Sign in to follow this  
Steven Mading

Mirror symmetry for triangular panels seems all wrong

Recommended Posts

These new Triangular panels made me pretty excited at first, until I tried using them in the SPH and found the mirror symmetry feels all wonky and wrong for them, making it hard to design certain things with them.

Look at these screen shots - the mirrored symmetry isn't making actual mirrored parts.  It's behaving just like radial symmetry:

The first two images are with Mirrored symmetry.  The last image is with radial symmetry to compare.  I don't see any difference.

Album 9HGrq will appear when post is submitted

Edited by Steven Mading
Imgur album link wasn't working

Share this post


Link to post
Share on other sites

Just tried and it is behaving like mirror except the part ist 180° rotated around the z axis, which it imho shouldn't be. You can see that if you put it a bit offcenter.

Edited by Hurzelgnurk

Share this post


Link to post
Share on other sites
On 5/19/2019 at 12:16 AM, EwingKang said:

This is very frustrating, considering how the bug makes the exterior shaping of my aircraft quite annoying. It should only needs a very simple fix.

Please help upvote this if anyone is having this issue. https://bugs.kerbalspaceprogram.com/issues/20762

Upvoted. IIRC it's a pretty simple config file fix.This problem also exists with the triangular armor parts in BDA.

Share this post


Link to post
Share on other sites
11 hours ago, FleshJeb said:

IIRC it's a pretty simple config file fix.

I haven't seen a config-file fix for this one. 
Adjusting the attachment orientation in the config-files somewhat fixes the similar problem with mk2 textures flipping, but I could not see a way for it to help these triangles.  IKSP depends on every surface-attachable part having mirror symmetry across  the x/y plane (the plane of an airplane wing in the SPH) when the part is placed before any rotation.  The M.H. credits list many freelance artists, and I can imaging this convention was not pointed out to them.

After thinking this through, it seems there probably is a config-file solution involving MODEL { rotation = ax, ay, az}   That would break old craft, so if Squad uses that method they would need to introduce a version-2 of the triangular parts, letting old craft use the version-1.

If anyone has a working config-file patch, please post it to the bug report

Share this post


Link to post
Share on other sites

Just ran into this issue, incredibly frustrating and makes the structural parts mostly useless for what I want to do with them. I would really like to see this fixed, especially considering they would be pretty handy for the upcoming robotics stuff.

Share this post


Link to post
Share on other sites
Posted (edited)

Rotating the models into the KSP coordinates so that they have an x-y plane of symmetry, like wing parts, breaks old craft that used the triangles
(because the craft file only stores how parts already in KSP coordinates are rotated onto the craft).

But new craft built with this MM patch have the triangles  follow mirror symmetry rules, most of the time.

Spoiler

@PART[Triangle*] {
  s = #$/node_attach[0]$
  @node_stack_top       = #   0, $s$, 0,  0,1,0, 0
  @node_stack_left      = # $s$,  0,  0,  1,0,0, 0
  @node_stack_topleft   = # $s$, $s$, 0,  0,1,0, 0
  @node_stack_lefttop   = # $s$, $s$, 0,  1,0,0, 0
  @node_stack_topright  = #-$s$, $s$, 0,  0,1,0, 0
  @node_stack_righttop  = #-$s$, $s$, 0, -1,0,0, 0
  @node_stack_bottomleft =# $s$,-$s$, 0,  0,-1,0,0
  @node_stack_leftbottom =# $s$,-$s$, 0,  1,0,0, 0
  @node_stack_edge      =     0,  0,  0, -1,-1,0,0
  @node_attach          = # $s$,  0,  0,  1,0,0, 0
  !s = delete
  @MODEL {
    %rotation = -90, 0, 0
    // drag cubes are automatically generated with the correct rotation
  }
}

@PART[EquiTriangle*] {
  !NODE,* {} // no need for variant nodes
  s = #$/node_attach[2]$
  @s *= 1.15
  u = #$s$
  @u *= 0.866
  v = #$s$
  @v *= -0.5
  %node_stack_left    = # $s$, 0,  0,   1,   0,  0, 0
  %node_stack_top     = # $v$, $u$,0,-0.5, 0.866,0, 0
  %node_stack_bottom  = # $v$,-$u$,0,-0.5,-0.866,0, 0
  %node_attach =  # $s$,0,0,  1,0,0, 0
  t = #$/node_stack_back[1]$
  %node_stack_back = # 0,0,$t$,  0,0,1,  0
  @t *= -1
  %node_stack_front= # 0,0,$t$, 0,0,-1, 0
  !s = delete
  !t = delete
  !u = delete
  !v = delete
  @MODEL {
    rotation= 120, 90, 90
  }
}

 

Node-attaching triangles to nodes of other triangles that are mirror-symmetry attached to a parent can sometimes confuse the symmetry.  Re-attaching the branch of parts to the parent usually corrects the errors in mirroring.  However, node-attachments along the line of the mirror symmetry, along long axis of a typical aircraft,  result in the child triangles failing to mirror.

Surface attaching triangles to triangles works fine.

It looks like the KSP mirror-symmetry system has some historical complications, and doesn't handle the generality we can build with all those nodes on the M.H. parts.  Having them surface attach like wing parts seems wiser in hidnsight.

Edited by OHara
simpler patch

Share this post


Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this