Part 1: Optical Process – Spherical vs. Anamorphic

(Featured image courtesy of Panavision.com)

What is the first element that your image will touch? The lens, right?! So, I think it would be fitting to makes this the first topic of discussion.

In this part we’ll be discussing optical format or process. To what do I refer? Basically whether you will process your image using spherical or anamorphic lenses.

For me this is the first and most critical consideration, even before we get to,  “Film vs. Digital or ALEXA vs RED Dragon/Weapon that!”. Why?

There are techniques now that allow you to emulate the look of film or match the look of different cameras, that to the general viewer, it is difficult to discern one from the other. However, there is really no effective means to duplicate the look of one optical process to the other.

For me this is like choosing between oil or watercolor. They are both used to create images, but the characteristics of each and their implications are so different, that choosing between the two is as much choosing a viewer’s experience, as it is a technical process.


Basics

If you are newer to photography/cinematography, it is critical to become familiar with the basics concepts of imaging. Here is a brief explanation of some of the terms that will appear in this article (there are so many resources already available that explain these so well that I won’t even try attempt to explain these at length).

FOV (Field of View) The observable expanse visible through an optical instrument at a given time. In simple terms, its the how much a lens can see.

Focal Length Technically speaking, the focal length is the measurement (expressed in mm) of the distance between the lens and the capturing device (film or sensor) when the subject is in focus.

Practically speaking, the focal length of a lens lets us know the FOV (field of view) of a lens – the lower the focal length the wider the field of view, the high the focal length, the narrower the field of view.

DOF (Depth of Field) The distance from the closet and farthest objects that appear to be in focus is referred to as the Depth of Field.

The amount of DOF is a result of 1) lens focal length, 2) lens aperture, and 3) subject distance to lens.

Aperture Opening which allows light to pass through a lens. This opening can be expanded (t0 allow more light) or constricted (reduce amount of light),  altering the exposure (brightness) of an image.

Lenses that have wider apertures are often referred to as being “faster”, meaning they are able to handle low-light situations. Another impact of aperture is that the wider it is, the less DOF a lens will have.

The size of the opening is often expressed numerically as “f-stops” or “t-stops”. This is not a physical dimension but a ratio expressing the ratio of the lens’s focal length to the diameter of the aperture. Practically speaking, the smaller the number, the wider the opening thus allowing more light to pass through lens.

And here is an awesome video discussing the fundamental properties of lenses.

 


Spherical and Anamorphic – what is it? 

Basically refers to differing optical processes (both in mechanical construction and visual characteristics).

Spherical lenses are by far the most common and readily accessible optical format. Think of any photo you’ve ever seen or taken and more than likely, they were all taken with spherical lenses.

Anamorphic lenses were designed primarily to serve the motion picture industry, though one could theoretically use them to take still photos as well. Many films have used this optical process – Blade Runner (1982), Braveheart (1995), Dark Knight (2008), just to name a few.

All camera lenses are constructed using multiple lens elements. Spherical lenses use spherical elements, a spherical lens element has curvatures on two axis – vertical and horizontal. This symmetry results in an image that is uniform. How does anamorphic format differ?

AFNikkor80200f28_cross_zpsb7a91631
Cross section of a lens.

Like their spherical counterparts, anamorphic lenses are also composed of multiple lens elements. The main mechanical difference between spherical and anamorphic lenses is the inclusion of a cylindrical lens element, a cylindrical lens element is only curved on the vertical axis. What impact does the cylindrical element impact have?

lens17
Diagram of a cylindrical lens.

Vertically, it bends light similar to a spherical lens. Horizontally, however, it bends light quite differently – it has a wider FOV horizontally than it does vertically. It “squeezes” the image on the horizontal axis and effectively gathers more information. Essentially, an anamorphic lens is similar to having two focal lengths but in a single lens – one for the vertical axis and another for the horizontal axis (Source:http://www.filmanddigitaltimes.com/wp-content/uploads/2013/08/Anamorphic-Math-FDTimes.pdf).

Initially, the image will look “smushed”, images look tall and objects appear narrow and slender. To properly view an anamorphic image it must first be “de-squeezed”. During film projection, this is accomplished by projecting the image through another anamorphic lens that “de-squeezes” the image. Digitally, you would need to change the images native pixel aspect ratio (Source: http://www.4kshooters.net/2015/11/25/desqueezing-anamorphic-footage-in-davinci-resolve-12/).

Screen Shot 2016-03-17 at 10.24.52 AM
Anamorphic image before “de-squeezing”.
Screen Shot 2016-03-17 at 10.25.01 AM
Anmorphic image after “de-squeezing”.

There are two basic “flavors” of anamorphic lenses, 2.0x and 1.3x lenses (we’ll discuss more in detail the differences between each “flavor” a little later in the blog). The numbers refer to the “squeeze” factor of the lens (there are also 1.25x and 1.5x but are less common). For example, a 2.0x anamorphic lens with focal length of 50mm will have double the FOV on the horizontal axis (it will actually be closer to a 25mm) as that of a 50mm spherical lens.

 


Spherical and Anamorphic – Why? 

It really all starts with Aspect Ratio. Simply put, aspect ratio is the ratio of the width of the image to the height. This can be expressed as two numbers like 4×3 or 16×9 or as a decimal such as 1.85 and 2.35 (4:3 = 1.33, 16×9 = 1.78, & 21×9 = 2.35) – though these can be written as a ratio as in 2.35:1. (Source: http://filmmakeriq.com/lessons/the-changing-shape-of-cinema-the-history-of-aspect-ratio/).

In filmmaking there are two ratios that we are most concerned with –  1) the aspect ratio of the capture device (shape of the film frame or digital sensor) and 2) the final or presentation ratio (shape of what the audience will see). Below is a video that discusses in detail the origin of aspect ratios.

As discussed in the video above, the goal of filmmakers was to achieve a wider aspect ratio or widescreen. How is this achieved when filming with spherical lenses?

When capturing spherically, to achieve a wider presentation ratio, the image needs to cropped. While achieving a wider frame, this also wastes significant portions of the capture area. What if you could achieve a wider frame without cropping?

Enter anamorphic format. The goal of anamorphic lenses is to achieve a wider aspect ratio without cropping the image and maximizing the capture area. How is this achieved?

When filming anamorphically, the final or presentation aspect ratio is the result of multiplying the capture aspect ratio by the squeeze factor of the lens. Let’s take a 2.0x anamorphic lens as an example.

If you pair an 2.0x lens with a capture device with an aspect ratio of 1.78, what do you get? 3.56 (1.78 x 2.0). This is too wide. To achieve the target ratio of 2.40 (rounded), you would need to crop the ends. However, if you pair that same lens with a capture device with a ratio of 1.33, what is the result? 2.66 (1.33 x 2.0). This is much closer! You still have to trim the ends but not by much.

The same principle applies to 1.3x lenses. If you pair a 1.3x lens with a ratio of 1.78 you get 2.3 which is close to the target of 2.40.

In this example we also see the primary purposes for the two “flavors”. 2.0x lenses are really designed for capture devices with a ratio of 1.33, while 1.3x lenses are for those with ratio of 1.78 (Resource: http://www.premiumbeat.com/blog/anamorphic-101-how-to-shoot-and-edit-anamorphic-footage/).

Another advantage of anamorphic lenses is that it captures images with higher resolution  How so? Anamorphic format doesn’t add more pixels it merely economizes what’s already there.

Let’s say you have a sensor with a resolution of 1920 x 1080 (which is also a 1.78 aspect ratio). If you where to film using spherical lenses, what would be your end resolution and aspect ratio? 1920 x 1080/1.78, right?!

Now, if you were to crop the top and bottom of the image to achieve a wider ratio of 2.40, what would be the final resolution? 1920 x 817ish (817 x 2.35 = 1920ish).

Let’s try this now with a 1.3x anamorphic lens. What is your captured resolution and aspect ratio before “de-squeezing”?  1920 x 1080/1.78. Now let’s “de-squeeze” the image (1.78 x 1.3). What’s your final ratio? 2.35 (rounded from 2.31).And your final resolution? Still 1920 x 1080.

