[PD] pix_set / pix_gain weirdness

[Matteo Sisti Sette]

On 01/31/2011 08:55 PM, Mathieu Bouchard wrote:

> In any case, such complications are one good reason to use GridFlow.

Well I may be wrong but I am under the impression that such 
complications are indeed mere gem bugs, rather than being intrinsic in 
the way GEM is conceived. That is, a pix object is expected to "inform" 
the objects below it when the pix is modified, and in these cases it is 
simply failing to do it.


> Seriously, in GridFlow, an image is more similar to a normal message,
> and the data actually flows from outlet to inlet,

Yeah I know, and I LOVE that, ideally.

However by trying it a little bit I got the impression that any 
real-life image processing of even a minimum of complexity is completely 
unfeasible in practice because it immediately becomes too slow. Or isn't 
it so?

Is it possible, for example, to do movement detection and blob detection 
in gridflow with an efficiency even remotely comparable to that you can 
achieve with [pix_movement] and [pix_blob]?

thanks
m.