Jump to content

Understanding aspect ratio


Recommended Posts

Ok, I feel like a complete idiot for wanting to understand this so thoroughly and feeling like everyone else does but me, so what I do not understand is certain aspects of aspect ratio, no pun intended. So I am hoping you guys can clear this up for me. So like most DSLR shooters, I shot at 16:9 and to make things appear more cinematic people have suggested adding adjustment layers in post to create a letterbox on the top and bottom to give it a 2.39 look , but I've learned that this is not good for final output because of compression problems and viewing this export on larger monitors than the standard 16:9. So the alternative that I have heard is to change the sequence settings in premiere and keep the 1920 but change to 803 instead of 1080. Ok so this is where the really stupid question comes. Once that setting is changed why does the preview monitor when looking at a clip become wider than it was before at 1920x1080? I haven't changed the horizontal pixels, only the vertical amount, so why does it do this?

Second this is a general question on aspect ratios, why do wider aspect ratios, such as 2.39 reduce their vertical size? compared to lets say 1.85 or 1.78? why not keep the same vertical amount and just extend the horizontal amount?

And lastly why even use aspect ratios that will have letterboxing ? I understand the philosophy of having different emotional feelings with different ratios but as cinematographers wouldn't you want the most amount of image on the screen? If you want the shot to appear wider why not use a wider lens or pull back during filming, if possible.

 

I really appreciate the patience of anyone who answers these questions, And please correct me if I am totally wrong about anything that I have said. I really want to understand this as much as possible, whereas right now I am just going along with these things because it's what I have been told, but I don't ACTUALLY understand it that way I want to.

 

Thanks everyone.

Edited by Eric Soto
Link to comment
Share on other sites

why do wider aspect ratios, such as 2.39 reduce their vertical size? compared to lets say 1.85 or 1.78? why not keep the same vertical amount and just extend the horizontal amount?

 

To extend the picture horizontally, rather than cropping vertically, would require pixels that do not exist in the original frame. The only way to do this would be to shoot with an anamorphic lens, which stretches the horizontal field of view.

 

 

And lastly why even use aspect ratios that will have letterboxing ? I understand the philosophy of having different emotional feelings with different ratios but as cinematographers wouldn't you want the most amount of image on the screen? If you want the shot to appear wider why not use a wider lens or pull back during filming, if possible.

 

Letterboxing on a TV screen or computer monitor is necessary because the screen is a fixed size and shape. If you use an aspect ratio which does not match that of the screen then letterboxing or pillarboxing is inevitable.

 

In the cinema, it is possible to change the shape of the screen to match the film's aspect ratio, so you don't see any letterboxing.

Link to comment
Share on other sites

To extend the picture horizontally, rather than cropping vertically, would require pixels that do not exist in the original frame. The only way to do this would be to shoot with an anamorphic lens, which stretches the horizontal field of view.

 

Right, so then its not actually wider, since you are only cropping vertically right?

Link to comment
Share on other sites

Right, it's not actually physically wider, it just feels wider because of the shape of the frame.

 

ok great. Its surprising how no one ever mentions that in their explanations about aspect ratios. That it's purely a feeling thing. At least the ones I have seen.

 

Thank You!

Link to comment
Share on other sites

Another little interesting fact about aspect ratio's: The 2.39:1 cinema-scope (widescreen as we think of it today) was designed to make movies stand apart from Television.

 

I'm also of the opinion that aspect ratios should be selected according to the material being filmed. 2.39:1 rarely works for a drama or comedy, while it works great for epics and fantasy films.

Edited by Landon D. Parks
Link to comment
Share on other sites

 

ok great. Its surprising how no one ever mentions that in their explanations about aspect ratios. That it's purely a feeling thing. At least the ones I have seen.

If a sensor is 4000 pixels across, that's all it's ever going to be, no matter how 'wide' you make it by cropping vertically.

Link to comment
Share on other sites

  • Premium Member

Movies were 4x3 for a long time -- 1.33 : 1 for Silent Era films and then 1.37 : 1 (Academy Aperture) after sound-on-film prints came out.

 

There was some attempts during the time before the widescreen revolution of the 1950's to make movies more impressive on the big screen. In the Silent Era, there was a process called Magnascope that enlarged the 35mm projected image suddenly during the movie to fill a much larger screen, probably cropped a little (due to limitations of architecture) to around 1.63 : 1.

