Digital versus film photography

From Wikipedia, the free encyclopedia - View original article

 
Jump to: navigation, search

Digital versus film photography has been a topic of debate in the photography world, as well as the film industry since the availability of digital cameras towards the end of the 20th Century. Both digital still photography as well as Digital cinematography versus film and motion picture film photography have advantages and drawbacks.[1][2] 21st century photography is now dominated by digital operation, but the older photochemical methods continue to serve many users and applications.

Image quality[edit]

Spatial resolution[edit]

The quality of digital photographs can be measured in several ways. Pixel count is presumed to correlate with spatial resolution.[3] The quantity of picture elements (pixels) in the image sensor is usually counted in millions and called "megapixels" and often used as a figure of merit. Digital cameras have a variable relationship between final output image resolution and sensor megapixel count.[4] Other factors are important in digital camera resolution, such as the number of pixels used to resolve the image, the effect of the Bayer pattern or other sensor filters on the digital sensor and the image processing algorithm used to interpolate sensor pixels to image pixels. Digital sensors are generally arranged in a rectangular grid pattern, making images susceptible to moire pattern artifacts, whereas film is not affected by this because of the random orientation of its grains.[5]

The resolution of film images depends upon the area of film used to record the image (35 mm, Medium format or Large format) and the speed. Estimates of a photograph's resolution taken with a 35 mm film camera vary. More information may be recorded if a fine-grain film, combined with a specially formulated developer, are used. Conversely, use of poor quality optics or coarse-grained film yield lower image resolution. A 36 mm x 24 mm frame of ISO 100-speed film was initially estimated to contain the equivalent of 20 million pixels,[6] although this estimate was later revised to between 4 and 16 million pixels depending on the type of film used.[7]

Many professional-quality film cameras use medium format or large format films. Because of the size of the imaging area, these can record higher resolution images than current top-of-the-range digital cameras. A medium format film image can record an equivalent of approximately 50 megapixels, while large format films can record around 200 megapixels (4 × 5 inch) which equates to around 800 megapixels on the largest common film format, 8 × 10 inches, without accounting for lens sharpness.[8] Medium format digital provides from 39 to 80 megapixels.

Thus film and digital work each provide a wide range of performance in this regard, overlapping but with film tending to higher resolution. Resolution of both film and digital are subject to the quality of lens fitted to the camera. The medium which will be used for display, and the viewing distance, should be taken into account. For instance, if a photograph will only be viewed on an old analogue television that can resolve approximately 0.3 megapixel or modern HDTV set of 1080p with 2 megapixels, the resolution provided by high-end camera phones may suffice, and inexpensive compact cameras usually will. Similar or more expensive hardware may also fill the screens of computer displays, though those few that show tens of megapixels is currently out of reach of low-end film photography and all but specialized scientific or industrial digital cameras.

Noise and grain[edit]

Thermal noise, produced by heat and manufacturing defects, degrades shadow areas of electronic images with random pixels of the incorrect color. Film grain becomes obvious in areas of even and delicate tone. Grain and film sensitivity are linked, with more sensitive films having more obvious grain. Likewise, with digital cameras, images taken at higher sensitivity settings show more image noise than those taken at lower sensitivities.[6]

However, even if both techniques have inherent noise, it is widely appreciated that for color, digital photography has much less noise/grain than film at equivalent sensitivity, leading to an edge in image quality.[9] For black-and-white photography, grain takes a more positive role in image quality, and such comparisons are less valid.

Noise is a particularly critical issue with digital cameras, often producing color distortion or confetti, occurring most severely on the blue component and least severely on the red component. Nearly all digital cameras apply noise reduction to long exposure photographs to counteract thermal noise. For very long exposures, the image sensor must be operated at low temperatures to prevent noise affecting the final image. Film grain is not affected by exposure time, although the apparent speed of the film changes with lengthy exposures, a phenomenon known as reciprocity failure.

Auto Focus And Auto Exposure Systems[edit]

In regards to compact cameras, digital compact cameras have the advantage of using phase detection or contrast detection on the actual image captured. This is significantly more advantageous than compact film cameras that use active infrared auto focus systems which then estimate the lens position. Similarly exposure levels can be detected directly from the image sensor rather than through another light meter used in film cameras. On the budget end of the scale, digital compact cameras tend to give better image quality than their budget film counterparts due to these more superior focusing and exposure mechanisms.

Dynamic range[edit]

Dynamic range is a complex issue.[10] Comparisons between film and digital media should consider:

Dynamic range is of considerable importance to image quality in both the digital and emulsion domain. Both film and digital sensors exhibit non-linear responses to the amount of light, and at the edges of the dynamic range, close to underexposure and overexposure the media will exhibit particularly non-linear responses. The non-linear dynamic response or saturation qualities of emulsion film are often considered a desirable effect by photographers, and the distortion of colour, contrast and brightness varies considerably between film stocks. There is no limit to the number of possible levels of colour on emulsion film, whereas a digital sensor stores integer numbers, producing a limited and specific possible number of colours. Banding may be visible in the unusual case that it is not obscured by noise, and detail may be lost, particularly in shadow and highlight areas.

According to Eastman Kodak in 2007, digital sensors of the time lacked the extended dynamic range of film. In particular, they tend to 'blow out' highlights, losing detail in very bright parts of the image.[11] If highlight detail is lost, it is nearly impossible to recapture in post-production.[12] Therefore, film can be underexposed and overexposed, retaining detail and information in the camera negative.[11]

Some amateur authors have performed tests with inconclusive results. R. N. Clark, comparing a professional digital camera with scans of 35 mm film made using a consumer level scanner, concluded that "Digital cameras, like the Canon 1D Mark II, show a huge dynamic range compared to [scans of] either print or slide film, at least for the films compared."[13]

Carson Wilson informally compared Kodak Gold 200 film with a Nikon D60 digital camera and concluded that "In this test a high-end consumer digicam fell short of normal consumer color print film in the area of dynamic range."[14]

The digital camera industry is attempting to address the problem of dynamic range. Some cameras have an automatic exposure bracketing mode, to be used in conjunction with high dynamic range imaging software. Some CCDs including Fujifilm's Super CCD combine photosites of different sizes to give increased dynamic range. Other manufacturers use in-camera software to prevent highlight overexposure. Nikon calls this feature D-Lighting.

Presentation technology is also relevant, as different color printing methods, cathode-ray tubes, LCDs and other displays all have different dynamic range limits and degrees of linearity.

Effects of sensor size[edit]

Drawing showing the relative sizes of sensors used in most current digital cameras.

Almost all compact digital cameras, and most digital SLRs or ILCs, have sensors smaller than the 36 mm x 24 mm exposure-frame of "35 mm" film. The smaller sensors found in DSLR cameras affect:[15]

  1. Depth of field;
  2. Light sensitivity and pixel noise;
  3. Relative cropping of the field of view when using lenses designed for 35 mm camera;
  4. Optimizing lens design for smaller sensor area;
  5. Increased relative enlargement of the captured image.

Depth of field is often quoted as being greater for digital cameras than for film cameras. The maxim packages several counterintuitive aspects of photography into a single (largely correct) theorem. Depth of field, for a given lens focal length, at a given f-number will scale with sensor (film/chip) size. In effect, a smaller sensor will increase the apparent depth of field because it magnifies the portion of the image that is in focus.

Manufacturers are increasingly using (especially in the budget digital camera market) "35 millimeter equivalent" focal lengths for lenses. This gives rise to the "depth of field is greater for digital cameras" myth: the shorter the focal length of a lens, the greater is its depth of field (at fixed F-stop). Therefore, if a sensor that is one-fourth the width and height of a 24 x 36 mm frame of film is exposed to an image through a lens that is correspondingly one-fourth the focal length, the depth of field increases 16x (scaling per the square of focal length) on an absolute scale, but 4x from a comparison-of-images perspective (the imaging dimension is 4x smaller).

This increase in relative depth-of-field may have advantages for taking snapshots; more image will be in focus than with a larger sensor and autofocus system accuracy is less critical for producing an acceptable image. Contrarily, photographers wishing to decrease depth of field to create certain effects, such as isolating subjects from their background need to increase aperature for sensors smaller than 36 mm x 24 mm to achieve the same degree of selective focusing.[16] Depth of field can be minimized by use of large format cameras, which are very rarely digital.

Light sensitivity and pixel noise are both related to pixel size, which is in turn related to sensor size and resolution. As the resolution of sensors of a specific format increases, the size of the individual pixels naturally has to decrease. This smaller pixel size means that each pixel collects less light and the resulting signal must be amplified more to produce the final value. Noise is also amplified and the signal-to-noise ratio decreases, and the higher noise floor means that less useful information is extracted from the darker parts of the image.[15] Countering these effects of digital-signal noise are advances being made in sensor technology itself. As of 2012, the top-end of digital sensor sensitivity is at ISO 204,800 (in both Canon and Nikon DSLRs), whereas less expensive prosumer DSLR and ILC cameras offer sensitivities up to ISO 6400 or even higher, often with good noise performance at one-quarter maximum sensitivity. In recent years larger sensor digital compacts have become available. However, they still are bigger and heavier than the smallest 35mm cameras and are not full frame.

Some digital SLRs use lens mounts originally designed for film cameras. If the camera has a smaller imaging area than the lens' intended film frame, its field of view is cropped. This crop factor is often called a "focal length multiplier" because the effect can be calculated by multiplying the focal length of the lens. For lenses that are not designed for a smaller imaging area whilst using the 35 mm-compatible lens mount, this has the beneficial side effect of only using the centre part of the lens, where the image quality is in some aspects higher.[citation needed] Only expensive digital SLRs and very rarely expensive 'compacts' have 36mm × 24 mm sensors, eliminating depth of field and crop factor problems when compared to 35 mm film cameras.[citation needed]

In compact digital cameras, the size of the sensor is often several times smaller than the standard 36 mm x 24 mm film, with the area being typically 20 to 40 times less than that of a frame of film.[17] This difference gives film compacts a substantial advantage when it comes to image quality and the ability to take pleasing portraits. In the standard consumer market film's advantage over digital in the compact market is often negated by operator error, the generally poor quality of the cameras or because of poor quality processing of films. The smaller sensor size of digital compact cameras means that prints are extreme enlargements of the focused image, and that the lens must perform well in order to provide enough resolution to match the tiny pixels on the sensor.

To manufacturers, large lenses are very costly to produce, smaller sensors in digital cameras enable the use of smaller and more compact arrangement of lenses. Affordable superzooms cameras that can magnify images 50-60 times are now available. These kinds of magnification are virtually impossible to achieve in 35mm film cameras. Compact cameras such as the Lumix LX-7 with a maximum aperture of f/1.4 is achievable with smaller sensors.

Convenience and flexibility[edit]

Flexibility and convenience are among the reasons for the widespread adoption of digital cameras. With film cameras, a roll is usually completely exposed before being processed. When the film is returned it is possible to see the photograph, but most digital cameras incorporate a liquid crystal display that allows the image to be viewed immediately after capture. The photographer may delete undesired or unnecessary photographs, or reshoot the image if required. A user who wants prints can quickly and easily print just the required photographs.

Photographic film is made with specific characteristics of Color temperature and sensitivity (ISO). Lighting conditions often require characteristics different from those of the film specifications, requiring the use of filters or corrections in processing. Digital photography allows color temperature and sensitivity to be adjusted at each shot, either manually or automatically.

Digital images may be conveniently stored on a personal computer or in off-line storage such as small memory cards. Professional-grade digital cameras can store pictures in a raw image format, which stores the output from the sensor rather than processing it immediately to form an image. When edited in suitable software, such as Adobe Photoshop or the GNU program GIMP (which uses dcraw to read raw files), the user may manipulate certain parameters, such as contrast, sharpness or color balance before producing an image. JPEG images can be similarly manipulated, though usually less precisely; software for this purpose may be provided with consumer-grade cameras. Digital photography allows the quick collection of a large quantity of archival documents, bringing convenience, lower cost and increased flexibility in using the documents.[18]

Modern film cameras are not as power thirsty as modern digital cameras, and can last longer on smaller batteries. Some film cameras, especially older ones, can operate without batteries: some will function completely without batteries while others may lose some functionality such as metering and some shutter speeds. Batteries that only have to power light meters are often very small and can last a long time. This can be a boon for those who may be spending a long time with little or no access to power. Film cameras may also be carried as backups for this reason.

