Reptillian

Filters that outputs multiple layer is not going to work on PDN because filters are restricted to one. @G'MIC David, could you offer a workaround for PDN users via export?

I can do that, but I'm thinking this would imply that either I should fill with nearest nonexisting colors if number of areas exceed a palette, or refill it with the current palette, or to do with random colors. It would take some time for me to implement, but right now I'm having bad case of insomnia, so it might not be possible any time soon on my end. That, and I do have some plans to fix my filter or implement a new filter.

Thorny Fractal - Secant Sea filter is on G'MIC. Also, I wanted to share this. That was made with the Thorny Fractal - Secant Sea filter. Here's the setting Escape - 1000000 Loop - 353 Subsampling Level - 8.929 Distortion X Formula - Petallian Distortion X - 7 Distortion Y - -8.3 Random Color Map - Colours - 6 Seed - 2116 After that, it's just modifying the color. I did applied it to a much bigger picture though.

Thorn Fractal - Secant Sea Filter is coming soon - Pull Request : https://github.com/dtschump/gmic-community/pull/210 Some images of the filters - It's basically a different version of Strange Attractor made by @MadJik

Deleted because it's not necessary anymore.

He probably had those code way before, but was finishing them. Am I right, @Xhin?

Deleted because it's not necessary anymore.

Delete this please.

@lynxster4 No worry about the influence, all I wanted was a bit of feedback. What I do is up to no one. I fixed Alpha Normalization bug on G'MIC Fragment Blur and made a pull request. Also new filter which is on the same pull request. @Eli @Iota16 Thank you so much for encouragement for this filter.

Paint.NET doesn't support batch processing. In any case, I'm sending you a code that works with g'mic qt as it support batch processing.

I didn't made a PDN plugin, but a g'mic qt solution. Is this what you're looking for? You won't see a repeating color here. EDIT: Also have an alpha-based solution so you can just paste

@lynxster4 There's two filters that I don't see useful enough to be worth fixing, but I really like to know if you think Vibrato and Channel Overblur... filters are useful enough for me to extend them. I could add anti-aliasing, and blending mode options to vibrato as well as make the gradient continuous. Next filters I'll be fixing Color Modulo Texture (Tricky) Restore color manipulation toolbox for Transfer Colors [Reduced] filter

After a huge amount of effort, I have solved the fragment blur. Now I know how it works. Details here - https://github.com/dtschump/gmic-community/pull/207 Image that was created with G'MIC Fragment Blur - Please ignore the border, I'm not sure how to fix that pasting bug.

People usually post them in threads. I'm guessing you're asking if the process would be automated from Paint.NET itself. This is a question to @Rick Brewster

I'm not seeing that it looks like glass, but it looks very interesting. Rather artistic. Also, the edge of the drawing, it looks dark. If you could add boundary condition, that would fix that problem. I have no idea how to do that with code lab. Neumann boundary would be appropriate here.

You know the OP could be referring to frames of video, right? To the OP, you need to use a plugin that supports CLUTs. As far as I know, G'MIC-QT is the only viable answer other than gradient map and blending modes. 'Colors / Color Presets' in G'MIC-QT offers 500+ CLUTs.

I'm working on improving Fragment Blur for G'MIC. Here's some improvements. Speed is over 500 times faster though not as fast as PDN version, but it's usable in many more cases. From a test I did with 50 iteration with a image that has 4000x3000. It goes from 30-90 minutes to 6 seconds. That's how fast it is now. Better handling of picture info. If some info are not needed to be processed, they will not be processed. In some cases, some options are taken out because it doesn't match expectations. Added bug handling option, in case of image going over-range or under-range as a result of the filter. You probably are not going to noticed, but just in case, you have this option. It behaves almost 100% proper to 100% proper in all cases. The current one has issues with certain images. The new one will fix it to the point where it's usable in every cases. So, with color space option and speed and bug handling options, it's now getting to be a go-to option if you want a little extra. Note - Bug handling option has another side which is that it can improve the result. So, you could try that when it gets here. But for now, I'm working on it.

The reason why I didn't provide you screenshot earlier is that it was part of the filter development. I had posted the thread in order to generate the link as the link isn't as simple as URL one can make to direct to this thread once ready. There is a number on the URL, and that's not something I can guess. After the link is generated, I posted it into the g'mic filter I have made so that users can read this tutorial when the filter gets ready. You can click on the link provided by the filter, and then you'll see that it gets you to this thread.

It doesn't help that the source code is unavailable, and the author wanted it to be open-sourced. Could it be possible to decompile it and fix it for 4.2 and release the source code which is what the author wanted? I'm assuming original text means source code. By the way, this might be something I'll take a crack in with G'MIC. Edit: I am staring at this part of the gmic reference and thinking of eval math-lib. Yeah, I can cook it up on gmic. - draw(_#ind,S,x,y,z,c,dx,_dy,_dz,_dc,_opacity,_M,_max_Mdraw(_#ind,S,x,y,z,c,dx,_dy,_dz,_dc,_opacity,_M,_max_M) - I'll get on it soon.

Okay, thread's opened and the filter arrived in G'MIC QT 2.7.2.

G'MIC 2.7.2 is now considered stable. @null54 Now, I think that tutorial thread can be unlocked soon so I can work on it.

I wonder if you can modify that to use 256x1 or 1x256 colors clipboard image as table. It would go well with my recent g'mic filter which basically reverse engineer gradient map from grayscale image and color image of the same size.

My changes that are coming to 2.7.2 after refresh - Fixed Color Existence Distribution - [RGB-8] a bit. Improved Interface. Added Reverse Engineering Gradient Map Filter. Obligatory proof here - https://github.com/dtschump/gmic-community/pull/204/commits/c7e7d3944f1ac1deafe012f19c5940c7a458438f

You have to change the color option to Alpha, then you move the left dot up. I'm not going to walk you toward every steps of this. If it doesn't work : The best way to do it is to understand how healing works, and cloning works. Then you'll be able to fix the issue.