Squidge said:Well, if the MMSP2+ is chosen, you'll be able to upload your fpu code and arguments to the processor, it'll run through the code and give you the result. You got the usual stuff like floating point add / divide / multiply / reciprocal / square root / etc.A_SN said:Here's what I don't get, what will we be able to do with that pseudo FPU we'll get?
Are you seriously considering this "way" as a valid way to "emulate" FPU operations ? that would be too slow anyway ! and will conflict with Open GL which also needs to compute with this "embedded" FPU in GPU.
For float operations, a true FPU is only the right one to have or better a VFPU9 would probably be the best. Any other tricks are just a pain in the ass (compiler are not aware of them) and I bet most developers wouldn't even know how to use it efficiently.
Just think about VFPU on PSP, it is actually not used by most of homebrews because the homebrew version of gcc doesn't even handle this coprocessor. FPU or VFP9 on ARM9 are handled by gcc.
EDIT:
OMG ! page 90 ! I didn't read 2/3 of all pages... :/ sorry this post may sound old...
Last edited by a moderator: