ARM NEON

I have regularly used qemu to emulate Cortex-A8 and later ARM devices under Linux (those devices are pretty much guaranteed to have vfs3 and NEON copros). I know qemu is also available for Microsoft Windows™. You could try that. or you could use the Google to find an alternative emulator of your choice.

Even if they work I don't think an emulator is good for any sort of performance measurement really.

Right. The whole purpose in using NEON is performance, not compatibility. If your goal is compatibility, you are doing it wrong. It is specifically meant to be processor-specific. Convert the code to SSE* using macros to select which code path to follow.

L. Spiro

Well In the most cases I'm only porting the code form btMatrix3x3 and btVector3 to 4 dimension version of the classes.

I beleve in the people who wrote bullet math. Of course I will pay attention in the situations where the performance is not good enough.

I'm not really into NEON (just can't find good papers). I'm thinking that the most optimal version using NEON should look like the SSE version of the code, is that correct?

Well In the most cases I'm only porting the code form btMatrix3x3 and btVector3 to 4 dimension version of the classes.

I beleve in the people who wrote bullet math. Of course I will pay attention in the situations where the performance is not good enough.

I'm not really into NEON (just can't find good papers). I'm thinking that the most optimal version using NEON should look like the SSE version of the code, is that correct?

For basic ops like add/sub, then yes they should be similar. For dot/cross and other operations based around shuffles/unpack/movehl, they aren't going to look too similar at all....