoritative familiarity in any way.

Just for clarity's sake, I agree that the aspect ratio is highly questionable from a verifiable, factual point of view (although I'm less convinced that it's consequential). And I agree that Vitali's statements on the matter seem bizarre. I must repeat what I've said before - I don't find this conversation to be without merit.

But beyond that? Pure speculation and/or simple personal preference. The living person who would normally be considered the most authoritative on the subject (Vitali) has been judged to be a crackpot, so what we're left with are just guesses in the absence of more definitive evidence than a handful of screenshots. And barring the emergence of such evidence, it seems that interested consumers have enough information at their disposal to make an educated decision as to whether or not to buy the disc.

[matrixschmatrix]

"Vertical pan and scan" is inaccurate- cropping is not inherently the same as pan and scan, and you're repeating the same objection twice (as the cropping, or at least the loss of information, is due to the change in aspect ratio.) I would also argue that give the number of Kubrick releases that were open matte for all releases previous to Blu, it's easy to believe that all previous releases of Barry Lyndon were a slightly expanded aspect ratio- or at least, were composed so as to fit several different ones.

As I said, I do agree that looking at the releases head to head, it's the blu that looks (very slightly) distorted- Ryan O'Neal is looking a little puffy in the cap- but I will also say that I'd seen it at least half a dozen times on DVD, and I did not notice the difference until comparing them directly. I can say that there did not seem to be any awkward cutting off of feet and foreheads on the blu.

Again, if there are any other points in the movie you guys want me to compare- something with a more clear circle, for instance- I'd be happy to.

[MichaelB]

Cinema conditions vary so widely and for so many reasons that even numerous theatrical screenings - even provided that people have uncommonly accurate memories of what they saw (I suspect that most people, though not all, are frankly full of pure shit when they talk about color schemes, framing, etc., at screenings a few days ago, much more so when we're dealing with several decades) - can hardly be considered to provide authoritative familiarity in any way.

I've seen Barry Lyndon twice in the cinema. The first time was in Paris in the 1980s, and I wouldn't have the first clue what the aspect ratio was. I suspect if I'd been an ordinary punter at the second screening, I'd have been equally unable to recall it - but in this case I had the advantage of actually working at the cinema in question, and so I knew for a fact that it was 1.66:1. (I remember checking, because I'd already heard Geoff Andrew's anecdote about being berated by a Kubrick spy for showing it in 1.85:1).

But that clarity comes with the advantage of priviliged information. Would I have been able to state with absolute certainty that it was in 1.66:1 without that? Very possibly not.

Just for clarity's sake, I agree that the aspect ratio is highly questionable from a verifiable, factual point of view (although I'm less convinced that it's consequential). And I agree that Vitali's statements on the matter seem bizarre. I must repeat what I've said before - I don't find this conversation to be without merit.

But beyond that? Pure speculation and/or simple personal preference. The living person who would normally be considered the most authoritative on the subject (Vitali) has been judged to be a crackpot, so what we're left with are just guesses in the absence of more definitive evidence than a handful of screenshots. And barring the emergence of such evidence, it seems that interested consumers have enough information at their disposal to make an educated decision as to whether or not to buy the disc.

What I'm finding amusing about all this is that none of this sound and fury has budged me from my two core positions, which are:

1. that the projected aspect ratio of the film should be 1.66:1, and;
2. that despite this, the Blu-ray looks absolutely fine, and I have no regrets about purchasing it.

[Nothing]

Brian, this is compelling circumstantial evidence, not just speculation.

Schmatrix, according to numerous online reports, the BD is vertically pan & scanned - ie. they have adjusted the framing up and down from shot to shot to maximise visibility within a 1.78:1 crop. This is very different from simply matting the film because:
1/ It alters the way we perceive each edit, the way in which each framing has been matched by Kubrick and Alcott to the next. The film has, to use those immortal words, "been altered to fit your screen."
2/ If the film has to be vertically pan and scanned to fit a 1.78:1 frame then clearly this is not the original aspect ratio...

And it seems to me that the folks who have already spent their money on the BD are more likely to suffer from confirmation bias.

[jedgeco]

according to numerous online reports, the BD is vertically pan & scanned - ie. they have adjusted the framing up and down from shot to shot to maximise visibility within a 1.78:1 crop.

But what you describe does not involve "panning" or "scanning." This is not an instance where the telecine operator is panning the frame up or down during a single shot (or at least nobody that I'm aware of has made that allegation). Rather, it's just as you say: framing is adjusted shot-to-shot, or at least was adjusted in a handful of screen caps when compared to older transfers that themselves involve some amount of cropping.