 

There was also Polyvision, which involved three 35mm prints projected side-by-side on a wider screen, used for a sequence in "Napoleon" (1927). I saw a presentation of the restored film and when that sequence came up, the audience started clapping, it was thrilling.

 

There was a brief fling with 70mm film with the Fox Grandeur process in 1930, which had an aspect ratio of 2.00 : 1.

 

But the widescreen revolution really started with Cinerama in 1952. It was designed by inventor Fred Waller, who had built a system that projected five images on a half-dome screen for gunnery training during WW2. Pilots told Waller that it would be great to see movies made that way, so he set about to refine it.

 

The general idea is to present a movie larger and with higher resolution so that you can sit close enough to create an immersive experience. IMAX does it with a nearly 4x3 screen but a very large one. But Waller knew that most theaters couldn't be modified to install a very tall screen so he decided that it was more important to increase the width of the screen so that at least your peripheral vision could be engaged if you sat close enough. He also made the wide screen curved and shot and then projected three 6-perf 35mm images side-by-side to create a 2.66 : 1 image that was very sharp. It was the IMAX of its day.

 

FA0025.THIS-IS-CINERAMA.IMG.SC.016.jpg

 

"This is Cinerama", independently produced, was one of the biggest box office successes in 1952 with lines forming around the block. Hollywood, fearing the loss of audiences to TV, jumped to replicate the success of Cinerama with a series of new formats.

 

CinemaScope (2X anamorphic lenses on 4-perf 35mm) came out in 1953, then VistaVision (8-perf 35mm, usually shown in 4-perf 35mm reduction prints) in 1954, Technirama (VistaVision with a 1.5X anamorphic squeeze to get a 2.35 : 1 CinemaScope aspect ratio) in 1955, as well as Todd-AO / Super Panavision (5-perf 65mm spherical 2.20 : 1, printed to 70mm). Then with a new projection format available, 70mm, you had variations like MGM Camera 65 / Ultra Panavision in 1957, which used a 1.25X squeeze on a 5-perf 65mm camera to get a 2.7 : 1 aspect ratio).

 

Also, soon after CinemaScope came out, Hollywood started vertically cropping 4-perf 35mm spherical to a wider shape, from 1.66 : 1 to 1.85 : 1.

 

Traditionally a movie theater showed 35mm anamorphic on a wider screen than they did their 35mm 1.85 prints, the idea that the widescreen image was "bigger".

 

It was with TV and its fixed screen shape, 4x3 and now 16x9, that widescreen movies had to be shown shorter with letterboxing to preserve their theatrical composition.

 

Most material shot for TV today is composed and presented in 16x9 full-frame other than some artistic commercials and music videos going for a 2.35 : 1 letterbox look to seem more cinematic. But a few shows have been shot for letterboxed presentation, like the 2.00 : 1 Netflix series like "The Crown" and "House of Cards".

Link to comment
Share on other sites

Most material shot for TV today is composed and presented in 16x9 full-frame other than some artistic commercials and music videos going for a 2.35 : 1 letterbox look to seem more cinematic. But a few shows have been shot for letterboxed presentation, like the 2.00 : 1 Netflix series like "The Crown" and "House of Cards".

 

Ok thank you for your thorough explanation. I guess I have a few more questions to follow :

 

So when it comes to resolution, lets say 2K and 4K which are both 1.89, if I'm correct, that means that obviously 4k is of higher resolution, but they both have the same aspect ratio correct? In terms of size of frame, the only difference is the number of pixels within the frame?

 

Also, I guess when it comes to order of understanding this, would it go like this. First you have your camera sensor which has its aspect ratio. Let's say the Red Epic Dragon which was used on some of the seasons of House of Cards, its sensor is 6144x3160, or 1.94. So anything that is filmed is captured at this aspect ratio but it is framed for 2.00 aspect ratio ? And this is done for purely aesthetic reasons, to make it appear more cinematic? To make it appear like 70mm ? And with all of that is aspect ratio comprise of camera sensor, aesthetics , and which monitor it will be viewed ?

 

Thank you very much, I apologise if these questions may seem redundant.

