#khtutorials

TRANSPARENCY

I always have Transparency checked. Does it take a slot up in the color count? Yeah, but it also makes a gif take up considerably less space! And unless you SPECIFICALLY have a 100% transparent pixel in your gif, your gif will still be absolutely opaque, I promise. Below is a palette from a 500p gif.

PALETTE

I also try to color my gifs so they have a specific palette. A lot of times the color count of my gifs look like one or two main colors. 

Now notice how many colors those two gifs above have? Not a lot. Seriously not anywhere near 256 colors. BUT, There are a few ways that this goes unnoticed. One, the specific color palette really helps. Because all the color slots are similar to each other, even if there’s half the color count, it’s not obvious. Here’s a gif from 2014, before I learned adjustment layers and how to give specific palettes to a gif:

This gif originally had all 256 colors, but lowering to half really makes it have a grainy feel :/ But here’s a similar gif I made in 2016, where I gave it a more specific palette:

The same amount of colors, but because the colors aren’t so widely different from each other, the graininess isn’t nearly as apparent.

I’ll even try to make it so the darker colors in a gif have the same/similar hues. Back in 2013 and 2014 I would have to use this to the extreme; I would use a purple color and put it on ‘lighten’ so that would cut out a few darker colors and lower the color count:

In the gif below, notice how the darker the color, the blue it is. It’s a similar technique to the top but it’s done by using the Selective Color adjustment layer (on ‘color’ blending mode) to change the black/darker colors to a more specific one. This saves color space as well as adds a sense of harmony to the palette.

I also use this technique when using gifs that purposely have different colors: 

These two are part of the same gifset. Notice how in both images the darker colors gradually become more purple in hue. Even though one gif is red and another is blue, the dark purple colors create a connection between the two gifs. (that last part doesn’t have anything to do with saving color slots but I thought i’d mention it anyway LOL)

DITHERING

I also have my dithering set to Pattern. I believe this actually makes a gif take MORE space, but it makes a gif look great. Sure it does add a sense of graininess, but every gif has that due to a 256 color limit, and pattern levels out the graininess in an organized manner. However, if a gif is too big, and I can’t afford to sacrifice the pixel size, frame rate, or anymore colors, clicking Diffusion or No Dither is my last resort. 

STATIC AREAS

Another thing I do that really helps me (but I understand not everyone wanting to do this too) is I use images that create a static, or still, area of the gif:

The first gif has a texture overlapping the entire gif on a slight opacity. Even doing this can help lower the gif size. There’s also a solid color text that even though it is blinking and changing, that large section of white in the gif will help reduce the size.

The second image has a completely opaque image, and this can reduce a gif size a LOT. You can still add psds and edit it to make it less boring.

And the third image has a smokey texture, an image and text all together so your gif can still look complex, but still reduce the gif size.

The use of static images and animated ones give a nice sense of different things to focus on too. The viewer can take a second to recognize both the gif and the image/text. 

CONCLUSION:

Part One: the two separate images

You can of course make your own animation to use the clipping mask on, but I get my textures from

motion addicts

motion a day

dvdp

bees & bombs

and here is the texture I am using in this tutorial, Comp by murAta Yuzi :)

and here are the images I will be using for this edit:

the reason why the second image is so small, is because it fits within the white circle of the texture we are going to be using. Make sure the images you’re using all have the same amount of frames. that smaller gif i’m using may not seem to have the same amount of frames as the larger one, but trust me it does ;)

Now that you have your images, you’re going to edit these now and THEN edit the texture for the clipping mask later. I may do a tutorial on my coloring techniques later, but for now I’m going to assume you know how to do them. When you’re finished, select all the frames in the Animation Window, and in it’s menu (

in the top right corner of the Animation window), select ‘Flatten Frames into Layers’. Here are the edited images I’m going to use:

Part Two: the animated texture

The next thing I do is work on the texture we’ll use for the clipping mask. We want the texture to be mainly black and white; not JUST grey-scale, but also mostly black and white the colors themselves. Because of this, it’s best to use textures with two main colors, ones with 3+ get more complicated.

As long as the texture has two solid colors, we can make it black and white ourselves. To explain this better we'll use this frame of TOHRU MiTSUHASHi’s GOOD BYE, which is blue and purple.

And to make it black and white, we open a new layer at the top of all the other layers, and fill it with white, grey, or black. Then we put that layer on the blend mode ‘Color’. Then open the Adjustments Window (window > adjustments), and click Curves.

