Home of Professional Cinematography since 1996
class="style7"> Super 35 2.35:1 Image Area
>Published : 6th April 2008
>My mind tends to glaze over when ever this stuff comes up, but I've had to order a ground glass and glow slide and am not sure which one to get.
>Obviously, there is more than one 2.35:1 Super 35 (4-perf) option, and while I'm sure there's a reason for the multiple formats, I don't know which one to pick.
>Option one is a 24.0x10.2mm area, and appears to be in the center of the ground glass.
>Option two is a 24.0x10.2mm area, and appears to be centered north by 1.35mm. I'm guessing the reason that it's above center is because you may not need much more headroom as leg room when re-composing for other formats, like TV or 1.85.
>Option three is a 24.8x10.5mm area, and also appears to be north of center by 1.35mm.
>I'm betting there are plenty more options as well, especially if you factor in covering multiple formats on one glass. Also, by doing some research with Arri paperwork (this is for a Moviecam), I'm noticing ANSI vs. DIN, and also centric vs. asymmetrical (the 1.35mm offset, no doubt). Of course, the Moviecam specs are not in English and not as well annotated as the Arri documents, which is complicating my efforts.
>Is there a "most common" version of Super 2.35? I'm interested in a rental house perspective, because it may have to go out with other cameras that I don't own, and these suckers are expensive.
>All the best,
Los Angeles DP
Super-35 shouldn't be off-centered; your choices in 4-perf tend to be 2.40 center extraction or 2.40 raised for "common top" -- although some put it higher near the top than others. I like common top slightly lowered from the top, more like sharing 4x3 TV transmitted headroom.
It's what is called the "Fincher" groundglass at Panavision, at least for the 3-perf Super-35 groundglass.
>I'm also curious whether center extraction is more popular or not.
>David Mullen, ASC
>I typically get requests for common top.
>I think I've only had one client ask for center extraction.
Camera Tech, NYC
class="style8">>>I don't understand... is this so there's never more headroom than what >>was originally intended?
>Yes. The problem with center extraction of 2.40 from a 4-perf Super-35 negative (1.33 native) is that if transferred to 4x3 video full-frame, the amount of extra headroom is ridiculous (plus the mic has to be farther from the actors) -- so you have to zoom into the negative in order to remove some of the headroom, but then you lose some of the sides, which was one of the reasons you probably shot Super-35 instead of anamorphic, to avoid losing the sides.
>But the problem of putting 2.40 too close to the top of the frame is that now your picture area is close to the framelines, so hairs in the gate, etc. become more critical, plus you are using less of the center of the lens. Hence why a "lifted" 2.40 frame is a good compromise, with the top of 2.40 being about 10% below the top frame line.
>With 3-perf, I think 2.40 common top becomes even more critical because more than likely you'll want to transfer full-frame (which is 1.78 native) over to a 16x9 full-frame version directly without having to make too many compositional adjustments. With 4-perf, you know that some vertical reframing will be necessary because 4x3 is just so different from 2.40.
>David Mullen, ASC
class="style8">>>...although some put it higher near the top than others. I like common >>top slightly lowered from the top, more like sharing 4x3 TV transmitted >>headroom.
>I, Robot also shot a "not common, not centered groundglass.
>It was sort of two thirds towards common top, but a little less displeasing
than some extractions:-)
VFX Photography and Supervision
LA based again starting tomorrow!!!
class="style9" >>I typically get requests for common top.
class="style9" >>I think I've only had one client ask for center extraction.
>Whichever format you choose, be sure to always shoot some film of your ground glass markings, to attach at the head of every roll of dailies projected or transferred, so the projection and transfer will have the same framing you wanted.
We would shoot 100 ft. or so during the camera checkout, and the Asst. Editor could order as much footage as he wanted from that negative, to splice onto the dailies rolls we were screening. The projectionist would use the closest matching mask in the projector, and frame to our "projection leader," to ensure accurate framing. Same for the video transfer.
>This goes for EVERY format, not just 2.35:1
>Hey Graham -
>I have always recommended using common center...I'm sure you're bound to get a hundred opinions from here but mine has always been to use the center of the lens- especially with Super. Some lenses that "mostly" cover S35 can vignette closer to the wide end.even if it's not a hard vignette you can sometimes get a stop or more loss from center to edge. If anyone knows of a lens with a sweet spot higher than center, I'd love to see it.
>Great point, Steve.
>Especially if you're using a Zoom. Although I also agree with David Mullen and others that it's better for video transfers to be able to have more room below the image rather than above, for reframing purposes.
>Still, what about the slight size difference between Option 3 (it's bigger) and options 1&2?
>And the difference between ANSI and DIN? And is Common Top usually based on this 1.35mm offset, or is there an even higher offset that puts it closer to the top of the frame?
LA based DP
>ANSI and DIN are two different standards. Don't loose sleep over the two.
ANSI measurements are slightly more generous than DIN...either way the
aspect ratio remains the same.
class="style9">>>also agree with David Mullen and others that it's better for video >>transfers to be able to have more room below the image rather than >>above, for reframing purposes
>I don't understand... is this so there's never more headroom than what was originally intended?
>I think the 1.35mm offset you keep referring to is based on the width of the frame....keep in mind that the aspect ratio is based on a simple mathematical equation. The width of the frame you are looking at is 24mm (ANSI S35) so regardless of if it's centered or common top, there will be equal distance either at the bottom of the frame (common top) or
>Arri's ANSI S35 2.35 + 1.85 centric vs. Arri ANSI S35 2.35 + 1.85 common top would be a good example of this.
class="style9">>>I, Robot also shot a "not common, not centered groundglass.
class="style9">>>It was sort of two thirds towards common top, but a little less >>displeasing than some extractions
>I shot a 1/3 - 2/3 extraction on "Monster's Ball" in 2001.
>I did it to try to avoid the nastiness often found in the extractions for home video and HBO who insist on a 4:3 version. I did the 1/3 from the top framing with my GG marked that way. And in order to keep the frame as close to that in the 4:3 version, I tried to include microphones and dolly track whenever possible. That way they would have to do a pan scan version which I actually prefer to the various alternatives.
Especially for TV where the 4:3 extraction from the 2:35 negative image area is not a problem for grain buildup or other "optical blow-up" type issues.
Roberto Schaefer, asc