post: ask the maker answers and tutorials
deternot asked:
Could you do a guide for how you work with the color red?
raiindust said:
A guide to monochromatic colouring would be fantastic, especially in relation to light and shadows and different tones in the colours.
sarisafari asked:
How do you colour, use light?
I like red. It's a little underused and a little hard to create, but it's fun and when you get it right I think it's really striking. There are three main ways that I use red, or techniques that I use to get red in my icons I guess. Which can really be applied to any color probably, umm, but this is how it works!
1. Inherently Red
AKA when there's enough of a red coloring or red element in the cap to guide the entire icon in that direction:
In the case of the first one, both Amy's hair and shirt are red, which made my inner monochrome junkie happy, and then just some initial coloring and a light pinkish red multiply layer over the whole thing and under a b/w soft light gradient turned the entire icon into a reddish-pink affair. In the second one, Chuck's shirt gives the icon enough red that I hardly did any tweaking to keep the coloring scheme that way. The wood behind her was even reddish, so all I did was up the vibrancy and the yellows a bit. The third cap was just shot in a very red/orange tone, so soft light layers and some vibrancy was all it took to intensify the color already present. Also Amy's hair and scarf again helped.
2. Gradient Map Red
AKA when the cap may not be red at all, but who cares because gradient maps exist:
The first icon was not red in the slightest to start with but lol it was a rainbow challenge and I wanted to make it work. Basically gradient maps can do anything, color-wise. A gradient map layer set to soft light (or even overlay or hard light sometimes) going from bright to dark red to white or to pink or yellow in a light shade changes every level of the cap to the color you want. It takes a little messing around to find the right shades, but after that and with multiple layers on top of each other, the entire tone of the cap will change. The same idea applies to the second icon, which I colored using dark red to white/pink gradient maps and some color fill layers on soft light at a dark reddish purple color. The third icon, because it was made up of three different caps, came together because I used the same gradient map to color all of the layers and, with some extra adjustments, give the icon a more uniform appearance.
3. Texture Red
AKA when the icon is a little red but not red enough, but your textures are red enough for everything:
Angel caps are pretty dark and boring most of the time, and with vibrance and soft light layers and even gradient maps the color doesn't infuse the cap as much as it would otherwise. So on top of the boring coloring I found a red light texture, set it to screen, and added reds and a little bit of yellow to the icon. The color on the top is balanced out with the text on the bottom. The second and third icons are very similar (might even be the same texture, actually!) - just a red texture with some sort of dark color, probably black, for the non-red part, set to screen and moved to suit the composition, frame the subject, and give the icons an overall red tone.
Monochromatic Coloring
This guide will kind of be to answer both raiindust and sarisafari questions (using the examples sarisafari gave), and extend from deternot, because a lot of my icons are monochromatic, at least to some extent, and I use a similar process even on ones that aren't. And lighting is always, always, always important. So, I guess the best way to do this is a little tutorial on a typical monochromatic coloring.
I'll start with this as a base, and attempt to go in the direction of a purple and reddish color scheme because there's some red in the cap but it's maybe not an obvious choice for color:
Start with the usual coloring steps: two screen layers and a soft light, all at 100% opacity:
Next, a gradient map layer set to Normal and going from a dark red (#1f0000) to a light purple (#edc1ff). This sets up the color, but it's pretty flat, and I like my monochrome icons to be rich and with a lot of depth. Move the gradient map layer under the top soft light layer for a bit more color:
Another gradient map on soft light going from dark purple (#590a45) to light blueish purple (#bebefa):
I don't really like the patch of light purple on the left, because it takes away the focus from her face and disrupts the lighting a bit imo, so I paint over that with a soft round brush on a new layer set to multiply using the color found on the section to the right side of her face:
Now her face is the most obvious focus, but I still want more light and depth, which is where my second favorite thing comes in: gradient layers! I love things having to do with gradients. I usually take a white to black gradient, put it on a new layer, and set the layer to soft light, adjusting the opacity (to 67% in this case) until the highlights aren't washed out and the icon has a bit more balance:
That's about it, for coloring! I might add some reds into the shadows with color balance and up the brightness and contrast a bit:
The coloring process usually takes care of any light work I do on the cap itself, and the rest is left up to light textures. For this icon I'll use this texture by nailbites and this one by raiindust, both set to screen, the first at 100% and the second at 60. Ctrl+shift+alt+e the layers together, filter > sharpen that and adjust opacity as needed to look normal, and:
Done!
Four of the examples sarisafari gave, and they were all made using exactly this method. Gradient maps for color, a b/w gradient for depth, and textures for light and, well, texture. The last one looks slightly more complicated, but only because I expirimented with a painty texture and ended up leaving it at 100% opacity.
These don't have b/w gradients, but the first does have a gradient of a dark gold to a dark red set to screen going diagonally from his right elbow to the top left of the icon for light and color in the shadows, and the second has, I believe, a dark purpleish blue fill layer on screen as well as a light texture.
That's pretty much it! Sometimes I do fancy things like selective coloring (gasp) and I used to use vibrance on everything but don't so much anymore, unless I really need to, or if I do I only use a little. Gradient maps man, seriously my best friend. You can do any color with them, and combining them in different modes (hard light can be fun) and masking away parts of them can give you a whole bunch of rich, vibrant colors on caps that maybe didn't have any to begin with. Also, they're great for all kinds of black and whtie and sepia coloring. I don't do anything super special with light except abuse light textures. I tend to use caps most of the time that have stunning light (or the potential for stunning light), so they don't take much work aside from the usual brightness/contrast stuff and gradient layers.
rowofstars said (seconded by fouroux and firstillusion):
I'd kill to know your thoughts on composition, how you approach complexity.
I approach complexity like I would a strange dog: gently and like we're old friends so as not to scare it off. I just pulled that simile out of my ass, but it's true. Most of the time, I go into PS not knowing what will happen, and without any sort of definitive image in my mind as to what I want my composition to be, because when I do have a "plan" or mental sketch, what I actually make is nothing like what I pictured. For example, of the icons you wondered about, only one of them was born of a thought-out concept, and only the concept remained the same, not the design itself. I'll go through each of the icons you wondered about and give a general process.
In my head, this icon was mostly black with a lot of negative space and Eleven cut out of his background and placed on either plain black or black with that orange stripe continued, and the Dalek in the middle as sort of a focus, but blended into the rest of the image. But I tried that and realized that with a normal-colored Dalek there was no way it would stand out against Eleven's jacket/the background, so I moved my crop on Eleven closer, kept the background how it was already, and scrolled through my textures, trying out random ones until I found that one, which not only framed the Dalek, but balanced out the orange square of the left side of the icon. The coloring wasn't much in terms of work, just a light orange layer on multiply to bring everything together (which apparently got erased a little over the texture, ugh, why did no one tell me I can see it on this monitor but I couldn't on my old one this is going to bug me now).
For both of these icons, I knew I wanted to blend with silhouettes (the DW one was for a silhouette challenge), but that's as far as my planning went. That's as far as it ever goes, really. Then it's just a matter of finding the right images to fit together, and arranging them in a way that's new and interesting. Like, that Eleven silhouette is a cap I've used dozens of times, so this time I brought the crop a little closer and rotated it. I added the TARDIS (set to screen and erased where it extended beyond the line of his face) to make it more than just a silhouette, and the blue texture at the bottom served to bring it from that cap's usual coloring of red-orange-yellow to more of a rainbow scheme. The silhouette of Luther has been used many times as well, so I combined it with the other image (set to screen or maybe lighten I don't remember) and added the text in the corner to balance it a bit and extend the lighter colors.
What I tend to go for in complex icons: visually interesting and with enough elements to intrigue the eye, yet with a clear focus, subject or subjects. Colors that are either similar or complement each other. Lighting as a way to bring blended images together and give a cohesive look to the icon. Textures as a means of connection between segments of an icon or between differently colored caps, subjects, etc.
rocketgirl2 said:
I'd also love to hear your process on Tumblr graphics.
And then I frantically tried to come up with an actual process for Tumblr graphics um okay I suppose I have three different general "methods" of coming up with larger graphics:
- Cap-based
AKA when I have a cap that I really want to use and I build my entire graphic around that cap.
This tends to result in a lot of simple graphics or close-cropped graphics (the first two examples). When I have a specific cap in mind when I make something I tend to forget about everything else, complexity, blending, etc, and if I can't find a cap on my hard drive that matches or blends well with the one I want to use, I simply don't use it. These types of graphics tend to end up either closely cropped, if the cap is good quality, or with negative space if it's not. Also I like to get lazy with these and go b/w or monochrome instead of trying to color something that big. The X-Files graphic is an example of cap-based but complex - I found a lot of caps of that episode that worked well blended together.
- Composition-based
AKA when I know I want to make a graphic with blocking or blending, or one with a colored image blended into a b/w image, or a graphic with small text, or a graphic with large text. If I want to go crazy with textures or keep it simple.
With these types of graphics, I come up with what I want to work on and then go searching for caps to fit my idea. The first one was me stubbornly committed to creating a graphic with some sort of painted on element, and it was a pain but I did it because I couldn't get the concept out of my head. The second one was one I knew I wanted to blend in (that shadowed cap was calling to me), and I also wanted to give it sort of a regal feeling. The third one echoes that regal element, but the graphic originated because I knew I wanted to use those lyrics and I knew I wanted fancy text.
- Concept-based
AKA when I have a super-cool idea!11!!!1!1 and then have to actually make it happen somehow.
I tend to spend more time on these than any other graphics, because when I want to convey a message I have to actually work at making that message come across. For example, if I'm ambitious enough to want to encapsulate the three facets of Mary Winchester as they appear in a Laura Marling song. Or if I want to make a Hush graphic focused on mouth crops, presenting speech as the primary method of communication and the obscurity and confusion that comes with the loss of speech. Or if I want to attempt to embody the extent of Sam and Dean's epic love in What Is and What Should Never Be. Etcetera.
As far as actual process goes, I usually color caps first (keeping in mind a color scheme if I want one and it doesn't just happen) then paste them into a canvas (my favorite default size is 500x600 px). From there I resize them as needed and blend them (screen and lighten are my bffs) and, if the colors don't go, smother everything in b/w or sepia gradient maps. Then I scroll through my giant-ass texture folder, in which there are thousands of large textures (I have a separate folder for my thousands of icon textures) and usually select 5 or so textures I think I might want to use, drag them into photoshop, and start pasting and playing with different compositions and effects and blending modes and opacities. Lately I've been using some filters like dry brush and colored pencil to give texture to duplicated caps or graphics which I then erase and make transparent as needed. The mosaic filter under pixellate is also a useful one I discovered recently for getting a bit of a geometric or pixellated look. Generally I have one or two textures that I use to further the composition, add lines and shapes or colors, and one that's a dark, black and white grunge texture pasted over everything, set to screen, and adjusted in the opacity. If I'm using text, that's the last step (sometimes before the textures, though, if I have a clear idea of what I want for text), and the text varies a lot, though I'm fond of small serif fonts with a kerning of 100 or so, and handwriting fonts with similar kerning. Big text can also be fun, especially for quotes. I don't know, it's just an expirimentation thing, which covers pretty much my whole process and how I learned to make big graphics in the first place. Try everything, and something will work!
That concludes my Q&A session! I hope this was helpful and whatnot. If you have any questions or want me to explain something else, leave me a comment and I'll see what I can do ♥