I have been super busy. Thought I might share something I played around with in August. Got a broccoli, sliced and photographed it for good old G’MIC fun.
If you are feeling stressed or depressed today, just take a moment with the broccoli!
An artistic filter you made?
Reminds me a bit of Cairo Pentagonal tiling.
Ghostify? Could be made as a new filter.
How did you know? Could be a filter as long as I keep the code, which I haven’t been doing. Another reason for the lack of filter production is code dependency: temp, changed or removed code makes it hard to replicate the good old results.
Pairs well with a soft light blend.
Indeed it works again [prawnsushi]
I’m trying to figure out why afre’s afre_contrast stopped working now, but I’m not very good at figuring out all the commands of GMIC and afre also uses his own addons in his filters, so I don’t know if the bug is in the afre_contrast part or in one of the others he uses. But I try to see if he uses those subroutines in other filters that do work, so I know it’s not that. like afre_orien, seems to be fine that one
And it’s not even a long bit of code, but I’m trying to decipher the bits, at least there is a chance I learn a bit from trying I figure, although it takes a lot of time.
I get an error that says:
*** Error *** Command ‘afre_contrast’: Undefined argument ‘$2’, in expression ‘$2’ (for 1 argument specified).
And there is a $2 in the code but the line is confusing to me because I’m still not at ease with the GMIC terminology and short forms.
And I’m very lacking in math skills, it is one of the things that really annoy me. If you have some math skills you should take a moment and feel good about it is my advise 
I now know that afre_y50 is OK too but his darksky filter has the same error as the afre_contrast one in gmic_qt, perhaps it’s a clue.
edit: Not much of a clue, the darksky filter uses afre_contrast, so it’s the same error and we are back at afre_contrast.
And his softlight filter has another error, a weird one, says it requires images with 1 or 3 channels, even when that is what it has. I’ll pass over that one.
ContrastFFT also has the 1 or 3 channel image required error.
Sorry about that. The filters are old in G’MIC time. I have been busy with life, so it may be a few weeks before I get to following up with you or fixing things if they are broken. People here are awesome. Perhaps, they can assist you in the meantime. I am flattered that you want to try my filters though; not sure if they are truly useful.
Don’t worry about it afre, I know the code is pretty old and also marked as ‘testing’.
I’m just curious what changed in 3.6 that borked it. David mentioned a few changes but I can’t directly link them to the errors myself.
But you are not required to fix it, or to try, it’s all just for fun.
I did use the afre_contrast filter from time to time though, I like how it fixes contrast issues without making the image too pale or dark like a normal contrast adjustment tends to do, or at least that’s how I perceive it.
I realize I’m doing it again, ‘hijacking’ a thread as it were with discussing something outside the scope really. The scope being prawnsushi’s stuff rather than the tangent of afre’s filters.
Sorry about that.
Moved to the original G’MIC fun page. 
Well, i did a little check and it seems @afre has added a new boolean parameter (ignore alpha) to afre_contrast around 2 months ago. But it works for me. The problem in Dark Sky is that afre_contrast is called without the second parameter ($2) on line 149 :
median. {$rd*2+1} n. $r1 fi c. 0%,0 +. 127 afre_contrast. {$2}
It should be:
afre_contrast. {$2},0 or afre_contrast. {$2},1
I’ve made a local copy of the script and it works if you change that line.
So if you are using afre_contrast in your scripts, you’ll have to add the second param too.
Are you my twin? 
Thanks prawnsushi, I already realized a while after my… well, idiocy, that I was being stupid.
You see in another thread a while before this (embarrassing) thing, I was discussing how I had trouble with the alpha channel and those filters, so he tried an update I could try and uploaded it, but then the update system was blocked a while for the roll-out of a new version, so he gave me a script I cold try locally.
Point being that the extra argument was freaking initiated by myself, and then I go and can’t figure why my old scripts had issues after GMIC was updated again…
Conclusion: I seem to be an idiot 
So not a twin after all (or the runt of the litter?)