Well, there's currently no way to get signals on a guaranteed boundary, but that's something I want to do in the future. I also want to allow objects to delay creating their input and output vectors (possibly avoiding promoting scalars to vectors for efficiency, and also allowing obejcts to create and/or deal with alternative vector sizes, for instance for multichannel signals).

I've also been looking at your polynomial cos~ approximations - I can't get them to run as efficiently as you seem to be able to, and couldn't immediately figure out if I needed to change compilers, or compile flags, or what. But that also is on the list :)

Miller

on it once and didn't go deeply into all the On Mon, Jul 02, 2018 at 10:04:09PM +0100, Claude Heiland-Allen wrote:

Hi,

Is there any guarantee about alignment of signal vector data?

If not, should there be, in the future?

Or should one conditionally `dsp_add()` their specific-alignment-needing kernels dependent on what the `dsp` method actually gets?

Context:

In some code unrelated to Pd, using GCC vector intrinsics (not CPU-specific) I got a near-2x speed boost by recompiling the same code to target a newer CPU, vs the binary compiled for an older CPU.

The old idea of compiling a machine-specific math~.pd_linux (or whatever) to speed up everything by overwriting internal objects is also on my mind.

Thanks,

Claude

https://mathr.co.uk

---

Pd-dev mailing list Pd-dev@lists.iem.at https://lists.puredata.info/listinfo/pd-dev