ronaldkr wrote: Kieryn wrote:
Just did an analysis of the average fluid level of the right-most 80 pixel columns since NP 191.
There is some inaccuracy due to the floating point math i used, but it's close enough to visualize what's going on.
(Spoiler: level up to newpix 647 is displayed)
Edit: the h-scale is exaggerated by a factor of 10.
finally we get an overwview of this phenomenon. Thanks a lot.
-- I saw another level rise at frame 247. Is it included in your 244 rise, or was the diffence too small to qualify?
-- Could you press the difference values into the graphic (for those who can do integrals in their head but not substraction)?
-- I assume your baseline is "bottom of frame". Maybe you could state that explicitly.
A graphic says more than 1E3 words.
-- The 247 rise I think is visible there, but it was small. I only tags frames with more than a >0.2 avg rise.
-- I will be able to do that tonight maybe, but seriously... google search is also a calculator.
-- Yes, the baseline is the bottom of the frame. Kind of.
The calculation I used was: Sum of the blue color values (0-255) in the right most 80 pixels of the image subtracted from maximum possible (=255 * 80 * 395) then divided by (255 * 80).
This gave levels of around 94-98-ish, which did not match with what I saw in the images, presumably because not every pixel is perfectly black or white.
Therefore I normalized the results to bring them closer to the actual ranges i saw, so it's not totally accurate but I think it's close enough.