[bitc-dev] Pools in lieu of GC

[Pal-Kristian Engstad]

Jonathan S. Shapiro wrote:
> What people tend to forget is that GC is just as fast as hand allocation.
>   
I wondered about this statement -- how can that be true? If we assume 
that the programmer did some careful analysis of the code, determining 
when memory can be reclaimed, how can that be as fast as having code 
(albeit with type information) that scans memory to see if it is still 
reachable? Mind you, I'm not arguing that it can be done with fairly low 
overhead, but there still must be some overhead? So can you try to 
substantiate that claim, Jonathan?

PKE.

-- 
Pål-Kristian Engstad (engstad at naughtydog.com), 
Lead Graphics & Engine Programmer,
Naughty Dog, Inc., 1601 Cloverfield Blvd, 6000 North,
Santa Monica, CA 90404, USA. Ph.: (310) 633-9112.

"Emacs would be a far better OS if it was shipped with 
 a halfway-decent text editor." -- Slashdot, Dec 13. 2005.