KeyFrameMovie Stretched Behavior

[fractalwizz]

So I've been testing out Kalles Fraktaler for a few days, and I thought one of my zooms would render well with a stretched palette. The images generated from the main program appear fine. I then migrated the zoom sequence to the movie maker, and set stretched palette there too.

However, in the Test Only preview, the stretching of the palette behaves differently than expected in some places. Further Analysis shows that the palette stretching goes from the minimum iteration of the picture to the largest encountered iteration count. In some places in the sequence, the max iter jumps from about 10000 to ~30000 over the course of one zoom. This results in large portions of the image covered by a single color. This anomaly occurs more than once through the sequence, to the point where it jumps by over 100k iterations, a point which isn't reached again for many zooms.

Is this intended behavior, or is there something wrong with the generated kfb files? Thanks.

p.s.: I can post screenshots of the test anim to illustrate what goes on if requested...

[Kalles Fraktaler]

Yes, because of the chaotic nature of fractals, one pixel may hit a minibrot and get a huge more iterations than all other pixels in a view.
I haven't done much on stretched palette, I am still experimenting and have not uploaded any new versions of KFMM in a long time...

[fractalwizz]

It just seemed weird that stretched palette in Kalles Proper and stretched palette in the MovieMaker behaved differently.

And I understand how the interior offset would mess with the coloring. But there's the thing; we're nowhere near the minibrot when it occurs first.

I have attached images illustrating before the event, after, and where it balances out before occurring again to illustrate how odd this behavior is. Apologies for the low resolution. It was a test render.

[Kalles Fraktaler]

It just seemed weird that stretched palette in Kalles Proper and stretched palette in the MovieMaker behaved differently.

And I understand how the interior offset would mess with the coloring. But there's the thing; we're nowhere near the minibrot when it occurs first.

I have attached images illustrating before the event, after, and where it balances out before occurring again to illustrate how odd this behavior is. Apologies for the low resolution. It was a test render.

The reason may be that there can be big jumps in max/min iterations per frame, and I tried to limit the jumps, by using lag in the iteration itervals and cutting off the most high-iteration pixels, and stuff. But I haven't got it to work very well, and not all locations can be rendered good this way.
So it is difficult

[fractalwizz]

I believe I understand. So is that why the max iteration becomes locked until another higher iteration count is present?