Trig version of GLCD routine closes circles

[Anonymous]

I had rather supposed that doing circles parametrically would automatically create closed circles, and it does.

Here's the code for my initial trial. I haven't generalized it yet, nor worked in the rounding and optimization for speed, but the results are very good. This demo simply prints a single, centered circle of radius 20. I'm using sine and cosine values to two decimal places.

See what you think. It runs on the simulator and the real LCD nicely.

;----- Configuration

#chip 16F88, 8                  ;PIC16F88 running at 8 MHz
#config mclr=off                ;reset handled internally
#config osc=int                 ;use internal clock

#include <GLCD.h>

;----- Constants

;Pinout is shown for the LCM12864H-FSB-FBW
;graphical LCD available from Amazon.

;       +5V                 ;LCD pin 1
;       ground              ;LCD pin 2
;       Vo = wiper of pot   ;LCD pin 3
#define GLCD_DB0 PORTB.0    ;LCD pin 4
#define GLCD_DB1 PORTB.1    ;LCD pin 5
#define GLCD_DB2 PORTB.2    ;LCD pin 6
#define GLCD_DB3 PORTB.3    ;LCD pin 7
#define GLCD_DB4 PORTB.4    ;LCD pin 8
#define GLCD_DB5 PORTB.5    ;LCD pin 9
#define GLCD_DB6 PORTB.6    ;LCD pin 10
#define GLCD_DB7 PORTB.7    ;LCD pin 11
#define GLCD_CS2 PORTA.0    ;LCD pin 12
#define GLCD_CS1 PORTA.1    ;LCD pin 13
#define GLCD_RESET PORTA.2  ;LCD pin 14
#define GLCD_RW PORTA.3     ;LCD pin 15
#define GLCD_RS PORTA.4     ;LCD pin 16
#define GLCD_ENABLE PORTA.6 ;LCD pin 17
;       Vee = pot low side  ;LCD pin 18
;       backlight anode     ;LCD pin 19
;       backlight cathode   ;LCD pin 20

#define GLCD_WIDTH 128
#define GLCD_HEIGHT 64

;----- Variables

dim index, tabVal as byte
dim i as word
dim sign as integer

;----- Program

InitGLCD
GLCDCLS

for i = 0 to 359
  cx = 63 - (20 * cos(i))/100
  cy = 31 - (20 * sin(i))/100

  Pset(cx, cy, on)
next i

;----- Subroutines

function ref(in arg1 as integer) as integer
  ;create reference angle (0 to 90) for the argument

  if (arg1 > 270) then          ;Quadrant IV
    ref = 360 - arg1
  else
    if (arg1 > 180) then        ;Quadrant III
      ref = arg1 - 180
    else                        ;Quadrant II
      if (arg1 > 90) then
        ref = 180 - arg1
      else
        ref = arg1              ;Quadrant I by default
      end if
    end if
  end if
end function

;-----

function sin(in arg2 as integer) as integer
  ;get sine of angle

  sign = 1                      ;assume positive first

  if arg2 > 180 then
    sign = -1                   ;negative in III and IV
  end if

  arg2 = ref(arg2)              ;get the reference angle

  index = [byte]arg2+1          ;index into the table
  readTable sineTab, index, tabVal
  sin = [integer]tabVal * sign
end function

;-----

function cos(in arg2 as integer) as integer
  ;get cosine of angle

  sign = 1                      ;assume result is positive
  if arg2>90 and arg2<270 then
    sign = -1                   ;but negative in II and III
  end if

  arg2 = ref(arg2)              ;get the reference angle
  arg2 = 90 - arg2              ;use cofunction identity
  index = [byte]arg2+1          ;index into the table

  readTable sineTab, index, tabVal
  cos = [integer]tabVal * sign  ;factor in sign
end function

;----- Data

table sineTab store data
  ;Sine data for 0 through 90 degrees, scaled up by 100.

