Fixed Point Arithmetic
vha3.github.io
Fixed Point Arithmetic
1–10 of 61 posts
Re: Fixed Point Arithmetic
#2Prepared as a lecture supplement for ECE 4760 (Digital System Design Using Microcontrollers) at Cornell.
Re: Fixed Point Arithmetic
#3Fixed point was still normal instruction in mandatory classes when I was in school. It is a pity, but I feel like we stopped teaching it at some point.
Re: Fixed Point Arithmetic
#4That would be easier with Forth :).
Re: Fixed Point Arithmetic
#5Fixed point is critical reading for a lot of FPGA work.
Re: Fixed Point Arithmetic
#6Fixed point is critical reading for a lot of FPGA work.
For those of us unfamiliar with FPGA work, can you elaborate?
Re: Fixed Point Arithmetic
#7I used fixed-point for all the 3d code in The Virtual Reality Playhouse (1992). That was the last time I wrote any significant assembler code. Good times.
Re: Fixed Point Arithmetic
#8I'm using fixed point for real-time DSP on cortex-M. Faster than floating.
Using the CMSIS-DSP Q31 functions, in particular. On Rust `i32`s.
Re: Fixed Point Arithmetic
#9I used fixed-point for all the 3d code in The Virtual Reality Playhouse (1992). That was the last time I wrote any significant assembler code. Good times.
By the way ... that was the first VR program one could run at home. I drove liquid crystal classes with the modem control lines. Switching left/right was synchronized to the monitor's vertical retrace.
Even better times. :)
Re: Fixed Point Arithmetic
#10I'm using fixed point for real-time DSP on cortex-M. Faster than floating. Using the CMSIS-DSP Q31 functions, in particular. On Rust `i32`s.
On the PIC32, fixed add is ~2 cycles and floating point is is ~60. And about a 2x increase for multiplies. On architectures with floating point hardware some of the advantage is lost, but it's fantastic for DSP on microcontrollers.