Make the 'Component Associations' UBO (on most devices) twice smaller - use the 'packed' layout and 'ivec2' in place of 'ivec4'.
This forces us to dynamically probe the 'Stride' - number of 4-byte entities that the UBO's 'ivec4' takes - and correct the already-constructed UniformBlockAssociation objects in case this turns out to be different than the expected 2, i.e. tightly-packed.
The point: this saves a lot of uniform space in th vertex shader, which lets us declare larger MAX_NUM_COMPONENTS - needed for the Gigaminx.
(and no, we cannot assume an UBO can hold at least 16KB since this is not the case on Samsung Galaxy J4+)
Make the 'Component Associations' UBO (on most devices) twice smaller -
use the 'packed' layout and 'ivec2' in place of 'ivec4'.
This forces us to dynamically probe the 'Stride' - number of 4-byte entities that the UBO's 'ivec4' takes - and correct the already-constructed UniformBlockAssociation objects in case this turns out to be different than the expected 2, i.e. tightly-packed.
The point: this saves a lot of uniform space in th vertex shader, which lets us declare larger MAX_NUM_COMPONENTS - needed for the Gigaminx.
(and no, we cannot assume an UBO can hold at least 16KB since this is not the case on Samsung Galaxy J4+)