RAW Myths & Facts
As the endless debate of RAW versus JPEG continues, more and more people appear to be confused as to why these two formats exist, what each is good for and when to use one or the other. First time DSLR users often wonder which format to choose and make a choice based on some random fact they heard. It does not help that a large number of incorrect information exists on the internet about the RAW and JPEG images. This feature dispels common myths encountered around the internet and in digital photography books.
Get it right in the first place is the mantra of those who advocate JPEG and Shooting JPEG is like throwing away the negative is the motto of RAW advocates. As usual, both sides are right and it is therefore very important to understand the facts before choosing sides.
RAW Myths & Facts
Myth: RAW IS AN IMAGE FILE FORMAT
False. RAW files contain data which includes some information for each pixel but not enough to represent a digital image. Even in cases that actually store full color information for each pixel, such as in cameras which use a Foveon sensor, the stored information is not in the same color-space or scale as normal computer images.
Myth: RAW IS BETTER THAN JPEG
False. RAW has advantages over JPEG but so does JPEG over RAW. There would not be so much debates on the subject if it was not! The same JPEG as the camera would produce can be obtained from a RAW file using the camera manufacturer's RAW development software or in-camera RAW conversion on select models. This means that a RAW can at least produce an image of equal image quality.
Myth: IMAGES PRODUCED FROM RAW HAVE MORE DYNAMIC RANGE
False. Once a RAW file has been converted into an image, either explicitly by someone or implicitly by an image browser, it is limited by current display technology which cannot display more dynamic range or colors than a JPEG can hold. Nearly every display in existence uses 8 bits per channel (24 bits per pixel) which is what JPEG files store. All common operating systems are also incapable of displaying more than 8-bits per channel, even for the two monitors that can support more.
Myth: RAW FILES ARE DULL AND LOW CONTRAST
False. With all the hype about RAW quality, a huge number of people's first observation about RAW is that they look bad when compared to JPEG images. They only do by accident. As mentioned earlier, a RAW file contains data not an image. Therefore when a preview is for a RAW file is shown, software must produce an image from the RAW data. If it does that poorly, which is commonly the default for many image browsers, what is shown is a poor quality image. Usually the image looks dull and has low local contrast because it shows the entire tone curve uniformly.
Myth: RAW LOOKS EXACTLY LIKE JPEG
False. Several RAW files actually include an embedded JPEG. This embedded data allows fast previews and obviously looks exactly like a JPEG. Additionally, using the default conversion from camera maker's software normally does the exact same RAW to JPEG conversion as the camera. So there are cases where the RAW is converted into the exact same JPEG. However, the benefit of RAW lies in its ability to produce different results. Shooting RAW and then using the default conversion defeats its purpose. Benefits of shooting RAW are only seen when there is time to apply different processing.
Myth: JPEG IMAGES ARE SOFT AND LACK DETAILS
False. All DSLR cameras have settings that influence sharpness and noise reduction. Manufacturers seem to believe people prefer soft images over noisy images because the defaults are often high noise-reduction and middle sharpness. Those are only defaults and most digital cameras can produce images with finely tuned characteristics. Take the time to set up your camera to produce the most pleasing images possible.
Myth: JPEG FILES HAVE HORRIBLE ARTIFACTS
False. JPEG uses something called lossy compression which means some information is removed to make the file smaller. In practice the highest quality setting affects less than 2% of pixels by less than 0.4%. The data removed is also chosen to be the least perceptible to the human visual system.
Myth: EDITING JPEG DEGRADES IMAGE QUALITY
True. Anytime a JPEG is modified and then saved, compression may reduce image quality by a tiny amount. If done repeatedly it can add up to a substantial loss of image quality. The trick used by film artists to work with lossy formats is to do all the work in a lossless format and then save back the final version in a lossy format once all the work is done. Therefore it is better to do repetitive edits in PNG, TIFF or PSD and then save a final JPEG for printing or online use.
Myth: SHOOTING JPEG WITH A DSLR IS LIKE USING A POINT & SHOOT
False. DSLR cameras have a huge number of advantages over point & shoot digital cameras, including higher image quality and speed. Even DSLRs tend to shoot longer continuously in JPEG than in RAW.
Myth: REAL PHOTOGRAPHERS SHOOT RAW
False. The professional industry is just as divided on the subject as everyone else. They are pros that exclusively shoot JPEG, ones that exclusively shoot RAW, ones that always shoot JPEG+RAW and ones that use each format part of the time.
Myth: DNG IS A NON PROPRIETARY RAW FORMAT
False. DNG belongs to Adobe. Adobe grants a license to use its patents to those who wish to use the DNG file format. While it is not owned by any camera manufacturer, it remains a file format that belongs to profit oriented corporation.
One of the reasons for DNG is to have a standard format that can be processed by future software long after the camera maker stops making a particular camera or goes out of business. While with is true in principal, the interpretation of the sensor data is specific to each camera and therefore that knowledge must be built in somewhere and kept up to date with new generations of software for legacy support. The one hope for longevity is to process RAW to a commonly understood DNG interpretation. Adobe covers part of what is needed for that in its 2009 DNG update which supports embedded profiles.
Myth: DNG CAN STORE RAW DATA FROM ANY DIGITAL CAMERA
True today. The DNG format was updated in 2009 to make sure it can hold the data produced by any current digital camera. It does contain special information to treat the information differently if it comes from a Fuji SuperCCD since the arrangement of pixels is different than conventional images. If other new technologies appear that does not fit the current DNG mold, Adobe has to change the DNG format to support those as well. Therefore more than one type of DNG files will exist and software will have to cope with the difference, just like they have to cope with different RAW formats.
Neocamera Blog is a medium for expressing ideas related to digital cameras and photography. Read about digital cameras in the context of technology, media, art and the world. Latest posts links:
Nikon D810 Review
Professional DSLR with anti-alias-filter-free 36 MP CMOS sensor. Ultra-low ISO 32 to 51200. 5 FPS and 1080p @ 60 FPS. Large 0.7X magnification 100% coverage OVF. All new processing-pipeline and Highlight-Weighed metering.
Fuji X-T1 Photographer Experience
Photographer Experience report on using the Fuji X-T1 along with the Fujinon XF18-135mm F/3.5-5.6R LM OIS WR and Fujinon XF10-24mm F/4R OIS lenses.
Olympus Stylus 1 Review
Premium compact with bright F/2.8 constant aperture stabilized 10.7X wide-angle optical zoom lens. Full manual-controls with dual control-dials, plus a huge 1.15X EVF with 1.4 MP and an Eye-Start sensor. 3-Stop ND-Filter and WiFi built-in.
Canon Rebel SL1 Review
The smallest DSLR yet packs a 18 megapixels APS-C CMOS sensor with hybrid Phase-Detect and Contrast-Detect AF. Captures images at 4 FPS and 1080p HD video.
Lenovo Thinkpad X1 Carbon 2014 Review
The lightest 14" ultra-book features a high-resolution 2560x1440 QHD non-glare display in a carbon-fiber body with illuminated and spill-proof keyboard. WiFi, WiDi, 4G and Gigabit Ethernet all in one sleek design.
Nikon D4s Review
All-new Nikon flagship professional DSLR with a 16 MP sensor capable for ISO 50-409,600, 11 FPS continuous drive for 200 JPEG or 78 RAW, full 1080p HD @ 60 FPS with clean HDMI out, Time-Lapse Video, Interval Timer. Built-in HTTP and FTP servers, plus Gigabit Ethernet and more.
Nikon D3300 Review
The newest entry-level Nikon DSLR features a 24 MP APS-C CMOS sensor without Anti-Alias filter. 5 FPS Drive, full 1080p HD and 11-point Phase-Detect AF in a simple and compact body.
Olympus OM-D E-M1 Review
16 MP Micro Four-Thirds mirrorless without anti-alias filter. Built-in 5-Axis stabilization and 37-point Phase-Detect AF. 10 FPS drive plus full 1080p HD. Freezeproof body with dual control-dials, a 2.4 MP EVF and 3" tilting touchscreen LCD.
Exclusive Fuji Finepix S1 Review
Weather-proof ultra-zoom with 50X optical zoom stabilized along 5 axis. 16 megapixels sensor delivers 10 FPS drive and full 1080p @ 60 FPS video. 3" rotating 920K pixels LCD and 0.2" 920K EVF plus plenty of controls.
Panasonic Lumix DMC-LF1 Review
World-smallest camera with built-in EVF. Full and direct photographic controls including dual control-dial in a compact body. Packs a 12 MP high-speed CMOS sensor capable of 10 FPS drive and a bright F/2 wide-angle 7X stabilized optical zoom lens.