Multithreading not working for static class functions

It seems that if my formula uses a static class function, it's rendered only on 1 CPU using "Render to Disk" when at arbitrary precision depth. If I write the formula inline it works fine.

There are no special rules for static functions: it depends on from where you call them. Can you share an example?

class Test {
  static complex func f(complex z, complex c)
    return z^3/(z+.3) + c
  endfunc
}

Testbrot {
init:
  z =  0
loop:
   ;z = z^3/(z+.3) + #pixel
   z = Test.f(z,#pixel)
bailout:
  |z| <= @bailout
default:
  title = "TestBrot"
  center = (0, 0)
  float param bailout
    caption = "Bailout value"
    default = 1e16
  endparam
}

Fractal1 {
fractal:
  title="Fractal1" width=800 height=600 layers=1
  credits="user;6/17/2018" antialiasing=yes
layer:
  caption="Background" opacity=100
mapping:
  center=-1.1116464547149054924890175/0.4896797411711562585172589
  magn=1.9321599E16
formula:
  maxiter=19811 adjust=yes filename="Test.ufm" entry="Testbrot"
  p_bailout=1e16
inside:
  transfer=none
outside:
  transfer=linear filename="Standard.ucl" entry="Basic"
  p_type=Iteration
gradient:
  smooth=yes rotation=-63 index=0 color=13331232 index=104
  color=16777197 index=193 color=43775 index=279 color=512 index=-64
  color=6555392
opacity:
  smooth=no index=0 opacity=255
}

Thanks for the clear example, I could easily reproduce it myself. It's a side-effect of how parameters are passed in arbitrary precision mode, which causes the different threads to wait for each other all the time. I'll try to fix it for a 6.02 update.