Moreover, I would imagine that this sort of shot-to-shot adjustment is de rigueur in modern film transfers. Notably, the older full-frame transfers of The Shining and Full Metal Jacket were not simply "full frame open matte," but were zoomed and cropped on a shot-by-shot basis depending on the needs of the shot in question.

[Nothing]

The film itself is still pan and scanned, whether or not frame adjustments are made within indiviual shots. And I have no problem if a director and/or DoP decides to do this in the D/I, but that's a very different thing from a technician imposing such changes without the consent of the artists after the fact. Also, unlike the distortion, we can comfortably take the taller transfers as the benchmark here as, with the whole of the hard-matted image exposed, it would not have been possible to shift the position of the frame around in the earlier telecines.

[MichaelB]

And it seems to me that the folks who have already spent their money on the BD are more likely to suffer from confirmation bias.

You could just as convincingly argue that those who've paid good money for the Blu-ray might be more miffed if it had problems.

(For what it's worth, for various off-topic reasons I paid virtually nothing.)

In fact, it's probably worth noting that both the old DVDs look pretty terrible on my setup, largely because the only way I can get the whole picture displayed without distortion is to run them in 4:3 - i.e. with massive black bars at the sides, noticeable ones at the top and bottom, and a decidedly small picture. So I'm actually much more conscious of the limitations of those discs when actually watching them than I am of the Blu-ray.

according to numerous online reports, the BD is vertically pan & scanned - ie. they have adjusted the framing up and down from shot to shot to maximise visibility within a 1.78:1 crop.

But what you describe does not involve "panning" or "scanning." This is not an instance where the telecine operator is panning the frame up or down during a single shot (or at least nobody that I'm aware of has made that allegation). Rather, it's just as you say: framing is adjusted shot-to-shot, or at least was adjusted in a handful of screen caps when compared to older transfers that themselves involve some amount of cropping.

Moreover, I would imagine that this sort of shot-to-shot adjustment is de rigueur in modern film transfers.

It's certainly been standard practice at the telecines I've personally sat in on - though in the vast majority of cases the filmmaker and sometimes the cinematographer have also been present and only too happy to be given the opportunity to make small tweaks to the framing. That said, this is clearly impossible with Barry Lyndon - Doug Milsome would probably be the closest to a bona fide authority in the absence of Kubrick and John Alcott, and I'm not sure if he was consulted at any stage.

[mckeldinb]

Just wanted to weigh in as someone who has seen BARRY LYNDON three times in a theater (including during its original run) and has purchased and viewed the Blu-Ray in its entirety. My gut reaction is that the film has not been cropped beyond what was shown theatrically. My memory of the original showing at a poorly run theater in Knoxville, TN called the Studio One is that the sides were not properly masked giving it that odd home-movie look with the fuzzy sides, this lends credence to the notion that Kubrick had instructed theaters to exhibit the film at 1.66 (though it's also possible that the theater had the masking set for 2.35). The aspect ratio for my subsequent theatrical viewings (at revival houses in New York City) are not memorable, so I assume whatever the ratio was it was properly masked (whether the image was cropped beyond what Kubrick wanted I don't know).

As for the Blu-Ray, I am, on the whole, very happy with it; but I agree that it is a flawed telecine. I suspect what happened was that it was matted to 1.66 and then geometrically stretched to 1.78. Doing an eyeball crop (not very scientific, I grant you) of the DVD capture above so that it matches the screen information in the BD cap, I got a 1.67 AR; whereas the BD cap itself is 1.78. Since there is obviously a difference in the geometry of the two images, I suspect that the proper 1.66 matte was used for the BD transfer, but then stretched to fill the 16x9 screen. Mind you, I'm completely guessing here, but this idea makes sense to me.

Is this practice acceptable? To me, no. Are the results in this case acceptable? Well, I have to admit that had the issue not been raised here I never would have noticed (I do admit that being aware of the potential stretching issue before I watched the disc had me questioning the image occasionally as I was watching, but I don't think I would have even thought about it otherwise... and I am very sensitive to geometry issues with my TV).

I appreciate those who will boycott the disc out of principle, but I'd be lying if I didn't say that despite a flawed telecine, this is the best I've ever seen BARRY LYNDON look on home video and I'm very happy to have the disc. I also have to say I honestly prefer the color-timing on the BD to the DVD (the grass in the DVD is a blue-green; whereas on the BD it looks far more natural... if a bit luxuriously heightened, as would be appropriate for the painterly images).

[reno dakota]

Stanley Kubrick's letter to projectionists on Barry Lyndon

[Gregory]

"Please be sure you project it at [1.66:1]" -- and yet Warner says that by presenting it at 1.78:1 they followed his wishes expressed here. Seems pretty clear that it was composed for 1.66:1, that was his strong preference, and 1.75:1 was the limit beyond which the ratio would cross the line into the totally unacceptable.

I respect the practically-motivated responses of MichaelB and others, saying that it looks acceptable to them, it's not a huge difference etc. But I think there's something more in the background of all this, which is a tendency toward widescreen revisionism in the HD age. So that's a large part of what I find so irksome about all this. It's the principle of the thing as much or more than how it looks to me personally, as a viewer.

[mckeldinb]

Well that's that, as far as I'm concerned. Vitali is wrong, and WHV is covering their collective ass. However, that still begs the question: is the BD overly cropped or geometrically stretched? I think it's the latter -- either way, one has to admit that it's a flawed telecine. (Still looks gorgeous though).

[Tom Hagen]

Ha! Even Kubrick's missives were in Futura!

[andyli]

geometrically stretched

Any proof of that?

[mckeldinb]

geometrically stretched

Any proof of that?

Not a shred, as I admitted (quoting two words of mine with no context is unfair). However, my eyes tell me that one of the two captures posted above suffers from geometric distortion. And here's what I originally wrote:

I suspect what happened was that it was matted to 1.66 and then geometrically stretched to 1.78. Doing an eyeball crop (not very scientific, I grant you) of the DVD capture above so that it matches the screen information in the BD cap, I got a 1.67 AR; whereas the BD cap itself is 1.78. Since there is obviously a difference in the geometry of the two images, I suspect that the proper 1.66 matte was used for the BD transfer, but then stretched to fill the 16x9 screen. Mind you, I'm completely guessing here, but this idea makes sense to me.

So my supposition is based on the fact that when I cropped the DVD cap to approximate the same frame as the BD cap, I got a 1.67 AR. Since it has now been proven that Kubrick's desired ratio was 1.66 and since the geometry of the DVD and BD caps clearly differ, I'm *guessing* that someone, somewhere at WHV got the bright idea to matte the frame according to Kubrick's wishes but then geometrically stretched the frame to fill in a 16x9 TV screen. The operative word here is "guessing."

What is not a guess is that somebody screwed up. Admittedly, a screw-up that doesn't affect my personal enjoyment of the movie, but as Gregory has already said, it's not a good idea to start down the slippery slope of reformatting for the TV screen.

BTW, I was initially all set to dismiss charges of telecine mistakes, but I feel the proof is incontestable at this point (the image has either been cropped beyond what Kubrick would have liked, or has been stretched... or some combination of the two).

[matrixschmatrix]

Well that's that, as far as I'm concerned. Vitali is wrong, and WHV is covering their collective ass. However, that still begs the question: is the BD overly cropped or geometrically stretched? I think it's the latter -- either way, one has to admit that it's a flawed telecine. (Still looks gorgeous though).

It certainly seems clear that Vitali was wrong in saying that the movie was never 1.66, but as I recall the aspect ratio he actually specifies is 1.75, which is within Kubrick's range- Vitali just argued that the difference between 1.75 and 1.78 was negligible. To some degree, I think the missive validates both sides: 1.66 is and always was optimal (though some of the DVD releases were specified as being in the 1.5 range, that doesn't match what is on my release), but the BD is pretty well within the acceptable parameters (I would challenge anyone, anywhere to distinguish between 1.75 and 1.78 by eye.)

I think it's pretty certain that it's not strictly a stretched 1.66 image- the loss of information at the top and the bottom makes that clear. Perhaps it's actually a 1.75 image minutely stretched to 1.78?

[mckeldinb]

I think it's pretty certain that it's not strictly a stretched 1.66 image- the loss of information at the top and the bottom makes that clear. Perhaps it's actually a 1.75 image minutely stretched to 1.78?

I would have thought that too, except that the DVD image is opened up to 1.58 (I just remeasured it, based on the DVD cap above). I think the difference in missing information is accounted for by going from 1.58 to 1.66; and the 1.66 to 1.78 is accounted for by a horizontal stretch. Though it could be the 1.58 isn't opened up, but a vertically stretched 1.66. Oy! I've got a headache. At any rate something's funny somewhere, and I wouldn't care that much (since for all intents and purposes the BD looks fabulous) except for the ramifications of possible future transfers.