Hacker News
points by varispeed 3 days ago | 0 comments
Why is it still called SIMD-accelerated, if these are now just normal instructions. You just have to take a little bit more effort into restructuring your algorithms to make use of them.
[–] anthomtb 3 days ago
How does using instructions that operate on multiple data elements, which make your software 3 times faster, not fit the definition of SIMD-accelerated?
[–] varispeed 3 days ago
Because these are normal instructions.

If you do multiplication using mul instead of loop and add, is it acceleration too?

[–] anthomtb 3 days ago
Yes.

The article would then be "MUL-accelerated multiplication". Obviously any modern compiler should not require a developer to make that optimization.

What does "normal instructions" mean? I don't see (or missed) a heading with "Normal" in the ESP32 docs.

Edit: Or are you saying the instruction called out by the author is not SIMD?

[–] vardump 2 days ago
The compiler does not generate SIMD instructions on this platform. So they're not "normal" instructions in that sense.

Or perhaps you want to argue that everything is running "normal" instructions. Be it a hardware video decoder or a GPU.

[–] boomlinde 3 days ago
What does "normal" have to do with it? SIMD means single instruction, multiple data, which is how these instructions are designed.