The director of a MV I am have been hired to shoot desires the scope and resolution of Anamorphic or Super-35. Having never shot in either of these mediums I would appreciate any advice on my approach to this project. I am mainly concerned with the increase of resolution and whether this warrents going with either of these formats rather than composing for 1.85.
Anamorphic and 2.35 super-35 have different looks and different practical considerations.
There are good reasons to choose either one....How about giving a little more information about the film and the concerns you have...
"I am mainly concerned with the increase of resolution "
If that is your only concern, with excellent lenses on the camera, anamorphic could be said to have more resolution than 2.35 super-35. However lots of variables come to mind.
>The director of a MV I am have been hired to shoot desires the scope and resolution of Anamorphic or Super-35. I am mainly concerned with the increase of resolution and whether this warrents going with either of these formats rather than composing for 1.85
If this is a music **video**, and will be finished and released on that medium, I really don't see what you're going to gain by using anamorphic. S35 is used in television every day, albeit with a 1.77:1 extraction rather than 2.35. This is done for the aspect ratio, not for increased resolution. I suspect Bill Bennett will take issue with this, but I really don't feel the increase in resolution apparent on film prints has any real significance when transferred to a medium whose resolution is much lower than that of film and fixed. But I would also urge you to not take mine or anybody else's word for it - shoot tests, if you can. And let us know the results.
Mike Most VFX Supervisor Los Angeles
Well, the main reason to shoot in 2.35 should be to COMPOSE in 2.35, shouldn't it?
These comments are really about shooting for 2.35 projection:
In terms of resolution, while some would say that spherical lenses are sharper than anamorphic lenses, once you crop Super-35 to 2.35 : 1 and blow-up what's left optically to an anamorphic dupe negative, you lose a lot of quality and get an increase in graininess. However, an anamorphic print uses a larger image area than matted 1.85 projection, so you have some improvement in graininess in the print since it is generally less enlarged to fill the same screen vertically as 1.85 projection. In the end, I think it sort of cancels out and Super-35 films blown-up to anamorphic end up looking similar in resolution & grain structure to 1.85 films -- the only difference is that you get the 2.35 composition.
Anamorphic photography mainly gains from using a larger negative & print area than 1.85, hence you get less grain & more detail. Anamorphic lenses generally aren't as perfectly sharp as the best spherical lenses, but if you stop down to at least T/4, you can overcome some of the distortion problems.
So resolution isn't enough of a reason to shoot for a 2.35 anamorphic print release, since the issue is more complicated than that. You really should WANT the 2.35 aspect ratio first, and then shoot in the format that gives you the right look in that aspect ratio. ___
Now since I've just realized that when said "MV" you meant "music video" and not "movie"....
35mm already exceeds the resolution of standard & high-definition television -- you're not really going to SEE a noticeable improvement in resolution by shooting in Super-35 nor anamorphic nor even IMAX if it's getting shown on standard definition television for broadcast. It would make more of a difference by using slower-speed film stocks, good lenses, and lighting with some contrast in terms of getting a cleaner, sharper image for TV. Music videos that have been shot in anamorphic in the past, and letterboxed, have used the format FOR its focus distortions, its lens flares, etc. -- in fact, they probably went out of their way to find older, more distorted anamorphic optics to get the classic anamorphic artifacts.
I think Bill Bennet would say that he shoots Super-35 mainly for the ability to adjust the image size & framing in the telecine transfer rather than to just use a bigger negative and use all of it for the final picture information.
David Mullen Cinematographer / L.A.
Agreeing with what has previously been said, I would like to add that by shooting SuperTV and transferring full ap (which can mean a variety of things) you will certainly gain resolution in the same regard that you gain resolution in shooting 35mm vs. 16mm. The advantage here is that you can shoot with faster stocks and fatter stops and buy back the compromises with the increase in resolution. So it allows more shooting options. So the final broadcast product may be the same but the process of getting there doesn't have to be. Right? The bigger negative always gives you greater flexibility.
Hi all involved in these two issues.>
Firstly, shooting anam. you only get a 2:35 framing thus keeping the TV vers. in mind most DoP's will concentrate the expensive actors in the middle of the frame knowing what happens in a pan-scan situation. There is a significant loss of resolution in the anam. lens and distorsions that are very difficult to match for CGI.
Shooting s35 you get about 2 more F-stops to work with and you can use slower film. Also using a proper ground glass that indicates both say a 1:78 (or 1:66 or 1:85) and the scope "slice" you can in fact frame for both versions and get away from pan-scan.
You might argue that the anam. lens used in tha lab. will also cost resolution but several tests (we have been involved in 4) show that a digital intermediate route with a digital anam. if by far superior the optical route (this is from 2K). Also it allows for an adjustable scope slice.
As for HD it has by all means a lesser latitude than film but one advantage the HD users have is a monitor on the set which means they can expose more correctly (providing it is all set up right) than a "not so trained" film Dop. Still the artifacts produced by trying to change the look of HDcam originating material are quite bad. This is why I suppose all this talk about "grading in the camera" is around.
Kris, Digital Film lab.
>You might argue that the anam. lens used in tha lab. will also cost resolution but several tests (we have been involved in 4) show that a digital intermediate route with a digital anam. if by far superior the optical route (this is from 2K).
The original post specifically mentioned that the questions were being asked relevant to the shooting of a music video. I believe he's looking for information related to video finishing only. I would think that any pan and scan issues are also irrelevant, since the video will, I assume, be finished letterboxed.
Good points about matching CG elements, though.
Mike Most VFX Supervisor Los Angeles
>Agreeing with what has previously been said, I would like to add that by shooting SuperTV and transferring full ap (which can mean a variety of things) you will certainly gain resolution in the same regard that you gain resolution in shooting 35mm vs. 16mm.
The difference in negative size between 1.37 Academy / TV and 1.33 Full Aperture (Super-35) is not as dramatic as the difference between 35mm and 16mm, so the gain in resolution exists, but I don't think it is significant when viewed in standard definition video later. I think the ability to zoom in and make framing adjustments when necessary is the bigger reason to shoot in Super-35 for commercial work -- what do you think Geoff?
David Mullen Cinematographer / L.A.
David Mullen wrote :
>I think the ability to zoom in and make framing adjustments when necessary is the bigger reason to shoot in Super-35 for commercial work -- What do you think Geoff?
I can't imagine not being able to alter framing in post
That's why I prefer to shoot S35 centered, there's the same space around to be played with.
As the stocks get better I find I'm framing a lot looser at times and then playing with image moves in post.
I shot a music video at the weekend, thanks Justin, for a TK colourist directing for the first time and we shot a lot of locked off stuff that he would scan at 2K and move around later.
I'm not sure what aspect ratio we were working to! he'll decide that later as well
It really does make a big difference to the way you shoot if you know that you have a 2K TK and unlimited time in Inferno/Flame/DS etc
I guess I've wandered from the topic a bit here but David is absolutely right, it's the ability to alter the framing that makes S35 so attractive for commercials, more room to play.
Geoff Boyle FBKSTS
Director of Photography EU based CML List Owner & Sysadmin
David Mullen wrote :
I didn't say that it was the same as the difference between 16 and 35, and it shouldn't take much imagination to understand what I was talking about, which was flexibility and options. If you loose a little resolution in a faster film stock (for instance) why not get it back with a bigger negative.
Shooting SMPTE Full Aperture, 1.33:1 gives you 50% more image area than standard TV transmission. That is significant enough for me. It isn't as though it's difficult or expensive to do. The 'conservative' Panavision SuperTV GG gives you 10% more image area, which you might say 'isn't much' but they did make the ground glass, so someone must have thought it was worth while. 'SuperTV' is just that, a larger transfer area, utilizing more negative, which is never a bad thing no matter how inferior the presentation medium is.
I agree that framing flexibility is a great reason to shoot Super in some situations, even ideal in many, but there are plenty of other ways and reasons to shoot Super 35.
Aside from the larger negative size used in origination and display of an Anamorphic project, another key element is the reduced depth of field offered by anamorphic shooting. Loosely translated, for an equal angular field of view / focal lenght, the depth of field will be reduced by 50%, due to the X2 factor of the anamorphic process, making your 50mm lens equal to a 100mm lens. This can be used as part of the cinematographic language, if that's what you wish, or it can be quite a nuisance, if you are only trying to get the 2:35 format. Also, good anamorphic lenses are heavier and cost more. For a TV finish, anamorphics may be more problems than they are worth. For a film finish, they offer a unique look that no spherical system can emulate.
Danys BRUYERE Dir. Operations Groupe TSF, Paris
>,,,,David is absolutely right, it's the ability to alter the framing "that" makes "20" S35 so attractive for commercials, more room to play.,,,, Geoff Boyle FBKSTS
I shot an MOW this summer in S35 and I think I noticed an improvement in the image structure. It's one of those things where you hear, "Wow, that looks great!", a few more times than the usual. It's a subtle improvement, but there none the less.
Does it make any sense that I thought the difference was a bit more noticeable with 5279?
May I suggest that you take a close look at the HAWK ANAMORPHIC compact lenses, manufactured by Vantage Films / Rodenstock.
