'convert to effect' other than node by node?

[randallr]

I've been diligently working on moving 19k ch 2015 LOR sequences to 105k ch Xlights for 2016. Existing props all lined up and import perfectly, and have built .txt file to import old 32ch rgb tree to 64x200 pixel tree.

Now I'm starting to make those existing props like the 50 rows of RGB yard lights better in xLights.

I've gone through smaller props using 'conver to effect' on each node, but I see that's going to be a nightmare with several larger ones like yard grids etc, plus each sequence I want to 'rework' in xLights.

Did I miss something; is there a way to convert more than a node at a time to 'effect' so that it can be  moved etc?

[Gilrock]

Sounds like you are bringing in the LOR stuff as a Data Layer.  Another option is to use the menu option "Import->Import Effects" and then you create a mapping file for how to line up your LOR channels with xLights.  Once you get the mapping file created and saved you can easily bring in additional sequences.  You will be ending up with a lot larger files than if you had created your sequences in xLights so the performance you experience will not be optimal because it will be generating thousands of effects from the LOR data.  Something that could have been done with a single effect in xLights might be hundreds of effects when you import raw data.  I can't point you any particular videos but I know there are tons of them linked around here all the time if you search and also look at Clyde Lindsey's video's.  Not sure if he posts here but definitely on Facebook where you'll find the most xLights help.

[randallr]

I had only used the "import effect" on the my new MegaPixeltree. I made a file with only the first Strand line mapped for each new pixel string from the old RGB tree:
MegaTree   Strand 1      Old MT88 p1   #000000
MegaTree   Strand 1   Node 1      #ffffff
MegaTree   Strand 1   Node 2      #ffffff
........
xLights took that data and put it on every node on the pixel strand, which was a perfect start. But it didn't create them as 'effects'.

I'm assuming that if I create a similar text file for existing props to identical models, it would because I'd be mapping node to node.

I'll try this. Thanks

[Gilrock]

Not sure what you mean because "Import Effects" only creates effects it just creates them at the node level.

[randallr]

When I put the mapped channel on the 'Strand' only, and not a node, (as in the file previously) It comes in on the strand. Then it renders it on to each node on that strand.
Each node doesn't have it as an actual effect. (attached a screenshot)

I went back and made a xml file and imported several models, with the channel mapped to the 'node', and the effect  comes in on the node level just fine. Just have to be build a xml mapping file for the whole setup. (The prior problem only surfaced because I was trying to adapt the 32rgb tree to 64x100, which worked for replicating each old RGB string to a pair of whole pixel strings. Which worked but left the individual nodes as rendered only.)

First part of new import mapping file with import channels on node level.
false
3
Yard - CtrWest
Yard - SideEast
Yard - SideWest
Yard - CtrWest            #ffffff
Yard - CtrWest   Strand 1         #ffffff
Yard - CtrWest   Strand 1   Node 1   Yard West 83 p1   #000000
Yard - CtrWest   Strand 1   Node 2   Yard West 83 p1   #000000

[Gilrock]

The nodes always show the rendered result of any effects you place at the model or strand level or any data layer data.  You can't eliminate that.

[randallr]

I'm learning!
I guess what I should have done to convert the 32rgb to 64x100- In the xml import file, I should copy the LOR channel to each node on the each strand on the new model. I was just too lazy to copy and paste that much. But ended up writing a program to generate the xml file. Just needed a few more lines to replicate it on the 200 nodes on each strand instead.

Thanks for the patience!

[Gilrock]

One of the toughest decisions for people to make is to decide when to throw away their old work.  I've thrown away literally hundreds of hours of LOR programming because it just wasn't worth it to try to save it.  It's so much easier to sequence in xLights and the effects don't get hard-coded into each channel like they do in LOR.  I just worry the technique you are using is going to cause you to not like xLights as much because its not the best way to use the program.  Do you have any videos of a sequence you are bringing over so we can see how complicated the effects are?

[randallr]

I agree 100%.. Am I'm hooked on xLights now. Only things that were complicated was SS effects which are a breeze to bring in. Most of 'my' work has been very basic and crude in LOR.
After an onslaught of traffic in 2014 from our "God Bless the USA" having 800k shares on FB, coupled with being on GCLF, we began the move to a new public site for 2015. Didn't get the okay until end of October, so I didn't have time to build our tree there, but added new 16x100 pixel trees. So I'd totally rebuilt sequences in a couple of weeks, so not very fancy. In testing after New Years, I realized a couple of Pi players outperformed my water-cooled supercomputer running LOR. Now I've got the city's blessing to expand to more than an acre, and got some funding to go from 19k channels to over 100k.  My goal was to start working on getting the last few years sequences brought in early, so I can spend time to enhance with the new effects.

Here's one from 2014, only cool thing I did was the color on yard during choruses-
https://www.youtube.com/watch?v=aMlVYFnOqYc

Here's one from this past year in our new site.
https://www.youtube.com/watch?v=Z0NbkBhvatA

[Gilrock]

Cool!  So it sounds like you were on the right track with creating the mapping file.  The nice thing is once you have the mapping file you can delete the sequence and try again if you need to make changes to the mapping.  I'm not sure how best to handle the tree doubling size.  You could create mapping that duplicates the data.  I might would try to start with a 32x100 tree in your layout and import onto that.  Then try promote the effect up to the strand and then model level.  If it detects the same effect on all nodes it will allow a promote and same with strands.  Then once all effects are at the model level you can change the model size.  It might take some experimenting because it depends on the type of effects as to how well they will promote.

[randallr]

Awesome. (more to learn...) I built a map file with the 32 RGB's mapped to all 12800 pixels. RGB 1 of original to Strand 1 nodes 1-200 & Strand 2 nodes 1-200 etc. xLights went to 3774mb on mem usage and ~ 20% CPU. After about an hour, it finally crashed. I noticed that it only was pulling up map file down to a spot on Strand 14 every time, so I tried it with just Strands 1-10 with all the pixels. It built those in the model, but stayed at 1770mb on mem usage, so I did a 'save'. After 20mins, it's down to 1120mb and 1.2% on cpu.

Gonna try the route you brought up tomorrow!

Although, it really may not have been a big deal with it mapping the strands, and then rendering the nodes. I just couldn't figure out how to clear and area if I wanted to do another effect in that spot. Couldn't select a region and delete. (something else to learn that I missed most likely?)

[Gilrock]

If you delete the actual effects and then re-render the data on the nodes should go away.  Placing another effect would fill in that data.  This is all assuming you left the render mode at the default of erase mode and the data is not data layer data.

[randallr]

That Render thingy again! I'm just not a good student. My ADHD kicks in too much!

BTW, the 'new' megatree we're building will not only be fitted with the 64x 200 pixel strings, but we're also going to have 32 pipes, 7' above the ground with a T on top to connect to a pair of strings each. A walkway will then come through the tree so people can be inside. (48' diameter at the base. The new tower is a Rhon 55G, 80'.) Playing with effects like the snowflakes or snowstorm with light blue flakes plus sparkles looks pretty awesome on preview.

It was more awesome with strings going to the ground. Made it impossible to walk inside during effects. It was cool when Sabrina went inside during judging and lost her balance when she looked up. But it just seemed to be easier, logistically, to keep the strings further away from people's hands by elevating all of them.

Hence, this is why I'm starting early to get all this figured out!