You are hereBlogs / Phil-Phlash's blog / "Expose to the Right"
"Expose to the Right"
This is a followup to last month's program on shooting your photos in RAW format, NOT .jpg. OK, shoot both if you like. As mentioned in the program, RAW images are captured and saved as 16-bit data files. (The actual picture data is contained in 12 or 14 of the 16 bits available, depending the camera.) The 12-bit images can have 4,096 tones of Red, Green and Blue which translates to over 16 million colors. The 14-bit images can have 16,384 tones of Red, Green and Blue, a lot more for only two additional bits. For this discussion, let's assume 12-bit data.
Digital cameras "see" a dynamic range of about 5 stops of light where each stop is twice as bright as the one preceeding it from pure black to pure white. So, one might logically assume each of these 5 stops contains 819 (4,096 divided by 5) tones of picture data recorded by the sensor. However, this is totally illogical so if you assumed this, you would be wrong. CCD and CMOS sensors are linear devices so each stop of dynamic range, beginning with the highlights, records twice the tonal range data of the succeeding stop. I bet that's clear as mud so let me explain.
Beginning at the white end of the scale, the first stop contains 2,048 of the total 4,096 tones, leaving 2,048 for the remaining 4 stops. The second stop contains half the remaining 2,048 tones or 1,024 tones. That leaves 1,024 tones for the next 3 stops. The third stop contains 512 tones, the fourth stop contains 256 tones and the fifth stop contains only 128 tones. The following table summarizes these data.
| Highlights | 2,048 Tones |
| Second Stop | 1,024 Tones |
| Third Stop | 512 Tones |
| Fourth Stop | 256 Tones |
| Shadows | 128 Tones |
For 8-bit .jpg data files, the picture (literally) gets much worse since there are only 256 tones possible. The first stop of highlights gets 128 tones, the second gets 64 tones, the third (middle) stop gets 32 tones, the fourth gets 16 tones and the fifth stop of shadows only gets 8 tones. While 256 tones may seem like a lot, when you see how they are distributed it's clear why there is so little data available for shadow detail even for a "perfectly" exposed .jpg image.
The message is clear! When you (or the camera) create a .jpg image file, you are discarding a lot of sensor data that cannot be retrieved. This considerable data loss severely degrades the quality of the image and all subsequent editing adjustments. Always shoot in RAW format and convert to .jpg format afterward (if you must).
So, the first stop of highlights gets the lion's share of tonal data while the fifth stop on the shadow end of the scale gets the least amount of tonal data. What practical significance does this have for photographers? A LOT!!!
When you are in the field taking pictures, you should record the greatest amount of tonal data you can for each and every image. This means shooting RAW format if your camera can do so. Since most of the data acquired by the image sensor is in the brightest 3 stops of dynamic range, always adjust your exposures so the camera's histogram is as far right as possible, even if a few blinkies show on the display. That is, "expose to the right" on every photo regardless of subject brightness or darkness.
When you read the data files into Adobe Camera Raw, you can reset the shadows and highlights where they properly belong. In the meantime, you have maximized data capture for every photo and significantly reduced noise and posterization in the dark shadow areas. By the way, make sure Camera Raw is set to process 16 bit images. This can be set by clicking on the line across the bottom of the Camera Raw window and selecting 16 bit processing.
In the unlikely event you cannot achieve the proper tonality of the original scene using Camera Raw processing, use Photoshop/Elements to make more adjustments. For example, if the image isn't dark enough, (a) open it in Photoshop/Elements, (b) make a copy of the Background layer (in the Layers pallette) by pressing Control+J, (c) click on the Copy layer to make it active and (d) build density and darken the image by changing the Blend mode (at the top of the Layers pallette) from Normal to Multiply. For even more density, repeat steps (a) thru (d). If the image becomes too dark, use the Opacity slider to adjust the Copy layer opacity. Similarly, if an image needs to be lightened, follow these steps except change the Blend mode from Normal to Screen and then adjust Opacity. Make additional adjustments using Photoshop tools as needed.
I hope this convinces you of the value of shooting RAW format and that every photo's histogram should be "exposed to the right" . Philippe
- Phil-Phlash's blog
- Login to post comments
Excellent discussion on the advantages of shooting RAW. Let's hope it motivates people to give working with RAW files a try.