[Bf-committers] Layer system

Martin Poirier theeth at yahoo.com
Fri Dec 21 23:23:29 CET 2007


--- David Bryant <aceone at bellsouth.net> wrote:

> I need to know what time the meetings are held in
> U.S. eastern time.

10 AM EST


> To add my two cents right now, bit fields are not
> the way to go.The old 
> system needs to be rebuilt.And even though it may
> seem like an insignificant 
> recode, it's the future whether we acknowledge it or
> not.Those that do 3D 
> for a living need more than 20 layers and named
> layers at that.

Named layers are not incompatible with using bitfields
as layer status indicators.

The advantages of bitfields (or an array of them) are
quick visibility tests, quick insertion/removal,
constant localized memory usage.

Using linked list is definitely not a good option for
the reasons that Brecht alluded to. Memory locality
penalties, linear check time (assuming sorted lists),
bigger memory footprint, ...

Again, additional layer features (names and all that)
don't have much to do with the actual back end used to
indicate layers status. So those two discussions can
run in parallel.

Martin


      ____________________________________________________________________________________
Never miss a thing.  Make Yahoo your home page. 
http://www.yahoo.com/r/hs


More information about the Bf-committers mailing list