ffs you're missing the point
>Initial opening of the image is fine:
>Once I do any edits, it looks like this:
but then you post
>t is supposed to create a chromatic aberration effect which is why the RGB channels are split up
and
>forgot to mention that the displayed image after the edit is not what it looks like. It shows way too much separation and shows shifts of bands that I didn't even touch yet.
Can't you properly isolate the issue by temporarily avoiding/disabling *any* modification of the image - including the chromatic aberration stuff.
Right now I can't tell which parts of the fucked upedness are intended and which are not.
/dpt/ daily programming thread - Go edition
Brandon Adams
Owen Perez
Is Android studio necessary to develop for Android? I want to keep muh Vim but I want a custom-made IRC client to flamewar in class.
Aiden Reed
so you're saying that fp imposes that data shouldn't be limited by anything?
so tell me why immutability is needed, please
tell me why fp langs generates so much garbage objects
Lucas Edwards
Here's a benchmark of Rust and Haskell, where Haskell uses rapidly allocating linked lists and Rust uses in-place mutating allocators:
fpcomplete.com
Haskell still manages to have respectable performance, thanks to a combination of good compiler output and good GC performance.
Caleb Miller
I see uncaffeinated all the time on r/progamminglanguages. There's this guy Jon Goodwin over there working on a language called Cone which is pretty neat.
Benjamin Hernandez
>fpcomplete.com
Leo Green
I'm writing a pure functional language (proof of concept) with a borrow checker, homeboy.
Isaac Sullivan
i need python guys
Levi Walker
alright, managed to fan up a decent shitstorm
time to fold'em
Hunter Lee
Oh ok. I misunderstood what you were asking. I commented out the scaling and chromatic aberration stuff. It seems like when it converts from PIL to wx.Image, it gets fucked up. So the code in this post: