02-27-2024, 12:51 AM
If you are using any image precision higher than 8-bit, the export PNG is 16bpc by default.
However you would expect this to only double the size of the PNG. But the PNG compression can only compress neighboring pixels of the exact same channel value, and with 16-bit or floating point the probability that these are equal is somewhat smaller so high precision images aren't as much compressed.
However you would expect this to only double the size of the PNG. But the PNG compression can only compress neighboring pixels of the exact same channel value, and with 16-bit or floating point the probability that these are equal is somewhat smaller so high precision images aren't as much compressed.