PDL Developers-

I had the pleasure to meet zowie in
person for dinner this week and a
chance to talk PDL.  That makes 2
core developers that I have met.
It was nice to hear some of the
ways that he has been using PDL
in his work and some of the context
from the earlier days of PDL.

Now, that PDL-2.015 is out, I think
it is time for us to shift focus to
the new PDL projects:

* PDLA which splits up the existing
   monolithic PDL distribution into a
set of core functionality and
   external (non-core) modules that
   provide the rest.  Ed and Zaki
   have been leading the charge here.

* PDL Next Gen (PDLng): which is a
   completely re-engineered PDL core
   computation and data engine to
   replace and supercede the existing
   PDL-2.x core functionality.

I plan to give a more in-depth update
on my ideas, hopes, and plans for
PDLng by the New Year.  I expect that
Zaki and Ed might have some updates
of their own.

To whet your appetite, here are some
URLs tied to various ideas that have
been floated for PDLng that Craig and
I discussed and could be the seeds
for some of the next generation PDL
capability:

   ArrayFire GPGPU parallel computing:
http://www.arrayfire.com/docs/index.htm
This could give PDLng a quick start for
   high performance using GPUs.

   COS, the C Object System:
https://github.com/CObjectSystem/COS/blob/master/doc/cos-draft-dls09.pdf
https://github.com/CObjectSystem/COS
   This could give PDLng an object oriented
   interface that maintains a C FFI so that
   PDLng could call or be called from other
   languages.  Matplotlib anyone?

   HDF5 Attributes:
http://www.hdfgroup.org/HDF5/doc1.6/UG/13_Attributes.html
   I was thinking that our PDLng implementation
   should cover the bases from FITS, HDF5, named
   dimensions,..  Ingo has already done some
   work along these lines.

In the meantime, I'm going to try to get
Perl OpenGLup to modern OpenGL APIs and
allow for improved and more flexible
graphics options for PDL-2.x and PDLng.

--Chris



------------------------------------------------------------------------------
Go from Idea to Many App Stores Faster with Intel(R) XDK
Give your users amazing mobile app experiences with Intel(R) XDK.
Use one codebase in this all-in-one HTML5 development environment.
Design, debug & build mobile apps & 2D/3D high-impact games for multiple OSs.
http://pubads.g.doubleclick.net/gampad/clk?id=254741911&iu=/4140
_______________________________________________
pdl-devel mailing list
pdl-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/pdl-devel

Reply via email to