For large format and ultra large format photography, film may have some advantages over digital cameras, such as price and flexibility, when used outside the studio environment. Digital rotating line cameras provide similarly high performance, but scan mechanically rather than use a single sensor. Thus they cannot scan anything that moves, and are expensive, large, and rarely moved.

Film speed[edit]

Digital cameras are capable of much higher speeds (sensitivities) than film, can perform more desirably in low light situations at night or indoors, and are more useful for ultra-fast photography. In addition, on digital cameras the speed can be adjusted at any time, while a film camera requires changing the film to change the film speed.

Cleanliness[edit]

Dust on the image plane is a constant issue for photographers, and especially so in digital photography. DSLR cameras are especially prone to dust problems because the sensor remains in place, whereas a film advances through the camera for each exposure. Debris in the camera, such as dust or sand, may scratch the film; a single grain of sand can damage a whole roll of film. As film cameras age, they can develop burs in their rollers. With a digital SLR, dust is difficult to avoid but is easy to rectify using a computer with image-editing software. Some digital SLRs have systems that remove dust from the sensor by vibrating or knocking it, sometimes in conjunction with software that remembers where dust is located and removes dust-affected pixels from images.[19]

Compact digital cameras are fitted with fixed lenses, which makes it harder for dust to get into the image area. Similar film cameras are often only light-tight and not environmentally sealed. Some modern DSLRs, like the Olympus E-3, incorporate extensive dust and weather seals to avoid this problem.

Integrity[edit]

Film produces a first generation image, which contains only the information admitted through the aperture of the camera. Trick photography is more difficult with film; in law enforcement and where the authenticity of an image is important, like passport or visa photographs, film provides greater security over most digital cameras, as digital files may have been modified using a computer. However, some digital cameras can produce authenticated images. If someone modifies an authenticated image, it can be determined with special software.[20] SanDisk claims to have developed a write-once memory stick for cameras, and that the images once written cannot be altered.[21]

Nikon film scanner, right, which images 35 mm film for digital input.

From an artistically conservative standpoint, some practitioners believe that the use of film offers a more authentic mode of expression than with easily enhanced digital images. As with the earlier transition from oil painting to photography, or from photographic plates to film photography, older methods are more expensive, thus encourage more selectivity and additional consideration.[22]

Cost[edit]

Film and digital imaging systems have different cost emphases. Digital cameras are significantly more expensive to purchase than film equivalents. Prices are however dropping rapidly due to intense competition. Film cameras, on the other hand, are quite inexpensive to purchase, especially used equipment. But film and development costs are ongoing.[23] However, in the digital realm, it could be argued that the constant state of technological change will cause a digital user to keep upgrading and buying other equipment once their digital camera becomes quickly obsolete.[24] Other costs of digital photography include specialist batteries, memory cards and long-term data storage. The emergence of very high quality phone cameras since the early 2010s are making lower end, small sensor digital cameras redundant, almost as quickly as they grew in the last decade. Manufacturers are focusing attention to premium models such as compact system cameras and large sensor compacts. Mobile phones such as the iPhone 5S, Samsung Galaxy S4 and the Nokia Lumia 1020 are capable of images that can rival or beat cheaper dedicated cameras. Inkjet printers can make low-quality prints cheaply and easily from digital files, but high-quality printing has high costs regardless of image source.

Film industry[edit]

There are film industry specific arguments in the film vs. digital debate.

Most digital cinema is displayed in 2K resolution, which is only a small amount more resolution than the consumer-oriented 1080p HD format.[25]

The decline of the use of 35mm prints directly contributed to the bankruptcy of motion picture film manufacturer Eastman Kodak Company.[26]

Roger Ebert publicly criticized the use of DCPs after a cancelled film festival screening of Brian DePalma's film Passion (2012 film) at New York Film Festival caused by a lockup due to the coding system.[27]

High profile film directors such as Christopher Nolan,[28] Paul Thomas Anderson[29] and Quentin Tarantino have all publicly criticized digital cinema and digital cinematography, and advocated the use of film and film prints. Most famously, Tarantino has suggested he may retire because (although he can still shoot on film) he cannot project on 35mm prints in most American cinemas, because of the rapid conversion to digital.[30] Paul Thomas Anderson recently was able to create the most 70mm film prints in years for his film The Master (2012 film). There also are many film directors such as Peter Jackson, Guillermo del Toro, George Lucas, and James Cameron who are adamant supporters of digital cinema and the potential for higher frame rates that it brings.

