The compiler it was written for is MikroC if that is any help.
Max.

 

I am assuming this is just simple 32 bit arithmetic functions.?
Max.

; 
unsigned char rpos
unsigned long bres

// 2. updates reference freq generator, records its position
  bres += 102400;                 ; add nS per interrupt period (512 insts * 200nS)
  if(bres >= MOTOR_PULSE_PERIOD)  ; if reached a new reference step
  {
    bres -= MOTOR_PULSE_PERIOD;
    rpos++;                       ; record new xtal-locked reference position
  }

 

Yes.

 

The compiler it was written for is MikroC if that is any help.
Max.

So are you trying to optimize your program, port it to another platform, or what?

 

Max, what is wrong with using the existing program and XC8 (free)? The changes for the compiler will be minimal, possibly limited to setting up the some lines for the config fuses and the program details what the values should be.

 

What would the assembly version of this be?
Max.

;  2. updates reference freq generator, records its position
  bres += 102400;                 ; add nS per interrupt period (512 insts * 200nS)
  if(bres >= MOTOR_PULSE_PERIOD)  ; if reached a new reference step
  {
    bres -= MOTOR_PULSE_PERIOD;
    rpos++;                       ; record new xtal-locked reference position
  }

Nothing esoteric. Perhaps some of the C operators confuse.
a += b is the same as a = a + b,
a -= b is the same as a = a - b
a++ is the same as a = a + 1

I think you've got the comparisons figured already.

I do find the use of bresenham's line drawing algorithm somewhat clever in it's simplicity but I think a simple PID approach could be better (smoother) though probably a lot more code. As it is, his approach is basically only I as he controls the error accumulation.

 

Anyone done any motor control design using Roman's methods here https://www.romanblack.com/onesec/DCmotor_xtal.htm
I am toying with the idea of variable control, but have to convert his program to Assembly if possible.
(was a forum member The_RB).
Max.

Wow... this guy is using Bresenham's algorithm to derive different frequencies from a master one! ... it's genius! ... Why didn't I think of that before? ... this could've come in handy in a previous project of mine. I'm gonna start studying his code and see if I can adapt it to my purposes.

Thanks for posting, Max.

 

So are you trying to optimize your program, port it to another platform, or what?

Intending to convert to assembler.
Max.

 

Wow... this guy is using Bresenham's algorithm to derive different frequencies from a master one! ... it's genius!
Thanks for posting, Max.

Yes I thought it neat, got Roman to thank also.
Max.

 

Bresenham 's basic algorithm has a lot of applications. The use of accumulated error to get within 1 discreet point via add/subtract is brilliant in it's simplicity. I use it to do temporal transitions between colors. A compare and 2 adds per step is all it takes. Sadly, he never got a dime for it.

 

Sadly, he never got a dime for it.

Yeah, but his name's gone down in history... now, if I could just come up with a Martinez' algorithm to be remembered by...

 

I think this application could be made easier by replacing the 16F628a with a INT on-input type, and even go to the 18F series and replace the quadrature reading head method.
Also the possibility of making it more flexible in setting various RPM's.
Max.