Speed optimization (framebuffer scrolling)

I'm working on my MAX7219 display code and I'm in optimization mode. Have managed to get from 22ms to 6.5ms for redrawing a screen of 81 matrices (at 2MHz SPI clock) and I don't think I can squeeze more performance out of that particular section of the code at this moment. Next in line is the scrolling of the framebuffer, here's the code for it:

Code:

---

NumberOfDisplays  CON 36
FrameBufferSize  CON NumberOfDisplays * 10 + 20

FrameBuffer      VAR BYTE[FrameBufferSize]
Row              VAR BYTE
Col              VAR WORD
Offset            VAR WORD
FrameBufferByte  VAR BYTE
MAX7219_Value    VAR BYTE


ScrollLeft:
    ' Shifts content of the screenbuffer one column to the left. It does not redraw the display.
    ' Input: None
    ' Output: None   
    ' This version: 3.5ms for 36 displays @64MHz

    FOR Col = 0 to NumberOfDisplays + 1                        ' 8 columns on left and right side, outside of display area
        For Row = 0 to 9                                                  ' One invisible row top and bottom of display area to facilitate vertical scrolling
            Offset = Row * (NumberOfDisplays + 2) + Col
            MAX7219_Value = FrameBuffer[Offset] >> 1

            ' As long as the next byte is NOT the last byte in the row:
            IF Col < (NumberOfDisplays + 1) THEN
              FrameBufferByte = FrameBuffer[Offset + 1]
              MAX7219_Value.7 = FrameBufferByte.0
            ENDIF

            FrameBuffer[Offset] = MAX7219_Value
        NEXT
    NEXT
RETURN

---

Any ideas on how to make this run faster?

Without getting into asm or anything, how about this (untested):

Code:

---

OFFSET_INCR CON NumberOfDisplays + 2
ScrollLeft:
    ' Shifts content of the screenbuffer one column to the left. It does not redraw the display.
    ' Input: None
    ' Output: None   
    ' This version: 1.30ms for 36 displays @64MHz

    ' do for Col < NumberOfDisplays+1
    FOR Col = 0 to NumberOfDisplays                        ' 8 columns on left and right side, outside of display area
        Offset = Col
        For Row = 0 to 9                                                  ' One invisible row top and bottom of display area to facilitate vertical scrolling
            'Offset = Row * (NumberOfDisplays + 2) + Col
            MAX7219_Value = FrameBuffer[Offset] >> 1

            FrameBufferByte = FrameBuffer[Offset + 1]
            MAX7219_Value.7 = FrameBufferByte.0

            FrameBuffer[Offset] = MAX7219_Value
            Offset = Offset + OFFSET_INCR          ' basically Row * (NumberOfDisplays + 2)
        NEXT
    NEXT

    Col = NumberOfDisplays + 1                        ' 8 columns on left and right side, outside of display area
    Offset = Col
    For Row = 0 to 9                                                  ' One invisible row top and bottom of display area to facilitate vertical scrolling
        'Offset = Row * (NumberOfDisplays + 2) + Col
        MAX7219_Value = FrameBuffer[Offset] >> 1
        FrameBuffer[Offset] = MAX7219_Value
        Offset = Offset + OFFSET_INCR          ' basically Row * (NumberOfDisplays + 2)
    NEXT
RETURN

---

That cuts it down to 1.3ms

try some asm

untested but should be close
bcnt var byte bank 1
dcnt var byte bank 1
dspbuff var byte[380] ;36 displays + 2 outside of display * 10

Code:

---

rotate_r:  ' pic18
asm
    banksel _bcnt    ;ROW
    movlw  10      ;10 rows 
    movwf  _bcnt
    clrf    FSR0H
    movlw  low  (_dspbuff)
    movwf  FSR0L
    movlw  111        ;BUFFER SIZE low byte  (380-1) 36 displays + 2 outside of display * 10-1
    ADDWF  FSR0L,F
    movlw  high (_dspbuff)
    ADDWFC  FSR0H
    movlw  1        ;BUFFER SIZE high byte  (380)
    ADDWF  FSR0H
    banksel _bcnt    ;note place bcnt and dcnt in same bank
    MOVLW 38    ;36 displays + 2 outside of display 
NROW     
    movwf  _dcnt
    bcf    STATUS, C
Ncol
    rrcf  POSTDEC0 ,f    ;PER COLUMN
    DECFSZ  _dcnt ,F
    BRA    Ncol
    BNC    NBNC
    bsf    PLUSW0,7    ;max displays would be 127 to use plusw
NBNC
    DECFSZ  _bcnt ,F
    BRA    NROW
    banksel 0
endasm

---

this is my 4 panel code for reference

Code:

---


rotate: ;pic18
asm
    banksel _bcnt    ;ROW
    movlw  8
    movwf  _bcnt
    clrf    FSR0H
    movlw  low  (_dspbuff)
    movwf  FSR0L
    movlw  31        ;BUFFER SIZE
    ADDWF  FSR0L,F
    movlw  high (_dspbuff)
    ADDWFC  FSR0H
    banksel _bcnt
    MOVLW 4
NROW   
    bcf    STATUS, C
    rrcf  POSTDEC0 ,f    ;PER COLUMN
    rrcf  POSTDEC0 ,f    ;PER COLUMN
    rrcf  POSTDEC0 ,f    ;PER COLUMN
    rrcf  POSTDEC0 ,f    ;PER COLUMN
    BNC    NBNC
    bsf    PLUSW0,7
NBNC
    DECFSZ  _bcnt ,F
    BRA    NROW
endasm

---

Richard, thank you. My problem is I suck as ASM and I don't understand half of what's going on there. It might very well work, in which case I'd be lucky but if it doesn't or I need to modify it I basically have no idea what I'm doing. I should use it as a learning exercise though as I do feel the need to better understand ASM.

tumblweed, wow, that was an easy yet effective modification (and it works perfectly fine) - just what I was hoping for.
Went down from 7.6ms to just above 3ms for a 81 matrix wide display. More than double the speed, cool!

One thing threw me off on a tangent for a while...

Code:

---

OFFSET_INCR CON NumberOfDisplays + 2

---

Since NumberOfDisplays is already a CONstant (with the value 81 in this particular case) I replaced

Code:

---

Offset = Offset + OFFSET_INCR

---

with

Code:

---

Offset = Offset + NumberOfDisplays + 2

---

expecting the compiler to replace both versions to

Code:

---

Offset = Offset + 83

---

It did not...your version with, the constant defined executes at 3.03ms, the other one in 3.45ms

I took 15 minutes for the penny to drop and now I need to go thru my code and see where else I've made that silly mistake.

Thanks again, both of you guys!

/Henrik.

Quote:

---

I should use it as a learning exercise though as I do feel the need to better understand ASM.

---

its worth the effort

this just under 100uS for a iteration

Code:

---

'****************************************************************
'*  Name    : UNTITLED.BAS                                      *
'*  Author  : richard                                          *
'*  Notice  : Copyright (c) 2022 caveat emptor                  *
'*          : All Rights Reserved                              *
'*  Date    : 11/10/2022                                        *
'*  Version : 1.0                                              *
'*  Notes  :                                                  *
'*          :                                                  *
'****************************************************************
#CONFIG
  CONFIG  FOSC = INTIO67
  CONFIG  PLLCFG = ON
  CONFIG  PRICLKEN = ON
  CONFIG  FCMEN = OFF
  CONFIG  IESO = OFF
  CONFIG  PWRTEN = ON
  CONFIG  BOREN = SBORDIS
  CONFIG  BORV = 190
  CONFIG  WDTEN = OFF
  CONFIG  WDTPS = 32768
  CONFIG  CCP2MX = PORTC1
  CONFIG  PBADEN = OFF
  CONFIG  CCP3MX = PORTB5
  CONFIG  T3CMX = PORTC0
  CONFIG  HFOFST = ON
  CONFIG  P2BMX = PORTB5
  CONFIG  MCLRE = EXTMCLR
  CONFIG  STVREN = ON
  CONFIG  LVP = OFF
  CONFIG  XINST = OFF
  CONFIG  DEBUG = OFF
  CONFIG  CP0 = OFF
  CONFIG  CP1 = OFF
  CONFIG  CP2 = OFF
  CONFIG  CP3 = OFF
  CONFIG  CPB = OFF
  CONFIG  CPD = OFF
  CONFIG  WRT0 = OFF
  CONFIG  WRT1 = OFF
  CONFIG  WRT2 = OFF
  CONFIG  WRT3 = OFF
  CONFIG  WRTC = OFF
  CONFIG  WRTB = OFF
  CONFIG  WRTD = OFF
  CONFIG  EBTR0 = OFF
  CONFIG  EBTR1 = OFF
  CONFIG  EBTR2 = OFF
  CONFIG  EBTR3 = OFF
  CONFIG  EBTRB = OFF
#ENDCONFIG
   




