After chasing down some of the references from the responses to your
question, I noticed this thread in comp.lang.c:
https://groups.google.com/forum/#!topic/comp.lang.c/Z0mycsYvPyI/discussion

But I do not think that that "weighting vector" is a good idea. On
modern chip architectures, the occasional multiply is free, or almost
free, and the cost of maintaining the premultiplied weighting vector
would be noticeable.

That said, this is a hardware specific opinion.

Thanks,

-- 
Raul


On Tue, Aug 11, 2015 at 4:35 PM, Dan Bron <[email protected]> wrote:
> Pascal wrote:
>> just wanted to add that your custom security level suggestions are well 
>> thought out.
>
> Thanks, if I can ever get my head wrapped around the structure of the J 
> source, I might be able to contribute to making such ideas a reality.
>
> In that vein, having not used C in eons, I was confused by the definition of 
> AF:
>
>      typedef A (*AF)();
>
> so I asked about it on StackOverflow:
>
>      http://stackoverflow.com/questions/31948392/what-does-typedef-a-af-mean 
> <http://stackoverflow.com/questions/31948392/what-does-typedef-a-af-mean>
>
> and learned a bunch of interesting things. For one:
>
>> when you have the 3 fields f, g, and h, whose function I haven’t determined 
>> yet.
>
> Turns out the type V is also used to represent adverbs and conjunctions, so 
> f/g/h represent *their* arguments.  For another, Roger’s 1993 BAA 
> presentation is available at Vector, and covers a lot of the design of the J 
> engine:
>
>      http://archive.vector.org.uk/trad/v094/hui094_85.pdf 
> <http://archive.vector.org.uk/trad/v094/hui094_85.pdf>
>
> and, of course, I was informed that AF is a type designating a function 
> pointer, which makes sense, given that V uses them to  capture the monadic 
> and dyadic definitions.
>
> I also learned cdecl, which translates obscure C type declarations to English 
> and vice-versa, is available as an online service:
>
>       http://cdecl.org/ <http://cdecl.org/>
>
> I’m sure this’ll come in handy if I keep diving into the J source.
>
> -Dan
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to