I remember a past lecture where it was mentioned that actually calculating light fall-off to the order of ~1/r^2 resulted in very dim and unnatural-feeling lighting (likely due to ambient reflection/lighting that was unaccounted for). Is the modified empirical drop-off of ~1/r still considered here for computer graphics, and if so, are we considering other forms of light reflection and ambient sources or just assuming them to be negligible?
waleedlatif1
I found this concept of Irradiance falloff particularly interesting, and while reading more about it I was introduced to the concept of vignetting, which involves darkening the edges of the frame. Is there a reason that humans are inclined to increase the vignette on their images or apply this filter to images after they are taken? Moreover, thinking about the difference in vignetting during the day and during night left me wondering, can we apply the same drop-off formulas during the day and the night if the image is already significantly darker at night?
joeyzhao123
An interesting thing IMO is how this relates to other stars in the galaxy and how we get light on earth from stars millions of lightyears away. I think it used the same equation for calculation too.
saltyminty
@Shruteek My understanding is that properly accounting for reflection/other effects produces similar effects to that of a 1/r dropoff, so in many cases its easier to just use 1/r as an approximation instead of using 1/r^2 and doing all of the other reflection/lighting effect calculations (so I'd presume its one or the other).
kkkhanl
I think this is more generally referred to as the inverse-square law and I read some more about it https://energyeducation.ca/encyclopedia/Inverse_square_law along with an experiment demonstrating it
I remember a past lecture where it was mentioned that actually calculating light fall-off to the order of ~1/r^2 resulted in very dim and unnatural-feeling lighting (likely due to ambient reflection/lighting that was unaccounted for). Is the modified empirical drop-off of ~1/r still considered here for computer graphics, and if so, are we considering other forms of light reflection and ambient sources or just assuming them to be negligible?
I found this concept of Irradiance falloff particularly interesting, and while reading more about it I was introduced to the concept of vignetting, which involves darkening the edges of the frame. Is there a reason that humans are inclined to increase the vignette on their images or apply this filter to images after they are taken? Moreover, thinking about the difference in vignetting during the day and during night left me wondering, can we apply the same drop-off formulas during the day and the night if the image is already significantly darker at night?
An interesting thing IMO is how this relates to other stars in the galaxy and how we get light on earth from stars millions of lightyears away. I think it used the same equation for calculation too.
@Shruteek My understanding is that properly accounting for reflection/other effects produces similar effects to that of a 1/r dropoff, so in many cases its easier to just use 1/r as an approximation instead of using 1/r^2 and doing all of the other reflection/lighting effect calculations (so I'd presume its one or the other).
I think this is more generally referred to as the inverse-square law and I read some more about it https://energyeducation.ca/encyclopedia/Inverse_square_law along with an experiment demonstrating it