Shifted Matrix Problem

[justsilviu]

Hello,
i have recently discovered this library and have tested it a bit, on 2 x 64x32 p4 matrix.
i have it setup as a 128x32 matrix.

my problem is that there is a one pixel shifting on the lower 16 rows of the matrix, when scrolling text.
this happens on all setups that i have tried. SmartMatrixv3, and SmartMatrix_GFX.

i have tested this with 2 different ESP32 dev boards, with and without level shifters, multiple color setups, fonts.
i have 2 pictures showing how it looks, it is very hard to catch on camera, but it is clear to the naked eye, especially if the font is bigger.
if the scrolling speed is slow or if it stops, there is no notable difference
it looks like the lower half of the matrix is getting the data a few milliseconds faster.
any ideas where is this from? and how to fix it?

also another question is:
when do i have to use the swapbuffers() method, this is not clear for me.

thanks for your help

[marcmerlin]

@justsilviu

SmartMatrixv3, and SmartMatrix_GFX.

Note that SmartMatrix_GFX maintains a framebuffer that is pushed in one swoop (swapbuffer) to SmartMatrix, so whether you use it, or not, it should not affect your results.
I do have tiled matrices, but I use 96x64, which is tiled the other way from you.
If you find that SmartMatrix updates the top and bottom part of a 32 pixel panel out of sync, that is likely a problem with the underlying driver, especially as the top 16 and bottom 16 lines are supposed to be updated at the same time.
Are you able to reproduce is you use 64x32 instead of 128x32?

@justsilviu

when do i have to use the swapbuffers()

If you use SmartMatrix_GFX, you do not, it will do it for you when you run matrix->show()
If you are using raw SmartMatrix, you need to call it every time you make an update you want to have visible on the panel

[Louis]

@justsilviu

any ideas where is this from? and how to fix it?

Horizontal update tearing

I think this is caused by the multiplexing in the panel. There are two lines driven at a time, spaced 16 rows apart, and the code scans from top to bottom. If the line is going from right to left and moving one pixel left per frame, pixels (31,15) and (31,31) are lit just before pixels (30,0) and (30,16) are lit. Try rotating the screen in software before running the same code, you’ll likely see this go away or at least look different at rotations 90 and 270. Response to a similar report: Ma…

[justsilviu]

yes, this issue is reproduced also in on the solo 64x32 matrix.
i’m not sure what the driver is, (i will check today when i get home), but the driver is not defective, because i have used the same 2 matrices with another library, the one written by 2dom, that communicates on only on color wire, and there is not problem there.
The problem there is that because of only one wire, the colors that are not only Red or Green or Blue, create a very strong flicker.
… so i’m thinking about an incompatibility between the library and driver, not necessarily a defective driver.
i will come back with the driver model, maybe i can find something in that area.
… or maybe the problem is my Chinese ESP32 model’s, which is neither the one you use or the ones used by jason on his dev’s boards. i have this one:

ESP32 Development Board WiFi+bluetooth Ultra Low Power Consumption Dual Cores...

[justsilviu]

the driver on the matrix is CHIPONE ICN2012.

i will try the sugestions from Louis and the solutions described in the mentioned post.
thank you for your help