Rachel's Phone (Tutorial)
TUTORIAL: Rachel's Phone
Program: Photoshop 7
Level of Difficulty: Idk, easy, I think.
Tools: Smudge tool, blending modes, layer masks, brushes, curves, selective coloring, filters, variations, etc.
Translatable? Nope. Uses some selective coloring.
'Rachel's Phone'-titled as such because both Will and Rachel are using Rachel's cell/mobile phone in the icon which was supposed to showcase an object-was made for a 20 in 20 set where the category was 'objects'. It was actually a pretty difficult category for me since it isn't so much about what's in the icon per say, but what the icon says. And in the episode Funk of Glee (or, you know, Glee in general) which I was assigned, there aren't that many objects with ~emotional impact~ save for the egg in the egging scene (which also appeared in the category set). So I forced myself to go looking for caps with objects in them, and I pulled out a whole bunch just to see what I could do with them really.
I'd played around with the cap of Rachel more than once without success, and once I found the cap of Will, I was determined to use it somehow. Unfortunately, my first efforts (just cropping Will) hadn't gone down too well. So I saved the cap, and went on to make the rest of the set before coming back to make my last 'object' icon.
Looking through Funk caps, I had a sudden epiphany. As lame as it sounds, I pretty much went "WAIT! A PHONE IS AN OBJECT! AND THEY ARE BOTH USING RACHEL'S PHONE! AWESOME!". I know, I know. Don't judge me. Anyway, I'd already made this Rachel/Will icon, and in case you didn't know I ship everyone in Glee (almost), including Will and Rachel. So I decided putting the two images together would create something cool.
And that's where the actual works starts, lol.
Note: For the purposes of this tutorial, please keep in mind that I'm using a 300x300 canvas for the majority.
Cap Source: Gallery @ rawr-caps.net | Specific caps: RACHEL and WILL
STEP ONE: CREATING CAPS I CAN ACTUALLY USE
As you can see from the original caps (versus the final icon) there's a lot of extra stuff going on in the caps that needs to be gotten rid of before the caps will be useful. I'd already decided I was going to layer the two images, one on top of the other, and I liked the fact that they were on opposite sides (Will on the right, facing left, and Rachel on the left, facing right). So I wanted to eliminate the extra details that might clutter up the focus.
I'll start with Rachel because she is by far the easiest.
First off, I wanted to get of the little bit of extra space to Rachel's left. There's a faint line where the window ends and Rachel's shoulder begins. That seemed like a good place to cut off; so I used the MOVE TOOL to move the cap over so that it ends on that line. I ended up with something that looked like this which is fine and dandy except for the TV logo in the corner and that pesky red banner. I could cut out the middle man, and simply extract Rachel from her background. But, eh, if I can avoid doing that, I do. I have a love/hate relationship with the lasso tool.
So I needed another way to get rid of the extra details that were distracting from Rachel. Right before the red banner starts, there's a strip of blank wall. And it would be really easy to use that to cover up both the logo and the banner! First I tried using the free transform tool because that would be easiest, but I was having trouble getting the selection just right. I kept accidentally catching Rachel's jacket. So I ended up using the SMUDGE TOOL, and I smudged the portions of blank wall all the way to the right.
When I was finished, the logo and banner were gone, but everything to Rachel's right looked streaky. So I duplicated my layer, and used GAUSSIAN BLUR (Filter >> Blur >> Gaussian Blur). Then I added a LAYER MASK to that layer and masked away the blur from Rachel so that the only thing blurred was the wall to the right. The final product looked like this. Then I resized the image so that the width was 300 pixels.
Then it was time to force the cap of Will into submission. We're going to be using more or less the same steps, it just takes a lot more time because there's a lot more clutter in the cap of Will.
Anyway, the cap of Rachel had a yellow wall, so I went with blue as a contrast, and would you look at that? Finn is conveniently wearing a blue shirt! So, using the SMUDGE TOOL again, I smudged the blue of Finn's shirt so that it covered up everything to the left of Will. Now it was a little tricky because I didn't want to accidentally smudge Will's arm/hair/face. So I smudged everything except the area right beside his arm/hair/face. It looked something like this.
To cover up the area right next to Will, I created a new, empty layer, and using a soft, round brush (and the eye dropper tool to pick the main blue color from the cap), I covered up what I hadn't already smudged. And the final result looked like this. In the original base, it looks like I had to duplicate the blur step from before, but this time I've smudged/brushed pretty cleanly, so I won't bother this time. If it had been streaky though, a blur layer would have been necessary. Anyway, I then resized this cap so that the width was 300 pixels as well.
PUTTING IT ALL TOGETHER
Now I have two caps that I can use. They have both been cleaned up and resized. So then I needed to arrange them in a way that pleased me. Since I worked with Rachel first, I put her on the canvas first, and I have this habit of putting the first image at the top. That's where Rachel ended up.
I put Will on the bottom and because I wanted to get rid of the logo around his shoulder, I put him under the image of Rachel (i.e in my layer palette, Will's layer would be at the bottom and Rachel's would be at the top). I still wasn't happy though because there's still clutter behind Will. To get rid of that, I moved the cap of Will to the right, and then using the FREE TRANSFORM TOOL I extended the width of the cap so that the left side (the blue) still met the edge of the canvas.
