You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have the source building and running and am following the examples to learn how to write an EXR image. I notice that RgbaOutputFile hard codes the format of each RGBA channel to type HALF.
I was wondering if there is a way to set it to FLOAT to use full 32-bit precision for each channel?
Thanks.
The text was updated successfully, but these errors were encountered:
Apologies for the very late response. In short: there isn't a way.
The RgbaOutputFile class only accepts half-float pixel data, which is why it uses HALF storage within th file. For RgbaOutputFile to support float data, there would also have to be a 'float' version of the Rgba type to provide the pixel data.
For anything other than writing half-float RGBA in a single part, the General Interface for Scanline Based Files should be used, via the OutputFile or MultiPartOutputFile classes. See the Reading and Writing OpenEXR image Files document for details. These interfaces are more versatile. For example, you can omit the A alpha channel completely, write it as half float and RGB as 32 bit float, or separate it into a separate part so it can be skipped when it isn't required.
Providing float support in RgbaOutputFile (and presumably RgbaInputFile as well) seems like it would duplicate a lot of effort for a small use-case without providing any extra functionality.
I have the source building and running and am following the examples to learn how to write an EXR image. I notice that RgbaOutputFile hard codes the format of each RGBA channel to type HALF.
I was wondering if there is a way to set it to FLOAT to use full 32-bit precision for each channel?
Thanks.
The text was updated successfully, but these errors were encountered: