I used "
texconv.exe" to batch convert a bunch of images from PNG to DDS.
With this code in the CMD:
Code: Select all
texconv -r "C:\png images\*.png" -o "C:\dds images" -m 1 -f DXT1
"-m 0" - means default number of mipmaps for the image,
"-m 1" - means one image
"-m 4" - means the image and 3 mip map levels.
-r is read from
-o is output
When not specyfying "-f DXT1" the default compression will be used. The output file will be big (useful for small UI icons).
"-f DXT1" is DXT1 compression method. It gives the smallest output possible (I think). No alpha is stored.
"-f DXT3" is good for images with sharp transitions from no-alpha to maximum transparency, like in GIF's.
"-f DXT5" is good for store all levels of transparency in an image. But the size of the file is the biggest of all the 3 lossy compressions.
I'm not sure, about this, but the width and the height has to be divisible by 4(?)
I'm not an expert about compressions. I'm only a user of the available tools.
XnConvert has no options when exporting into DDS.
DDS has a lot more compression options, but these are the basics. Just look how many exporting properties for DDS GIMP has. You don't need any extra plugins (since 2020).
Thanks to your detailed description I can reproduce