Sam Hocevar
8bca639606
math: add missing operators (==, !, <= etc.) to the "half" class so that
it can at last be put into a vector.
hace 14 años
Sam Hocevar
54f74a1b20
win32: some compilation fixes here and there.
hace 14 años
Sam Hocevar
4a16c072f2
core: new combinations of integer vectors (unsigned, 8-bit, etc.), dot and
cross product, normalize, etc.
hace 14 años
Sam Hocevar
4b9bd58747
core: reactivate half denormals for the PS3.
We know we will not have denormal floats on the PS3, but we should still
create denormal halves in case the other end (maybe the GPU?) knows how
to handle them.
hace 14 años
Sam Hocevar
cb001ddbcc
core: tune the half precision code so that the best variants are being
used on the PS3 platform.
hace 14 años
Sam Hocevar
e47928e8d1
test: improve the benchmark code to measure a lot more half precision
number conversions.
hace 14 años
Sam Hocevar
205fc5fc22
core: rename half::isnan() etc. to half::is_nan() because "isnan" can be
a C macro on some systems (Android NDK, PS3...).
hace 14 años
Sam Hocevar
4fea7a091d
core: implement array versions of the float / half conversion routines.
hace 14 años
Sam Hocevar
2d85e76ec8
core: minor refactoring in the float / half conversions to accomodate
for future array versions.
hace 14 años
Sam Hocevar
8b71ddb30c
core: implement new half / float operations and the associated tests.
hace 14 años
Sam Hocevar
c3b23d1f4d
core: add a half to float conversion routine and utility floating point
functions such as isnan(), isinf() etc.
hace 14 años
Sam Hocevar
96007d26c4
core: start working on the "half" type.
hace 14 años