Noticed this before, and that I'm doing more testing, I'm seeing this:

[Stormyblade]

So I'm currently doing testing on the following: 8 strings, 70 pixels each, each one set as a mini-tree. I noticed this happening last holiday season in my show and now that I'm doing even more testing, I noticing that there many times a random pixel will flash red instead of the color it is assigned. I even have a couple pixels stay lit on red when the sequence ends. It happens randomly -- when I'm telling xLights to do spirals, color wash, or even the simple on/off effect. This is the same setup I used for the last holiday season, so I can't see anything wrong with that...I run my displays/sequences at 20fps, so I don't think it's a timing issue. (Well, unless the timing for the SanDevices E682 shouldn't be set at 300ns for a standard WS2811 12mm bullet pixel???)

Any ideas on what causes this? It's happened across several sequences, and when I've sequenced the color to be a light yellow and I see flashes of red, it is far too noticeable and gets annoying rather quickly. If I need to add files, or setup screens, please let me know.

[RobbySchwarzbach]

Yes. I have the same thing happening. I'm a newbie just starting out with my 20 string, 50 pixel each, quasi mega tree. On the last string, one or two pixels will stay a color not assigned. Not always the same color. That would be too easy. First, I thought it was a power issue. So I put a new injection just at the last string. Nope. Then I thought it was data. But my volt meter says everything was fine. So I went ahead and replaced the entire string. Nope!
I am trying to change the way I view this.  "It's a feature, not a bug!" I tell myself. It's a G**D*** feature!

[needbrew]

Typically for me I see this when a pixel is going bad, or if there is an issues with the data wire.  If it is a long run that can cause issues as well depending on what controller you are using.  Are you doing a simple color sweep or some effect?  Are they staying on at the end of a sequence?  Are you playing from xlights or a pi player?

[RobbySchwarzbach]

Hi.  Thanks for the interest. I'm using a Falcon 4v3 running from a Mac and playing using xSchedule. I would consider one megatree: 20 stands 50 pixel each connected to the controller by a three foot cord to not be a long run. And I replaced the entire strand where the errant pixels were, so... Can't be bad pixel. It does happen during complex sequences. But not every sequence. One pixel (not always the same pixel) will turn some incorrect color (again, not always the same color) and stay on while the other pixels around it move on in the effect. Then it will start behaving correctly. Then another pixel in this strand will do something weird. Since I have a rotation of 190? and it's the last strand really no one notices but me.
Still it's a mystery..

[needbrew]

To isolate the issue I would do the following in this order to help

• Use the Falcon color wash test to see if they are all the correct color.
• If they are all the correct color I would create a simple sequence of chase using single strand and slow it down so you can see it.  Run that over a cycle to see if that misgehaves.
• if it is a complex effect check that you do not have layers bleeding into each other.  there are a few setting that do this to allow for different effects.

[/list]

[RobbySchwarzbach]

Done the color wash test already. Never shows up there. Nor does it show up with a simple line down like meteors. Only when lots of things are changing. So even if I had layers bleeding into each other, shouldn't it effect other pixels on other strings? It only happens on the very last string. And I've triple check power and data, and I think they are fine. I really got stumped when I replaced the entire strand and it still happened. 

Jeez. Thinking about it, it has to be a data issue. Right? Can't be anything else, because it happens on the last strand only even with a new strand. So it's gotta be that. Right?
I get a little overwhelmed thinking about tracking down a data issue like this...

[needbrew]

Yeah, it is something to do with the program.  If a simple effect does not do it I would check that your mappings are correct and that allow blending between models is not checked in the sequence settings.  Also if you are using layers, check that you dont have something overriding them.  Have you done the "Tools->Check Sequence"

[allknowing2012]

Have you pushed the configs to the controller? Could the controller be set at 512 universe, and xlights set to510?

[RobbySchwarzbach]

OMG. This is Waaaay over my head. "Push the configs to the controller." Huh??
I assume there must be some sort of "How to for Dummy's" out there, if you could be so kind as to point me in the right direction. A newbie still has a lot to learn...

[allknowing2012]

go  to xlights.org and hit the zoom room .. we can walk you thru it..

[RobbySchwarzbach]

Oops. Sorry. Can't do zoom. I am unable to speak due to an accident that severely disfigured my face.  I can "text to Speech" it.  Sometimes my typing is a little slow, but maybe you could speak then I can listen and respond via text. Would that work?  I don't have a camera on my computer. But I have speakers, of course

[Stormyblade]

I truly hope that you all can get Robby helped out, but I'm coming back to this to see if there's any thoughts on what I originally posted. It does not happen to the same pixels on the same strings, but randomly across all 8 strings. I noticed this happening this last Christmas season when playing from FPP -- now during my testing I'm seeing it happen while playing the same sequences directly out of xLights. Just to ensure that I didn't screw up effects in a particular area, I deleted the effect and entered in a new one, then rendered, saved, and told xLights to play the sequence with Output to Lights turned on. Still happening, randomly. And, still getting the random pixel on a random string to stay lit up red when the sequence ends. As soon as I start another sequence that pixel goes out and proceeds to behave normally.

Any thoughts on this issue?

[RobbySchwarzbach]

I am sorry that I hijacked your thread. Kind of got caught up in the moment. I'll start a separate thread to deal with my concerns. Again, sorry about that!

[allknowing2012]

zoom room doesnt need video or speech .. text in chat  and share screen.

[Stormyblade]

Robby, don't worry about it - if you get help with your items, I'm happy to share this thread as needed. 

I was just re-visiting in the hopes that someone might know what is causing the issue with my strings.