SPICING THINGS UP
Or coloring. Whichever you prefer. ;D
Before I start, I have a confession to make. raiindust wrote a tutorial in which she taught how to use actions in Photoshop, and I have this one action saved that I like to use on Glee caps a lot. For the record, I can do everything in the action manually as well; it is just easier to click the action button. Anyway, the next...four steps come from that action (which has a total of six adjustment layers in it; two of which I won't use here), and it seemed like a good idea to confess that, lol.
Steps 1-3 are one CURVES layer duplicated three times. The first layer is set to NORMAL at 100% opacity, the second is NORMAL at 60% opacity, and the third is set to SOFT LIGHT at 100% opacity. These three curves layers are there to brighten up the cap (1-2) and add some contrast (3) to the image. So I only used the RBG channel, and it looks like this.
Step 4 is a SELECTIVE COLORING layer. It decreases the amount of yellow in the reds (-26) and increases the amount of yellow in the neutrals (+32). In this particular image that increases the yellow/green quality of everything.
At this point, everything is looking rather flat. There's no real light and shade in the image. To fix that I:
Copy merged everything and pasted it as a new layer at the top. I set this layer to SOFT LIGHT. Then I created a new (empty layer) at the top, and used a soft, round brush (in a light beige color) to brush over places I wanted to lighten up, etc. I then used Gaussian Blur to blur the brushing a little bit.
But I wasn't happy with the depth of the shadows, so I duplicated the soft light layer once. At full opacity, it was too much; so I lowered the opacity of the layer until I was happy (40%). This looked fine everywhere except their faces where it was just a bit too much. So I added a layer mask to the layer, and I masked away the parts over their faces (and I blurred the layer mask a little as well which has the same effect as using gray instead of black on your mask or using a black brush at a lower opacity).
At this point, I wanted to see what it looked like small, so I copy merged everything, pasted it as a new layer at the top, and sharpened that layer (Filter >> Sharpen >> Sharpen). At full opacity, the sharpen was too strong. So I lowered the layer opacity to 50%. I resized to 100x100, but it definitely needed some more work. So I hit undo (or Step Backward which is Alt+Ctrl+Z), and kept working.
I thought, maybe, the yellow was bit too much in the end, and I wanted to cool the icon down a smidgen. So I copy merged everything, pasted it as a new layer at the top, and then opened VARIATIONS (Image >> Adjustments >> Variations). I made sure I had 'original' selected so that 'original' and 'current pick' were both the same. Then I hit MORE BLUE once. I only wanted this change to be subtle, so I lowered the opacity of the layer to 50%.
At this point, we have color, but it all looks rather flat. To jazz up everything, I used PAINT DAUBS (Filter >> Artistic >> Paint Daubs). I set it to SIMPLE with a brush size of 50 and sharpness of 0. It creates the effect of a really funky light texture. I set the layer to soft light at full opacity. You'll notice that it adds depth to some places, but not to others at this stage. To even that out, I used Gaussian Blur on the Paint Daubs layer. And now the colors have a little more kick (and the image it a bit brighter as well). I loved what it did to the colors and the brightness of their faces, but I didn't love how it washed out the negative space. To fix that I added a layer mask to the layer and masked away bits that I didn't like (and I blurred this mask a little as well before going back and masking a little more). The mask looks like this. The lighter, gray portions are where I've blurred. The darkest black bits were additional masking after the blur.
But, eh, the blue in Will's section isn't popping enough for my tastes. So I copy merged everything, pasted it as a new layer at the top, and opened up VARIATIONS again. I made sure I had 'original' selected like last time, and then I proceeded to up the blue/cyan (I might have also clicked on 'lighter' once or twice) until I was happy with how the wall looked (I ignore everything else, and you'll see why in a moment). Once I was happy with the level of blue/cyan tones in the wall, I added a layer mask to the layer, and I masked away everything except the portion of the wall.
I probably could have finished here. I mean it is perfectly fine as it is, but, I don't know, I thought it was a bit boring. So I opened up my completely unorganized textures folder and proceeded to look through everything. I sort of have to stare at everything in my textures folder before I make up my mind, and even then I usually open about four or five textures that I might use. Anyway, this time I opened up a folder of watercolor textures by realproof, and there was a texture that actually had my color scheme (yellow/orange and blue) in it! Lucky me. ;D
I had to rotate the texture so that the orange half of the circle was on the top of the canvas, and the blue half was on the bottom of the canvas. After that, I copy and pasted it, as it was, onto my icon canvas. I set the layer to SOFT LIGHT after I scrolled through all the blending modes first, and I masked the texture off of just their faces because it obscured them too much.
>>
Now we are in the final steps of the icon! As it stands, the icon is lacking contrast. I could just use a soft light layer, but I wanted to preserve the brightness as well. So, to compromise, I used HIGH PASS on a new layer (copy merged and pasted at the top again), and I set that to SOFT LIGHT. Now all of the shadows really stand out from everything else, and the icon has a lot more depth and wow factor than it did before.
The only thing is that it left Will's face a bit too dark for my tastes. So I created a new (empty) layer at the top, set it to soft light, and used a soft, round brush in a light beige color to brush over Will's face.
Resize and that's it!
END NOTES: You can link this anywhere you please as long as the link leads back to me (i.e Don't redistribute this tutorial saying that you wrote it). I'm probably not going to share the .psd, sorry. If you have any questions and/or want to discuss any portion of the tutorial with me, feel free!
Next Up: Still working on the Text Guide. It is super hard to write; who knew?