>From my experience no compilers do not optimize or reorganize structures
or classes


The structuring of and orderng of types is nothing new its something that
many programmers never have to think about though , but in something like
OSG were you can be passing thousands of objects about then it can make
and does a difference

I learned about the bFrom my experience no compilers do not optimize or
reorganize structures or classes data apoart from padding

The structuring of and ordering of types is nothing new its something that
many programmers never have to think about though , but in something like
OSG were you can be passing thousands of objects about then it can make
and does a difference

I learned about the benefits this back in the days of Performer 1. and GL
times and how the Performer team really pulled their hair out optimizing 
all the internal structures and how adding say an int would cause a large
performance impact etc

I have tried to practice this as much as possible since then

We have metrics were I work now and it it does make a difference
unfortunately I'm not at libertity to make these public


>From my own experience I would recommend every one make the ordering of
their data members part of the common work practicesenefits this back in
the days of Performer 1. and GL times and how the Performer team really
pulled their hair out optimizing  all the internal structures such and how
added say an int would cause a large performeance impact

I have tried to practice this as much as possible since then

We have metric were I work now it it does make a difference unfortunaly I
cannot release these


Regards

Gordon.


>>    5.11 suggests reordering structs/classes by the size of
>> their atomic members -- doubles, then floats/longs, shorts,
>> bytes to avert this misalignment (using padding where necessary).
>
> I'd suspect that a decent compiler could perform this optimization at
> compile time.
>    -Paul
>
> _______________________________________________
> osg-users mailing list
> [email protected]
> http://openscenegraph.net/mailman/listinfo/osg-users
> http://www.openscenegraph.org/
>

_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/

Reply via email to