    DEFINE OSC 64
    DEFINE DEBUG_REG PORTB
    DEFINE DEBUG_BIT 7
    DEFINE DEBUG_BAUD 9600
    DEFINE DEBUG_MODE 0
    LATB.7=1
    bcnt      VAR BYTE  bank0
    dcnt      VAR BYTE  bank0
    BUFF      VAR BYTE[32]
    iteration  VAR word
    OSCCON=$70
    OSCTUNE.6=1
    while ! osccon2.7 :WEND    ;wait for pll
    ANSELB=0
    ANSELC=0
    ANSELA=0
    dspbuff var byte[380]
    clear
    trisc = %01111111
    TRISB = %01111111
    TRISA = %11111111
    pause 1000
    debug 13,10,"ready",13,10
    LATc.7=0
    ;38x8 bits wide 10 bits high
    dspbuff[379]=$c0
    dspbuff[37]=$c0
main:
LATc.7=1; trigger oscilloscope
gosub rotate_r   
LATc.7=0
debug 13,10,dec iteration
debug 13,10,bin8 dspbuff[379]," " ,bin8 dspbuff[378]," " ,bin8 dspbuff[343] ," ",bin8 dspbuff[342]
debug 13,10,bin8 dspbuff[37] ," " ,bin8 dspbuff[36] ," " ,bin8 dspbuff[1]  ," ",bin8 dspbuff[0]
pause 200
iteration=iteration+1   
goto main   
   
   
   
   
rotate_r:  ' pic18    100uS
asm
    banksel _bcnt    ;ROW
    movlw  10      ;10 rows 
    movwf  _bcnt
    movlw  high (_dspbuff)
    movwf  FSR0H   
    movlw  low  (_dspbuff)
    movwf  FSR0L
    movlw  123        ;BUFFER SIZE low byte  (380-1) 36 displays + 2 outside of display * 10-1
    ADDWF  FSR0L
    movlw  1        ;BUFFER SIZE high byte  (380)
    ADDWFC  FSR0H
    MOVLW 38    ;36 displays + 2 outside of display
NROW 
    movwf  _dcnt
    bcf    STATUS, C
Ncol
    rrcf  POSTDEC0 ,f    ;PER COLUMN
    DECFSZ  _dcnt ,F
    BRA    Ncol
    BNC    NBNC
    bsf    PLUSW0,7    ;max displays would be 127 to use plusw
NBNC
    DECFSZ  _bcnt ,F
    BRA    NROW
    banksel 0
    return
endasm

---

Thanks Richard,
What device did you run this on?
After being unseccessfull in integrating your version into my code I tried to get your code (I removed some oscillator specific stuff to match my device) to compile for the 57Q43 but it won't.
It can't fit bcnt/dcnt into bank 0 so I tried various banks. Bank 5 works but I still get [ASM WARNING: ASM-SCROLL.ASM (411) : Invalid RAM location specified]

Line 411 in the .asm listing is banksel 0 in the NBNC routine.

I know it says warning and not error but the description sounds bad enough :-)

See, this is what I mean....

For reference, here's what I'm trying to compile for the 57Q43:

Code:

---

DEFINE OSC 64
DEFINE DEBUG_REG PORTB
DEFINE DEBUG_BIT 7
DEFINE DEBUG_BAUD 9600
DEFINE DEBUG_MODE 0

LATB.7=1

bcnt      VAR BYTE  bank0
dcnt      VAR BYTE  bank0
iteration  VAR word

ANSELB=0
ANSELC=0
ANSELA=0
dspbuff var byte[380]
clear
trisc = %01111111
TRISB = %01111111
TRISA = %11111111
pause 1000
debug 13,10,"ready",13,10
 LATc.7=0
 ;38x8 bits wide 10 bits high
 dspbuff[379]=$c0
 dspbuff[37]=$c0


main:
LATc.7=1; trigger oscilloscope
gosub rotate_r   
LATc.7=0
debug 13,10,dec iteration
debug 13,10,bin8 dspbuff[379]," " ,bin8 dspbuff[378]," " ,bin8 dspbuff[343] ," ",bin8 dspbuff[342]
debug 13,10,bin8 dspbuff[37] ," " ,bin8 dspbuff[36] ," " ,bin8 dspbuff[1]  ," ",bin8 dspbuff[0]
pause 200
iteration=iteration+1   
goto main   
   
