16mm & Super16 Aspects Ratios

I spoke recently with a camera assistant who seemed to think 16mm and Super16 both share the aspect ratio of 1.66:1. I told him I thought he might be wrong, that I thought 16mm had an aspect ratio of 1.33:1 and Super16 an aspect ratio of 1.66.1. Also, that 16mm and Super16 share the same gate height but that Super16 is wider because of the loss of perforations on one side of the negative. I said I thought the whole point of shooting Super16 was twofold; one because it allowed for a wider image which could facilitate a blowup to 35mm (and a 1.85:1 aspect ratio), and two because negative space is gained (horizontally only) allowing for a finer grain when doing such a blow up.

As I understand it, when shooting regular 16mm to achieve a blowup to 1.85 or some similarly wide aspect ratio, one needs to crop the top and bottom of the 16mm frame. Also, such cropping happens MORE in 16mm than it does with the Super16 frame, which given the aspect ratio (as I understand it) of 1.66 for Super16 is already closer to 1.85.

On page 36 of Dominic Case's book "Film Technology in Post Production" I read the following :

"The image on 16mm film has an aspect ration of 1.33:1 - the same as for standard television." and "... The Super 16 frame is wider [than 16mm], using single perf 16mm negative exposed almost to the edge of the non-perforated side, and so already has an aspect ration of 1.66:1."

This would lead me to believe that my previous assumptions, at least about the respective aspect ratios of 16mm and Super16 images, are correct. However, my friendly camera assistant called up a large and reputable New York City camera rental house just to make sure he was correct about his assertions and then called me back today to say his suspicions had been confirmed by a camera tech there... that 16mm and Super16 cameras have the same aspect ratio, 1.66:1. Apparently the camera tech said Super16 enlarges the width of the frame BUT also the height of the frame, resulting in the same aspect ratio for both 16mm and Super16. He said the gate size for 16mm is 7mm x 11.5mm and for Super16 is 7.5mm x 12 1/3mm. If my math is correct, this would result in an aspect ratio of 1.643:1 for both 16 and Super16.

Dominic Case's book (mentioned above) states on page 37 that the 16mm frame is 0.295" x 0.404" and the Super16 frame is 0.292" x 0.493". Again if my math is correct, this would indicate that 16mm has an aspect ratio of 1.369:1 and Super16 an aspect ratio of 1.688:1.

Now it has been more than ten years since I worked as an camera assistant...having been blessed with a suspect memory from birth I'm not relying on past experience to figure this out. So I come to you, CML, to clear this all up.

Does anyone know if there is a DEFINITE answer here? Are different cameras setup differently when it comes to the above mentioned 16mm/ Super16 debacle?

Also, can someone explain to me why we speak of 1.66 and 1.33 when my calculations above come up with numbers like 1.688, 1.369 and 1.643 which don't even come close to 1.66 or 1.33 ?

Are 1.66 and 1.33 just ROUGH numbers? Do numbers exist which give the EXACT aspect ratios for what is commonly referred to as 1.66 and 1.33 and do these numbers remain the same for EVERY 16mm or Super16 camera?

Contrary to what the camera technician at the reputable New York City camera rental house ostensibly said, Mr. Case's book shows that the frame height in 16mm is slightly taller than Super16. If Mr. Case's book is correct, why design the gate to be smaller in Super16? Why not leave it the same height and simply change the width? Is there a good reason for this?

Please excuse my laborious writing, but I'm trying to be very exact here so as to avoid any confusion.

Cheers,

Piotr Jagninski
Gaffer / New York City

Ok,

Arriflex and Aaton formats are slightly different, but not by a lot.

Here are the specs for Arri std / super 16 formats (I'll not mention any others)

style="font-size: 13px">STD 16 :

Camera aperture 10.3mm x 7.5mm ( 1:1.37 )

1:1.66 and 1:1.85 etc will be cropped from this.

Standard TV Transmission : 9.35mm x 7.0mm ( 1:1.33 )

style="font-size: 13px">Super 16 :

Camera aperture 12.35mm x 7.5mm ( 1:1.65 )

Super 16 projection 11.75mm x 7.05mm ( 1:1.66 )

Super 16 1:1.85 11.75mm x 6.35mm

Super 16 1:1.78 Transmitted 11.95mm x 6.72mm

Hope this helps,

Andy Taylor
Camera Engineer
Arri Media
UK
www.arrimedia.com
www.arri.com

Piotr --

To put it short & sweet, you are right and your friend is wrong. Regular 16mm is 1.33 and Super-16 is 1.66. When cropping to 1.85, S-16 yields a 47% greater negative area than R-16. The only possible misunderstanding that I can imagine here is that most cameras available today are switchable between 16/S-16 but the physical gate is left unchanged so that you are exposing a part of the negative that is unused in R-16 photography. That may be the birth of the misunderstanding from the AC who believed that R-16 is still 1.66. Super-16 was invented by Rune Ericcson (sp?) in Sweden when he carved out the 1.33 gate on a 16mm camera, expanding one side by two millimetres into the area where the soundtrack normally is placed on a print. The height was not changed only the width, resulting in a new gate ratio of 1.66.

