Skip to content

haydenbeames/FastMultiplication

Repository files navigation

FastMultiplication

Fast Multiplication algorithm in SystemVerilog by using 4 to 2 compressors combined with Half adders and CSA (Carry Sum Adders)

8 Bit 4:2 Compression Tree Multiplication Now Functional!!!!!

  • fix cin bug on removal of '{default:'0} for synthesis error

  • Now to make 16 bit, 32 bit and 64 bit : )

  • but first need to make a signed 8 bit multiplier

    • making the 8 bit tree took me over 24 hours of work lmaoo bc stupid bugs and complicated tree
        • on the plus side, the code is much more readable than other trees I have seen online
  • Of several 4:2 compression techniques, this should be the most optimal routing to ensure fastest result, however, I acknowledge that modern Synthesis tools may be better at optimization

  • at end of multiplication tree, just use CPA (Carry Propogate Adder) to get final results

Currently exploring Hybrid 4:2 Compression tree with 3:2 CSA for Booth Recoding multiplication

About

Fast Multiplication algorithm in SystemVerilog by using 4 to 2 compressors

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published