Click the white eye-drop, and then click the lightest area in the gif Then click the black eye-drop, and then click the darkest area in the gif. This makes it so the gif is almost completely black or white, perfect for clipping masks.

The next part will take patience. Select the Background Eraser tool. Have the settings like my own below. The tool doesn’t have to be as big as 2200px, but be sure it’s large enough to contain the canvas of the texture.

Next, on each layer and corresponding frame, you’re going to (with out a tablet; use your laptop for this) click the solid black parts of each layer. this will erase the percentage of blackness in the image, and leave only a white vector- PERFECT for clipping masks :D If you want, you can do the reverse, and erase the white to leave a black vector. I just have the habit of erasing the black haha. (it IS transparent, the red is there so you can see it better)

Part Three: combining the images with the texture

Now we’re going to select/open the gif you want to have appear on the texture. Like I said earlier, be sure that you selected all the frames, then ‘Flattened Frames into Layers’. Once again select all the frames, then select ‘Copy Frames’. Then, back in the texture image, select all frames again. Then select ‘Paste Frames’. The select ‘Paste over selected layers’.

Now if needed, select all the layers that were just recently inserted into the image, and move them to where you want them to be (over the solid white part of the texture). This isn’t really necessary if the texture and image you’re using are the same sizes haha.

Now this part will really take patience lol. Like I said before, there may be a quicker way to do this, but I don’t know what it is sO WE’RE JUST GONNA DO THIS LMAO. On the first frame, grab the first layer of the set that was recently pasted in. Then drag it all the way down to the matching layer of the transparent vector. Then right click it, and click ‘Create Clipping Mask’. Do this for all frames and their corresponding layers!

*swipes sweat* WOOOH! Then select all frames, then flatten frames into layers once again (once again, the red represents the transparency)

then you’re going to select all frames and copy them once again, and then paste them over the second image you have.

Conclusion

PART 1

So start off with your image. If it doesn't have a transparent background then duplicate the image (with the original’s visibility off) and erase the background on the copy. If it does have a transparent background, then let’s get started! full-sized image:

Try to imagine your image in 3D. Which parts will be further to the back, and which parts will be closer to the front. Some parts will be obvious, but some may may need a little more concentration (at least it does for me lmaoo). I tried to illustrate this on the left. It may help you to illustrate it too, but you don’t need to :) ((I have changed some parts later on, like mickey’s other pocket))

You can do more or less, but I’m going to split this image up into five layers: the way back, the back, the middle, the front, and the way front. And with each one I’m going to duplicate the image’s layer, so we have the original image (invisible) with five more layers. Make sure the back layers are at the bottom, and the front layers are on top, like shown.

Go through each of the five layers and erase everything other than the parts that belong. For instance, in the Way Front Layer, I erased all but the glove, and in the Way Back Layer, I erased all but the back of mickey’s weapon. TIP: If you mess up and erase too much and ctrl+Z won’t help, you can duplicate the original image and use that to gain back the erased part. When it comes to the three middle layers, you don’t have to erase 100% exactly; you can be a little lose and then later on you can erase more :)

Then, with the Animation Window open, duplicate the frame eight times to get nine frames in total. The “main” frame is going to be the middle frame, or Frame 5. This frame will look exactly like the original image, with no “rotation” effect added.

PART 2

Okay, this is where the rotation effect actually happens lol. Select the Way Back Layer, and select all the frames on the right of the Middle Frame (so Frames 6-9). Now with the left arrow key on your keyboard, move the Way Back Layer to the left only about three or four pixels. (larger images I suggest four pixels). Then, select all the frames two frames away from the Middle Frame (so Frames 7-9), and move it over three or four pixels. Then select all frames three frames away from the middle (Frames 8 & 9), then four frames away (Frame 9), etc, each time moving the layer over until you don’t have any frames left.

Then, in the Animation window’s menu, ‘Select All Frames’ then ‘Reverse Frames’. Then repeat that last paragraph, except instead of moving the Way Back Layer to the left, move it to the right. Then ‘Select All Frames’ then ‘Reverse Frames’ once again. Press the play button and hopefully the Way Back Layer will be seen moving across the gif repeatedly.