See also[edit]

References[edit]

  1. ^ Mark Galer and Les Horvat (2005). Digital Imaging. Elsevier. ISBN 0-240-51971-X. 
  2. ^ Glenn Rand, David Litschel, Robert Davis (2005). Digital Photographic Capture. Elsevier. ISBN 0-240-80632-8. 
  3. ^ Marvin J. Rosen and David L. Devries (2002). Photography & Digital Imaging. Kendall Hunt. ISBN 0-7575-1159-7. 
  4. ^ Jurij F. Tasič, Mohamed Najim and Michael Ansorge (2003). Intelligent Integrated Media Communication Techniques. Springer. ISBN 1-4020-7552-9. 
  5. ^ Issac Amadror (2009). "3". The Theory of the Moiré Phenomenon. Springer London. ISBN 978-1-84882-180-4. 
  6. ^ a b Langford, Michael (2000). Basic Photography (7th Ed.). Oxford: Focal Press. ISBN 0-240-51592-7. 
  7. ^ Digital vs. film, executive summary R. N. Clark, may 2002 - march 2008. Retrieved May 2010.
  8. ^ Resolution Test Area 2: trees and Mountains R. N. Clark, 8 April 2001. Retrieved 2 September 2006.
  9. ^ Shannon information theory, noise and perceived image quality Norman Koren, 2000/2010, retrieved May 2010.
  10. ^ Dante Stella. "More is Less is More: The Devil is in the Dynamic Range". DanteStella.com online article, retrieved Jan 2013. 
  11. ^ a b http://www.motion.kodak.com/motion/uploadedFiles/US_plugins_acrobat_en_motion_education_film_info.pdf
  12. ^ http://www.roguepaddler.com/camcope.htm
  13. ^ R. N. Clark. "Dynamic Range and Transfer Functions of Digital Images and Comparison to Film". ClarkVision.com online article, July 3, 2005. [self-published source]
  14. ^ Carson Wilson. "Real World Test: Kodak Gold 200 vs Nikon D60 Dynamic Range". apples.carsonwilson.com online article, September 13, 2008. 
  15. ^ a b Bob Atkins. "Size Matters". Photo.Net Equipment Article, 2003. 
  16. ^ Bob Atkins. "Digital Depth of Field". BobAtikins.com.  []
  17. ^ Vincent Bockaert. "Sensor Size". 
  18. ^ http://www.physorg.com/news139751840.html,Accelerated research using a digital camera
  19. ^ "How does 'Image Dust Off' work?". Nikon. 1 Sep 2003. Retrieved 2010-09-04. 
  20. ^ Nikon image authentication
  21. ^ http://www.engadget.com/2008/07/15/sandisk-introduces-write-once-worm-sd-cards/
  22. ^ Carbone, Kia M. 2009. "Making Contact: The Photographer's Interface with the World." Student Pulse. http://studentpulse.com/articles/57/making-contact-the-photographers-interface-with-the-world
  23. ^ Virtual Photography Studio
  24. ^ http://www.wired.com/gadgetlab/2010/01/five-reasons-you-should-ditch-your-dslr/
  25. ^ http://www.reduser.net/forum/archive/index.php/t-2398.html
  26. ^ http://www.bleedingcool.com/2012/01/19/as-kodak-file-for-bankruptcy-future-of-cinema-unveiled/
  27. ^ http://www.technologytell.com/entertainment/6748/digital-projection-mishap-derails-premiere-of-de-palmas-passion-at-new-york-film-festival/
  28. ^ http://www.joblo.com/movie-news/christopher-nolan-discusses-why-he-prefers-film-to-digital-his-approach-to-cgi-and-much-more
  29. ^ http://www.youtube.com/watch?v=7npmu9Ge_OY
  30. ^ http://www.digitalspy.co.uk/movies/news/a441960/quentin-tarantino-i-cant-stand-digital-filmmaking-its-tv-in-public.html

External links[edit]