Report by Donald Becker of dbCreativeImages.com (OMP Member #155)
JPEG vs RAW: Which is best for you?
Yes, that is a huge digital question, debated amongst photographers of all skill levels. At many of the workshops and photography classes for serious photographers, the instructors will tell the students: “Shoot RAW! You get better quality images!” Students I teach ask me what I do, and then when I tell them, they say “but all of their other instructors tell them to shoot RAW!”
What I do is a balance of JPG and RAW, depending on the situation. I shoot about 80% of my images as high level JPGs. I only shoot RAW images when I believe that I will need to make significant, global modifications to the captured image, or when the image has special meaning for me. Examples of those types of changes would be: major exposure change; major color balance or white balance modification; large contrast change; etc. Examples of shooting situations in which these kinds of corrections are likely to be needed include: outdoor portraiture, especially when light conditions are changing while shooting; shooting in direct sunlight; high contrast situations, like a dark skinned bride in a white wedding dress; shooting under less than optimum lighting; etc. Under studio conditions, where the photographer has absolute control over the light, such global modifications should rarely be needed.
So, to justify to myself that what I am doing is actually giving me the best images under the conditions I am shooting, as well as providing substantial benefits to my workflow and storage procedures, I ran a controlled test on the question: Which is best (better?): JPG or RAW? What this actually means is: am I losing any important sharpness, resolution, and/or tonal information and gradation if I shoot JPG under my normal conditions?
One additional note, to describe my workflow: once the image is opened in Photoshop™ (PS), for both RAW and JPG captures any further SAVE is as a PS document (.psd), and all subsequent manipulations are completed and saved in PS. The final images for this article were then saved as jpgs.
To evaluate the question I took a series of exposures with my camera set on JPG + RAW, of a high contrast resolution chart, and of OMP model Hootie (OMP Model #313203) taken both in open shade (should not need RAW capture) and in direct sunlight (probably needs RAW capture for best quality). I used my primary shooting camera, a Nikon full frame D3 with a Nikon 28-200 f3.5-5.6 G zoom lens, with no in-camera image modifications other than sharpness set at “Normal”. The results of these tests are as follows.
Figure 1 is the entire area captured for the high contrast resolution charts, showing the small area selected in black, which was then enlarged to 400% for the RAW and JPG sections. (Note: the exposure for the resolution test was 1/125 second at f 8 and an ISO of 200 using a studio flash system.) Figure 1a is the enlarged section for the RAW capture, and Figure 1b is the enlarged section for the JPG capture. No post-capture sharpening or modifications of any kind were applied to either enlargement, except for small corrections with the Photoshop™ levels tool to give both versions approximately the same image tone.
It is clear from these images that there is little if any discernable differences in the sharpness and resolution of the high contrast chart for these two enlarged portions. The RAW image has some apparent color fringing in the black lines of the chart, which reduced the black image density, and thus required some darkening via PS levels to make it approximately equivalent to the JPG version (shown without modification). If anything, the JPG version seems a little sharper and clearer, possibly due to the “Normal” in-camera sharpening for the JPG version, and/or to the blacker lines of the JPG version due to the absence of color fringing.
The same general procedure was applied to an image capture of a model in open shade, actually taken just inside a building right next to a large bank of windows viewing the open sky. Figure 2 is the entire area captured, with the enlarged area encircled.
Figure 2a is the enlarged section for the RAW capture, and Figure 2b is the enlarged area for the JPG capture. Both enlarged images were lightened and had minor corrections which I would normally do to make a final image, but no smoothing or sharpening changes were made to either image. These two enlargements are virtually identical, and I see no reason to prefer one over the other. (Note: exposure information for the shade image was 1/250 second at f 8 with an ISO of 800 and 120mm focal length.)
The difficult test is shown in Figure 3, where the image capture of Hootie is in direct sunlight. Substantial modifications were required to lighten the dark shadows on the shadow side of her face in the image, but again, no smoothing or sharpening was applied to either enlargement. In this case of needing substantial global modifications, the image quality improvement in the RAW image over the JPG capture is apparent.
The tonal qualities of the RAW image are superior, and the higher image quality especially of the deep shadow area is easily visible. Remember, these image enlargements as shown are the equivalent of a final 12 inch x 18 inch print at 240 dpi. At smaller sizes (such as 8”x10”) the JPG capture would look significantly better, but likely still not as good as the RAW image. (Note: exposure information for the sun image was 1/1000 second at f 16 with an ISO of 800 and 135mm focal length.)
So there you have it, MY testing to determine what works best for ME. I feel that these results support my decision on how I usually handle my images… JPG only capture for “normal” lighting and conditions (e.g., hundreds to perhaps a thousand images in a studio shoot), JPG + RAW for “difficult” or special lighting and conditions. Keep in mind that there are many different, proprietary RAW formats out there, and that different RAW formats provide differing results, including variable results when used with different RAW converters – i.e., equipment manufacturers’ converter; Photoshop™ converter; Phase One converter; etc. (Note: I used the PS converter.) The bottom line is that each photographer should test their own system and decide what procedures will provide them with the quality they want in their photographs.
As a parting shot, in Figure 4, I have included a final image, my interpretation of the shade image of Hootie, a beautiful young lady (Note: made from JPG capture).