(Resource: http://www.cinematography.com/index.php?showtopic=68271)

So, as discussed, historically, achieving a wider aspect ratio and retaining as much of the captured image as possible were the primary motivations between choosing spherical or anamorphic. Today, however, the primary consideration would probably be aesthetic rather then technical (Source: http://www.theasc.com/site/blog/thefilmbook/the-elastic-frame/)

 


How do spherical and anamorphic lenses differ visually?

Below are two images. The top image was filmed using spherical lenses. The bottom image was filmed using an anamorphic lens. Both were filmed using the same camera (ARRI Alexa which is capable of recording both in 1.78 and 1.33 formats) and under the same exact conditions, so any visual differences should be directly related to the difference in optical process. Can you see the differences?

TSF_PARIS_2K_2048x858-40mm-Cooke-S4i-640x267TSF_PARIS_2K_2048x858-75mm-Cooke-Anamorphic-640x268

Now that you’ve had a chance to examine the images yourself, let’s discuss the main differences.

Aspect Ratio.  You’ll notice that both images have the same aspect ratio. What you don’t see is that the top image, shot spherically, needed to be cropped to achieve the wider ratio and thus has less vertical resolution.

Focal Length.  Notice the technical notations at the bottom left corner of each image. The spherical image was captured with a 40mm lens while the anamorphic a 70mm lens. As discussed, to achieve similar horizontal FOV as a spherical lens, you have to use an anamorphic lens that is double the focal length, if using 2.0x lenses. What is the impact visually?

Generally speaking, a longer focal length lens has will have shallower DOF which can “isolate” or separate objects from the background.

For me, this is one of the most appealing characteristics of anamorphic format. It really allows you to focus attention onto a specific object without sacrificing the environment because you can still achieve wider shots.

Bokeh. Refers to the way a lens renders out-of-focus points of light. In the images above the spherical lens produces round bokeh while the anamorphic lens produces oblong or oval bokeh. The oval bokeh is a product of the “squeezing” effect.

DOF. To my eye the anamorphic image appears to have less depth of field. The background appears “softer”. As discussed above, this could also be an effect of the different focal lengths.

However, even if we were to compare same focal lengths, there would still be a difference in DOF. Why? Because the anamorphic lens will exhibit the attributes of two focal lengths. Here is a link that explains this even further:

http://www.cookeoptics.com/techdoc/3E4F343E911D932885257CF200688631/61FDTimes2.0-300m-june2014-pg24.pdf)

Dimensionality. If you look at the models face, in the anamorphic image, her face appears more dimensional (interesting that spherical format is sometimes referred to as “flat”). Personally, I think it is very complementary to this subjects face.

Another characteristic of anamorphic lenses, not seen in the examples above, is the unique way they flare (streaks of light produced by stray sources of light). Anamorphic lenses are known to create long horizontal and often bluish flares. This unique flaring has been used (or over-used depending on who you ask) for aesthetic effect, the most recent Star Trek films come to mind.

3529_2-1024x429
Anamorphic lens flaring. Image courtesy of Presynkt.com
ZZ33961791
Star Trek lens flare.

Yet another characteristic, if you can call it that, are known as aberrations or distortions. These are actually optical issues (here is a resource discussing the various types: http://www.digitalcameraworld.com/2012/10/12/lens-distortion-everything-every-photographer-must-know/).

All lenses, both spherical and anamorphic can experience these issues, however, the more complicated design of anamorphic lenses can often make them more susceptible to these. That being said, at times it is these same issues that make anamorphic lenses appealing to some since it can add “character” to the image. An example is a bending or bowing of vertical lines.

Anamorphic lenses that are 40mm and wider are known to exhibit a particular distortion that “bends or bows” horizontal lines. This can be used as a aesthetic element. Director Wes Anderson has employed this effect in a few of his films. Image below is an example. Notice blue wall above actors head and how it “bows” down at edges of frame?