This reminds me of a somewhat similar misunderstanding noted here a few months ago where a producer misunderstood the difference between lens coverage in 16 v. 35 and it was claimed that a camera rental house tech explained it improperly. We all know that it was the producer who simply didn't understand what was said. But this case is an AC and a camera house tech, who are both people who should know better. Honestly, if I found out that my AC or camera tech didn't understand the difference between 16 and Super-16 I don't think I'd ever work with them again. It's their job to know this sort of information.

Mitch Gross
NYC DP

Super-16 camera image area is specified by standard SMPTE 201M.

Regular 16mm is specified by SMPTE 7.

Standard SMPTE 96M-1999 specifies the scanned image area for display in on both 4:3 and 16:9 video systems.

Standards may be purchased from the SMPTE :

http://www.smpte.org/smpte_store/standards/

John Pytlak
EI Customer Technical Services
Research Labs, Building 69
Eastman Kodak Company
Rochester, New York 14650-1922 USA
http://www.kodak.com/go/motion

Mitch Gross writes:

>The height was not changed only the width, resulting in a new gate ratio >of 1.66.

I don't suppose there was anything he could have done to *increase* the height anyway, short of eliminating the frameline altogether. ...which would have gained... what? A half-millimetre?

That would have been useless anyway, since the specified thickness of the frameline protects against the normal frameline-position variations between individual cameras.

Dan "has seen LOTS of those variations" Drasin
Producer/DP
Marin County, CA

>Dominic Case's book (mentioned above) states on page 37

Thank you Piotr : . . . in the interests of precision, I should point out that you refer to the second edition (2001). Early buyers who still have the first edition will need to look at pp 24-25.

The dimensions that I quote in the book are identical to those quoted in my copy of the American Cinematographer Manual. SMPTE standard 201M (type W - or super 16) shows a slightly different width (.486" instead of AC's .493"). Andy Taylor quotes slightly different dimensions for Arri, and Aaton is different again. All within 2% though.

I tend to quote aspect ratios for the corresponding print dimensions (which differ only in the second decimal place, not a distinction that I lose a lot of sleep over). The actual ratio for standard 16 are 1.37:1 for the negative, 1.33:1 for the print. For super 16, the negative is 1.66:1 or 1.69:1 depending whether you back SMPTE or ASC, or 1.65:1 if you take Andy's real world of Arri cameras.

In practice, the camera gate dimensions are less important than the print (or TV) masking, as that's what defines what is in and what is out.

In passing, I also choose to show ratios as 1.66:1 rather than 1:1.66. It's against the convention, I know, but it's easier to say. English and Australian CMLâ€™ers will recognise a similar clash of conventions in quoting
cricket scores.

Dominic Case
Atlab Australia

>I can't say I understand why any camera manufacturer would deviate >from a standard. How hard could it be to make a film gate exactly match >the specs?

But what practical value would that be? What matters more is that framing lines in the viewfinders are standardized and projection apertures are standardized. The fact the some edges of the negative are not used is less important -- for example, NO 35mm projection format uses the entire 35mm negative area except perhaps the Silent Era aperture. In fact, many people have found it useful that the entire negative area is not used in projection -- to hide splices, hairs, etc. So I don't understand what difference does it make if some S16 cameras have a 1.66 gate and others have a 1.69 gate, since there is no such thing as full aperture S16 print projection (other than at some labs.)

The answer for these subtle variations tend to be historical -- standards evolve over time. First someone invented the S16 format and then someone else decided to start writing some standards for it. And then those standards are adjusted. But like I said. not being a projection aperture, the standards for the S16 camera aperture are not as critical.

This isn't about maintaining the integrity of a composition, which is a separate issue from what the full aperture is. Plus I can't imagine a composition that is visibly different in 1.66 versus 1.69 anyway.

David Mullen ASC
Cinematographer / L.A.

David Mullen ASC wrote :

>...Plus I can't imagine a composition that is visibly different in 1.66 >versus 1.69 anyway.

Yeah, don't be a ratio-ist!

Wade K. Ramsey, DP
Dept. of Cinema & Video Production
Bob Jones University
Greenville, SC 29614

Thank you to all who replied to my query.

I can't say I understand why any camera manufacturer would deviate from a standard. How hard could it be to make a film gate exactly match the specs?

Dominic Case's statement that a 2% deviation from the standard is probably not worth losing sleep over makes sense to me. That said, someone had to CHOOSE to do things differently. I would have thought there would be a reason for that choice.

Cheers,

Piotr Jagninski
Gaffer / New York City