Optimization of Multiplication in FPGA

Jun 23, 2006 2 Replies
A

hi, there



in my project, i need to do some multiplication. and i optimized the multiplicand to numbers like 0.75, 0.625. I wonder would the synthesizer automatically detect them like 0.75 = 1 - 1/4, 0.625 = 1/2



  • 1/8, or i need to code it manually in this way?

does it differ across different synthesizer? i am using xilinx synthesizer.



thanks Zhaoyi



Vote
Twitter Facebook LinkedIn
F

agou schrieb:

You need to code "semi" manual anyway, since most synthesizers do not support floting point multiplications. So when you make a x 0.75 multiplication, you need to multiply by 192 when you need 8 bit precision. A multiplication with an constant is cheap anyway.

Regards Falk

Vote
Twitter Facebook LinkedIn

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required