S35 vs FF sensors?

[sbcooler]

Is the Depth of Field for the S35 different than the FF35 sensor Red will sell? In other words will the FF35 allow for even shallower depth of field compared to the S35 Red One and S35 Scarlet?

And if so, what does Hollywood use as far as film cameras for the sensor? Is it the S35 size or the FF35 size. I am trying to emulate the Depth of Field that big budget Hollywood films produce and curious about this.

Now still cameras that are top notch Digital SLR's I believe are FF35 sensors. Am I generally correct here?

Thanks all. I'm learning so much here and wish I could find the words to show my appreciation.

 

[David Mullen ASC]

There is no single depth of field look for Hollywood films because even though they usually use 35mm film (the S35 size, not the FF35 size), they shoot it at different f-stops depending on the light level.

Also, whether they shoot S35 or 35mm anamorphic will also affect depth of field characteristics. For all intents and purposes, you could say that FF35 depth of field characteristics are closer to what 35mm anamorphic gives you more or less.

The FF35 format is the same as the VistaVision format in motion picture work. VistaVision is basically 8-perf 35mm horizontal, as opposed to S35, which is 4-perf or 3-perf vertical.

The magnification factor between S35 and FF35 is 1.5X more or less, which also corresponds to the difference in depth of field characteristics -- in other words, if you match field of view in both formats, then FF35 shot at f/4 is similar to shooting S35 at a f/2.0-2.8 split.

35mm anamorphic, because the anamorphic element doubles the horizontal view, causing you to use twice as long a focal length compared to 35mm spherical, has roughly a 2-stop difference in depth of field -- so FF35 is actually not quite as shallow-focus as 35mm anamorphic can be.

So FF35 will still fall roughly into the general look, depth of field-wise, of many Hollywood movies, though it is a little closer to the shallower depth of field of 35mm anamorphic than it is to S35.

By the way, FF35 (in the form of VistaVision) was used on a number of Hollywood movies in the 1950's such as Hitchcock's "Vertigo" and "North by Northwest" and John Ford's "The Searchers."

 

[sbcooler]

That was fascinating. I'm going to reread it again tomorrow to try to absorb as much as I can. But basically what I am hearing tonight is that Hollywood uses both sized sensors depending on the film. I realize one can change (increase) the focal length to get a shallower depth of field as well.

I'm wondering what pop movies of the last 20 years used which size film...like say classics like, The Godfather, Jaws, Rocky and others interestingly shot movies like Winged Migration...and say what was Rattle & Hum filmed on (U2 concert)?

 

[Jonathan Stevenson]

By the way, FF35 (in the form of VistaVision) was used on a number of Hollywood movies in the 1950's such as Hitchcock's "Vertigo" and "North by Northwest" and John Ford's "The Searchers."

You are the king, Mr. Mullen! I saw you interviewing in "Cinematographer's Style" last week, discussing the black and white American flag in "Northfork." I am always excited to read your posts here, wondering what great new fact I'll learn next! Hat's off to you, sir...

Sorry sbcooler, I have no 2 cents for you... But rock on anyway

 

[David Mullen ASC]

Well, VistaVision (FF35) isn't really used anymore except for visual efx plates. But as I said, 35mm anamorphic is still used and has the shallower-focus characteristics / issues that FF35 would be subject to.

I'd say that over half of Hollywood releases are 2.40 and about a quarter of that half were shot with anamorphic lenses, maybe less. Last summer you had the western "Appaloosa" shot in anamorphic, there was the recent horror film "The Unborn", the upcoming "Friday the 13th" remake, the romantic comedy "He's Just Not That Into You", etc. Most big Hollywood action movies though have opted for the easier-to-shoot (and create efx for) spherical Super-35 format, except for "Transformers" (Michael Bay being a die-hard anamorphic fan.)

Because FF35 / VistaVision only had its heydey in the late 1950's, and then was resurrected by ILM for the efx on "Star Wars" (1977), there have never been decent sets of cine lenses built for that format, just adapted still camera lenses.

For the past 40 years, up until recently, most movies were shot in either

(1) Sound aperture: 4-perf 35mm matted widescreen (1.85) or
(2) Sound aperture: 4-perf 35mm anamorphic-lensed widescreen (2.40), plus
(3) Silent / full aperture: Super-35 (4-perf or 3-perf) framed for cropping to either 2.40 or 1.85.

The first two are also standard (sound) projection formats and allow contact printing to create release prints.

In the past, Super-35 had to go through an optical printer step to convert it into one of the two 35mm sound aperture formats (1.85 matted widescreen or 2.40 anamorphic widescreen) in order to be able to make contact prints for release.

Now with D.I.'s able to do the conversion digitally, most movies are shot in Super-35, some in 3-perf instead of 4-perf to save money, framed for either 1.85 or 2.40 (or 16x9 for TV transfer), and then digitally converted to one of the two standard 35mm projection formats.

"Rocky" and "The Godfather 1 & 2" were shot in standard 35mm for 1.85 projection ("Godfather 3" was shot in Super-35 framed for 1.85). "Jaws" was shot in 35mm anamorphic 2.40.

Of course, there were also a few movies shot in other formats, from Super-16 to video, and now, HD and higher digital formats.

 

[David Mullen ASC]

Super-35 is nothing new -- in fact, 4-perf Super-35 is basically what Silent Era movies used, but instead of calling it "Super", you'd call it the Silent or Full or Edison Aperture. Being 4-perf tall, it was naturally 4x3 (1.33 : 1.)

Then when sound-on-film came along with the Fox Movietone process, they put the optical soundtrack on the left side of the print, which shaved the width of the image and made the aspect ratio 1.20 instead of 1.33.

That looked a bit too square-ish for some, so the studios proposed not only having the mask in the projector gate hide the optical soundtrack on the left, but also trim a little off the top & bottom, creating the smaller Academy or Sound aperture. That aspect ratio was close to 4x3 again (Academy is technically 1.37 : 1.)

Academy Aperture was used by most everything shot in Hollywood from 1932 to the early 1950's. Then with the release of popular widescreen processes like 3-projector Cinerama (2.66 : 1) followed by CinemaScope (anamorphic-lensed 35mm widescreen), standard Academy movies started being cropped further top & bottom to make them also look widescreen, what has been called matted widescreen, usually cropped to 1.85 : 1, though in Europe, cropping only to 1.66 : 1 was popular for years.

So now all 35mm print projection is 4-perf and either matted widescreen (called "flat" prints) or anamorphic widescreen (called "scope" prints.) Usually the cropping to 1.85 happens by use of a 1.85 mask in the projector gate, though some prints also have some degree of a matte printed into the image.

Back to Super-35... basically there was a brief attempt in the mid 1950's to shoot with the slightly bigger Full Aperture (Silent), frame a widescreen image, and reduce this in an optical printer to a Sound Aperture format for release printing. First attempt actually involved shooting the smaller Academy Aperture, framing for cropping to 2.00 : 1, and optically printing this into the middle of a CinemaScope release print. This was called SuperScope. Then someone thought that better quality could result by using the Full (Silent) Aperture, plus the slightly larger negative could allow them to crop further to 2.35 for CinemaScope release -- this was called SuperScope-235. But it never really caught on, there is even some controversy about whether any SuperScope-235 movies ever made it to the theaters.

Then jump to the early 1980's and John Alcott was about to shoot "Greystoke" for Hugh Hudson. Hudson and the studio wanted a 2.40 release but Alcott did not like anamorphic lenses, the only way at the time to get a 2.40 image. He proposed using Full Aperture cameras and framing the negative for cropping to 2.40, then optically blowing this up to 35mm anamorphic and 70mm as well. He called it "Super Techniscope" because it reminded him of the old 2-perf 35mm format that Sergio Leone used to get widescreen images.

Super Techniscope slowly caught on, used for movies like "Top Gun" and "Silverado" as well, but eventually the more generic name "Super-35" was used instead since Techniscope was connected with Technicolor Labs, plus Techniscope was a 2-perf 35mm process and Super-35 was a 4-perf 35mm process.

Since 4-perf Super-35 is naturally the same as the Silent Era format and is 1.33 : 1 (4x3), at some point, people started suggesting that some cameras be retooled to only pulldown 3-perfs instead of 4-perfs of film, since everyone was composing the 4x3 Super-35 negative for cropping to some widescreen format or another and thus wasting a lot of film. 4-perf is naturally 1.33 and 3-perf, being shorter, is naturally 1.78 -- exactly the same shape as 16x9 HD. So 3-perf first took off mainly for people shooting 35mm for TV.

Now with D.I.'s for theatrical movies becoming the norm, 3-perf has also caught on for theatrical features.

 

[M Most]

Since 4-perf Super-35 is naturally the same as the Silent Era format and is 1.33 : 1 (4x3), at some point, people started suggesting that some cameras be retooled to only pulldown 3-perfs instead of 4-perfs of film, since everyone was composing the 4x3 Super-35 negative for cropping to some widescreen format or another and thus wasting a lot of film. 4-perf is naturally 1.33 and 3-perf, being shorter, is naturally 1.78 -- exactly the same shape as 16x9 HD. So 3-perf first took off mainly for people shooting 35mm for TV.