   
rotate_r:  ' pic18    100uS
asm
    banksel _bcnt    ;ROW
    movlw  10      ;10 rows 
    movwf  _bcnt
    movlw  high (_dspbuff)
    movwf  FSR0H   
    movlw  low  (_dspbuff)
    movwf  FSR0L
    movlw  123        ;BUFFER SIZE low byte  (380-1) 36 displays + 2 outside of display * 10-1
    ADDWF  FSR0L
    movlw  1        ;BUFFER SIZE high byte  (380)
    ADDWFC  FSR0H
    MOVLW 38    ;36 displays + 2 outside of display
NROW 
    movwf  _dcnt
    bcf    STATUS, C
Ncol
    rrcf  POSTDEC0 ,f    ;PER COLUMN
    DECFSZ  _dcnt ,F
    BRA    Ncol
    BNC    NBNC
    bsf    PLUSW0,7    ;max displays would be 127 to use plusw
NBNC
    DECFSZ  _bcnt ,F
    BRA    NROW
    banksel 0
    return
endasm

---

18f26k22

i have not tried much with Q43 yet , but
they have no ram below bank 5.
i dont know what bank pbp defaults to for them bank0 makes no sense at all , before them pbp kinda assumes bank0 before and after calls.
the instruction set is slightly different. movfffl or whatever
there may be other weird shit with indirect addressing

i have some q43's i just lack the enthusiasm to learn about them when the tried and trusty k22 serves my needs

try them in bank6
and just remove the banksel 0 at the end


i added a rotate left to and a few tweaks

Code:

---

'****************************************************************'*  Name    : UNTITLED.BAS                                      *
'*  Author  : richard                                          *
'*  Notice  : Copyright (c) 2022 caveat emptor                  *
'*          : All Rights Reserved                              *
'*  Date    : 11/10/2022                                        *
'*  Version : 1.0                                              *
'*  Notes  :                                                  *
'*          : pic18f26k22                                                  *
'****************************************************************
#CONFIG
  CONFIG  FOSC = INTIO67
  CONFIG  PLLCFG = ON
  CONFIG  PRICLKEN = ON
  CONFIG  FCMEN = OFF
  CONFIG  IESO = OFF
  CONFIG  PWRTEN = ON
  CONFIG  BOREN = SBORDIS
  CONFIG  BORV = 190
  CONFIG  WDTEN = OFF
  CONFIG  WDTPS = 32768
  CONFIG  CCP2MX = PORTC1
  CONFIG  PBADEN = OFF
  CONFIG  CCP3MX = PORTB5
  CONFIG  T3CMX = PORTC0
  CONFIG  HFOFST = ON
  CONFIG  P2BMX = PORTB5
  CONFIG  MCLRE = EXTMCLR
  CONFIG  STVREN = ON
  CONFIG  LVP = OFF
  CONFIG  XINST = OFF
  CONFIG  DEBUG = OFF
  CONFIG  CP0 = OFF
  CONFIG  CP1 = OFF
  CONFIG  CP2 = OFF
  CONFIG  CP3 = OFF
  CONFIG  CPB = OFF
  CONFIG  CPD = OFF
  CONFIG  WRT0 = OFF
  CONFIG  WRT1 = OFF
  CONFIG  WRT2 = OFF
  CONFIG  WRT3 = OFF
  CONFIG  WRTC = OFF
  CONFIG  WRTB = OFF
  CONFIG  WRTD = OFF
  CONFIG  EBTR0 = OFF
  CONFIG  EBTR1 = OFF
  CONFIG  EBTR2 = OFF
  CONFIG  EBTR3 = OFF
  CONFIG  EBTRB = OFF
#ENDCONFIG
   


    NumberOfDisplays  CON 36
    NumberOfRows  CON 10
    DEFINE OSC 64
    DEFINE DEBUG_REG PORTB
    DEFINE DEBUG_BIT 7
    DEFINE DEBUG_BAUD 9600
    DEFINE DEBUG_MODE 0
    LATB.7=1
    bcnt      VAR BYTE  bank0
    dcnt      VAR BYTE  bank0
    BUFF      VAR BYTE[32]
    iteration  VAR word
    OSCCON=$70
    OSCTUNE.6=1
    while ! osccon2.7 :WEND    ;wait for pll
    ANSELB=0
    ANSELC=0
    ANSELA=0
    dspbuff var byte[380]
    clear
    trisc = %01111111
    TRISB = %01111111
    TRISA = %11111111
    pause 1000
    debug 13,10,"ready",13,10
    LATc.7=0
    ;38x8 bits wide 10 bits high


main:
    iteration=0
    dspbuff[379]=3
    dspbuff[37]=3
    WHILE  iteration<610
        LATc.7=1
        gosub rotate_l
        LATc.7=0
        debug 13,10,dec iteration
        debug 13,10,bin8 dspbuff[379]," " ,bin8 dspbuff[378]," " ,bin8 dspbuff[343] ," ",bin8 dspbuff[342]
        debug 13,10,bin8 dspbuff[37] ," " ,bin8 dspbuff[36] ," " ,bin8 dspbuff[1]  ," ",bin8 dspbuff[0]
        pause 50
        iteration=iteration+1
    WEND
    iteration=0
    dspbuff[379]=$c0
    dspbuff[37]=$c0
    WHILE  iteration<610
        LATc.7=1
        gosub rotate_R
        LATc.7=0
        debug 13,10,dec iteration
        debug 13,10,bin8 dspbuff[379]," " ,bin8 dspbuff[378]," " ,bin8 dspbuff[343] ," ",bin8 dspbuff[342]
        debug 13,10,bin8 dspbuff[37] ," " ,bin8 dspbuff[36] ," " ,bin8 dspbuff[1]  ," ",bin8 dspbuff[0]
        pause 50
        iteration=iteration+1
    WEND
goto main   
   
   
   
   
rotate_r:  ' pic18
asm
    banksel _bcnt  ;ROW
    movlw  _NumberOfRows    ;10 rows 
    movwf  _bcnt
    movlw  high (_dspbuff)  ;BOTTOM OF BUFFER
    movwf  FSR0H   
    movlw  low  (_dspbuff)
    movwf  FSR0L
    movlw  low((_NumberOfDisplays + 2)*_NumberOfRows-1) ;TOP OF BUFFER
    ADDWF  FSR0L
    movlw  high((_NumberOfDisplays + 2)*_NumberOfRows-1)   
    ADDWFC  FSR0H
    MOVLW  _NumberOfDisplays + 2;36 displays + 2 outside of display
RROW 
    movwf  _dcnt        ;display
    bcf    STATUS, C
Rcol
    rrcf    POSTDEC0 ,f  ;PER display
    DECFSZ  _dcnt ,F
    BRA    Rcol
    BNC    RBNC
    bsf    PLUSW0,7    ;max displays would be 127 to use plusw
RBNC
    DECFSZ  _bcnt ,F
    BRA    RROW
    banksel 0
    return
endasm
 rotate_l:  ' pic18
asm
    banksel _bcnt              ;ROW
    movlw  _NumberOfRows
    movwf  _bcnt
    movlw  high (_dspbuff)    ;BOTTOM OF BUFFER
    movwf  FSR0H   
    movlw  low  (_dspbuff)
    movwf  FSR0L
LROW
    MOVLW  _NumberOfDisplays + 2;36 displays + 2 outside of display
    movwf  _dcnt        ;display
    comf    WREG
    incf    WREG
    bcf    STATUS, C
Lcol
    rlcf    POSTINC0 ,f  ;PER display
    DECFSZ  _dcnt ,F
    BRA    Lcol
    BNC    LBNC
    bsf    PLUSW0,0      ;max displays would be 127 to use plusw
LBNC
    DECFSZ  _bcnt ,F
    BRA    LROW
    banksel 0
    return
endasm

---

with my suggestions it compiles for 18f47q43, whether it works is not tested . a quick browse of the q43 instruction set and access bank looks ok

Bank 6 (or 7) doesn't work at all. I don't know what it's doing exactly. It does display the text but it does not scroll it.

Bank 5 (or 25) works - sort of.
I can scroll it left OR right continously without issues but when I change direction whatever is in the non visible part the buffer comes back on to the display shifted down by one row - at least I think that's what happening.

But it's fast, that's for sure.
* Mine was like 7ms or something like that
* tumblweeds was more than twice as fast at 3ms.
* Richards comes in at 0.2ms - amazing.

Mine and tumbleweeds simply scrolls the display, anything that's shifted outside of the buffer is lost while Richards rotates it so they're not exactly the same.

/Henrik.

On most of the Q family the SFR registers are in banks 0-4, and the start of user ram is in bank 5.
That's where the access bank is (0x500-0x55F). The other part of the access bank is at 0x460-0x4FF, but that's for the fast SFR registers so you can't use that.
So, the "default" setting for 'banksel' should be 5 and not 0.

If you use the MOVFF instruction both the source and destination must be in the first 4K of ram space.
If either are outside that you need to use MOVFFL, which is a three-word instruction so it's more code (and slower) than MOVFF.

