It ought to crash regardless of environment.
Just sh[0] 3
would suffice.
It ought to crash regardless of environment.
Just sh[0] 3
would suffice.
I’m guessing it crashes on CLI, well, I’ll be fixing that then. I’m not sure if I did copy and pasted wrong though I really doubt it. I would like to have a log during the meantime, I am using shared on the binary colors textures.
While developing the shared version, I am confuzzled by something.
channels 0
f 1 +f. 1 . a c hsi2rgb
Why am I seeing that the range is on 0,765? Is this a bug? I guess I’ll just use cut 0,255 for this case.
The equivalent to your command is
gmic 100%,100%,1,3,1 hsi2rgb
HSI(1,1,1)
isn’t a (normal) colour, so it isn’t surprising that the corresponding RGB would be out of bounds. Although I don’t understand why
gmic 100%,100%,1,3,1 hsi2rgb rgb2hsi
would have the stats
..., std = 3.332e-008, coords_min = (0,0,0,0), coords_max = (0,0,0,1).
See std
. Speaking of stats, what are coords_min
and coords_max
? @David_Tschumperle
Never mind about earlier comment. I would like to know where does Picture Mosaic would go in? Patterns or Rendering or Array and Tiles.
I haven’t categorized my filters before. David made my community *.gmic file.
Personally, I would dump everything into Testing
→ Reptorian
because you make changes all of the time and having it in one folder helps in discovery too.
The trouble with “Testing” is that some users might avoid that category because it sounds tentative or risky. I wonder if we could simply have a user category at the top… but I would be hesitant to do that without asking David first.
Not so surprised, this is probably due to rounding errors occuring because of the double conversion hsi<->rgb. Almost all transforms that operate on image values share this behavior ($ gmic sp lena fft ifft
is a typical one).
coords_min
and coords_max
are simply the coordinates of the min/max value in your image.
In your example, it seems clear that the green channel after the double conversion becomes a bit higher (but with a very low difference, as std
tells us).
So maybe “Testing” is not a nice name ? “Users” may be more appropriate ?
I would agree with calling the category “Users” since the filters under that category are not all highly-experimental ones.
Chiming in here to support the name change.
Name change is probably good, whether “Users” is the best I’m not so sure… perhaps there are other options (“community” or something?). It just reminds me of the old IT joke - two lines of work refer to people as “users”; personal computing and drug rehab.
Users sound less risky, and might promote more users to develop filters. Community is also another option. What about G'MIC Users' Created Filters
. A bit too long, but that sounds more like it.
I am fine with Community
or G'MIC Community
.
I disagree. Not all filter devs are typical users and not all of their filters are in this category.
Community
is the best IMHO.
Now, what is the best ?
Community/
Community/Reptorian/
Community/Afre/
...
or
Community/Artistic/
Community/Deformations/
...
Also I think I’ll keep the Testing/
category alive, with filters that are not stable or finished.
It would be nice if filters dev could decide reasonably whether their filters go to Community/
or Testing/
.
The part about unfinished/stable filters, I don’t know how to solve that issue. I imagine some sort of oversight to have users to verify that they are indeed finished or stable. If not, then it should go right back at Testing.
Voting feature for G’MIC?
Thinking about situations where categories are used, somebody might be browsing or trying to find a filter to suit a purpose. In that case, category by type rather than filter creator seems more useful. If you’re searching by filter maker you probably have an idea of the filter name anyway (and can use the search).
If “Testing” remains, I’d probably use that for filters in progress - then when I’m done tweaking it for a while, move it elsewhere.
So in that case, why not moving the ‘finished’ filters done by community directly into the existing category ?
Just as it was intended at the beginning
Sometimes that should probably happen, but I like the idea of making it obvious which filters are official. It’s good to have a high standard for those. Quite often I don’t get around to (or forget) the final “polish” - things like handling opacity, giving good error messages for number of layers, preview options and so on…
I propose that Testing
becomes
Development → name
and old or buggy filters go here
Maintenance → name
Why not Upcoming -> name
?
Maybe it’s a more “positive” name ?
I have some bit of news. Right now, I’m back to revisiting my construction material texture. I’m going to attempt to try to get close to a real road texture as I’m adding secondary turbulence and noise filtering alongside with breaking them into shapes, and adding distance transform. That way, users will have the option to stimulate road texture. After that, I’ll be releasing these new filters into g’mic:
Binary Alternating Quaddro Texture [Basic] - Description : A filter based off integrating Asmageddon texture generators found in this thread source code - Asmageddon Tools Pack v3 - Plugin Packs - paint.net Forum
The filters that have been combined into this one basic filter is Quadro Bumperoo,Quadro inverto,Quadro Sharper
Binary Alternating Quaddro Texture [Multichannel] - The multichannel version of the basic code with multiple color space supports, and 8-bit processing as a option. RGB(A), CMYK(A),LCH(A)… are all supported. You can even swap channels in cases you switch into something like HSL, and LCH, but just want to reorder luminosity channel to fit the color space change.
Sinosoidal Water Distortion - A filter that can degrade image with sinusoidal image. This gives mostly glitch-like effect.
Logarithmic Distortion - Just as title said. Though, it doesn’t yet offer pinching function, but it does support Kaleidoscope output. Heavily extended version of Asmageddon Prince distortion tool. Name been changed for a more appropriate name.
Squareroot Logarithmic Distortion - Similar to above, but actually use squareroot function into it.
Strange Bulger - Integrated Strange Bulger A, and Strange Bulger B with mirror function, and more interpolation options.
Strange Bulger Kaleidoscope - The Kaleidoscope version of above.
Other than that, what’s possible is a fix to nebulous (I actually hate how some color space setting does not align well with black and white, and it would be nice to use luminosity of color results too with contrast setting), and spiral matrix transformation since that issue is solved.
Here’s something I want to show. That’s going to be used as another layer for construction material texture. Does it look realistic? I’m thinking of using it as a aid to create road textures.
Also, pinning @hover for more updates.
You’d need some kind of multi-scale control to allow for much greater realism. There’s often sporadic bumps on a road which are pretty much just localised patches of coarseness with a bit of finer grit on top. Also consider trying to model the potholes on a road and maybe the cracks too.