Autodesk 62205-051408-9001 User Guide - Page 604

22 Working with Bit Depth The Combustion architecture supports five bit depths for color information. Use higher bit depth when working with film for greater control over image quality. About Bit Depth Bit depth is the number of bits used to describe the primary colors for each pixel in an image. The number of bits determines the number of color variations for red, green, and blue. When you use a higher bit depth, results may improve in keying, color correction, and other tasks where smooth color gradients are affected. With the increased number of possible colors comes an increased demand for memory and processing time; you should always consider the costs and benefits of increasing bit depth beyond the very capable 8 bits per channel. Usually, the bit depth of your footage dictates your choice of bit depth in Combustion. If film images are digitized and saved as 10 or 16 bit image files, for example, you want to keep all the color information the file format holds. For film work, the entire flow from footage to composite to output can be processed at the higher bit depth. Bit Depth Basics To use bit depth, you should understand how the number of bits determines the number of color variations for red, green, and blue. If two bits are used to describe a pixel, there are four possible combinations: 00, 01, 10, and 11. To calculate the number of tones that can be represented by two bits, you raise 2 to the power of 2 (22), which equals 4. If you continue this exercise, you see that each additional bit in a binary number doubles the number of possibilities. The most common bit depth in compositing is 8 bits, which produces 256 (28) levels for each primary color. If you multiply 256 by 256 by 256 (for the three colors), you get 16, 777, 216 (224) possible colors for each pixel. In Combustion, you can work at 8, 10, 12, or 16 bits per channel, or you can use Float. (The Particles operator is only 8-bit.) Bit Depth 8 10 12 16 Float Number of Levels per Color 256 (28) 1024 (210) 4096 (212) 65536 (216) 4294967296 (232) The float option uses a floating point number between 0 and 1 with the incredible precision of over 4 billion intermediate steps. The bit depth used for the color channels is also used for the alpha channel to describe the transparency of each pixel. For more information on the alpha channel, see "About Channels" in Chapter 9, "Working with Image Channels''. When describing a file format, the total number of bits is often used instead of the bit depth. For example, if you import a PNG image sequence, the footage details in the Footage Controls panel (or in the Footage Library view) may show "32 bit". This means the images use 8 bits per channel, and they include an alpha channel (24+8=32). If the details show "24 bit", the images do not include an alpha channel. If the details for an RPF image show "48 bit", the file has a bit depth of 16 bits, and it does not include an alpha channel. 588