My Bloody Valentine

[Keith Walters]

I've been some surprised at how little discussion there has been about what for many of the population will be the RED's first mainstream cinematic outing.

 

A combination of a wet Saturday afternoon with everybody else out doing their own thing, combined with curiousity about this apparent lack of interest led me to sample a 3-D digital screening at my local multiplex.

 

First off, if you're looking for something that showcases what the RED can do, this movie isn't it. Everything looks like it was shot either at night under balloon lights or in the early evening, or on dull, overcast days. The whole film has a dull "day-shot-for-night" look about it, which reminded me of a drive-in move shown using an underpowered projector when there's a full moon :lol:

 

Everything also looked quite soft, almost like the film was shot on SD cameras or super-16.

 

(If you want to suggest that the 3-D projector was somehow at fault, before the main feature they had a number of trailers of upcoming fully animated 3-D features, and they were superb. While I can see that there is a potentialy huge market for 3-D, I rather suspect it will be limited to cartoons and CGI; live action may always be too hard to get right).

 

I also suspect that, rather like Jurassic Park, this film was originally more an engineering range-finding exercise than a specific attempt at producing a box office winner, because there was an obvious learning process involved.

 

I think the first thing they discovered is that for 3-D live action to be convincing, you need an extremely shallow depth of field. The resultant fast focus falloff gives the eye extra information for correctly synthesizing the 3D illusion. Obviously on some shots they had the iris stopped down too much, giving the the characteristic "3-D" look of two cardboard cutouts talking to each other. But in other, similarly setup scenes, they appear to have learned their lesson and shot with a fully-open aperture and NDs, which may explain the apparent softness of most of the film. The "cardboard cutout" scenes were definitely much sharper-looking, which tends to clash with the rest of the film. I suspect that if live action 3-D ever catches on, they will need to shoot with 65mm-sized sensors.

 

Overall, the 3-D effect was quite convincing, far better than all the earlier cinematic efforts I've seen.

 

As for the movie itself, it appears that as far as this genre goes, nothing much has changed in 40 years!

 

While most folks will find the special effects sufficiently gory, it seemed to me that the old adage "less is more" seems to hold true, in that the graphic 3-D closeups tended to highlight the fakeness of the effects rather than enhancing them.

 

Paying loyal homage to the film's heritage, it had the obligatory and totally gratuitious extended full-frontal nude scene, although the softness of the image artfully concealed complete expression of the actress's full Brazilian glory :lol:

 

Actually this is kind of bizarre. The nude scene started fairly demurely with the actors either facing away from the camera, or casually holding bedclothes in strategically placed positions, Days of Our Lives style, and then it was almost like the director said "oh stuff this, let's just let it all hang out!"

 

After that there was no attempt at concealing anything.

 

I'm still none the wiser what the actual plot was supposed to be. The identity of the culprit for all the mayhem is eventually revealed, without there ever being a single hint beforehand to give the audience a clue that he might have been responsible. Apparently some of the earlier scenes he was in that tended to prove his innocence, turned out to only be hallucinations on his part, but there was nothing that gave any hint that that was the case.

 

At one part the local sherrif was lamenting that all the sensationalist news coverage was going to give the rest of the world the impression that the town was populated by inbred retards. Given the unbelievable ineptitude demonstrated by the victims in trying to escape the killer, to me it would have seemed a pretty fair call :lol:

 

Another interesting quirk was that the credits make no mention of what cameras were used, and even though this was a digital 3-D projection, it says "Kodak Film".

Edited February 14, 2009 by Keith Walters

[Matthew Rogers]

I still don't understand why Hollywood thinks most movies are going to go to 3D in the future. I personally have no desire to wear glasses while watching a movie. That's why I wear contacts instead of glasses!

 

Looking at an HQ Flash trailer the picture quality (as far as coloring) looks pretty bland. But, it's a horror movie so maybe it's supposed to be. I'd say the lighting in general is kinda flat and not contrasty enough for my taste when you consider it's a horror movie.

 

Matthew

[Stephen Williams]

I'd say the lighting in general is kinda flat and not contrasty enough for my taste when you consider it's a horror movie.

 

Matthew

 

Hi Matthew,

 

No doubt they followed of the light if flat & fix it in post technique, that simply does not work.

 

Stephen

[Jaron Berman]

I believe there was a bit of SI-2K footage mixed in, perhaps explaining the deeper DOF and sharper picture in the "cardboard cutout" scenes. From everything I've seen, it does make a significantly better picture overall though it lacks the shallow DOF.

