A bit of context.
I know that, in a way, x86-64 architectures use CISC, “Complex Instruction Set”, which means it has more commands than ARM, which is RISC, “Reduced Instruction Set”. In a way, CISC is “halfway” into adding more functions to the language, though Assembly is still significantly less human readable than C
SectorLISP and SectorForth are interesting ideas that left me wondering, is it possible to make an architecture whose bare metal language is easier to understand and closer to higher level languages, rather than some variation of Assembly?
Thank you, everyone, for the answers. Had no idea it was tried before, nor that it turned to be very far from a good solution. Today I Learned
If you write a program once and it executes millions of times, which process of the two is more important? The writing of the program or the execution of the program? Which should be optimized?
The answer to your question is it’s possible and has been done, but it’s a bad idea.
Photonic tensor CPUs and analogue iterative machines show a lot of promise. However I think human assisted AI or even AI alone are more likely to take over Assembly level programming before photonic computing has that much of an impact
Intel sorta tried, twice. iapx and itanium come to mind. Two attempts to kill x86 :)
