|
|
|
|
|
|
by seoulbigchris
588 days ago
|
|
|
Yeah, I get that. Existing hardware supports double-width integers for multiplication (`u16 = u8 * u8`, etc ). But for addition, typical ALUs don't support oddball widths like (`u9 = u8 + u8`, etc). And then what do you about chaining calculations? Sum up a huge column of numbers that way and you could get a u41593 value. Maybe this could be solved if there was a concept of a floating width integer, which is on paper just a special case of floating point. Maybe. My remarks have been about integers -- the typical programming language floating point rules (`f32 = f32 * f32`, etc) make sense to me. EDIT: Markdown editor didn't like the asterisk in my example equation. |
|
|