Post Processing Settings
Once Continuum has rendered out the final image, complete with the various interactions with the virtual "camera" system, the user has a certain level of control over the final image, by way of these post processing settings. While they don't necessarily give the user control over individual elements of the scene, they do give the user control over the representation of the scene, at a very high level. Settings such as Vibrance, Saturation, Gain, Lift, Contrast, White Balance and a full L.U.T (Look Up Table) give the user an unprecedented amount of control over the final image, without being overwhelming.
Vibrance & Saturation
Vibrance & Saturation are two very similar settings. In essence, they are multipliers that affect the saturation of the final image. Both settings are zero centered, where a value of 0.0 denotes no change in the final image. Conversely, a value of 1.0 would denote a doubling of saturation in the final image, where a value of -1.0 denotes a complete desaturation of the final image.
Where the settings deviate, however, is in how they are performed. Vibrance is performed in a way that preserves, luminance; that is, vibrance preserves the effective brightness of the image. Where saturation does not care about the luminance. Due to this, they can result in two very different results, depending on the circumstance.
Lift allows the user to manipulate the black point of the final image, pulling it up or down. This can have an adverse effect on the image: raising it to 1.0 pulls the black point up, adding a gray tone to the whole image, where lowering it to -1.0 pulls the black point down, bringing out the low lights a lot more.
Gain is effectively a brightness multiplier, allowing the user to manipulate the brightness of the final image, without affecting the virtual "camera" system. The input to gain is clamped between 0.0 and 1.0, meaning that a gain value of 1.0 would denote a doubling of the brightness for every pixel in the image. Conversely, a gain value of -1.0 would denote a halving of the brightness for every pixel in the image.
Contrast determines the scale between high lights and low lights in the final image, and as such is probably a setting that most users are familiar with. Contrast is zero centered, meaning that a value of 0.0 denotes no change in the image. A value of 1.0 would make high lights brighter, and low lights darker, where a value of -1.0 would crush both the high and low lights, shifting the image more towards a middle point.
White Balance influences the overall temperature of the final image, measured in Kelvin. The higher the White Balance value, the more the image is shifted towards blue. 6500 Kelvin is perfectly balanced, resulting in no change in the temperature.
The L.U.T, or Look Up Table, is arguably the most powerful tool in Continuum's arsenal of post processing related settings, allowing the user to map one color space to another. This allows the user to perform color grading to an incredibly extent, never has a shader given the user this much control over the whole tone of the final image.
The L.U.T, in essence, is a special RGB texture that stores a 2D table of tiles. Each tile contains a 2D gradient, which maps the three color channels (red, green & blue) to a full RGB color. It is through this that Continuum allows users to change the entire tone and color of the final image.
Use cases of the L.U.T cover many different possibilities, including emulation of actual color grading profiles used by the industry, as well as usage of completely custom color grading profiles.
LUT disabled (default)
Kodak film stock LUT
Neutral LUT texture
Fujifilm Eterna film stock LUT
Temporal Anti-Aliasing (TAA)
Temporal Anti-Aliasing, or TAA, is a particular approach to anti-aliasing that results in a smoother and higher quality final image. TAA not only mitigates aliasing, but also smooths out a lot of the sampling artifacts, such as noise and jittering, that naturally come from many of the effects that Continuum produces.
Anti-aliasing is a type of graphical effect which removes a phenomenon referred to as aliasing, where edges of geometry that are drawn to pixels on screen are drawn with jagged edges. There are many types of anti-aliasing, however TAA is a fairly recent approach to anti-aliasing that is fairly cheap, effective, and helps smooth out other parts of the image.
TAA works by jittering, or moving the world within individual pixels. The world is moved in amounts much smaller than the size of a pixel, and in a random direction, every frame. TAA then takes the previous rendered frames, and blends them in with the current frame. The end result is very similar to that of a very small blur, but without the overhead of a blur. This smooths out edges of geometry, mitigating any aliasing.
Where TAA shines in Continuum, though, is in its ability to smooth out sampling artifacts. Many of the effects that are in Continuum naturally produce sampling artifacts, such as noise, dithering, jittering, clipping, etc. TAA by itself can smooth out any jittering or clipping, but if we apply a jitter to the noise or dither functions, then TAA can also smooth those out.
In the vast majority of cases, TAA should remain enabled as it helps to prevent aliasing and sampling artifacts, though if for any reason you wish to disable TAA, the option is there.
Click on one of the images below to view a larger comparison for best effect