got a S4-Plus-MSM8227 here in a windows phone. i'm not gonna start changing it to android nor my devenv.
problem at hand is that visual studio doesn't support half float aka float16 intrinsics which the cpu allegedly supports. reason being i wanna use them to increse particle generation/"simulation" thruput or in general increase data thruput and storage. so...
is there a document somewhere that allows me to see the instruction opcode encoding to add the f16 instructions to the default arm_neon.h supplied with vs and the windows phone sdk.
Hello, fp16 datatype is available on CPUs based on Arm Architecture v8.2A and later. I believe you are using an v7A compatible CPU. See:https://developer.arm.com/architectures/instruction-sets/floating-point
yo. i found the instruction encoding. unfortunately it does not do the arithmetics. atleast it should be able to convert the data if i add the opcodes. fine. i'll figure this out.
View all questions in Arm Development Platforms forum