LPM_mult

Max+Plus II Help Notes


If LPM_WIDTHP is too small, the LSBs of a*b + sum will be discarded. (Reference lpm_mult.tdf library file).

LPM_WIDTHP = MAX(LPM_WIDTHA+LPM_WIDTHB, LPM_WIDTHS)

If pipelined, instantiate "clock", otherwise leave for default.

Error is strange when instatiation assignment arrow is backwards (error message will be included later).


Resource Usage Tables

-- lpm_mult instatiated with lpm_ff as input and output registers
Notes Width Pipelines Optimization Logic Cells Speed
Constant input, Sum input 16-bit 1 5 75=6% 69.5ns
" 16-bit 1 6 86=7% 64.8ns
" 16-bit 1 10 91=7% 52.9ns
Constant input 16-bit 1 10 30=2% 29.7ns
" " 1 0 24=2% 27.3ns
16 & 8-bit inputs 16-bit 0 0 345=29% 92.4ns
" " 0 0, fast 255 42.3ns
" " 1 " 256 24.6ns
" " 3* " 260 15.8ns
16 & 19-bit inputs 18-bit output 0 0 843 124.7ns
" " 4* 0 860 82.2ns
" " 4* 0/10,fast 629 19.0ns
18 and 19-bit inputs 20 bit output 4 0,fast 743 27.2ns
18 and 19-bit inputs 20 bit output 5* 0,fast 771 18.6ns


Note: Output width specified as the sum of the input port widths unless noted.

Sum input width = Output width

* indicates optimum performance/utilization as given by Altera compiler Info.

Return to LPM hints


Last Updated March 25, 1999 by Crystal Jorundson, 243412