3 ms·

Anyone up to count sprinkles? Let's see if this simulation worked in real life!

• sargstuff 4y ago
  Are simulation results limited to data not ingested? For non-automated, real life counts, would be good to have the standard deviation of ingested data / time to arrive at total count.
• rob74 4y ago
  I'm just imagining that someone really did this and after all that effort realized how spectacularly off the value they got is, then proceeded to doubt and redo their calculations, only to later realize that the method is at fault...
• glapworth 4y ago
  I love this, but I thought that the sprinkles need to be the same length in order for the calculation to work? Looking at the image in the article, the length of the sprinkles seem to vary with some being just dots and the others are of a similar length. I wonder if it would work if you discounted those smaller dot sprinkles?
  • sargstuff 4y ago
    Think the follow up article addresses the sprinkle variations using lambda calculus.
• eurasiantiger 4y ago
  Just take a better picture and code it up
  • sargstuff 4y ago
    Way to much time/effort ;-) 1) use sprinkles of all one color shade (other than frosting color). 2) take picture with filter color of sprinkles. 3) either a or b a) print out the picture so that sprinkle spots allow light to shine through & non-sprinkle area blocks light shinning through. b) show picture full screen. 4) Use lens to focus light from 3a or 3b onto photo diode (generate sound or light intensity) 5) Figure out a chart showing the average size / number of sprinkles based on light intensity / generated sound. chart calibration test case: use tape/paper to block more and more of a cake size screen holes in order to calibrate the light reduction ratio. aka 3 meters of sprinkles gives 90% reduction in light for given cake size. ~1 screen hole (1mm average sprinkle size) over 3m -> 3000 sprinkles.