Technical specs are as follows: 30mm T2.2 mod 3'6" / 1 m 4.2kg 35mm T2.2 mod 3'6" / 1 m 4.2kg
40mm T2.2 mod 3'6" / 1 m 2.1kg 50mm T2.2 mod 3'6" / 1 m 1.9kg 60mm T2.2 mod 3'6" / 1 m 2.1kg 75mm T2.2 mod 3'6" / 1 m 2.4kg 100mmT2.8 mod 3'6" / 1 m 2.8kg
135mmT2.8 mod 3'6" / 1 m 4.9kg 180mmT2.8 mod 3'6" / 1 m 6.4kg 250mmT2.8 mod 3'6" / 1 m 6.8kg
There is also one compact zoom lens (55-165 whhich weighs only 2.1kg at T4. Very nice for hand-held and steadycam. A few other zooms complete the set, but are predictably larger: 46-230mm T3.6 300-900 T4
So your main lens set, from the 40 yo 100, can be carried in two small lens cases. If you need closer focus, the Hawk V-Series lenses (30mm, 35mm, 40mm, 50mm, 60mm 75mm 100mm and 135mm) offer close focus down to 60cm, alowing you to have objects in focus about 15cm in front of your matte box.
These lenses are available exclusively in Europe from Iris Camera (Groupe TSF), Arri Media, EPC, and Vantage Films.
Flaring is a real issue with anamorphics. The Hawk lens system performs very predictably and consistently throughout the range of lenses, so you know how things will look.
>Claimer: I work for one of the above companies, not all of them... Do not hesitate to contact me directly if you want more info.
Danys BRUYERE Dir . Operations Groupe TSF / Iris Camera Paris.
Is this with an anamorphic scan (in effect, moving the CCD across the neg at half speed) providing anamorphic images straight out of the scanner, or is it by scanning flat, and then doing a vertical digital scale x2? If it's the latter, then I find it hard to believe that this could ever look all that sharp, given that the height of the cinema screen is going to be filled with only about 870 pixels of information. Or am I missing something? As I see it, doing super35 for 'scope with 2k flat scans is a pretty marginal exercise - I just saw a feature in this format with 2k vfx shots in, and I have to say, they didn't look very sharp against the non-vfx shots...
Well I saw an example of the work that DFL has done in anamorphic on Friday and was blown away by it!
I know that Gary & Simon from Red were impressed as well.
A beautiful looking film from a Finnish DP who shot with F400 in order to exploit the process.
No question of soft pictures there.
Geoff Boyle FBKSTS
Director of Photography EU based CML List Owner & Sysadmin
It seems to me that there is more of a difference between Anamorphic and Super 35mm 2.35 films looks than can be explained by either the distortions inherent in anamorphic taking lenses or the depth of field. They seem to render things in a much more dimensional way. I shoot frequently with anamorphics ( I own a russian set ) and whenever I investigate doing a project in Super 35, I put up the lenses, look through the finder and instantly feel somewhat disappointed with what I see.
Hypothetically say you shot with a: 50mm anamorphic at T8-5.6 and a 24mm flat prime lens at T2 in the super 35mm 2.35 format both focused at 10 feet.
The lenses / formats would produce a very similar horizontal and vertical field of view. The depth of field would also be very similar.
Would these images be nearly identical? I know that those might not be any ones real world stops for the same scene... But what would account for any differences?
Well, I'm not sure that this answers your question, but:
The perspective will not be the same. The angles may match but your relationship (the cameras relationship) to the angle will not be the same. That is to say, if you picture a wedge representing your field of view, in the anamorphic scenario you will be further inside of the wedge than in the non anamorphic. In the non anamorphic you will be closer to the 'point' of the wedge. Therefore the sense of peripheral vision as will not be the same. Some would argue that this is why anamorphic looks more natural or realistic than non anamorphic.
A subtle one is that the out of focus areas are (on the projected image) twice as soft vertically as they are horizontally. We're not looking at circles of confusion here, but ELLIPSES of confusion. On screen, like I said, the softness is twice vertically as it is horizontally, which means that on the neg (or digital image) it is four times.
How to spot digital vfx in anamorphic movies - look at the out of focus areas of the image. Very few people seem to make the effort to simulate this right.
"A subtle one is that the out of focus areas are (on the projected image) twice as soft vertically as they are horizontally. We're not looking at circles of confusion here, but ELLIPSES of confusion. On screen, like I said, the softness is twice vertically as it is horizontally, which means that on the neg (or digital image) it is four times.
How to spot digital vfx in anamorphic movies - look at the out of focus areas of the image. Very few people seem to make the effort to simulate this right."
My question is: How do you simulate the elliptical COF's? (Or is it OOF?)
Jim McKinney wrote:
>> My question is: How do you simulate the elliptical COF's? (Or is it OOF?)
Two choices - most decent compositing software (eg Cineon, Shake) allows you to apply a circular "blur" filter to an image. This should turn small sharp dots of light into hard circles, as in a lens, and works much better with an extended dynamic range colourspace such as 16 bit linear (MUST be applied in the linear domain - it doesn't work with native log images). One could adapt the filter to apply an elliptical filter instead of a circular filter (need a programmer!) Or, more pragmatically but less correctly, one could first scale the image by .25 along the vertical axis, apply a circle blur, and the scale it back to the original proportions. Other similar options would be, for best quality, to scale the image by 4 horizontally, blur it, and then scale back (expensive), or, as a decent compromise, scale it by .5 vertically, and 2 horizontally before the blur.