boost::simd

[Sandro Wenzel]

Yes, there are a couple of competitors on the market:

a) boost::simd

b) Vectorclass
http://www.agner.org/optimize/#vectorclass

c) ...

A student of mine has  played with them all. His conclusion was that
boost::simd was lower in performance than Vc ( but this should be carefully
examined ) and there might be a difference in supported  platforms and
instruction sets. I can look up the examples he studied and send them to
you ...



Best

Sandro





2014-04-01 11:04 GMT+02:00 Engelbert Tijskens <
[please enable javascript to see the address]>:

> hi all,
> I recently came across a library boost::simd - not a part of boost - but
> of nt2 (github.com/MetaScale/nt2). It has a similar scope but a
> different approach in that it is a header only library relying heavily
> on  expression templates. I assume that you are aware of it. Anyone
> willing to share his opinion on this with me?
> kindest regards,
> bert
>
> _______________________________________________
> Vc mailing list
>[please enable javascript to see the address]
> https://compeng.uni-frankfurt.de/mailman/listinfo/vc
>



-- 
Dr. Sandro Wenzel
PH / SFT
CERN
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://compeng.uni-frankfurt.de/pipermail/vc/attachments/20140401/23ea8f40/attachment.html>