Richie-T
Royal Tenebaums “bending” distortion.

[Update 5/18/2016: If you really want to see the bowling effect, Steven Spieldburg’s “Bridge of Spies”, really employed this (especially in jailhouse meeting room).]

Now, how does a anamorphic 2x lenses differ from a 1.3x lenses? Both lenses will exhibit the characteristics discussed above. However, since 2.0x lenses are a “stronger” application of the process (it distorts light more), I personally feel that the effects are more pronounced and thus have a deeper impact.

Below are two samples, top is a 1.3x lens and the bottom a 2.0x lens If you examine the “bokeh”, the 2.0x is more oval and the 1.3x is more “almond” in shape. Additionally, for me, in 2.0x image, there is less depth of field and more separation from the background.

AnamorphicTest_1
Hawk Anamorphic 1.3x
ana3
Hawk Anamorphic 2.0x

 


What are some considerations when choosing an optical format?

Now that we know the basic difference between spherical and anamorphic formats and the history, the next questions is, “How to choose?”

Story. I feel, and I think most other cinematographers will agree, that the primary drive for any decision should be the story. What decision best supports the story and the way we want to tell it?

Personally, their are two primary considerations when discussing optical format and how it relates to the story – Aesthetics and Perspective.

Aesthetics. Personally I love the look of anamorphic.  I feel its very “painterly” and “organic”. Even when looking at simple, every-day, or even mundane things – it simply “beautifies” the image. It is very “Cinematic”!

These all sound like superlatives –  and they are! Unless, that is exactly what you don’t want!

There are certain stories where you want to elevate the imagery beyond what is realistic. Some stories, however, are best told as honestly and realistically as possible. For those stories, the “honest” way spherical lenses capture the world may be a better option.

It is for this reason some cinematographers abstain from using anamorphic lenses, not because they produce bad images, but because they don’t fit the stories they are trying to tell.

Perspective. In photography there is a concept refered to as compression. The common thought is that the focal length is the determining factor. In truth, compression is more a result of distance from subject (Resource: http://www.digitalcameraworld.com/2014/08/15/debunking-the-myth-that-focal-length-affects-perspective/).

It is this aspect that I refer to. Why? Subconsciously, the viewer can “feel” the distance from subject. They know if they’re close of far. This can be used effectively to effect viewer experience – the closer they are to the action, the more participatory the experience; the farther away, the more observational.

Thus many cinematographers like using wider lenses so that they can get close to the subject, and let the audience “feel” like part of the action, without completely losing sense of environment.

As discussed already, when using anamorphic lenses, you often times use longer focal lengths which will increase the distance from subject which can lead to a more observational experience.

[Update 6/23/16: Quick article discussing how focal length affects viewer experience. http://www.premiumbeat.com/blog/various-focal-lengths-for-images/

Below are two examples (follow link for more examples: https://www.slrlounge.com/lens-compression/). Can you sense the distance of the camera? How does it affect you?

UntitledUntitled2

Cost. Practically speaking, shooting anamorphic will be a more expensive endeavor. For example, looking to buy anamorphic lenses? That’ll be $20-30,000…per lens!

[Update: SLR Magic has just released a 3-lens set of 1.3x lenses. While not  “cheap”, they’re between $6000 – 65000 a lens, they are more affordable that previous options. Here’s a look at this lenses – http://www.eoshd.com/2016/03/slr-magic-anamorphot-cine-pl-set-355070-a-shoot-in-london-and-brighton-with-the-sony-a7s-ii-and-samsung-nx1/.]

[Update 3/28/2016: SLR Magic has also released a 3-lens set of 2.0x lenses. While able to utilize a 1.33 capture ratio these ONLY have an image circle of 21.6mm, suitable for the M4/3 format  – http://www.slrmagic.co.uk/slr-magic-3-piece-2x-anamorphot-cine-lens-set-mft.html.]

