Hub75 P10 1/2 Scan 16x32 Outdoor Panel Setup?

[nospamsam2011]

Good day All, Can anyone provide some direction as to how I can modiffy Henner Zeller's excellent library/api to work with a yet another odd ball LED Panel? My panel is a P10, 1/2 Scan, 16x32 Outdoor panel with the HUB75 Interface. My HUB75 interface includes the usual signals except that A and B are only indicated. I tried every combination of Demo's command switches and was unable to get any of the demo images to display properly. Reviewing the doc I see that there is no mention of 1/2 scan and so I suspect that my panel is yet another version... So...I reversed engineered the display to decode how the various control signals affect the individual LEDs. the results are as follows:

1. Address lines A and B

   • Only Address A is used. Address B must be Low
   • Where A=0, Every odd row is displayed and when A=1 Every even row is displayed

2. The LED positioning is really odd... at least to me. The LED's appear to wrap to the next line every 8 LEDs and this continues across the Columns.

Attached are a few pics of the panel and the HUB75 connector

As an example, I have included a PDF of the LED positioning.

P10 16x32 1_2 Scan Outdoor Decoding.pdf

These positions are referenced as from viewing the panel head on with the "HUB75 data in" located on the right side. Row 1 Column 1 is the top right Corner. here I get the following:

• LED 0 (R1-G1-B1) (A=0) is located at Row 3 Column 1
• LED 7 (R1-G1-B1) (A=0) is located at Row 3 Column 8
• LED 8 (R1-G1-B1) (A=0) is located at Row 1 Column 1
• LED 15 (R1-G1-B1) (A=0) is located at Row 1 Column 8 etc.

With this arrangement to display a single row one would need to clock 128 times and only set the corresponding bits:

e..g To display Row 1 (A=0), then a total of 128 clock pulses are needed with:

• Clock's 8-15, 24-31, 40-47, 56-63 with a suitable value of R1-G1-B1, all the other clock's will have the R1-G1-B1 set to 0 (off).
• Next Set A=0 and Latch, and then Set OE low

With that said, I am still a newb to the library and so any suggestions on how to modify to use with this panel would be greatly appreciated.

Thanks in advance!

Cheers,

Sam

[tammaaziz]

Hello,

I do have a similar issue with a P10 panel !! Any help please

[Robycop58]

Hello, same problem. The multiplexer HUB75 IC is 5020. I had try all the combination of multiplexing, scan-mode and row-addr-type options but the pannel don't work properly. Please hzeller up and help us..... Thank's

[kasperv79]

Any luck with this panel? I’m also stocked... :(

[nospamsam2011]

Sadly, no progress or luck...That being said I have not had time to dig deep into this issue, as I was hoping that there would have been some info on how best one could tackle the problem.

[felaniMT]

Hi, it seem this thread still open..i also facing problem with p10 rgb 1/2 scan panel.. anyone have any idea how to check scan pattern?

[WernerFettke]

Hi All,

Also having the same problem, also tried hours of command line options... the best I can get is (using the word 'hello' as my test ):

which is the result of: sudo ./text-example -f ../fonts/8x13.bdf -C0,0,255 --led-slowdown-gpio=4 --led-parallel=1 --led-rows=16 --led-cols=64 --led-chain=1 --led-cols=32 --led-multiplexing=9

And at a different font size: sudo ./text-example -f ../fonts/6x10.bdf -C0,0,255 --led-slowdown-gpio=4 --led-parallel=1 --led-rows=16 --led-cols=64 --led-chain=1 --led-cols=32 --led-multiplexing=9

The panel model number is: p10-2s-32x16-3535-v1 (1/2 scan)

I saw some articles suggesting creating a transform file, but I have not had any luck with that either (might be related to I'm not 100% code efficient).

If anyone can assist I would really appreciate it, even if you could just point me in the right direction.

Thanks in advance.

[davemaster]

When I try this great code for the first time, I had the adafruit panel (1/16) and other from china, a P10 but with 1/4 scan. check the option " --led-multiplexing ", the default one worked perfectly with the Adafruit panel, but for the chinese P10 32x16 1/4 scan, the option what I found by try-and-error was --led-multiplexing=8. Now, your call for test it with yours

Best regards

[WernerFettke]

Hi Dave, Many thanks for the feedback, I've tried using the multiplexing option already, I found the best result was option 9 (9=P10-128x4-Z), however the images in the original post are using this already. It's almost as if I'm almost there.... just need a small tweek somewhere. In terms of the chinese panels you using, are they also p10-2s-32x16-3535-v1 ? Did you need to make any other changes to your .h or .cc files in order for the panels to work ?

[davemaster]

Because I am using now a RPi 4, the slowdown option must be set with 2 minimum, try that too, check the rows, cols (those were my errors at the beginning)

El jue, 19 ago 2021 a las 4:19, Werner Fettke @.***>) escribió:

Hi Dave, Many thanks for the feedback, I've tried using the multiplexing option already, I found the best result was option 9 (9=P10-128x4-Z), however the images in the original post are using this already. It's almost as if I'm almost there.... just need a small tweek somewhere. In terms of the chinese panels you using, are they also p10-2s-32x16-3535-v1 ? Did you need to make any other changes to your .h or .cc files in order for the panels to work ?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/hzeller/rpi-rgb-led-matrix/issues/782#issuecomment-901751910, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAIGUX6IAJB2DJ3X23YULTDT5TEAHANCNFSM4HAWSKKQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email .

-- David Elias Flores Escalante TeleTracking SAC