How to configure - is this possible

[dirkblaze]

Hey Dan, what are the limitations of the software for LED array configurations?

I think I may have read that the number of columns and rows must be multiples of 8?

We have an array consisting of 12 strips of 50 LEDs at the moment. That was the length of the LED strips we had. The entire set of 600 LEDs are driven right now by from 1 pin of our Octo; we are not worried about the display "refresh rate" for our application.

Can your code drive a display of those proportions? If so, what would the correct HEIGHT, WIDTH and other settings be?

Thanks!

[i-make-robots]

It's been a while....

"Beware of image distortion from long LED strip lengths. During the WS2811 update, the LEDs update in sequence, not all at the same instant! The first pixel updates after 30 microseconds, the second pixel after 60 us, and so on. A strip of 120 LEDs updates in 3.6 ms, which is 10.8% of a 30 Hz video frame time. Doubling the strip length to 240 LEDs increases the lag to 21.6% of a video frame. For best results, use shorter length strips. Multiple boards linked by the frame sync signal provides superior video timing accuracy." -- the comments at the top of videoDisplay.ino

So the limit isn't really the number of LEDs, it's how fast they update. I drove 64x36 LEDs from three pins and got pretty good refresh rate. Could have used more pins and gone faster.

[i-make-robots]

send me a video when it's working!

[dirkblaze]

Thanks for getting back to me so quickly!

Yes, I get that 600 LEDs on one wire will take a "long" time to refresh. I'm just trying to get it working at all!

I gather that the teensy code and the processing code can drive a 50x12 LED matrix without problems? I still can't seem to get the COLUMNS and ROWS params correct to even display the test screen correctly.

[i-make-robots]

Maybe send a pic of what you're getting and we'll figure it out from there.

On Mon, Sep 26, 2016 at 6:58 PM, dirkblaze notifications@github.com wrote:

Thanks for getting back to me so quickly!

Yes, I get that 600 LEDs on one wire will take a "long" time to refresh. I'm just trying to get it working at all!

I gather that the teensy code and the processing code can drive a 50x12 LED matrix without problems? I still can't seem to get the COLUMNS and ROWS params correct to even display the test screen correctly.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/i-make-robots/LEDWall/issues/2#issuecomment-249746865, or mute the thread https://github.com/notifications/unsubscribe-auth/ABZYhiOJU1ECi0zVQWCgF8Dh2fwOXxpnks5quHg-gaJpZM4KHIv3 .

Dan Royer, Owner, Marginally Clever Robots https://www.marginallyclever.com/ Ph: +1 (604) 259-9564 Mo: +1 (604) 916-2281

[dirkblaze]

Here are some pics. The array is 50 leds x 12 strings. one data wire. It runs a sinusoidal test pattern nicely.

[i-make-robots]

The test screen image may be made for a 4:3 aspect ratio. You have a 25:6 aspect ratio. The plasma rainbow would probably work because it doesn't care about aspect ratio.

[dirkblaze]

So OK, we are just trying to get something cycling; could we fake the code out by telling it the matrix had more rows, possibly on more output pins? Its fine if the image truncates, we're just try to prove the basic components and technology now.

4:3 with 50 columns would be about 37 or 38 rows....

[i-make-robots]

I guess so. the LEDs themselves wouldn't care if you sent too many instructions, the signal would have no effect.

On Mon, Oct 3, 2016 at 12:36 PM, dirkblaze notifications@github.com wrote:

So OK, we are just trying to get something cycling; could we fake the code out by telling it the matrix had more rows, possibly on more output pins? Its fine if the image truncates, we're just try to prove the basic components and technology now.

4:3 with 50 columns would be about 37 or 38 rows....

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/i-make-robots/LEDWall/issues/2#issuecomment-251204126, or mute the thread https://github.com/notifications/unsubscribe-auth/ABZYhovQbKYDFTsF5LBze-o601jwnlnrks5qwVlagaJpZM4KHIv3 .

Dan Royer, Owner, Marginally Clever Robots https://www.marginallyclever.com/ Ph: +1 (604) 259-9564 Mo: +1 (604) 916-2281

[dirkblaze]

Do the imaging or teensy libraries care about the number of rows or columns? I seem to recall seeing a "multiples of 8" requirement for # of columns somewhere. Is 50 columns copacetic ?

[i-make-robots]

/Processing/screenCapture/screenCapture.pde

• SCREEN_WIDTH and SCREEN_HEIGHT will need to match your dimensions.
• led_map() will have to change to match your wiring plan. int led_map(int input); should return input (eg remove everything, you don't need the swizzling).

/Teensy/videoDisplay/videoDisplay.ino

• ROWS should match SCREEN_HEIGHT from processing
• COLUMNS should match SCREEN_WIDTH from processing
• PINS_USED should be 1
• leds.setPixel(led_map3(i),rainbow(j,k)); change led_map3 to led_map

That will probably help. I don't have your grid so I can't seriously test this stuff. If you find this help useful, please paypal a donation. I'll put it towards improving our local makerspace.

On Mon, Oct 3, 2016 at 12:49 PM, dirkblaze notifications@github.com wrote:

Do the imaging or teensy libraries care about the number of rows or columns? I seem to recall seeing a "multiples of 8" requirement for # of columns somewhere. Is 50 columns copacetic ?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/i-make-robots/LEDWall/issues/2#issuecomment-251207096, or mute the thread https://github.com/notifications/unsubscribe-auth/ABZYhvJoKINEkHxAHG2Cgw3sxjGipyQPks5qwVwwgaJpZM4KHIv3 .

Dan Royer, Owner, Marginally Clever Robots https://www.marginallyclever.com/ Ph: +1 (604) 259-9564 Mo: +1 (604) 916-2281