The price of a modern list architecture?
in [Ruby]
|
Prev: Null checks in Ruby
Next: Saving HTML as MHT
From: Jacko on 18 Jul 2010 19:10 http://groups.google.co.uk/group/comp.lang.forth/browse_thread/thread/ef39e59023a8ae4c/9eb9f6b99139a5df#9eb9f6b99139a5df Simple really, if Lisp's (X , Y) construct should have limits placed on assignment of Y for translation lookaside buffer vitual memory mapping thrashing and for easier garbage collection (provable), then are there any structures in Ruby that may be better adapted such that kernal poijnters may be used for faster code, and possibly extra memory bandwidth for no garbage collector gc stalls and faster code? I wish to call this a constrafucation. Cheers Jacko
From: Eleanor McHugh on 19 Jul 2010 07:22 On 19 Jul 2010, at 00:15, Jacko wrote: > http://groups.google.co.uk/group/comp.lang.forth/browse_thread/thread/ef39e59023a8ae4c/9eb9f6b99139a5df#9eb9f6b99139a5df I remember now why I stopped reading CLF... > Simple really, if Lisp's (X , Y) construct should have limits placed > on assignment of Y for translation lookaside buffer vitual memory > mapping thrashing and for easier garbage collection (provable), then > are there any structures in Ruby that may be better adapted such that > kernal poijnters may be used for faster code, and possibly extra > memory bandwidth for no garbage collector gc stalls and faster code? There could well be places in a Ruby runtime where such an approach would be possible although that may not be true for any of the current implementations. Do you have a link through to the original articles/sites describing this modification to Lisp along with some code examples? Ellie Eleanor McHugh Games With Brains http://feyeleanor.tel ---- raise ArgumentError unless @reality.responds_to? :reason
From: Jacko on 19 Jul 2010 08:19 On 19 July, 12:22, Eleanor McHugh <elea...(a)games-with-brains.com> wrote: > On 19 Jul 2010, at 00:15, Jacko wrote: > > >http://groups.google.co.uk/group/comp.lang.forth/browse_thread/thread... > > I remember now why I stopped reading CLF... > > > Simple really, if Lisp's (X , Y) construct should have limits placed > > on assignment of Y for translation lookaside buffer vitual memory > > mapping thrashing and for easier garbage collection (provable), then > > are there any structures in Ruby that may be better adapted such that > > kernal poijnters may be used for faster code, and possibly extra > > memory bandwidth for no garbage collector gc stalls and faster code? > > There could well be places in a Ruby runtime where such an approach would be possible although that may not be true for any of the current implementations. > > Do you have a link through to the original articles/sites describing this modification to Lisp along with some code examples? > > Ellie > > Eleanor McHugh > Games With Brainshttp://feyeleanor.tel > ---- > raise ArgumentError unless @reality.responds_to? :reason Not yet, currently getting feedback on a multitude of languages to see what 'structures' would need to be converted. As yet the structures have not been converted as it is a research initiative on if a new language would have to be created, an existant language could be adapted, or all or many languages could be fed through code translators.
From: Eleanor McHugh on 19 Jul 2010 10:57 On 19 Jul 2010, at 13:20, Jacko wrote: > Not yet, currently getting feedback on a multitude of languages to see > what 'structures' would need to be converted. As yet the structures > have not been converted as it is a research initiative on if a new > language would have to be created, an existant language could be > adapted, or all or many languages could be fed through code > translators. Well if you need a testbed at some point let me know, I have a virtual machine library in development as part of my effort to port Ruby to Go and would be happy to try your ideas out. Ellie Eleanor McHugh Games With Brains http://feyeleanor.tel ---- raise ArgumentError unless @reality.responds_to? :reason
From: Jacko on 19 Jul 2010 12:32
There's a section started on http://sites.google.com/site/jackokring and a you may see other things are happening too. Any feedback would be fine. |