The K42/K83 force you to use MOVFFL to access SFR registers outside the access bank since on those the SFR regs are at the very top of ram.
That's one reason to skip the K42 and use the Q43 instead.

Quote:

---

I can scroll it left OR right continously without issues but when I change direction whatever is in the non visible part the buffer comes back on to the display shifted down by one row - at least I think that's what happening.

---

i expected to have had the buffer cleared out and reset before a direction change . however that sort of behavior is unanticipated

Quote:

---

* Richards comes in at 0.2ms - amazing.

---

i get 0.1mS , maybe my oscilloscope is off



one of my clients sent me a test jig for q43 chips yesterday. i need to get into them soon i guess
no more excuses

@tumbleweed for confirming what i thought

just for interest i tried my code on the st7920 buffer for rows 8 to 15 for the full 16 byte width of course

https://youtu.be/5CVAvcELO6A

works as expected even when direction reversed midstream

or a window to scroll
[i just can't help myself]
https://youtu.be/IYPaKOO-0Pk

Quote:

---

i get 0.1mS , maybe my oscilloscope is off

---

Your scope is fine. All my quoted measurements are for my current configuration with 81 displays, so roughly double your measurement.

Quote:

---

works as expected even when direction reversed midstream

---

It clearly does. Do you have a non visible are in the buffer? I can't see how it would matter but again, in my system it does weird things with what's outside the actual display.

The scrolling window was cool!

Quote:

---

Do you have a non visible are in the buffer?

---

no , i'm using a pointer into the the actual frame buffer of the display so i can't hide any data.
its vital that -
the width is in whole bytes ie for a 128x64 display thats 16 bytes
if i had a hidden byte each end then it would be 18
if you are short or long then the data will appear to move up or down at the wrap around
this may be clearer and can be applicable anywhere

note bank dependence has been eliminated

Code:

---

 
    width  con 128    ;display buffer width  and
    height con 64      ;              height
    bcnt      VAR BYTE 
    dcnt      VAR BYTE 
    WindowWidth  CON 5    ;window is 5 bytes wide    scrolling window
    windowHeight  CON 8        ;window is 8 bits high            "    "
    @windowPad    =    _width/8 - _WindowWidth  ;number of bytes to get to start of next row
    @dspbuff =    _fbr+16*8+5      ;window start address row8 byte 5
;  fbr is the display frame buffer
rotate_r:  ' pic18
asm
    banksel _bcnt  ;ROW
    movlw  _windowHeight   
    movwf  _bcnt
    movlw  high (dspbuff)  ;BOTTOM OF BUFFER
    movwf  FSR0H   
    movlw  low  (dspbuff)
    movwf  FSR0L
    movlw  low((_WindowWidth )*_windowHeight-1) ;TOP OF BUFFER
    ADDWF  FSR0L
    movlw  high((_WindowWidth )*_windowHeight-1)   
    ADDWFC  FSR0H
    MOVLW  _WindowWidth
RROW
    banksel _dcnt
    movwf  _dcnt        ;display
    bcf    STATUS, C
Rcol
    rrcf    POSTDEC0 ,f  ;PER display
    DECFSZ  _dcnt ,F
    BRA    Rcol
    BNC    RBNC
    bsf    PLUSW0,7    ;max would be 127 to use plusw
RBNC
    banksel _bcnt
    DECFSZ  _bcnt ,F
    BRA    RROW
    ;banksel 0
    return
endasm






rotate_l:  ' pic18
asm
    banksel _bcnt              ;ROW
    movlw  _windowHeight
    movwf  _bcnt
    lfsr    0, dspbuff
LROW
    MOVLW  _WindowWidth
    banksel _dcnt
    movwf  _dcnt        ;
    comf    WREG
    incf    WREG
    bcf    STATUS, C
Lcol
    rlcf    POSTINC0 ,f  ;
    DECFSZ  _dcnt ,F
    BRA    Lcol
    BNC    LBNC
    bsf    PLUSW0,0      ;max  would be 127 to use plusw
LBNC
    movlw  windowPad
    addwf  FSR0L ,f
    bnc    Lrnc
    incf    FSR0H ,f
Lrnc   
    banksel _bcnt
    DECFSZ  _bcnt ,F
    BRA    LROW
    banksel 0
    return
endasm

---

ps window is rotate_l only . not done rr yet

Aha, then that might be it as I have 81 displays so buffer is 83 bytes wide.
But again, it does not move down as it wraps around, only as it changes direction.