parameters read incorrectly

[LegalizeAdulthood]

legalize[CodePlex]
From: jackoftradez@comcast.net Anyone who has seen my deep zoom viddies has noticed the occasional jerk frames. I do not know what causes this, but it seems that fractint reads the PAR file wrong and scrambles the numbers for some reason. Thus with off-parameters or coordinates the image(s) are a bit distorted or off center.

Now, attempting a zoom animation of Muth FOTD 3_15_07 I have encountered this again, only the error occurs in the magnification number - the first time I have seen this! I need someone to verify that I am not hallucinating. Attached are PARS for 3 images in the sequence.

You clearly see the mag factors are 1.35e+12, 1.37e+12, and 1.40e+12 respectively. The first and last render correctly, but the middle one ends up with a mag factor of 6.88e11 !

Why is this and where does the number come from? It is totally reproduceable too!

You dont have to render the entire image (they take awhile). Just a couple lines then hit TAB key to see the numbers.

A clue might be that this only occurs in the last third of the sequence - beyond e12. There are so many messed up frames an otherwize awesome zoom is not now good for posting!

Whoever solves the mystery gets a free DVD!

[LegalizeAdulthood]

zooming.par:

782.0000          { ;
  reset=2004 type=formula formulafile=allinone.frm formulaname=mandelbrotbc2 passes=1
  center-mag=-0.63982851388294900/+1.17265143805666000/1.35148269130032e+12
  params=1.5/0/85/0 float=y maxiter=5000 inside=0 periodicity=10
  }

783.0000          { ;
  reset=2004 type=formula formulafile=allinone.frm formulaname=mandelbrotbc2 passes=1
  center-mag=-0.63982851388294900/+1.17265143805666000/1.37763961568162e+12
  params=1.5/0/85/0 float=y maxiter=5000 inside=0 periodicity=10
  }

784.0000          { ;
  reset=2004 type=formula formulafile=allinone.frm formulaname=mandelbrotbc2 passes=1
  center-mag=-0.63982851388294900/+1.17265143805666000/1.40426842987625e+12
  params=1.5/0/85/0 float=y maxiter=5000 inside=0 periodicity=10
  }

frm:MandelbrotBC2   { ; by several Fractint users
  e=p1, a=imag(p2)+100, p=real(p2)+PI
  q=2*PI*floor(p/(2*PI)), r=real(p2)-q
  Z=C=Pixel:
    Z=log(Z)
    IF(imag(Z) > r)
      Z=Z+flip(2*PI)
    ENDIF
    Z=exp(e*(Z+flip(q)))+C
  |Z| < a }

[LegalizeAdulthood]

I couldn't reproduce this from the given par, so I'm closing unless more information shows that it still happens.