[Update 3/13/17: To clarify, 2.0x Anamorphic lenses were orginally intended to be used for Academy 4-Perf Film. This format included a magnetic strip that reduced horizontal width from 24.9mm to 21.95mm, thus changing ratio from 1.33 (for Full Aperture, no magnetic strip) to 1.375. This resulted in a final imaging area of approx. 21.95×18.6mm. In 1993, SMPTE standardized 4-perf format to approx. 21×17.5mm . The GH4’s 4:3 Mode has an approx. area of 17.3×13. This yields a crop factor of approx. 1.2.] 

Thus, most would probably rent. Still, your talking about $500 a day…per lens. Over the course of a shoot, that cost will add up.

A more “economical” approach would be to use anamorphic adapters, which basically adds a cylindrical element to an existing spherical lens (the Letus AnamorphX Adapter is one I’ve personally tried).

The issue with using adapters is that you essentially have two lenses, each with its own focusing mechanism, which can be challenging for certain shots.

For those interested in getting experience working with anamorphic format, adapters may still be the best way, despite the possible issues. EOSHD.com is a good resource for using anamorphic adapters.

Ergonomics. Due to the complicated nature of the construction of anamorphic lenses, there are some limitations that need to be considered:

Speed. Generally speaking anamorphic lenses are typically slower than their spherical counterparts. Which could technically and creatively be limiting. That being said, newer generations of anamorphic lenses, such as ARRI Master Anamorphic Lenses are quite fast – t1.9 maximum aperture.

Minimum Focus (minimum distance a subject has to be from lens to achieve focus). Most anamorphic lenses have a rather long minimum focus distance, feet as opposed to inches. To achieve closer focus distances, a diopter often used (Source: http://www.cambridgeincolour.com/tutorials/macro-extension-tubes-closeup.htm), which adds another variable into the equation.

So as you can see, there really is no “right or wrong” way to decide. Each will need to examine their project and take all factors into account.

However, sometimes it helps to see how someone else would consider matters. Here is a fantastic video of how one cinematographer, John Mathieson ASC (Gladiator, Kingdom of Heaven, The Man From U.N.C.L.E), deliberates between spherical and anamorphic formats.

 


Take-Away Points: 

I know this was a very long blog. So here are a few key take-away points.

  • Spherical lenses are the most common and accessible optical format.
  • Anamorphic format was designed to achieve a wider viewing experience without cropping the image. This is achieved by optically “squeezing” more information onto the capture area which, when “de-squeezed”, achieves the wider ratio.
  • Anamorphic lenses come in two “flavors” which define the squeeze factor – 2.0x and 1.3x. 2.0x anamorphic lenses are primarily designed for capture formats with a ratio of 1.33 while 1.3x lenses are designed for capture formats with a ratio of 1.78.
  • Visually, anamorphic lenses exhibit unique visual characteristics such as shallower depth of field, oval bokeh, lens flaring, and aberrations. These can be very aesthetically pleasing but may not be appropriate for all stories.

Now, if your not too tired, here are some additional resources:

http://www.fdtimes.com/pdfs/articles/anamorphic/FDTimes-Anamorphic-Special-May2015.pdf

http://nofilmschool.com/2013/10/get-started-shooting-anamorphic

Well, hope you enjoyed this article. Please leave comments below if you have any questions or recommendations!

So, you’ve now chosen or at least understand the difference between spherical and anamorphic formats. What’s next?

Well, after passing through the lens the next element the image will encounter is the capture medium – the actual device that will record your image. Stay tuned for second part of this series:  Part 2: Capture Medium – Film vs Digital!

Post No.04

 

Advertisements

12 thoughts on “Part 1: Optical Process – Spherical vs. Anamorphic

  1. Great informative and educational article clearly explained. I’m researching for a WWII short film I’ll be shooting this year. I’m wondering what your thoughts are on what if any anamorphic lenses were used for Sergio Leone’s or Hitchcock’s work. Thanks so much for your time.

    Like

    1. @robertcota For the time period, I would assume that all anamorphic lenses were 2x squeeze. As for exact lenses used, not sure, but probably Panavision. Quick research shows that Sergio Leone also used 2-perf spherical to achieve a wider frame periodically.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s