Edited by Eric Soto
Link to comment
Share on other sites

So when it comes to resolution, lets say 2K and 4K which are both 1.89, if I'm correct, that means that obviously 4k is of higher resolution, but they both have the same aspect ratio correct? In terms of size of frame, the only difference is the number of pixels within the frame?

 

2K and 4K are generally DCI terms. You’ll see these used when referring to a DCP package for cinema release. There is also UHD (Ultra HD) and HD, which are Television formats.

 

In terms of the “K” variety, 4K and 2K come in either scope or flat ratio; 2.39:1 and 1.85:1 respectively.

 

Here is a handy little chart:

 

2K Scope | 2048x858

2K Flat (1.85:1 crop) | 1998x1080

2K Native | 2048x1080

 

4K Scope | 4096x1716

4K Flat (1.85:1 crop) | 3996x2160

4K Native | 4096x2160

 

In the world of Television, you only have two really accepted 'standard' formats - HD and UHD, with UHD being closest to the cinema equivalent 4K. The TV aspect ratios are generally standardized at 16:9 and 4:3, though some odd aspects have been making their way in, like 2.00:1 used on some of the Netflix stuff like Stranger Things.

Link to comment
Share on other sites

 

2K and 4K are generally DCI terms. You’ll see these used when referring to a DCP package for cinema release. There is also UHD (Ultra HD) and HD, which are Television formats.

 

In terms of the “K” variety, 4K and 2K come in either scope or flat ratio; 2.39:1 and 1.85:1 respectively.

 

Here is a handy little chart:

 

2K Scope | 2048x858

2K Flat (1.85:1 crop) | 1998x1080

2K Native | 2048x1080

 

4K Scope | 4096x1716

4K Flat (1.85:1 crop) | 3996x2160

4K Native | 4096x2160

 

In the world of Television, you only have two really accepted 'standard' formats - HD and UHD, with UHD being closest to the cinema equivalent 4K. The TV aspect ratios are generally standardized at 16:9 and 4:3, though some odd aspects have been making their way in, like 2.00:1 used on some of the Netflix stuff like Stranger Things.

Ok so if I am understanding this right, when you are shooting lets say 4K Scope, when it is displayed in 16x9, the image will be cropped both vertically and horizontally? but you will only see letterboxing on the top and bottom right?

Link to comment
Share on other sites

It depends, do you want to keep the scope ratio?

 

Generally speaking, its rare for a film to be made with scope in mind, and then later make it 16x9 for release. More often than not, if a film is envisioned for 2.39:1, it is shot with that in mind - and released on ALL formats - including television, in that ratio. This done by 'letterboxing' the 16x9 frame to match the scope ratio.

 

There is a technique called 'pan and scan', which is what you would need to do if you where converting a 2.39:1 ratio into 1.78:1 - since cropping would be required. The 'Pan and scan' technique allows you to 're-frame' the film in the proper ratio. However, like I said, the more common way now is to release a 2.39:1 shot film in 2.39:1 across all platforms, in which case no cropping is needed at all - only letterboxing on consumer displays.

 

1.85:1, or flat release, can be different though. Since 1.85:1 is so close to 16x9 ratio, its just as common for the filmmakers to shave a few pixels off the top of each frame, and bring it natively into a 16x9 format so no letterboxing is needed on TV's. This works in this case because the difference between the two is so small. The difference between 2.39:1 and 1.78:1 are so vast that pan and scan is needed to really make it work, unless you are also framing for the 16x9 ratio when shooting.

 

Another important factor to remember here: as long as your aspect ratio is stating the same, you'll never need to cut or crop any part of your picture, no matter the final output size.

Edited by Landon D. Parks
Link to comment
Share on other sites

  • Premium Member

2K and 4K are theatrical digital projection formats, so you wouldn't generally show "DCP 4K scope" on a 16x9 display, you'd convert the 4K master to whatever home video master is needed, let's say. 1920 x 1080 HD with a letterbox, etc. And even the 4K DCP is made from a digital master of some sort that is the basis for all the up and downconversions, etc. for both theatrical and home video / broadcast.

 

So if your film was shot in a 2.40 format like 35mm anamorphic, then the digital master could be 4096 x 1716. Then your 4K DCP would be in that ratio, your UHD master would slightly downsample from 4096 to 3840 pixels with a letterbox because UHD is 16x9, and your HD master would downsample 4096 to 1920 pixels with a letterbox. You'd only crop the sides of the 4K master if your master was 2.40 and you had to make a 16x9 full-frame version from that (but there would be no need to crop top & bottom then.)

 

But many 2.40 movies are not shot in formats that are exactly 2.40, they have an original negative (like Super-35) or digital recording that is less widescreen than that and the image was composed for cropping to 2.40. So the digital master after color-correction usually retains something close to the full information of the original recording in order to make it easier to make all the different masters for delivery, whether widescreen theatrical, letterboxed home video, or full-frame home video (what is called "pan & scan").

 

For example some Red cameras record a 2.00 : 1 image and people sometimes compose for cropping top & bottom to 2.40, but if a 16x9 full-frame master has to be made for broadcast TV, they will use the full height of the 2.00 : 1 recording (assuming that was protected while composing) but have to crop the sides to create 1.78 : 1.

 

Two years ago I did an indie movie shot on an Alexa camera where we recorded 2K 16x9 ProRes4444 (2048 x 1152 downsampled in the camera from a 2880 x 1620 sensor area). It was finished to a 2K 16x9 master (2048 x 1152) for making a 2K scope DCP (2048 x 858), plus an HD master (1920 x 1080 w/ letterbox). I composed for cropping 2.40 from a 16x9 recording using a near common-top framing guide:

90M_framing.jpg

  • Upvote 1
Link to comment
Share on other sites

It depends, do you want to keep the scope ratio?

 

Generally speaking, its rare for a film to be made with scope in mind, and then later make it 16x9 for release. More often than not, if a film is envisioned for 2.39:1, it is shot with that in mind - and released on ALL formats - including television, in that ratio. This done by 'letterboxing' the 16x9 frame to match the scope ratio.

 

There is a technique called 'pan and scan', which is what you would need to do if you where converting a 2.39:1 ratio into 1.78:1 - since cropping would be required. The 'Pan and scan' technique allows you to 're-frame' the film in the proper ratio. However, like I said, the more common way now is to release a 2.39:1 shot film in 2.39:1 across all platforms, in which case no cropping is needed at all - only letterboxing on consumer displays.

 

1.85:1, or flat release, can be different though. Since 1.85:1 is so close to 16x9 ratio, its just as common for the filmmakers to shave a few pixels off the top of each frame, and bring it natively into a 16x9 format so no letterboxing is needed on TV's. This works in this case because the difference between the two is so small. The difference between 2.39:1 and 1.78:1 are so vast that pan and scan is needed to really make it work, unless you are also framing for the 16x9 ratio when shooting.

 

Another important factor to remember here: as long as your aspect ratio is stating the same, you'll never need to cut or crop any part of your picture, no matter the final output size.

ok so if you are shooting on a 4k camera and want to see it at 2.39 would you shoot at native resolution of 4096x2160, and then the footage would be cropped in post? Or would the camera offer a setting to shoot in 4k scope in which case, there wouldn't be any cropping until being displayed on a standard 1.78? I guess what I am still not understanding is that when you shoot for 2.39, you are assuming there is going to be cropping or letterboxing on the top and bottom, correct? Is it ever the case that one can shoot at 2.39 and it displays without cropping? ( Monitors or screens that have that aspect ratio?)

Link to comment
Share on other sites

  • Premium Member

Ok so if I am understanding this right, when you are shooting lets say 4K Scope, when it is displayed in 16x9, the image will be cropped both vertically and horizontally? but you will only see letterboxing on the top and bottom right?

 

As I said, if your master is 4096 x 1716 pixels because you had shot in a true 2.40 format like 35mm 2X anamorphic, then to make a 16x9 version for UHD (3840 x 2160) then you'd downsample (shrink, resize, etc.) from 4096 to 3840 and use letterboxing because the image was 2.40 so wouldn't fill 16x9 vertically. So no cropping horizontally unless someone just wanted to crop to get to 3840 from 4096 to avoid the slight resizing. If you plan for that, I can imagine a scenario where you record a 4096 image but compose for a slight trimming to 3840 because you are aiming to make a UHD master instead of a 4K DCP.