There's a bit more to that story, at least regarding television. In (as I recall) 1986, I was working in post production at Lorimar and we were about to shoot 6 episodes of a show called Max Headroom. We knew going in that there was going to be an awful lot of video playback, and we also knew that we wanted to be able to use live video feeds as well as pre-edited pieces for those video screens. At the time, 24 frame playback was not a "live" operation, as it basically is today. It required the material to be pre-converted and color corrected for display on specially outfitted monitors in order to work. That's what was done for the pilot. But for the series, it would not work due to much tighter turnarounds and a more abbreviated production schedule. So we decided the only way to make the show was to shoot film at 29.97 fps, allowing us to use standard NTSC video and lock the film cameras to the playback system (using 25FPS and PAL video was considered, but rejected due to the need for standards conversion for the network delivery, which was deemed unacceptable). The problem was that the show had already been budgeted for 24fps film, and to go to 30 frame shooting was going to inflate the production as well as the post budget, since it would impact film stock, processing, and video transfer. My boss at Lorimar (Matthew Knox) had heard about a cinematographer in Europe named Rune Ericson, who had not only developed the Super 16mm format, but was experimenting with 3 perf pulldown on 35mm. Shooting with 3 perf would save 25% of the film stock, allowing us to shoot the show at 30 fps using the same amount of stock and processing that we would for 4 perf at 24fps. Since we were already posting on videotape (Lorimar was the first studio to adopt nonlinear electronic editing and video post production), the nonstandard nature of the format in the film world wasn't a deal breaker. Panavision specially outfitted some cameras for us, Rank Cintel came up with a 3 perf modification and a special, 12 tooth sprocket version of their 35mm gate, we built 3 perf synchronizers for the lab, and Max Headroom became the first 3 perf production in the US - in 1986. The system proved very workable, and within 2 years all Lorimar film series were shot on 3 perf, even though they were shot at 24fps. Viacom - who was the other early adopter of video post production for film shows - adopted the format as well, although all of the major studios remained on 4 perf, in part because they were still editing their programs on film. When Warner Bros. bought Lorimar in 1989, they had a policy of assembling negatives for all film originated programs, which Lorimar, of course, had not been doing. So 3 perf was abandoned at that point. But in less than 2 years after that, Panavision built 2000 foot mags for the G2's that were commonly used on multicamera sitcoms, which, combined with the 3 perf format, allowed a film camera to run more than 25 minutes without a reload - a big advantage when you're working in front of a live audience. This quickly became the format of choice for almost all film sitcoms to this day (although "Two And a Half Men" is now the only remaining multicamera sitcom that is shot on film), and migrated into single camera television when 16x9 came into the picture about 2 years prior to the introduction of HD broadcast in 1997.

Isn't history fun?

 

[sbcooler]

Very VERY interesting. So now I am thinking most of the stuff I see is based off of S35 sensor technology (Film or digital). So if one was to "step it up" to FF35 I suppose the larger sensor would be able to do better in low light situations...perhaps more dynamic range as well.

Now with great glass and using both from RED (which is not out yet) I wonder if we could see the difference or would both look great and the same. If you did not know which was which could you pick out which cam had the FF35 sensor and which had the S35?

I suppose that might be possible if you could freeze frame and look frame by frame and perhaps blow it up or magnify it so you could see fine detail and thus which looked like it had more detail...I am just wondering how much if at all the difference would be.

And is this meaningless until Red comes out with both brains and test sample are run showing the comparisons?

Thanks again for incredibly interesting info-

 

[David Mullen ASC]

Well, it can be hard to see some differences, just like it isn't always easy to spot 4:2:2 from 4:4:4, or Kodak from Fuji, or Cooke from Zeiss, or uncompressed from mildly compressed -- it can be shot dependent, and it depends on how you color-correct that shot.

Certainly some extra dynamic range would be welcomed in terms of color-correction, but not every shot pushes the limits of dynamic range. Even a camera with poor dynamic range can handle certain subjects or lighting set-ups just fine.

 

[Douglas Underdahl]

Well, 2 perf, or Technovision, pre-dates 3 perf. Do a Google on these terms and you will see that it goes back to the late 50's and some really influential films were shot with this money and time saving format. We shot the underwater sequences of TITANIC in 2 perf to get longer run times on deep dives. Lots of ways to skin the cat.