Subtitles of the Movie

This movie will discuss bit depth and color spaces. Bit depth is a way of describing how many values show the range of colors in an image. It's defined as the number of color values equals X to the second power, where X is the number of bits. In a one bit image, you have two values; black and white, which is 1 to the second power. A two bit image gives 2 to the second power, or 4 color values per channel. Now, this would be for red, green, and blue. Higher bit depths mean more colors and smoother transitions from one color to the next. We have more steps of color from one color to the next. 8 bit works well for most non film composites, but for smoother transitions between colors, you may need to go for a 16, 24, or 32 bit color depth. Now this doesn't mean just converting from one color bit depth to another using an other Bytes node or by adjusting your bit depth in your globals. What it does mean is that you need a higher color bit depth to begin with in order to see smooth color transitions, such as by shooting on high quality stock or rendering out from your 3D program at 16, 24, or 32 bit color bit depth. Shake works with a color range of 0 to 1 in RGB linear space. You can work in different bit depths with 0 as black and 1 as white. A Color Space node allows you to switch from one color space to another, and by color space, I mean RGB, being red, green, blue, CMY, which is cyan, magenta, yellow, HLS, or hue, luminance, saturation, HSV, which is hue saturation value, or the YIQ, or YUV color spaces, YUV being something that's used for DV footage for instance. And the Color Space node is very useful when you're doing color corrections. For instance, if you apply a Color Space node right after a color wheel and then an Add node and make some adjustments there, and then reapply the Color Space node, the Add node having made things red, actually made things cyan because going RGB to CMY meant that the red channel corresponded to cyan. Doing the identical color correction without the Color Space node conversion before and after, and the Color Space node conversion went RGB to CMY and then CMY back to RGB, just doing that Add node after the color wheel gave us a completely different result from doing it with Color Space nodes in between. Let's go over the parameters of the Color Space node right now. The In space allows you to specify what the colors coming into the node are at the top. The Out space allows you to specify what the output of the colors should be. The Luminance Bias lets you weight the three channels, red, green, and blue, for the luminance calculations when you're working in an out of hue luminance saturation. Luminance calculates your brights based on our eyes perception that green is brighter than an equal value of blue. The LogLin node is for handling log film plates like Fineon files coming from film or when doing an output for film. A conversion happens here to go from the log color space to a linear color space for compositing. Then at the end of your tree, you would apply another LogLin node to convert back to log space for film output. In the LogLin node, we have the option of going log to lin or lin to log. Then we have our red, green, and blue offset, which is to offset your color channels. Our black sets your black point, and the default is 95, and our white sets your default white point. If you're noticing some funkiness when doing your conversion from log to lin or lin to log, you can make the adjustments here to compensate for those. The soft clip is for your roll off values on the white point and the RN gamma and RD gamma should really be left at their defaults for best results. To wrap up, color bit depth determines the quality of your image by determining how many colors you have to work with. Working with a higher bit depth gives you lower incidents of visible banding between colors such as here. Color spaces are the various ways of looking at and changing the colors in an image. By converting color spaces, we can better target our colors during color correction.