  • Upvote 1
Link to comment
Share on other sites

It depends on the camera. As David said above, some cameras can actually record ratios different from the standard DCI and HD standards. However, for the sake of example, lets say you are shooting and recording at 4K DCI (4096x2160), and you're goal is for a 2.39:1 release. Yes, you would crop off the top and bottom of the DCI image to get the ratio you need. You'd generally 'frame' for this on the set with frame markers on your monitors.

 

This of course doesn't take into account some cameras that can shoot native scope (though they still tend to just crop the sensor), and some cameras and lens combinations that allow anamorphic recording on a 4x3 sensor, in which case you wouldn't crop anything, you'd squeeze it into the correct ratio. Ratios and formats are rather complicated subjects, because there are so many variables involved.

Edited by Landon D. Parks
Link to comment
Share on other sites

It depends on the camera. As David said above, some cameras can actually record ratios different from the standard DCI and HD standards. However, for the sake of example, lets say you are shooting and recording at 4K DCI (4096x2160), and you're goal is for a 2.39:1 release. Yes, you would crop off the top and bottom of the DCI image to get the ratio you need. You'd generally 'frame' for this on the set with frame markers on your monitors.

 

This of course doesn't take into account some cameras that can shoot native scope (though they still tend to just crop the sensor), and some cameras and lens combinations that allow anamorphic recording on a 4x3 sensor, in which case you wouldn't crop anything, you'd squeeze it into the correct ratio.

 

Awesome ok great, you guys have been a great help. I understand now. I guess a last more artistic question, why do you guys choose to shoot at 2.39 when you know that you will lose some of frame due to cropping? I understand that it gives a certain feel. But do you guys think that that is more an appeal to tradition and its just a way that we think things look cinematic? As cinematographers wouldn't you always want to present the most of the frame as possible ?

Link to comment
Share on other sites

  • Premium Member

ok so if you are shooting on a 4k camera and want to see it at 2.39 would you shoot at native resolution of 4096x2160, and then the footage would be cropped in post? Or would the camera offer a setting to shoot in 4k scope in which case, there wouldn't be any cropping until being displayed on a standard 1.78? I guess what I am still not understanding is that when you shoot for 2.39, you are assuming there is going to be cropping or letterboxing on the top and bottom, correct? Is it ever the case that one can shoot at 2.39 and it displays without cropping? ( Monitors or screens that have that aspect ratio?)

 

4096 x 2160 is 1.89 : 1 -- some cameras offer that recording option, I think maybe the Sony F65 does, and a Red camera can do that.

 

Yes, most people if composing for 2.40 using normal spherical lenses will record more picture height than 2.40 if the sensor offers that because that makes it easier in post to both do things like image stabilization, or gives a VFX person more room to move the image around, but mainly because you will have to make both a 2.40 master and a 16x9 "pan & scan" version for TV broadcast.

 

But some cameras like the Red will allow you to only record a 2.40 image area so you are basically stuck with that aspect ratio in your original master, and therefore any versions that are less widescreen than 2.40 will involve cropping the sides of the image rather than opening up the image top & bottom.

 

If you shot 4-perf 35mm with 2X anamorphic lenses (or 3-perf 35mm with 1.3X anamorphic lenses) or 2-perf 35mm with spherical lenses, you'd have a 2.40 original and your scan and digital master would be 2.40, but most other formats in film that you shoot are less widescreen in shape.

 

Monitors are almost universally 16x9 (1.78 : 1) and so are most home video masters.

Link to comment
Share on other sites

 

4096 x 2160 is 1.89 : 1 -- some cameras offer that recording option, I think maybe the Sony F65 does, and a Red camera can do that.

 

Yes, most people if composing for 2.40 using normal spherical lenses will record more picture height than 2.40 if the sensor offers that because that makes it easier in post to both do things like image stabilization, or gives a VFX person more room to move the image around, but mainly because you will have to make both a 2.40 master and a 16x9 "pan & scan" version for TV broadcast.

 

But some cameras like the Red will allow you to only record a 2.40 image area so you are basically stuck with that aspect ratio in your original master, and therefore any versions that are less widescreen than 2.40 will involve cropping the sides of the image rather than opening up the image top & bottom.

