On a similar note I was wondering what my options are if I want to manipulate layer pixels at close to realtime speeds... Perhaps I should explain:
I am doing imaging of samples from a microscope. Data from my hardware comes in as intensity values which can be either int32 of double (depending on the detector and the C# code responsible for HW interaction). Images are acquired in multiple color channels/detector channels (for instance using an optical camera and an IR camera on the same sample or something similar. Images are typically no bigger than 512 by 512 pixels and one pixel is typically acquired every milisecond, just to give an idea.
Right now I have my own application in C# to do this imaging with a very crude file format for data storage. However, I was thinking that it might be cool to offer the option of saving the image data directly to Pdn or to create a plugin that could channel the data to Pdn somehow, preferably in real time as data comes in. Most of my data needs to be post processed in a paint program anyway to prepare it for reports, publications, ...
Would this be feasible?
I guess it comes down to the question if
public ColorBgra this[int x, int y]
public ColorBgra this[Point pt]
of Surface offer good enough performance and if the rendering process in Pdn is fast enough...
Can anybody shed some light?