  0
  2
  3
  5
  7
  9
  10
  12
  14
  16
  17
  19
  21
  22
  24
  26
  28
  29
  31
  33
  34
  36
  37
  39
  41
  42
  44
  45
  47
  48
  50
  52
  53
  54
  56
  57
  59
  60
  62
  63
  64
  66
  67
  68
  69
  71
  72
  73
  74
  75
  77
  78
  79
  80
  81
  82
  83
  84
  85
  86
  87
  87
  88
  89
  90
  91
  91
  92
  93
  93
  94
  95
  95
  96
  96
  97
  97
  97
  98
  98
  98
  99
  99
  99
  99
  100
  100
  100
  100
  100
  100
end table

[Anonymous]

Oh, here's a pic.

[Anonymous]

Okay, things are coming along. I worked in the rounding now, and some error detection to keep within the bounds of the screen, and have also made a filled circle command. See the attached figure.

;Thomas Henry -- 4/16/2014

;----- Configuration

#chip 16F88, 8                  ;PIC16F88 running at 8 MHz
#config mclr=off                ;reset handled internally
#config osc=int                 ;use internal clock

#include <GLCD.h>

;----- Constants

;Pinout is shown for the LCM12864H-FSB-FBW
;graphical LCD available from Amazon.

;       +5V                 ;LCD pin 1
;       ground              ;LCD pin 2
;       Vo = wiper of pot   ;LCD pin 3
#define GLCD_DB0 PORTB.0    ;LCD pin 4
#define GLCD_DB1 PORTB.1    ;LCD pin 5
#define GLCD_DB2 PORTB.2    ;LCD pin 6
#define GLCD_DB3 PORTB.3    ;LCD pin 7
#define GLCD_DB4 PORTB.4    ;LCD pin 8
#define GLCD_DB5 PORTB.5    ;LCD pin 9
#define GLCD_DB6 PORTB.6    ;LCD pin 10
#define GLCD_DB7 PORTB.7    ;LCD pin 11
#define GLCD_CS2 PORTA.0    ;LCD pin 12
#define GLCD_CS1 PORTA.1    ;LCD pin 13
#define GLCD_RESET PORTA.2  ;LCD pin 14
#define GLCD_RW PORTA.3     ;LCD pin 15
#define GLCD_RS PORTA.4     ;LCD pin 16
#define GLCD_ENABLE PORTA.6 ;LCD pin 17
;       Vee = pot low side  ;LCD pin 18
;       backlight anode     ;LCD pin 19
;       backlight cathode   ;LCD pin 20

#define GLCD_WIDTH 128
#define GLCD_HEIGHT 64

;----- Variables

dim index, tabVal as byte
dim i as word
dim sign as integer

;----- Program

InitGLCD
GLCDCLS

circle(10,10,10)
circle(117,10,10)
filled(63,31,10)
circle(63,31,20)
filled(10,53,10)
filled(117,53,10)

sub filled(cenX, cenY, rad)
  for i = 0 to 358 step 2
    cx = cenX - ((10*rad * cos(i))/100+5)/10
    cy = cenY - ((10*rad * sin(i))/100+5)/10

    edge = 2 * cenX - cx
    for j = cx to edge
    Pset(j,cy,on)
    next j
  next i
end sub

sub circle(cenX, cenY, rad)
  for i = 0 to 358 step 2
    cx = cenX - ((10*rad * cos(i))/100+5)/10
    cy = cenY - ((10*rad * sin(i))/100+5)/10

    if (cx>=0 and cx<=128 and cy>=0 and cy<=64) then
      Pset(cx, cy, on)
    end if  
  next i
end sub

;----- Subroutines

function ref(in arg1 as integer) as integer
  ;create reference angle (0 to 90) for the argument

  if (arg1 > 270) then          ;Quadrant IV
    ref = 360 - arg1
  else
    if (arg1 > 180) then        ;Quadrant III
      ref = arg1 - 180
    else                        ;Quadrant II
      if (arg1 > 90) then
        ref = 180 - arg1
      else
        ref = arg1              ;Quadrant I by default
      end if
    end if
  end if
end function

;-----

function sin(in arg2 as integer) as integer
  ;get sine of angle

  sign = 1                      ;assume positive first

  if arg2 > 180 then
    sign = -1                   ;negative in III and IV
  end if

  arg2 = ref(arg2)              ;get the reference angle

  index = [byte]arg2+1          ;index into the table
  readTable sineTab, index, tabVal
  sin = [integer]tabVal * sign
end function

;-----

function cos(in arg2 as integer) as integer
  ;get cosine of angle

  sign = 1                      ;assume result is positive
  if arg2>90 and arg2<270 then
    sign = -1                   ;but negative in II and III
  end if

  arg2 = ref(arg2)              ;get the reference angle
  arg2 = 90 - arg2              ;use cofunction identity
  index = [byte]arg2+1          ;index into the table

  readTable sineTab, index, tabVal
  cos = [integer]tabVal * sign  ;factor in sign
end function

;----- Data

table sineTab store data
  ;Sine data for 0 through 90 degrees, scaled up by 100.

