Numerical accuracy of SIMD primitives

Hardware vector arithmetic (SIMD)

Hardware vector arithmetic (SIMD)

Prev: | Supported SIMD instruction sets and operations |

Next: | Writing efficient SIMD code |

No guarantees are made that math.vectors.simd words will give identical results on different SSE versions, or between the hardware intrinsics and the software fallbacks.

In particular, horizontal operations on float-4 vectors are affected by this. They are computed with lower precision in intrinsics than the software fallback. Horizontal operations include anything involving adding together the components of a vector, such as sum or normalize.

In particular, horizontal operations on float-4 vectors are affected by this. They are computed with lower precision in intrinsics than the software fallback. Horizontal operations include anything involving adding together the components of a vector, such as sum or normalize.

This documentation was generated offline from a
`load-all`

image. If you want, you can also
browse the documentation from within the UI developer tools. See
the Factor website
for more information.

Factor 0.99 x86.64 (2203, heads/master-424edf64aa, Mar 8 2023 13:48:50)