This list is for discussion of the design and implementation of field-programmable gate array based processors and integrated systems. It is also for discussion and community support of the XSOC Project (see http://www.fpgacpu.org/xsoc).
|
Hello all, I am trying to use an IP-CORE Virtex multiplier block for my design. The CORE GEN is targetted to a Virtex-2 FPGA, has an output latency of 4 cycles, uses triangular packing, has input widths of 16 bits(signed), and an output width of 32 bits(signed). I obtain a maximum clocking frequency of 164.853 Mhz on instantiating a single multiplier. On instantiating a bank of 8 multipliers (in parallel), the max. clocking frequency drops to 83.8 Mhz. 1).Why is there such a large drop in the max. clocking frequency? 2).Is there any special way for generating multiple instances of a IP core or can it be treated like any other component instantiation? 3). Should any constraints be placed while generating multiple instances of IP-cores? Thanks Vikram |