1/8S Display wrong line/scan order

[James2k2]

Hi all,

I'm not quite sure where to go with this, but I've got a display listed as P5-3232-3528-8S-A2. It uses DP74HC138D as the main chip and then DP5020B.

Running the VirtualMatrixPanel example 3 produces an almost working result where the only issue is the odd and even sets of lines are the wrong way round. So lines 1-8 display on rows 9-16, and the lines 9-16 show on rows 1-8. It's the same issue on the second set of 16. So we're not a million miles away.

I've attached a photo of what I mean. I tried switching the ABC lines around, but no joy.

I couldn't find anybody with the excat same issue so apologies if this is repeated somewhere.

Cheers,
James

[board707]

Running the VirtualMatrixPanel example 3

What is the "example 3" ?

the only issue is the odd and even sets of lines are the wrong way round.

So Just swap "if" and "else" branches inside getCoords() method code ...

[James2k2]

Cheers, 707.

I presume you're referencing this statement:

`case FOUR_SCAN_32PX_HIGH:
/* Convert Real World 'VirtualMatrixPanel' co-ordinates (i.e. Real World pixel you're looking at
on the panel or chain of panels, per the chaining configuration) to a 1/8 panels
double 'stretched' and 'squished' coordinates which is what needs to be sent from the
DMA buffer.

           Note: Look at the FourScanPanel example code and you'll see that the DMA buffer is setup
           as if the panel is 2 * W and 0.5 * H !
        */

        if ((coords.y & 8) == 0)
            // 1st, 3rd 'block' of 8 rows of pixels, offset by panel width in DMA buffer
            coords.x += ((coords.x / panel_pixel_base) + 1) * panel_pixel_base;
        else
            // 2nd, 4th 'block' of 8 rows of pixels, offset by panel width in DMA buffer
            coords.x += (coords.x / panel_pixel_base) * panel_pixel_base;
			coords.y = (coords.y  >> 4) * 8 + (coords.y & 0b00000111);
        break;`

I'll give that a go.