Can anyone "de-mystify" how to use stackSymmetry on "splitter" type strut parts?

[SciMan]

The part.cfg file makes it possible to force the part symmetry settings to a specific value using the stackSymmetry value.

From the KSP wiki, we already know some of the values that work for stackSymmetry:

stackSymmetry = 1 : 2x symmetry

stackSymmetry = 2 : 3x symmetry

stackSymmetry = 3 : 4x symmetry

This much, most part and mod makers already know.

However, what about 6x and 8x symmetry?

for 6x symmetry:

do you set stackSymmetry = 4 to get 6x symmetry?

Or is it stackSymmetry = 5 to get 6x symmetry instead?

the same question applies to 8x symmetry as well:

do you you set the value to 5 in order to get 8x symmetry?

Or should you set it to 7 instead?

Another thing we don't know is why sometimes it fails to work when everything we know about how to use it says it should.

I will use a few parts from NovaPunch as examples. Specifically. both of the "3 meter to 5x 1 meter diameter" couplers have the value for stackSymmetry set to 1 in their .cfg, and changing that to 3 instead ends up turning OFF symmetry, instead of doing what was expected by setting 4x symmetry. Why would it do that? If anyone knows something about stackSymmetry that is not covered here, please post it so that the rest of the KSP modding community can better understand it, and so that we can have hex- and octa-couplers that actually work the way the tri-coupler does!

[EndlessWaves]

Looking in the dll that's actually a property of the general part class rather than the child strut class, which doesn't actually contain any new properties. It's an integer rather then an enum and I can't see any obvious method names relating to it in Part so I can't tell which values it can take.

Perhaps it's something to do with the model as to how the game interprets it?

There is a symmetryMode property, another integer, that looks like it might be related.

[SciMan]

yeah, stackSymmetry belonging to the general part class makes sense, because I've seen fuel tanks among other things that have working symmetrical attach nodes.

by the way, whats an "enum"? I know the integer (1,2,3,4,-54), float(3.14159), bool(0/1, true/false, on/off), and vector("1.4, 3.0, 5.2") data types, but an "enum" is new to me. care to explain?

Honestly, I was just trying to figure out if there was a way to fix the symmetry options for some of the NovaPunch parts, or other stuff with 6 or 8 circularly symmetrical nodes, because I have only ever seen symmetry forced to 4x, but never higher than that.

As for symmetryMode, I have no data, but I have a few guesses...

It could be something to tell the game to treat a part's symmetry differently depending on if its in the SPH or the VAB

Or, it could be some as of yet not-quite-implemented feature

Or, it could be just a dead end value that does not actually have any actions attached to it (meaning that it effectively does nothing no matter what the value)

[EndlessWaves]

but an "enum" is new to me. care to explain?

Enumerator. Basically a data type with a fixed but custom list of values. For example vessel status is an enum containing PRELAUNCH, FLYING, ORBITING, SPLASHED and so on.

[Nuke]

i noticed that to get 6x symmetry, a value of 5 seems to work, a value of 4 doesn't seem to work at all. after sciman brought to my attention that i was using the symmetry setting incorrectly in my half meter parts, my first guess was that stack symmetry values corresponded with the symmetry settings in the vab (supporting the enumeration hypothesis). i tried a value of 3 in my quad coupler and it worked. i inadvertently changed symmetry in my 19 coupler to 5 (theoretically it should have been 4 but when i tested it it worked as expected), and this allowed me to fill in the inner 6-ring with 1 engine placement and the outer 12-ring with 2 engine placements. id bet that 4 is reserved for some as of yet unimplemented feature (5 way symmetry?). id suggest trying a value of 6 or 7 for 8 way symmetry. i dont have an 8 way adapter so i cant test this.

Edited October 14, 2012 by Nuke