Aspect Ratio discussion for Magnificent Obsession

[yoshimori]

How long have we been emphasizing that the film was composed for both AR's-- the post about shifting viewpoints is breathtaking!

I think y'all've been at it for, uhh, maybe two weeks, since a day or two after the CC announcement. And Gregory is still suggesting Sirk "composed primarily for academy ratio". This is wrong, as I tried to explain before, under any common or proper understanding of "composed for". He composed for 2:1 and also made it look acceptable/passable/fine/great/whatever in Academy. At least we (except for Gregory) can agree on this. No? HerrSchreck's insistence that the Academy camp has always maintained Sirk "composed for both" ratios is pure fantasy, as I was scolded more than a year ago for even suggesting that 2:1 worked well for this film, and the American Cinematheque, which showed the film in 2:1, was laughed off as a fly-by-night operation (which it is, but that's another story) which had brutalized/amputated/etc the film.

In the great tradition of Universal totally fucking up 1.37 titles to "update" or in some way "improve" them (Touch of Evil, Jet Pilot, etc.), Magnificent Obsession is most definitely [incorrectly] masked/cropped to 2.00:1

There is plenty of head height throughout the movie, in charcter with Academy Ratio shooting but that's not a reason to excuse either misframing or extremely poor teaching ... Where do you kids learn this bogus shit!! At fucking University??

If solid evidence of the correct aspect ratio [the poster means 1.37:1] can be put in front of the right person, and if enough people e-mail the company as soon as possible to ask them not to repeat the mistake, maybe they could be persuaded to postpone that [2:1] release and do it over. It might sound like a longshot, but it worked with MGM's big Bergman set.

[2:1 is] improper AR tarnished

[Widescreen Sirk is] dead wrong.

NO ONE I know ever saw these 53 to 58 Sirks non Scope Sirks (for example) in anything like 2:1.

Universal and the fucking AFI can bray on all they like about 2.1 being the preferred ratio

etc etc

[mteller]

Sirk obviously REALLY meant for everyone to watch his movies at home on TV, where most of them probably couldn't afford air conditioning. So clearly Sirk's mise en scene demands that MO be watched in conditions that are slightly warm to uncomfortably muggy. I have no proof of this, but I understand Sirk's artistic intentions better than anyone, because I say so.

[Antoine Doinel]

How do we know that Sirk even directed the movie? Because his name is on it? Y'all are so naive.....

[HerrSchreck]

C'mon, Yoshi-- you're quoting David Hare. He stopped posting on the thread 9 months ago (he quit the board, in fact), and has nothing to do with what we've been repeatedly saying for the past couple of weeks to Bob, and poo-bear, and gringo and whomever else: that the film was composed with both Ar's in mind. Said it over, and over, and over, and over, and over again. And over and over and over again.

How this is rendered moot by comments made up to nearly two years ago and primarily by one defunct member, I dunno.

[mattkc]

By the way, is this from the full frame version of the film? or is it WS cropped?

Before this thread disintegrates even further, did anyone look at the clip? I can't believe anyone would want to watch that masked to 2:1... Someone please let us know if this is an example of WS cropped.

[yoshimori]

C'mon, Yoshi-- you're quoting David Hare. He stopped posting on the thread 9 months ago (he quit the board, in fact)

Wasn't just hare, but whatever. [And very sorry to hear he quit the board.]

What about Domino's opening comment - "Fuck fuck fuck fuck" - when the 2:1 correction of the original "Academy ratio" announcement came out? Was that meant to be read as "I really think the movie was composed in both ratios"? And there's still the language issue: it makes my skin crawl when someone says "it was composed for both ratios" as this is simply not possible or is a quite non-standard, non-technical use of "composed for". If Sirk had composed MO for 1.37:1 it would look quite different. No?

Still, fine. I think many of the 1.37:1 frames are quite interesting, often imaginative, beautiful. I'm sure Sirk and Metty worried about them. I also think, though, that the main argument of the pro-Academy group - that a consideration of the history of Sirk's mise-en-scene leads us to declare 1.37:1 was there preferred AR for this movie - is pure bluster.

