-
Notifications
You must be signed in to change notification settings - Fork 239
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Creating an additional option for creating texture arrays instead of deducing from layer count #596
Comments
Oops. |
Thank you @MarkCallow for attending to this. Well I guess at this stage we might have to do a wider community poll? |
Hi @MarkCallow, hope all is well. Just wanted to check-in and see if you've made a decision either way regarding the tool. If so, please let me know so that I can align my local fix towards your preference in advance. Much appreciate, |
I'm leaning towards using
to
I sent a survey to the Khronos WG responsible for KTX. I don't have a good place to make a general survey of users. I doubt most would see anything in the Wiki or GitHub questions. |
Perfect, thank you for the update: I'll assume that to be the new behavior. Hopefully, with a new update coming soon... I can just replace my build with yours. Much appreciated! |
I'm implementing this change. What behavior do you prefer when
|
Exit with an error, it was probably unintended. |
Fixes #596. Add checks for valid argument values for depth, layers and levels. Update toktx documentation. * Add note about how to disable 2-component normal map conversion. * Reformat some of the source.
Fixes #596. Add checks for valid argument values for depth, layers and levels. Update toktx documentation. * Add note about how to disable 2-component normal map conversion. * Reformat some of the source. * Move input_swizzle to scapp.h, support it in ktxsc and move common encode code to scapp.h.
Fixes KhronosGroup#596. Add checks for valid argument values for depth, layers and levels. Update toktx documentation. * Add note about how to disable 2-component normal map conversion. * Reformat some of the source. * Move input_swizzle to scapp.h, support it in ktxsc and move common encode code to scapp.h.
Fixes KhronosGroup#596. Add checks for valid argument values for depth, layers and levels. Update toktx documentation. * Add note about how to disable 2-component normal map conversion. * Reformat some of the source. * Move input_swizzle to scapp.h, support it in ktxsc and move common encode code to scapp.h.
Fixes KhronosGroup#596. Add checks for valid argument values for depth, layers and levels. Update toktx documentation. * Add note about how to disable 2-component normal map conversion. * Reformat some of the source. * Move input_swizzle to scapp.h, support it in ktxsc and move common encode code to scapp.h.
Fixes KhronosGroup#596. Add checks for valid argument values for depth, layers and levels. Update toktx documentation. * Add note about how to disable 2-component normal map conversion. * Reformat some of the source. * Move input_swizzle to scapp.h, support it in ktxsc and move common encode code to scapp.h.
Fixes KhronosGroup#596. Add checks for valid argument values for depth, layers and levels. Update toktx documentation. * Add note about how to disable 2-component normal map conversion. * Reformat some of the source. * Move input_swizzle to scapp.h, support it in ktxsc and move common encode code to scapp.h.
Currently toktx seems to create arrayed textures only when layers is specified to be larger than one:
This causes issues when compressed textures are generated in bulk and assigned to various slots on a variable count descriptor set of
sampler2DArray
s. The issue mainly surfaces on AMD drivers (where the.z
component oftextureSize()
ends up being undefined in the event the combined sampler is backed by atexture2D
and not atexture2DArray
).I propose a separate option of
--is-array
that sets the flag inside KTX metadata to ensure that single layer texture arrays are possible (when potentially provided alongside larger/taller arrayed textures).The text was updated successfully, but these errors were encountered: