Hi Tom,
a little addition...
Carsten Neumann wrote:
> Since the verbosity of the begin/endEditCP calls is indeed quite
> annoying we actually got rid of them for OpenSG 2, there is only the
> equivalent of endEditCP (now called commitChanges) which updates derived
> values after you have completed a number of related changes.
Just to make it clear: you only have to call commitChanges() after you're done
with your changes, so usually not very often (it's a little more complicated
than that, but that's the basic idea).
> that is more or less what OpenSG 2 does. Initially it was thought to be
> too expensive to perform this recording on each call to a mutating
> function. Internal changes in 2 now reduce this cost so that the
> convenience of not having to call begin/endEditCP outweighs the
> performance impact.
The main problem with the basic approach are loops like this:
for (int i = 0; i < normals->size(); ++i)
normals->setValue(i, normals->getValue(i).normalize())
For thread sync we need to store all changes that happen in a ChangeList.
Calling setValue() a million times with either creating a ChangeList entry or
at
least checking whether an entry already exists is going to be prohibitively
expensive.
The solution that was done for 2 is to have read-only and read-write iterators,
and only do an entry when a read-write iterator is created. That removes the
problem, and we can get rid of the explicit begin/endEdits.
Bottom line: if you're looking into OpenSG right now, start looking at 2. It's
still in development, but definitely usable (there are a few groups using it
productively for a while already).
Hope it helps
Dirk
------------------------------------------------------------------------------
The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
production scanning environment may not be a perfect world - but thanks to
Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
Series Scanner you'll get full speed at 300 dpi even with all image
processing features enabled. http://p.sf.net/sfu/kodak-com
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users
