Hi, I am trying the image processing examples from this late 2008 blog post:
http://blog.wolfram.com/2008/12/01/the- ... rocessing/
My image is only 640x480 so quite small. I am on a RPi 2.
I notice even the simple example of manipulating contrast/brightness in the notebook UI, using sliders, is very very slow. (Too slow to be really usable.)
While that occurs, I see in the CPU meter that it isn't so busy, only up to 25% max from idle of under 10%. This leads me to believe it is limited to 1 (of the 4) CPU cores.
I can hardly believe Mathematica can't handle more cores for this. Is that simply disabled for the RPi edition, or they just don't have it working well on ARM yet?
I think with all 4 cores in use it could be a lot better, at least for small images like this. Or, using the GPU for computation. Again, Mathematica can do this I understand, so is it just not enabled for RPi or for ARM in general?
Is this performance expected to improve? Right now, I'm trying to get my oldest kid (grade 7) interested in this sort of stuff, but it's unusable.