Post ApMEPImgFOvqTR88Ei by slembcke@mastodon.gamedev.place
 (DIR) More posts by slembcke@mastodon.gamedev.place
 (DIR) Post #ApLyOhMxkwm7AQn7cO by RL_Dane@fosstodon.org
       2024-12-23T23:49:16Z
       
       0 likes, 1 repeats
       
       Reading through "Introducing Go," by Caleb Doxsey.I'm surprised that Go only supports 32 and 64-bit floats. I remember in the olden days, computers (without FPUs) had *crazy*-huge floating point numbers. I'm guessing if you're going to slog through doing it all in software, you might as well go whole hog.Oh found it! It was called SANE: Standard Apple Numerics Environment. It supported up to 80 bit floats, and that's on the 6502!!#Go #Golang
       
 (DIR) Post #ApLyyOh0fmjjfSBsbg by theorytoe@ak.kyaruc.moe
       2024-12-24T00:12:03.821029Z
       
       0 likes, 0 repeats
       
       @RL_Dane math/big is basically arbitrary-precision numbers
       
 (DIR) Post #ApMEPImgFOvqTR88Ei by slembcke@mastodon.gamedev.place
       2024-12-24T02:34:57Z
       
       1 likes, 0 repeats
       
       @RL_Dane The x87 floating point coprocessor did 80 bit as well. It’s still officially supported in modern x86 w/o the external x87, but kinda slow I think. It’s what you get with “long double” in C for instance.