Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Is C# best ideal for any chess programming?

Author: Eugene Nalimov

Date: 21:54:01 11/27/01

Go up one level in this thread


On November 27, 2001 at 17:38:58, Dann Corbit wrote:

>On November 27, 2001 at 17:34:25, Eugene Nalimov wrote:
>
>>Actually, you don't need GC for the Basic. Reference counting is sufficient
>>there. And it definitely can be used in the real time code.
>
>That's very interesting.  I've never heard that before.  It seems to me that
>with functions like mid$(), etc. which allocate hunks of stuff on the fly, you
>would need some way to do GC.  But obviously, there is something I have not
>thought of.
>
>At any rate, if it can be used for BASIC, I suspect it might also be possible in
>Java.

Wrong. In Java you need full GC. Of course you can use Smalltalk's approach --
reference counters for the non-curcular structures, with GC calls sometimes to
free dead curcular structures, but that still means nondeterministic behavior.

Eugene

>
>BTW, Microsoft's C# should really be D#, because it looks a *lot* more like
>Delphi than C or C++.



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.