The manual explains in detail why in some cases you get a huge number of copies. That number of copies has to be a common multiple of the H and V cycles, and the H and V cycles are each a common multiple of the step and the layer dimension. So with random number things can get big.
I was playing with this script and as much as I tried different configurations, the result was always a non-constant animation.
I varied the number of steps, created a seamless layer, etc. I always got an oscillating gif, as if it stuck in some parts.
So I saved the .xcf and decided to check with version 2.10.18 and voilà.
In summary, at least with my Samj portable version 2.10.12 I was unable to generate results with continuous transitions and without bumps.
With version Samj's Portable 2.10.18 I didn't have a single problem of this type.
There should be no jumps if your layer is seamless, and I don't thing the Gimp version comes into play. Otherwise please share the XCF (before the animation is generated).
[attachment=4316 Wrote:Ofnuts pid='17813' dateline='1587672701']There should be no jumps if your layer is seamless, and I don't thing the Gimp version comes into play. Otherwise please share the XCF (before the animation is generated).
05-04-2020, 08:59 PM (This post was last modified: 05-04-2020, 09:13 PM by Ofnuts.)
Taking both pictures, exporting all layers as TGA (PNG includes a timestamp so identical images would produce different files), then taking the MD5 of each file for easy comparison:
So you have several clusters of files where the image is the same which is why you have this choppy animation. To be accurate you have two 5-frames breaks:
14-15-16-17-18
29-30-01-02-03
They are spaced exactly 15 frames apart...
And of course, I find these identical frames in the pre-animation version your posted. But I still have a very hard time believing that this would be due to the version of Gimp... Do you have the same problem with different steps (-4,-5,-6,-12)?
05-04-2020, 11:13 PM (This post was last modified: 05-05-2020, 12:04 AM by Krikor.)
I used the same .xcf to generate both animations.
After applying the scroll-layer + interleave layers and generating the animation 01 (Gimp 2.10.12), I opened the .xcf in Gimp 2.10.18, deleted the layers created with the scroll-layers in gimp 2.10.12 and created new layers this time using the scroll-layer with gimp 2.10.18 and then applying the interleave layers.
In common both versions had two initial layers:
001 copy
Base
(05-04-2020, 08:59 PM)Ofnuts Wrote: Taking both pictures, exporting all layers as TGA (PNG includes a timestamp so identical images would produce different files), then taking the MD5 of each file for easy comparison:
Code:
...
So you have several clusters of files where the image is the same which is why you have this choppy animation. To be accurate you have two 5-frames breaks:
14-15-16-17-18
29-30-01-02-03
They are spaced exactly 15 frames apart...
And of course, I find these identical frames in the pre-animation version your posted. But I still have a very hard time believing that this would be due to the version of Gimp... Do you have the same problem with different steps (-4,-5,-6,-12)?
Sorry, I don't understand what steps that would be (-4, -5, -6, -12).
(Some minutes after...)
Oops ... do you say steps in the scroll-layer script? (horizontal and vertical)
I haven't touched it again for maybe 10 days, but vaguely I remember that I tried several steps and usually always gave this problem, but I have a vague reminder that when the number of layers was quite large I didn't notice the oscillation (dysrhythmia / jumping ) in the animation.
![[Image: TNn94Sm.gif]](https://i.imgur.com/TNn94Sm.gif)
![[Image: url]]](https://i.imgur.com/d68vYMt.gif[/url])
![[Image: url]]](https://i.imgur.com/fVTatvQ.gif[/url])
![[Image: o6OxjTV.gif]](https://i.imgur.com/o6OxjTV.gif)
ScrollLayerReady.xcf (Size: 670.03 KB / Downloads: 244)
- Scroll-layer With 2.10.12
