On Thu, Dec 30, 2004 at 07:02:04PM -0500, Nathanial Hendler wrote:
> On Thu, Dec 30, 2004 at 10:03:50PM +0100, Jan-Erik Soderholm wrote:
> > For the "multipl-LED" case, set up 8 registers (bytes) to hold the
> > "image" of what should be seen on the 8x8 LEDs. Then setup
> > an timer-interrupt routine to do the regular multiplexing of the
> > LEDs.
> > 
> > Finaly add whatever logic you need to update the 8 regs with
> > whatever "image" should be displayed on the LEDs.
> > 
> > In that way, you separate the logic to calculate the image from
> > the locig that continously updates the LED. Since the update is
> > running from an timer-int, it shold be possible to make it
> > flicker-less.
> 
> I guess the above is what I was aiming at having; i.e. 8 registers that
> represent the image.  I can understand how to translate 8 registers to an 8x8
> LED, using TMR0.  What my question still is, regards how to manipulate the 8
> registers in the ASM code.  Say I have a pixel anywhere on the 'screen'.  How
> do I light the pixel under it.  In other words, how do I turn on LED x,y-1?  How
> do I setup my registers so that they are addressable in such a manner that I
> can do the following...
> 
>   bsf 5, ?????????
> 
> ... to set LED 5,3?
> 
> If I'm just being dense here, I'm sorry.  Thanks for everyones' help so far.

Not dense at all. Let me take a high level walk through with it.

Since right now you only have 8 rows and columns let's use 4 bits to represent
the row and column. And just to keep it simple let's use W to carry the
parameters. So the call would be something like:

MOVLW 0x53
call  setled

To set LED 5,3

Now let's presume you have 8 contiguous registers to store the bitmap as you
specified. There are a few keys to making this work:

1. Use a jump table to convert the column to a bit number.
2. Use the FSR/INDF register to set the row.
3. Don't use the BSF and BCF instructions because there's no way to set a
   variable bit number. In order to use BSF/BCF you'd have to use a second
   jump table with an entry for each bit.

To keep it short I'm going to shortcut and presume that the target is saved
in a coord register and in W. The logic is something like:

call col2bit  ; convert column to bit number. Saved in register column
swapf coord,W ; get the row in the lower 4 bits of W
andlw 0x0f    ; clear the top 4 bits
addlw display ; add the display register base
movwf FSR     ; setup row in FSR
mowf column,W ; get the column value
iowf INDF,F   ; Set the column bit in the specified row.
return        ; Bye.

Hope this gives you some insight.

BAJ
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist