I hope this isn't a thread hijack, but just following up on this topic a bit.
I think what Sean is saying is that if you wire it basically in the shape of a "top down" view of a megatree (i.e. each node starts int he middle and goes out in order) ... then it'll benefit from the effects that would work on that, without having to do a custom model ... you could call it a vertical matrix? I know he is showing you a custom model and you could use it ... but I'm thinking the idea here is ... if you wire it in this configuration (top down view of megatree sorta) ... that you could skip the custom model?
I did the custom model for mine and learned a number of things about the power of the custom models. I really could not leverage Seans' ideas because there are numerous shapes on my topper ... and while sequencing them all is ok, its not the best way to control it, imho. To be honest, most of my sequencing in the last 2 years has been against all the nodes and not the shapes -- and I know I'm missing the true power of whats possible.
One of my challenges was that my father decided to "reuse" some nodes for different shapes in the super topper. So meaning, some nodes which are say part of the big CROSS, were also part of the original 89 node star-cross. So he basically made some nodes dual use ... and few others were even used by 3 different shapes. Thats what happens when an engineer builds something -- lol. So obviously I could map out all the nodes into one big model (and did) ... but if I wanted the rgb effects to be seen on just the CROSS (nodes) versus the STAR-CROSS (nodes) ... I would need to make a custom model for each. So I did that. But one of the issues that came up was how to make a STAR-CROSS model which combined the nodes from SSC1 and SSC2 and only used certain nodes from each?
What I did was to create a custom STAR-CROSS model which starts with the channels in the first SSC and which ends with the channels at the end of the second SSC. But was was truly confusing to me was how to refer to these nodes within the custom model itself. The nodes are relative to what you mapped out for the model. So say I mapped out SSC1 for 3 nodes land SSC2 for 3 nodes, and both have a hybrid (string mode) channel also. And say in this example my customer object only includes SSC1 Node1,Node2 and SSC2 Node2,Node3 ... just four nodes for this example. You see this excludes the string mode channels for each SSC and the nodes not included (SSC1 Node3, SSC3 Node1) in the custom model shape.
PN#1,2,3 model node#1 SSC1 string mode (hybrid)
PN#4,5,6 model node#2 SSC1 Node1
PN#7,8,9 model node#3 SSC1 Node2
PN#10,11,12 model node#4 SSC1 Node3
PN#13,14,15 model node#5 SSC2 string mode
PN#16,17,18 model node#6 SSC2 Node1 / object Node#4
PN#19,20,21 model node#7 SSC2 Node2 / object Node#5
PN#22,23,24 model node#8 SSC2 Node3 / object Node#6
So if I mapped out the custom model for this to start with PN ch#1 and go for 24 channels total.
I then would map out only model node#'s 2, 3, 7 and 8. So on the custom model grid I'd only put
in 2,3,7 and 8 into the locations on the grid, where I wanted them to go.
The tricky part was keeping track of what model node#'s nutcracker would call them ... so I put these things in an excel spreadsheet ... which would show me that the "relative node #" for the model would related to which pixelnet ch# ... and they have to be in groups of 3, since 3 PN channels per RGB node. ANd I also needed to included the string mode channels, since they are know to the SSC as valid PN channel#'s.
I even asked Sean what order these would be sequenced if I say threw in several models into the same sequence and threw effects against them ... how would it determine the order of sequencing when collisions occur? I will say that my initial tests are very encouraging ... meaning sequencing against multiple models on the same topper ... (think of a holdman star with its 3 distinct areas -- inner cross, inner star, flares) ... and overall it works. Esp. since my original 89 nodes did not include any "shared" nodes between shapes. So right now I've basically avoided combing the big cross with the star-cross ... but I think it would work if I combined them.
Anyways, I find the ability to create these models extremely powerful and really takes me out of my LOR way of sequencing and into a whole new mindset.
I will find a picture of my topper and post it here.