Revision as of 03:17, 4 February 2014 edit J caraball (talk \| contribs) 89 edits No edit summary ← Previous edit		Revision as of 12:46, 4 February 2014 edit undo SGGH (talk \| contribs) 49,689 edits m →Floating Point Capacity: clean up and gen fixes, typo(s) fixed: smaller then → smaller than using AWB Next edit →
Line 29: == Floating Point Capacity == Its 64-bit floating-point unit, and 128-bit VMX unit, can perform a theoretical 12 floating-point operations per cycle, as all IBM PowerPC floating-point units can do floating-point mulitply-adds, and come no smaller ~~then~~than 64-bits. That gives 3.2 billion clock cycles * 12 = 38.4 billion floating-point operations/second. ==References==

Power Processing Element: Difference between revisions