[Antoine Doinel]

I had no idea david hare left the board. Very sorry to see him go....

[Narshty]

By the way, is this from the full frame version of the film? or is it WS cropped?

Before this thread disintegrates even further, did anyone look at the clip? I can't believe anyone would want to watch that masked to 2:1... Someone please let us know if this is an example of WS cropped.

I doubt that's the full frame in that clip. Some of the left-right sides in those look very cropped.

[HerrSchreck]

If anyone could answer authoritatively it's greg, who has a copy of the film in academy to hand.

Yeah, David Hare left around a month ago. He's winding up a trip to Paris as we speak. As for why he quit the board, if he didn't say so on his own then I don't feel it's my business to state the reasons. But needless to say, it's a big loss to the board. He was quite a writer and was an amazing fount of knowledge. And few people love cinema as much as he does without working in the biz.

[Gregory]

By the way, is this from the full frame version of the film? or is it WS cropped?

Before this thread disintegrates even further, did anyone look at the clip? I can't believe anyone would want to watch that masked to 2:1... Someone please let us know if this is an example of WS cropped.

I doubt that's the full frame in that clip. Some of the left-right sides in those look very cropped.

Yes, in fact it is the full frame in the clip. I just compared it to the UK DVD, and the part at about 0:21 in the YouTube clip shows Merrick's head cut off right at the top of the ear. The part at 0:33 in the YouTube clip shows his head cut off through the middle of the ear and through his eyebrow. In the last few seconds of the scene shown here, the guy with the light brown hair crouches down to attend to Mrs Phillips. On the UK DVD, the bottom is cropped off, and so as the scene fades out you can't see his eyes looking down at her, only his forehead, ear and the top of his head.
I've read that this transfer is actually more like 1.9:1, so if Criterion does an actual 2:1 transfer, it could look even more severe, unless they include more visual information on the sides to get this ratio.
For those of you who think it's "correct" to watch it this way, have fun, but I don't see why we should subject ourselves to something like this just because it has the same ratio as what the studio was promoting.

Yoshimori: To try and clear this up once again, I believe they had widescreen in mind when they composed the film, but I don't think it looks well composed in 2:1. Based on the UK DVD, at least, compared to my 1.37 version, I think it looks like crap. But I recognize that they knew it had to look at least acceptable to people in ratios up to 2:1. It apparently passed muster, but this doesn't contradict my reason for thinking it looks bad because I'm watching it in different ways than people did then. Most people then didn't give a second thought to all the subtle things about Sirk's films, as he knew. Many of the things I'm looking at were irrelevant to the studio executives and to original audiences.
And I'm not alone in my position, as you suggest. I've heard from a number of people on the forum who agree with my position, and when this UK set came out there was incredulity at the idea of being forced to watch MO in a 2:1 transfer not only here (mainly David, who took the time to have the discussion, but also myself and others who just didn't join in) but also at the DVD Beaver listserv and at a_film_by, including from people who had screened the 35mm film.

[GringoTex]

By the way, is this from the full frame version of the film? or is it WS cropped?

Before this thread disintegrates even further, did anyone look at the clip? I can't believe anyone would want to watch that masked to 2:1... Someone please let us know if this is an example of WS cropped.

I wish I could post the 2:1 of this scene on youtube, because I believe it's evidence for a 2:1 ratio. Look in the first shot how the characters are positioned unusually low in the frame for a medium close-up. Then when we cut to the reverse angle, there's a jarring "jump up" as they're suddenly positioned higher in the frame in another medium close-up. This is not Sirkian mise-en-scene. In 2:1, the reverse shot is a close-up of Wyman's character, not a medium close-up. The back of Hudson's head is used to frame the right side of the frame. So Sirk is cuttimg from medium close-up to close-up reverse shot, which was his standard practice.

