Difference between fixed point and floating point DSP

DSP Topics
Post Reply
SukhdeepMankoo
Lieutenant
Lieutenant
Posts: 92
Joined: Tue Oct 27, 2009 7:50 pm

Difference between fixed point and floating point DSP

Post by SukhdeepMankoo » Thu Apr 15, 2010 3:58 pm

Hi,
What is the difference between fixed point and floating point DSP.
Thanks
User avatar
Neo
Site Admin
Site Admin
Posts: 2642
Joined: Wed Jul 15, 2009 2:07 am
Location: Colombo

Re: Difference between fixed point and floating point DSP

Post by Neo » Thu Apr 15, 2010 7:15 pm

A fixed point DSP comes with one or more fixed point ALUs. All floating point calculations are re-arranged by compiler to work in fixed point ALU(s) which is considered very inefficient.

A floating point processor usually has two or more ALUs including one or few floating point ones. Executing floating-point calculations are extremely fast when compared to Fixed point processors.

Fixed point processor -> Software floating-point operations
Floating-point processor -> Hardware floating-point operations


See C6713 (Floating-point DSP) has following functional units.
Eight Independent Functional Units:
  • Two ALUs (Fixed-Point)
  • Four ALUs (Floating- and Fixed-Point)
  • Two Multipliers (Floating- and Fixed-Point)

See C6413 (Fixed-point DSP) has following functional units.
Eight Highly Independent Functional Units With VelociTI.2™ Extensions:
  • Six ALUs (32-/40-Bit), Each Supports Single 32-Bit, Dual 16-Bit, or Quad 8-Bit Arithmetic per Clock Cycle
  • Two Multipliers Support Four 16 x 16-Bit Multiplies (32-Bit Results) per Clock Cycle or Eight 8 x 8-Bit Multiplies (16-Bit Results) per Clock Cycle
SukhdeepMankoo
Lieutenant
Lieutenant
Posts: 92
Joined: Tue Oct 27, 2009 7:50 pm

Re: Difference between fixed point and floating point DSP

Post by SukhdeepMankoo » Mon Apr 19, 2010 3:40 pm

Thanks a lot Neo.
Post Reply

Return to “Digital Signal Processors”