Do this will all the other layers too (besides the Middle Layer; leave that alone). The Font Layers should move in the opposite directions as the Back Layers to give the rotating affect. The Way Front Layer should move three or four pixels as well, and the layers Front and Back should only move about one or two pixels. As I said though, this will vary depending on the size of the image you’re using. The result should look something like this:

Hey it looks like it’s rotating!! :D The next step is making it look less like layers moving, and even MORE like rotation. And we’ll do this on the very first and last frames because they show the largest difference in movement, which is what we’re going to ‘disguise’. And by disguising these two frames, we easily disguise the other frames at the same time! :D 

So click the last frame, and click The Way Back Layer. At the top of the Layers Window, click the checkered square that says “Lock” by it. That makes it so you can brush over the layer without going off the ‘edges’ of it. Focusing on the Way Back Layer, with the brush tool ‘paint over’ that visible difference. TIP: If you hold down the “Alt” button, you can easily eye-drop and pick colors on the image. Then click the first frame, and paint over any differences visible there too. Do this for all layers (even the Middle Layer, you may need to paint over it to). I’m sorry if this step is hard because you use the mouse pad or a something :( I have a wacom tablet which makes steps like this a lot easier ^^; I’ve heard though that the Clone Stamp Tool is also good for something like this, so try that too, if you’d like!

So, here’s what it looks like after I “erase all the differences”. Hopefully this too helps explain because sometimes i have no idea what i’m talking about: (it’s a big image too, open it in a new tab so see more detail)

In some areas I wasn’t really sure what to put, so I had to look at a reference for Mickey’s outfit. You may need that too, as sometimes painting over the differences takes more than your imagination ^ ^; Overall I’d say this is the step that takes the most work. But look at it now!

Nice! Mickey really came together, and he’s ready to kick some 3-dimensional ass! There’s only one final step: in the Animation Window, select all the frames except the first and last one. then click the ‘notepad’ at the bottom of the window. That should duplicate those layers. While still having the duplicates selected, drag them to the right, past the last frame. Then in the Animation Window’s menu, select ‘Reverse Frames’. That will make it so he not only rotates one way, but then rotates the other way as well. it makes a nice gif transition :) That’s about it! :D

Conclusion:

Here's the piece we'll be working with! Full size from khinsider ^^ Sora will be our Sharp Focus point, with everything else blurry.

To start off press the new layer button (the peeling square on the bottom of the Layers Window) and with a solid color (I prefer black, but in some cases I'll go with another color to see better) and a hard-edged brush around 5px-10px (I prefer diameter touch-sensitive, like with a tablet. If you don't know what I'm talking about just ask me ^^) slowly outline your focus. the outline can be thick, messy, whatever. It doesn't matter ^^

When you make the solid color outline, make sure that there's a 1-pixel wide gap between the edge of your sharp focus and the solid color outline. If they both end up touching, then later on you could get that slight shadow-blur from the above banner xP

Then with the Paint Bucket Tool (make sure the 'Contiguous' box at the top is checked) click outside the outline. Whether or not the 'Anti-Alias' box checked doesn't matter, and make sure the 'All Layers' box is un-checked.

Next, with same brush (once again around 5px-10px) and color from earlier, brush over that empty outline surrounding the Sharp Focus. The whole point was to surround the Sharp Focus with the solid color (on a different layer, remember), as seen below

Next, duplicate the first layer that features the image (right click, select 'Duplicate Layer'). Drag the copy to be a layer above the surrounding color layer. Now right click the copy and select 'Create Clipping Mask'. This makes it so the copy layer won't show in places where the layer below it is empty. If you don't know what I'm talking about, click the eyeball left of the original image layer (the most bottom), and see how there's the shape of your Sharp Focus missing!

Now right click the most top layer and select 'Merge Down'. Now in the Layers window you should see the original image layer (bottom), along with one missing the Sharp Focus (top).

Selecting the top layer, go to the top of the Ps window, select 'Filter' > 'Blur' > 'Gaussian Blur'. I like my radius around 3.5 pixels. I seriously seldom go over 4 to 5 pixels when doing this technique. Select OK

Whoo! Almost done!

Now next select the original image layer (bottom layer). Select the Blur Tool on your Tools window (below the Paint Bucket tool). Now use a brush with a soft edge, a size of 5px - 10px, and a strength around 75%. With this tool, go over leftover places close to the Sharp Focus that could use some blurring.

And that's it! If you want, merge the two layers together. Here's the full-sized blur image, available for use because why not. Below is a comparison to before and after the blurring technique.

To some this may have been obvious, but like I said this is for beginners ;) Thanks for your time, and I hope I was helpful. If I was unspecific in certain areas, give me some feedback and I'll edit this while also trying to improve in my next tutorial ^ ^