AASettings

The settings in the <AASettings> group control the way sampling points are adaptively determined for texture baking, and how those samples are combined together for each pixel in the final texture.

NOTE:The settings in this group are only used when rendering scenes with "classic" materials. They are not used at all for physically based rendering.

The following block lists the settings available in this group, with their default values. Descriptions of the settings are below.

<AASettings>
    <maxSampleRate>0</maxSampleRate>
    <minSampleRate>-2</minSampleRate>
    <contrast>0.1</contrast>
    <diagnose>false</diagnose>
    <clamp>false</clamp>
    <minValue>0</minValue>
    <maxValue>1</maxValue>
    <filter>box</filter>
    <filterSize>
        <x>1</x>
        <y>1</y>
    </filterSize>
</AASettings>

maxSampleRate

Determines the maximum number of samples per pixel, using the following formula: 4^<maxSampleRate>.

Accepts values in the range [-4..4]. The default value is 0.

This setting may be overridden by the <minSampleRate> value. See <minSampleRate> below for details.

minSampleRate

Determines the minimum number of samples per pixel, using the following formula: 4^<minSampleRate>.

Accepts values in the range [-4..4]. The default value is 0.

It is recommended to use at least one sample per pixel for production use (i.e. <minSampleRate> >= 0). Values less than 0 allow less than one sample to be used per pixel; this undersampling is most useful when doing camera renders or baking textures with big UV-patches.

This setting is interpreted in conjunction with <maxSampleRate> as follows to determine the minimum and maximum number of samples per pixel:

	-4	-3	-2	-1	0	1	2	3	4
-4	1/256	1/256 to 1/64	1/256 to 1/16	1/256 to 1/4	1/256 to 1	1/256 to 4	1/256 to 16	1/256 to 64	1/256 to 256
-3	1/64	1/64	1/64 to 1/16	1/64 to 1/4	1/64 to 1	1/64 to 4	1/64 to 16	1/64 to 64	1/64 to 256
-2	1/16	1/16	1/16	1/16 to 1/4	1/16 to 1	1/16 to 4	1/16 to 16	1/16 to 64	1/16 to 256
-1	1/4	1/4	1/4	1/4	1/4 to 1	1/4 to 4	1/4 to 16	1/4 to 64	1/4 to 256
0	1	1	1	1	1	1 to 4	1 to 16	1 to 64	1 to 256
1	4	4	4	4	4	4	4 to 16	4 to 64	4 to 256
2	16	16	16	16	16	16	16	16 to 64	16 to 256
3	64	64	64	64	64	64	64	64	64 to 256
4	256	256	256	256	256	256	256	256	256

contrast

If the difference in contrast between two adjacent samples is less than this threshold, Beast will consider the sampling good enough. If the difference in contrast is greater than this threshold, and if the <maxSampleRate> value allows it, Beast will place additional samples in the area.

The default value is 0.1.

diagnose

Enable this to diagnose the sampling. The brighter a pixel is, the more samples were taken at that position.

clamp

To work efficiently on LDR images, the sampling algorithm can clamp the intensity of the image samples to the range [<minValue>..<maxValue>].

When rendering in HDR, this is not desired. The <clamp> setting should then be disabled.

minValue

When <clamp> is enabled, this setting determines the minimum sample intensity.

maxValue

When <clamp> is enabled, this setting determines the maximum sample intensity.

filter

The sub-pixel filter to use when weighing together the samples for each pixel. This value, along with the filter size, may affect the final look of the lightmaps. Accepts the following values:

Box: Smoothing. Each sample is treated as equally important. This is the fastest filter to execute, but it gives blurry results. This filter is used by default.
CatmullRom: Smoothing.
Cubic: Smoothing.
Gauss: Smoothing. Uses the Gauss function as the filter kernel. This gives the best results (removes noise, preserves details).
Lanczos: Sharpening.
Mitchell: Sharpening.
Triangle: Smoothing. The filter kernel is a tent, which means that distant samples are considered less important.

filterSize > x

The width of the filter kernel, in pixels. Accepts values from 1.0 to 3.0 inclusive. The default value is 1.0.

filterSize > y

The height of the filter kernel, in pixels. Accepts values from 1.0 to 3.0 inclusive. The default value is 1.0.