Prompt #7: OpalI ended the last post by asking what the effect would be if a contrastive filter was applied to the "feathered" part of graphics derived from brown128.jpg. Before I get to that, I'd like go back to the enlarged version of the original background tile. (Well, a relabelled version.)
1. In
the third Opal post, the areas indicated in #1 had their colour muted/paled. In this post, though, they are the ones
not being neutralized. Although there are numerous bands in these sections (which may well react differently to the filter), they will remain coloured.
2. Areas that are dark in the original graphic will remain so.
3. The narrow brownish-grey band is
lighter than the surrounding zones. My first attempts at bicolours involved colourizing this part of the graphic. With some of the more highly derived variants, the result was vividly contrastive. However, in many of the variants that I have made, this stripe takes on essentially the same hue as the areas that surround it—though it remains distinctly paler:
Once I had figured out how to use contrastive filters to produce the sorts of variants in Opal Part #3, I then tried the same method to strip out all the colour from the
other parts of the graphic, turning them a true grey.
As you can see, in some of these background tiles, the narrow band (#3) is so pale as to appear white.
Naturally, I then thought of trying to get darker variants—ones in which there is a broad dark area (combining #3 with #2) that contrasts strongly with the coloured portions of the graphic. The first thing I did was simply darken the whole thing, using the Contrast on Microsoft Picture Manager to keep the coloured areas (#1 on the enlarged picture above) as distinct as possible from the dark areas (#2 and #3).
Since these variants have had not been affected by contrastive filters, the dark areas retain some colour. The #3 band is therefore tinted, as it was in the first examples up at the top.
It becomes more complicated if one wants these areas
not to be tinted. In particular, any contrastive filter used to counter the hue of the #3 band will also affect the rest of the graphic, lending unwanted colour to the dark zones in #2, which often have a different underlying hue. It requires that the variant to be tweaked be carefully selected; and, even then, there's a little dance back and forth trying to get the ideal balance.
Nevertheless, I do have a few examples of black-and-coloured (or at least dark-and-coloured) variants where the #3 strip is grey
and surrounded by a broad very dark, almost colourless zone.
And, just to prove that what goes around can keep going round and round, I've sometimes overshot on the filter and accidentally
recoloured the #3 band, resulting in a new bicolour!
(In the middle one above, for example, I used a green filter to counter the redness of a brown zone. You can still see traces of the original colour—now turned to olive brown—if you look carefully.)
But it doesn't end here.
On 6 January 2021, it occurred to me to try using contrastive filters
not on dark colours (to produce dark grey to black) but instead on light colours (to produce light grey to white). For this I went back to the one of the gaudy multi-coloured flat variants that I'd created the previous fall. At this point, I'm not sure exactly which one it was, for there are several very similar ones. However, it may have been the one below. Certainly, it was basically purple with yellow bits; but there was also some red in it.
To it, I applied a blue filter. This turned the purple areas indigo-blue; but it completely countered the yellow, turning those areas off-white.
Blue filters have also been applied in the following four examples:
Given how very different the four initial graphics were, it may seem odd that the results are always blue & white (or blue & purple & white). However, the colour that best mutes pale is yellow. Other colours mute darker, often to quite a deep grey. In all cases, therefore, the original graphics I selected were ones with yellow in the feathering, regardless of the hue of the rest of the pattern. If you look at the original graphics of each pair above, you can see that—whatever colour the rest of it is—the part that I have turned white was originally a shade of yellow, golden orange, or chartreuse.
The contrastive colour to yellow is blue, which is why I used a blue filter. As the colour is well saturated, it must be applied fairly strongly. As a result, it winds up tinting the whole graphic. However, the effect of the filter on the other colours will vary, hence the (relatively subtle) differences in the end result.
After I first figured out what to do, I tweaked the result a bit, and then took the best graphic off to GRSites.com to run it round the colour wheel. The white areas always stayed white: it was the coloured bits that were altered by the colour-wheel-rotation doohickey. That gave me a full range of combinations of colour-and-white variants; and I then tweaked
those graphics some more.
Here are a further selection of graphics with areas of white—or at least pale tints—
within a dark zone. Some are obvious bicolours, while others have tonal variation.
One last point: as I described in
the Rose Quartz post, it is possible to
retint the white areas. Most often, when I do this, it is to add just a touch of yellow to give it a creamy colour. However, the variants below show other combinations:
Previous Days:
Prompts #1 and #2 (Amber and Topaz)Prompt #2 (Rose Quartz)Prompt #3 (Garnet)Prompts #3 and #4 (Moonstone and Hematite)Prompt #4 (Kyanite)Prompt #5 (Peridot)Prompt #5 (Bloodstone)Prompt #6 (Aquamarine)Prompt #6 (Amethyst)Prompt #7 (Sunshine Jasper)Bonus #1 (Turquoise)Bonus #2 (Smoky Quartz and Onyx)Bonus #3 (Rubellite)Prompt #7 (Opal) - Part OnePrompt #7 (Opal) - Part TwoPrompt #7 (Opal) - Part Two-and-a-halfPrompt #7 (Opal) - Part Three![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