[Gregory]

How can you call it a close-up of Phillips (Wyman) when it's a simple reverse two shot? They're equally positioned in the frame in the shot you're referring to as the close-up, starting at 0:11. Merrick (Hudson) is not off to the right, so I don't know why he would be used to "frame the right side of the frame." It switches back and forth from the front of Wyman's face/back of Hudson's being visible to the opposite: the front of Hudson's/back of Wyman's. The former is shot closer in, but that doesn't mean we're only supposed to see everything from Hudson's ear down. They're both simple reverse two shots, which can be seen in lots of Sirk's other films in academy and in films of the period in general.
I also wish this could be posted, although the fact that some would inevitably prefer what they see in 2:1 would challenge none of the points I've made.

[HerrSchreck]

Now this is a genuinely productive conversation.

Measure where Hudson's head in in the first shot. Now measure where Wyman's head is in the second shot. It's a simple reverse angle shot, compensating for the dramatic difference in height between the two actors, that's all. The difference in camera position (or lens) is merely to compensate for this.

The scene has no c/u's. Wyman's face hadn't even been seen yet, to establish her persona in the scene, for Sirk to just jump right in to a closeup. The second shot is an equivalent for Wyman as the preceding shot of Hudson. Sirk plays off of Wyman's back in the first shot, then off of Hudson's back in the 2nd shot. Then the sequence repeats. Another shot for Hudson, then another for Wyman.

Then the conflict ratchets as Wyman walks out of frame. What you're calling something like a failed closeup in academy is merely Sirk and Metty moving the setup in to get Wyman, because of her height, into proper speaking height for a matching composition vis a vis the previous shots. The headroom of the speakers in both setups is precisely identical. For the first shot, put your finger on the screen to establish where Metty & Sirk place the speaker Hudson in the vertical, and hold that mark... so that when the subsequent 2 shot appears to allow Wyman to speak, you can slide your finger horizontally over to see that they've simply brought the camera forward (or changed lenses) to bring Wyman up to the same vertical point, only over on her side of the screen. Hudson speaks-- boom, he's at the same vertical degree. WYman crosses the screen and walks away from hudson, keep your finger there... boom, she lands against the tree, at the same vertical plane. It's an incrdedibly consistent set of 2 shots with no closeups.

You'll see the top of the cab, then Wyman's head in the cab, will be directly under your finger. These setups are not closeups, in fact I don't see a closeup in the whole bunch.

[GringoTex]

How can you call it a close-up of Phillips (Wyman) when it's a simple reverse two shot?

In a reverse two shot where the actors are positioned relatively the same distance from the screen, the same angle has to be maintained. But it's not in the 1.37 framing here. The heads are much higher in the second shot than the first shot. So the cut is jarring. In 2.1, the second shot focuses much more on Wyman's face than the first shot does on Hudson's face. It's a definite "step in." And this is what Sirk did. He rarely used equidistant reverse shots.

Shreck- I have to run off for Halloween now, but I'll address your post tomorrow. I agree this is an excellent discussion.

[yoshimori]

These are classical over-the-shoulder medium-close-up reverses, as HS and G note. The sizes of the subjects match well enough (HS is right, GT is wrong on this point). But the reverses work perfectly well in 2:1. In fact, the "pop" that GringoTex notices is much less obtrusive in 2:1. These shots, then, work better, work more smoothly as a set of matching reverses, in 2:1.

[Gregory]

Personally I don't find it jarring, but if I did I don't understand how cutting off the top of Hudson's head in some shots but not others (which is what happens in 2:1) makes anything less obtrusive.

[yoshimori]

I don't understand how cutting off the top of Hudson's head in some shots but not others (which is what happens in 2:1) makes anything less obtrusive.

But this (cutting heads in one shot, not in the other) is standard procedure in matching over-the-shoulder reverses of people of different height in any aspect ratio. The point is, the less of the height difference of the foreground figures we see, the less obtrusive that difference in size.

