Welcome to TVPaint Community forums !

I have tried to solve the problem from this request viewtopic.php?f=21&t=5204
with a script of my own. and I think it works for my needs
the script work that way on current layer from start point it will look for the next new image and delete the instances between and put in exposures.
You can increase the accuracy by changing the split value in the top of the embedded george script. the split value is how many time in the width and height the script will look for a change. and change the sensitivity by changing the threshold value in the start of the script.
-Mads

interesting

Thank you. it actually works for my needs. to my big surprise. And it is faster than I expected(hats of to George). But get's a little slow ehen you run on 5000 frames.

-Mads

humm nice job but a little bit memory and config file intensive

why you don't use the spare buffer to store the previous image ?
-tv_copytospare
to copy the current image to the spare buffer
-tv_getpixellayer "s" x y
to read the spare pixel value

Hervé wrote:humm nice job but a little bit memory and config file intensive

why you don't use the spare buffer to store the previous image ?
-tv_copytospare
to copy the current image to the spare buffer
-tv_getpixellayer "s" x y
to read the spare pixel value

Cool I will do that

the other way is to use arrays in GRG
PixelArray(1,2) = "Pixel Array"
tv_warn PixelArray(1,2)

what do you think is fastest?

madsjuul wrote:what do you think is fastest?

Arrays are faster but use more memory