Some progress with full DXT1a support.
Move quantization settings to compression options.
This commit is contained in:
@ -94,11 +94,6 @@ void InputOptions::reset()
|
||||
m.textureType = TextureType_2D;
|
||||
m.inputFormat = InputFormat_BGRA_8UB;
|
||||
|
||||
m.enableColorDithering = false;
|
||||
m.enableAlphaDithering = false;
|
||||
m.binaryAlpha = false;
|
||||
m.alphaThreshold = 127;
|
||||
|
||||
m.alphaMode = AlphaMode_Transparency;
|
||||
|
||||
m.inputGamma = 2.2f;
|
||||
@ -262,21 +257,6 @@ void InputOptions::setKaiserParameters(float width, float alpha, float stretch)
|
||||
m.kaiserStretch = stretch;
|
||||
}
|
||||
|
||||
/// Set quantization options.
|
||||
/// @warning Do not enable dithering unless you know what you are doing. Quantization
|
||||
/// introduces errors. It's better to let the compressor quantize the result to
|
||||
/// minimize the error, instead of quantizing the data before handling it to
|
||||
/// the compressor.
|
||||
void InputOptions::setQuantization(bool colorDithering, bool alphaDithering, bool binaryAlpha, int alphaThreshold/*= 127*/)
|
||||
{
|
||||
nvCheck(alphaThreshold >= 0 && alphaThreshold < 256);
|
||||
m.enableColorDithering = colorDithering;
|
||||
m.enableAlphaDithering = alphaDithering;
|
||||
m.binaryAlpha = binaryAlpha;
|
||||
m.alphaThreshold = alphaThreshold;
|
||||
}
|
||||
|
||||
|
||||
/// Indicate whether input is a normal map or not.
|
||||
void InputOptions::setNormalMap(bool b)
|
||||
{
|
||||
|
Reference in New Issue
Block a user