[Gregory]

The cropped head in some shots and not in others calls attention to the changing distance from one shot to the next because it exaggerates the effect of being closer to Hudson.
But from the way I'm reading your words ("the less of the height difference of the foreground figures we see, the less obtrusive that difference in size") I'm not sure we're talking about the same thing. GringoTex was talking about the change of their position in the frame from one shot to the next. But it sounds like you're talking about the difference between Wyman and Hudson in the foreground.

[yoshimori]

The cropped head in some shots and not in others calls attention to the changing distance from one shot to the next because it exaggerates the effect of being closer to Hudson.

Just the opposite. The cropping of Hudson's head reduces the "effect of being closer" because it reduces our fixation on him. This phenomenon (the one you're complaining the 2:1 creates), btw, is typical of every matching reverse in every film. Also, there is no appreciable "changing distance from shot to shot" in terms of composition. The distance between camera and Hudson in the shot of Hudson is (very close to) the distance between camera and Wyman in the shot of Wyman (if the lens hasn't changed). That what makes them matching reverses.

But from the way I'm reading your words ("the less of the height difference of the foreground figures we see, the less obtrusive that difference in size") I'm not sure we're talking about the same thing. GringoTex was talking about the change of their position in the frame from one shot to the next. But it sounds like you're talking about the difference between Wyman and Hudson in the foreground.

GringoTex is talking about the radically shifting vertical position of the "non-subjects" (Hudson in the shot of Wyman, and Wyman in the shot of Hudson). This kind of compositional shift is precisely what filmmakers would try to avoid in matching reverses by reducing the prominence of the non-subject (by cropping him or her). 1.37:1 here emphasizes the size difference and the vertical shift, 2:1 reduces the obviousness of size differences and pretty much eliminates the sense of vertical shift.

[Gregory]

By the changing distances I was referring to the shots where they're framed between the two trees compared to the other shots.

Just the opposite. The cropping of Hudson's head reduces the "effect of being closer" because it reduces our fixation on him.

Again, I see how this could work in the shots where his body is largely out of frame, but not in those where he's centered equally with Wyman in the frame and all that's cropped off is his head from the ear off. I don't believe that's how "every film" does it, not by a longshot. The subsequent reverse shots look better, for the reason you say. And by the way, I am pretty familiar with the formal practice of shooting matching reverses and how Sirk goes against this.
And again, as I've said all along, partially cropped heads are not my main problem here, and I don't think the discussion should focus just on that.
As for the vertical position of "non-subjects," I understand what you mean better now -- I thought you might be talking about the height difference between Wyman and Hudson in the same shot.
What I'm not sure I agree with is that Sirk and Metty were trying to reduce the effect you describe. Why would they go against a convention -- viz. cutting medium shots with medium shots, medium close-ups with medium close ups. etc. -- at the same time as they were hypothetically trying to reduce the visual effect of this decision.

[Adam]

At least we (except for Gregory) can agree on this. No? HerrSchreck's insistence that the Academy camp has always maintained Sirk "composed for both" ratios is pure fantasy, as I was scolded more than a year ago for even suggesting that 2:1 worked well for this film, and the American Cinematheque, which showed the film in 2:1, was laughed off as a fly-by-night operation (which it is, but that's another story) which had brutalized/amputated/etc the film.

The American Cinematheque isn't "fly-by-night." It's been around for twenty years now I think.

[Bob Furmanek]

Did Sirk ever compose a shot where someones head is cropped in his pre-1953 films?

You may find the following of interest. On June 24 1953, Sirk and Metty began production of their first film composed for 2:1 widescreen under UI's new studio policy.

Here are some 1.37 ratio frame-grabs for the ongoing study of Sirk's mise en scene:

[reno dakota]

Criterion now lists their release of Magnificent Obsession at 1.33:1 on the new website.

[domino harvey]

I've got my fingers crossed. And lol at a random screencap with two background extras serving as "proof" of Sirk's mise en scene

[reno dakota]

I've got my fingers crossed.

Me too. I wondered if the new AR listing might be just the default setting for the templates on the new website, but the other new releases grouped around MO seem to have correct AR listings. Now we wait, I suppose.