One use for [separator] is to texture the same image or video to multiple surfaces.  Another is to control the center position of multiple geos that are offset from that center.

On Sunday, December 6, 2015, João Pais <jmmmpais@gmail.com> wrote:
My examples are very simple, so there won't be any problems with loops and no rendering orders have to be calculated.

Do I take from your comment that using [separator] does add more overhead than [gemhead]?

For my simple case, it seems that using both of them has the same result. But I would want to know what is exactly the difference.

Best,

Joao

The [gemhead] also controls the order of execution of chains while [separator] inherits that from the [gemhead].  One giant chain off a single [gemhead] with a bunch of [separator] will not give the same results as a bunch of different [gemhead] with explicit order of execution.  There is some overhead with [separator] that can become non-trivial if enough of them are used (like in a loop). 

One thing that helps is to always give each [gemhead] a value for render order.  This will avoid, or at least help understand, many problems with how rendering occurs.

On Sat, Dec 5, 2015 at 9:15 AM, João Pais <jmmmpais@gmail.com> wrote:
Hello list,

I don't work with GEM so regularly, so I had a question regarding independent processing chains.
I'm showing different [shpere]s, and wanted to do individual transformations on each, such as [color] and [translateXYZ]. Until now I used an individual [gemhead] for each [sphere], but I just learned that it can also be done using one common [gemhead] and [separator]s instead.

I would like to learn, what are the pros and cons of each solution? Which solution is more efficient and for which reasons?


Plus: [color] and [colorRBG] seem to have the same function. Besides the extra inputs, is there any difference/advantage in one object instead of the other?

Best,

jmmmp

_______________________________________________
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list