  0
  2
  3
  5
  7
  9
  10
  12
  14
  16
  17
  19
  21
  22
  24
  26
  28
  29
  31
  33
  34
  36
  37
  39
  41
  42
  44
  45
  47
  48
  50
  52
  53
  54
  56
  57
  59
  60
  62
  63
  64
  66
  67
  68
  69
  71
  72
  73
  74
  75
  77
  78
  79
  80
  81
  82
  83
  84
  85
  86
  87
  87
  88
  89
  90
  91
  91
  92
  93
  93
  94
  95
  95
  96
  96
  97
  97
  97
  98
  98
  98
  99
  99
  99
  99
  100
  100
  100
  100
  100
  100
end table

[MBB]

Thomas,

I tried to run your software (the second post) but it would not compile. It gave me two "Else outside End If block" errors and two "End If without If" errors. I traced this to the 'function ref" section. Since I couldn't find anything wrong with the code I removed the "ELSEs" and then it compiled.

I may have made some mistakes by removing these else statments because all I got on my display was a bunch of horizontal lines.

Any ideas?

[kent_twt4]

Check this line for the Table:

table sineTab store data

Should be:
table sineTab  'store data

[Anonymous]

MBB, unless I did something funny with cut and paste, I can't understand why it doesn't compile for you. Just to make sure, I've attached the program here. And I tested it again just now, and it compiles and runs for me.

Kent, that won't hurt anything to comment out "store data" option, but by leaving it in, the table goes to EEPROM. Commenting it out puts the table in program memory.

[kent_twt4]

Thomas, I was having a similar problem to what MBB had, only I got some random pixels.

My bad on the store data command, hadn't used that before. But "as is" code for me, the compile went O.K. sorta, as PicKit2 complained about the hex file being too big for my 16f1783, which shouldn't be the case. Commented out the "store data", thinking it was in need of commenting (hah), and all good for PicKit2. Output got some semblance of circles, but experiencing detached arcs of approx. 45 degrees.

I'm using Steini's GCB SYNWrite, with 0.9 22/9/2013. What are you using?

[Anonymous]

I'm using 0.9 17/02/2014, which I think is the version in the Hot Release package.

[MBB]

I tried your GLCD.gcb post but I still couldn't compile it.

I'm using version 1.0 21/9/2013.

Something else I had a problem with is this line in your program:

chip 16F88, 8 ;PIC16F88 running at 8 MHz

I get a Microsoft.NET Framework unhandled exception saying

Conversion from string "8 ;PIC1F88 run" to type Double is
not valid

If I delete the comment so I have only: #chip 16F88, 8
then GCB can open the file. This makes no sense.

I had the same problem when I downloaded your Sine program.

[Anonymous]

Hey guys,

I'm sorry to learn that this is causing you grief. I just got a real GLCD and tried it. It sure works for me. (See the attached pic). It really does sound like it's a GCB Version problem. I think Anobium manages to keep the versions straight; maybe he'll weigh in with a comment. Wasn't there a problem with integer arithmetic previously?

I've attached the code I used for the picture. It's very short and sweet, since it now calls upon the trig include file I've described in another thread. And I added some explanatory comments. I'm not sure who is maintaining the GLCD include file, but maybe this would be suitable for inclusion (circle and filled circle commands).

Since MBB seems to be having a PC related problem, I'll mention that I've done this on two separate laptops, both XP, and using both MPASM and GCASM.

I'll be working on some diagonal line and triangle commands next.

;Circle and filled circle commands on a graphic LCD.
;This uses the 2-place trigonometric routines found in
;the include file.