 

If you shot 4-perf 35mm with 2X anamorphic lenses (or 3-perf 35mm with 1.3X anamorphic lenses) or 2-perf 35mm with spherical lenses, you'd have a 2.40 original and your scan and digital master would be 2.40, but most other formats in film that you shoot are less widescreen in shape.

 

Monitors are almost universally 16x9 (1.78 : 1) and so are most home video masters.

I see, makes more sense now.

Link to comment
Share on other sites

  • Premium Member

 

Awesome ok great, you guys have been a great help. I understand now. I guess a last more artistic question, why do you guys choose to shoot at 2.39 when you know that you will lose some of frame due to cropping? I understand that it gives a certain feel. But do you guys think that that is more an appeal to tradition and its just a way that we think things look cinematic? As cinematographers wouldn't you always want to present the most of the frame as possible ?

 

What does "present most of the frame as possible" mean? You mean "use as much of the TV monitor as possible"? What if you are shooting for theatrical release?

 

Yes, the primary reason people shoot and compose for 2.40 is that they like the aspect ratio, for graphic reasons (they like the shape for staging action and composing subjects), for historical / nostalgic reasons (they like old CinemaScope movies, they are inspired by the scope compositions of Kurosawa, Leone, Lean, etc.) and because they are shooting for theatrical release and feel that a 2.40 image is more "epic" (again, for nostalgic reasons partly, since a lot of theaters today unfortunately use common width instead of common height for showing 1.85 and 2.40.)

 

Even back in the days of 35mm shooting, before D.I.'s, before 2000, if you shot for a 1.85 release, your negative was 4-perf and therefore 1.33 : 1 so you were composing for cropping to 1.85 during projection (what is called matted widescreen).

 

Cropping is only a "problem" if you feel that the viewer is going to notice some sort of loss in resolution, otherwise if no one notices that you cropped to achieve your desired composition, then why does it matter?

Link to comment
Share on other sites

This is the old framing guide used in the Super-35 cameras for "Terminator 2" to give you an idea of the framing & cropping issues for shooting a movie that was released in 35mm 2.40 anamorphic prints:

super35chart.jpg

 

 

What does "present most of the frame as possible" mean? You mean "use as much of the TV monitor as possible"? What if you are shooting for theatrical release?

 

Yes, the primary reason people shoot and compose for 2.40 is that they like the aspect ratio, for graphic reasons (they like the shape for staging action and composing subjects), for historical / nostalgic reasons (they like old CinemaScope movies, they are inspired by the scope compositions of Kurosawa, Leone, Lean, etc.) and because they are shooting for theatrical release and feel that a 2.40 image is more "epic" (again, for nostalgic reasons partly, since a lot of theaters today unfortunately use common width instead of common height for showing 1.85 and 2.40.)

 

Even back in the days of 35mm shooting, before D.I.'s, before 2000, if you shot for a 1.85 release, your negative was 4-perf and therefore 1.33 : 1 so you were composing for cropping to 1.85 during projection (what is called matted widescreen).

 

Cropping is only a "problem" if you feel that the viewer is going to notice some sort of loss in resolution, otherwise if no one notices that you cropped to achieve your desired composition, then why does it matter?

 

 

What does "present most of the frame as possible" mean? You mean "use as much of the TV monitor as possible"? What if you are shooting for theatrical release?

 

Yes, the primary reason people shoot and compose for 2.40 is that they like the aspect ratio, for graphic reasons (they like the shape for staging action and composing subjects), for historical / nostalgic reasons (they like old CinemaScope movies, they are inspired by the scope compositions of Kurosawa, Leone, Lean, etc.) and because they are shooting for theatrical release and feel that a 2.40 image is more "epic" (again, for nostalgic reasons partly, since a lot of theaters today unfortunately use common width instead of common height for showing 1.85 and 2.40.)

 

Even back in the days of 35mm shooting, before D.I.'s, before 2000, if you shot for a 1.85 release, your negative was 4-perf and therefore 1.33 : 1 so you were composing for cropping to 1.85 during projection (what is called matted widescreen).

 

Cropping is only a "problem" if you feel that the viewer is going to notice some sort of loss in resolution, otherwise if no one notices that you cropped to achieve your desired composition, then why does it matter?

 

Right ok, I understand, Thank you for all your help.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...