[Keith Walters]

I believe there was a bit of SI-2K footage mixed in, perhaps explaining the deeper DOF and sharper picture in the "cardboard cutout" scenes. From everything I've seen, it does make a significantly better picture overall though it lacks the shallow DOF.

Possibly, but I don't know why they would have done that, since the scenes in question were just ordinary 24fps "Drawing Room Drama" type footage, no high speed work involved.

[Jaron Berman]

No clue, just speculating (and probably accidentally starting rumors). I do know the SI-2K 3-d rig is TINY, about 10lbs total, so it is significantly faster and easier to setup. you never know, maybe they got into a time pinch or maybe they just shot it as a b/c camera and the editor just chose that angle as opposed to an angle shot on RED. I would find it hard to believe they varied the stop enough on RED to make it as deep as the SI, but again - all speculation.

[Keith Walters]

I still don't understand why Hollywood thinks most movies are going to go to 3D in the future. I personally have no desire to wear glasses while watching a movie. That's why I wear contacts instead of glasses!

 

 

Matthew

It's not quite the same thing though. In the real world, glasses can only ever be a compromize that you have to work around, but in a cinema you're only looking at one thing (the screen) at a fixed distance.

 

If you've only ever experienced 3-D through throwaway cardboard polaroid glasses, you could be forgiven for not being terribly impressed. The cinema I went to had ones marked "Dolby-3D" that you had to return at the end of the session and they were much better quality.

 

 

It would appear that Dolby® are being proactive in providing carefully thought out 3-D solutions for cinemas the same way they did for sound 30 years ago. If you have the right type of DLP projector, the additional cost of a 3-D upgrade is quite small.

 

You basically need an add-on polarizer wheel assembly, a box of re-usable glasses, and a special cleaning tank for them. Apparently competing solutions also need a special high-gain screen which looks terrible with 2-D (so Dolby say anyway). They also bring up the point that presently, the distributors heavily subsidize the cost of throwaway 3-D glasses, but there's no guarantee of how long that will continue.

 

If 3-D cinema does take off, I can see a market for high quality 3-D glasses for people who either want top quality or don't feel comfortable with "used" glasses.

 

Now there's a field that Jim Jannard could probably get his teeth into, but he's probably not allowed to :rolleyes:

[Matthew Rogers]

It's not quite the same thing though. In the real world, glasses can only ever be a compromize that you have to work around, but in a cinema you're only looking at one thing (the screen) at a fixed distance.

 

If you've only ever experienced 3-D through throwaway cardboard polaroid glasses, you could be forgiven for not being terribly impressed. The cinema I went to had ones marked "Dolby-3D" that you had to return at the end of the session and they were much better quality.

 

I've seen 3D shows where they use reusable glasses and I still don't like it. I just don't like having things on my face period so glasses, especially ones that can't adjust to your face, are a big no no for me. I'm also not crazy about the gimmick of 3D. I rather have theaters do the environment experience... you know, heat and cold, wind, spray water, smells, etc are much cooler and get you into the movie. If you've been to Disney's California Adventure park and rode the Wings over California ride you know what I'm talking about. IMAX dome movie that they "glide" you up into and then do all sorts of environment type things to you. It really makes you feel like you are handgliding over CA.

 

Matthew

[John Sprung]

. Apparently competing solutions also need a special high-gain screen which looks terrible with 2-D (so Dolby say anyway).

 

Yes, in the old days of 3D, we had those high gain screens. The one at the Four Star was 55 ft wide, IIRC something ungodly like a 3:1 gain. It's necessary to retain part of the polarization. A 1.0:1 screen gain would give the audience zero polarization, so no 3D. To do both normal and 3D, you'd have to fly one of the screens.

 

 

 

 

-- J.S.

[Hal Smith]

The best solution would be kerr cell glasses where the lenses can sequentially let light through to each eye in turn. The projector would send out a series of left/right/left/right... images in sync with the glasses. No special screen required and pretty much no light loss...but very pricey

[Keith Walters]

Yes, in the old days of 3D, we had those high gain screens. The one at the Four Star was 55 ft wide, IIRC something ungodly like a 3:1 gain. It's necessary to retain part of the polarization. A 1.0:1 screen gain would give the audience zero polarization, so no 3D. To do both normal and 3D, you'd have to fly one of the screens.

 

-- J.S.

I've been in that same cinema (#4) a number of times to see non-3D features and I didn't notice anything unusual about the screen.

I'll look more closely next time.

Maybe there is a way to make low-gain screens that also preserve the polarization.

I've seen numerous polaroid-type 3-D demos over the years and I don't recall any special screen being involved.

The biggest problem with all of them was the low screen brightness.

[Keith Walters]

The best solution would be kerr cell glasses where the lenses can sequentially let light through to each eye in turn. The projector would send out a series of left/right/left/right... images in sync with the glasses. No special screen required and pretty much no light loss...but very pricey

That's the only way it is ever going to happen with TVs at home. I presume modern LCD screens can respond fast enough, but you're going to be bringing back screen flicker.

 

I think it was Sanyo who brought out a 3-D camcorder about 20 years ago that recorded the left and right images as alternate fields of a standard video signal.

Playback was only available in the camera, and you also had to plug a set of active LCD glasses into it, and you also needed a 100Hz TV (120Hz for NTSC) to view the images.

 

Don't know why it never caught on...

Edited February 17, 2009 by Keith Walters

[Antti Näyhä]

I think the first thing they discovered is that for 3-D live action to be convincing, you need an extremely shallow depth of field. The resultant fast focus falloff gives the eye extra information for correctly synthesizing the 3D illusion. Obviously on some shots they had the iris stopped down too much, giving the the characteristic "3-D" look of two cardboard cutouts talking to each other.

That's strange. I've always felt that 3D needs as much DOF as possible to work well. My Bloody Valentine was no exception – I found the deep-focus scenes by far the most convincing (and the least headache-inducing).

Edited February 20, 2009 by Antti Näyhä

[Keith Walters]

That's strange. I've always felt that 3D needs as much DOF as possible to work well. My Bloody Valentine was no exception – I found the deep-focus scenes by far the most convincing (and the least headache-inducing).

Maybe different people interpret the 3-D illusion (and it is only an illusion constructed in the viewer's brain) in different ways.

 

Our brains make assumptions about the world around us from a variety of visual cues. One of these is the difference in focus between the front and back of an object, even when your eye has no control over the focus, (such as when viewing a photograph). That's why a model spaceship always screams "Fake!" unless you either shoot it from a long way off using a telephoto lens, or you stop the iris as far down as is practical to maximize the depth of field. (Or more likley, both). If the front of the model is in focus and the rear of it is not, your brain immediately works out it's only maybe a couple of feet long instead of hundreds of feet long.

 

In the case of My Bloody Valentine, the offset of the two cameras gave a very strong clue that the actors were only a couple of feet away from the lenses, but the large depth of field (and consequent lack of focus falloff) tended to contradict this. However since our brains are used to processing flat images (photographs etc) it sort of goes: "Jeez ... I dunno..." and concludes that the images of "actors" are actually closeups of animated flat photographs!

 

Just to be clear, the illusion that the two "animated cardboard cutouts" were physically separated from each other and the background was very strong, but it simply looked gimmicky, not anything like reality.

 

I personally found the more subtle shots to be far more convincing, and far less eye-straining. If the movie itself had been a tad less ridiculous, I probably would have gotten so immersed in it that I would have forgotten I was even wearing the glasses!

 

There is a future for this technology, but more work on shooting technique will be needed.

Edited February 21, 2009 by Keith Walters

[Robert Niessner]

It seems like they overdid the effect and put the distance between cameras too wide.

I think 3D can be a nice feature but it has to be stup carefully not to strain the watcher.

 

A friend of mine here in Graz, Austria is working on 3D since years and he showed me some really impressive footage e.g. a flight over a mountain in 3 - it's so real that you feel dragged in.

[Antti Näyhä]

Our brains make assumptions about the world around us from a variety of visual cues. One of these is the difference in focus between the front and back of an object

Thanks for elaborating, Keith. You are, of course, correct – I never thought about this aspect of 3D perception.

 

It could be my personal viewing habits. I guess I just like to let my eyes occasionally wander outside the focus point when looking at any image. That's OK to do on a 2D film: it doesn't feel uncomfortable in any way to take a quick glimpse at the out-of-focus foreground actor in an over-the-shoulder shot, for instance. Our brain will obviously notice that the actor is blurred – but as there is no double-image, the eyes will make no effort to change the physical focus distance (which remains at the distance of the actual screen). Hence, no sense of frustation or extra strain to the eyes.

 

However, if you take even a quick look at a blurred area in a shallow-focus 3D image, something completely different happens. When the brain registers a blurred double-image, it will immediately try to 1) make the two images overlap and 2) get them into focus. But unlike in the real world, only 1 is possible when watching a 3D film. Futile attempts at 2 drive the brain crazy, and causes strain on the eye muscles.

 

So, in order to have realistic 3D, I guess we should first record everything in deep-focus. Then we'd have to invent a presentation system that scans the viewer's eyes, finds out which point of the image they are looking at and synthesizes the shallow focus in real-time according to that information...

Edited March 3, 2009 by Antti Näyhä