;Thomas Henry -- 4/17/2014

;----- Configuration

#chip 16F88, 8              ;PIC16F88 running at 8 MHz
#config mclr=off            ;reset handled internally
#config osc=int             ;use internal clock

#include <GLCD.h>
#include <Trig2Places.h>

;----- Constants

;Pinout is shown for the LCM12864H-FSB-FBW
;graphical LCD available from Amazon.

;       +5V                 ;LCD pin 1
;       ground              ;LCD pin 2
;       Vo = wiper of pot   ;LCD pin 3
#define GLCD_DB0 PORTB.0    ;LCD pin 4
#define GLCD_DB1 PORTB.1    ;LCD pin 5
#define GLCD_DB2 PORTB.2    ;LCD pin 6
#define GLCD_DB3 PORTB.3    ;LCD pin 7
#define GLCD_DB4 PORTB.4    ;LCD pin 8
#define GLCD_DB5 PORTB.5    ;LCD pin 9
#define GLCD_DB6 PORTB.6    ;LCD pin 10
#define GLCD_DB7 PORTB.7    ;LCD pin 11
#define GLCD_CS2 PORTA.0    ;LCD pin 12
#define GLCD_CS1 PORTA.1    ;LCD pin 13
#define GLCD_RESET PORTA.2  ;LCD pin 14
#define GLCD_RW PORTA.3     ;LCD pin 15
#define GLCD_RS PORTA.4     ;LCD pin 16
#define GLCD_ENABLE PORTA.6 ;LCD pin 17
;       Vee = pot low side  ;LCD pin 18
;       backlight anode     ;LCD pin 19
;       backlight cathode   ;LCD pin 20

#define GLCD_TYPE GLCD_TYPE_KS0108
#define GLCD_WIDTH 128
#define GLCD_HEIGHT 64

;----- Variables

dim cx, cy, edge, j as byte
dim i as word

;----- Program

InitGLCD
GLCDCLS

circle(10,10,10)            ;upper left
circle(117,10,10)           ;upper right
filled(63,31,10)            ;center
circle(63,31,20)            ;center
filled(10,53,10)            ;lower left
filled(117,53,10)           ;lower right

;----- Subroutines

sub circle(cenX, cenY, rad)
  ;Center of circle = (cenX,cenY), radius = rad

  for i = 0 to 358 step 2                 ;every two degrees
    cx = cenX-((10*rad*cos(i))/100+5)/10  ;properly rounded x value
    cy = cenY-((10*rad*sin(i))/100+5)/10  ;properly rounded y value

    ;the following ignores the pixel if off the screen
    if (cx>=0 and cx<=GLCD_WIDTH and cy>=0 and cy<=GLCD_HEIGHT) then
      Pset(cx, cy, on)                    
    end if  
  next i
end sub

sub filled(cenX, cenY, rad)
  ;Center of circle = (cenX,cenY), radius = rad

  for i = 0 to 358 step 2
    cx = cenX -((10*rad*cos(i))/100+5)/10
    cy = cenY -((10*rad*sin(i))/100+5)/10
    edge = 2 * cenX - cx                  ;compute right edge

    for j = cx to edge                    ;fill entire line
      Pset(j,cy,on)
    next j
  next i
end sub

[Anobium]

I have watching.... and reading with interest. Well done. This is excellent work. I will just tried on the new GCLD code - compiles with no issues.

Then, I tried on the Simulator.... no issues. Worked as expected.

Then, the I tried on the new ST7920 GLCD driver code.... no issues. EXCELLENT!!

Great job. (I am please too... this is the first piece of external code to be tested on the ST7920 uber cheapo GLCD device...! It was £4.00 GBP)

My thoughts on the @MBB issues. The release from Sept 13 did have some stack issues where the stack was not handled correctly. I would advise using the Hot Release. Just backup you existing config. FYI: When I was using the Sept 13 release I was getting very odd results during runtime (when the Pic is running) with stack errors - this would show up as corruption on.... the LCD......

[kent_twt4]

I want to apologize to everyone for reporting some problems which were of my own doing. I'm going to sit in the corner with the dunce cap one.

When I moved the GLCD data lines over to PortC, so as not to conflict with programming, I had crossed two adjacent data wires, no wonder I was seeing goofy results.

I did upgrade to the newest Hot release, but may not have made any difference.

With the AGM1264F: