Start Channel/End Of Model

[bwinter]

Not sure what I'm doing wrong here, but when using Start Channel/End of Model, I seem to be getting odd results.

My Roof 2 is only a single string, 25 nodes, starting after Roof 1.

Roof 1 original End Chan was 9074, but I modified the number of nodes in Roof 1 (to see if Roof 2 would update properly).  Roof 2 Start is still stuck at 9075.

Also, I notice that Roof 2 End Chan is 18223, which is really bizarre since there are only 25 nodes.

Is there something I need to do to refresh (other than SAVE at the bottom of the panel)?

[keithsw1111]

If you save and then restart xlights does it correct itself.

[bwinter]

If you save and then restart xlights does it correct itself.

Nope...no luck there.

[Gilrock]

Well you can look at how you did Roof 4, 5, and 6 to see how to do it correctly.
It needs to be ">Roof 1:1".  You are telling it to offset 9075 channels from the end of Roof 1 instead of 1 channel from the end.

[bwinter]

Well you can look at how you did Roof 4, 5, and 6 to see how to do it correctly.
It needs to be ">Roof 1:1".  You are telling it to offset 9075 channels from the end of Roof 1 instead of 1 channel from the end.

4, 5 and 6 isn't working properly either.

For example, I just increased the # nodes in Roof 3 (and Roof 3 End Chan updated appropriately).  However, you can see that the Roof 4 (and 5 and 6) End Chan hasn't updated properly.

[Phrog30]

You can also hard type the start value in as well. That's what I've done. I don't really like the auto channel stuff.

Sent from my Pixel using Tapatalk

[Gilrock]

So its probably a refresh problem on the list...doesn't mean it's not working.  Do an action that forces a complete list update.  Otherwise you gotta provide files instead of pictures.

[jnealand]

I'm with phrog on this one.  I could never seem to get things right using that notation, so I just went with using absolute numbers.  It starts going haywire once you start making changes.  The original seemed to be ok, but I just gave up and went to absolute addressing.

[bwinter]

I'm with phrog on this one.  I could never seem to get things right using that notation, so I just went with using absolute numbers.  It starts going haywire once you start making changes.  The original seemed to be ok, but I just gave up and went to absolute addressing.

Kind of what I'm thinking now too. I was using absolute numbers before, but figured I would see if I could implement this method.

[Gilrock]

I have no problem using the model chaining you just gotta not do anything stupid.  If you have 10 models chained together and you delete the first one guess what's gonna happen?

[Phrog30]

If you delete a model and the software doesn't auto correct, then it's not worth using. What's the benefit versus just manually plugging numbers in?

Sent from my Pixel using Tapatalk

[keithsw1111]

In v51 I have added to check sequence more extensive validation which will highlight specific reference issues in start channels

[Ebuechner]

In v51 I have added to check sequence more extensive validation which will highlight specific reference issues in start channels

That won't log a warning if you intentionally skipped channels because of null pixels?

[bwinter]

I have no problem using the model chaining you just gotta not do anything stupid.  If you have 10 models chained together and you delete the first one guess what's gonna happen?

I guess it's just not obvious to me what I did that was stupid.

[pixelpuppy]

So its probably a refresh problem on the list...doesn't mean it's not working.  Do an action that forces a complete list update.

Which actions force a complete list update?  When I've had this problem, I just forced the individual model to update.  Being able to force a complete list update would be nice for times like this when it gets out of sync for whatever reason.

Relative-format addressing is a nice feature when it updates correctly.