Home Artists Posts Import Register

Content

Throughout these years, even within the current project, my way of handling graphics and how they look has changed a lot, and it'll probably continue to change as I try to improve and continue into looking for quicker/better ways to create the images for my games.

Especially since two of my tiers offer Outlines / High Res CGs, I thought it'd be interesting to show my current workflow, in this case to create a background, so here it is! I hope you will enjoy the read.


1. Gather references and create a level in 3D.

In this example I am creating a background for a parody game: Issei's room, so I hunt references of his room online, possibly from multiple angles so I can make out more details or have a whole 360 view of his room.

If I were to make an original map, I would simply have a quick layout made on paint to know where to place what.

Once the references are ready, I follow them to create a level in 3D as closely resembling the geometry of the references.
As long as it is a table or other furniture, it doesn't have to be perfectly the same as the reference, however, for some niche items, modelling my own might be necessary as I did with the Hostess of Fertility sign: https://www.patreon.com/posts/can-you-guess-51731042  some years back!


2. Camera placement

Once the scene has been created, I place the camera to an angle that looks good and edit the light around it.
It could potentially save some time to create the map around the view of the camera, but I think it's hardly worth the effort.
I learnt throughout the years that it's very likely at some point I will need different angles for CGs, so it's way better to simply create the whole map even if not all of it will be seen right away.


3. Outlining the map

Once I am happy with the lighting, I take a screenshot and call it "Base", then with post processing I will show the outline only with no colors, and save another one called "Outline".

I put those 2 images in a folder, and use AI to upscale them... With my 1660, I can work comfortably with a 8k image, I tried 16k in the past and my image editor would begin to struggle.

With my assets upscaled, I delete the original and open the "Base" in my image editor and start outlining everything.

I will use "Outline" to help myself outlining some objects in case the Base's lighting or some other factors do not make clear enough what needs to be outlined.
I do not use Outline directly because Post Processing will not outline everything, and the size of the outline depends on the distance, which I do not personally like.
I like to keep the outline perfectly uniform.

I also use this opportunity to draw in some extra components to the best of my abilities, such as the lamp attached to the bed.


4. Color

Once the Outline is done, I color each component with the color I want (Sometimes using the reference as a palette).
Each component is a separate layer, and I call them "Base color", the end result looks really simple.


5. Texture

Next step is giving them a texture, and I do so in two ways depending on my needs: I either use brushes or use texture images.

Even if the Mesh is simple enough, I do not UV Map it but rather overlay the texture image by following the perspective of the camera within the Image Editor, that's because I do not want lighting to affect it nor mipmapping due to the camera's distance.

The end result looks better, but it is bland.


6. Shading

To give more depth, I start lighting each component.
I do so by making two layers called "Sharp Shadow" and "Sharp Lighting" which as the name suggests, they have a sharp edge.

A pure darker color of the "Base Color" to simulate the contrast in most anime, and thanks to the "Base", I know where light and shadows hit.
It doesn't have to be perfect as long as the direction is close enough, sometimes I may put shadows/lights in places where it doesn't realistically make sense only because I think it looks good.

Once I am happy with them, I create two other layers called "Soft Shadow" and "Soft Lighting" those two use the same color as the sharp ones, but the brush opacity is set to 10% and soft, then I just start painting around the sharp one or in areas where there should be shading, but not as strong as the one given by the sharp ones.

If this background had other sources of light (such as the sun coming from the window) or it was a hentai scene with skin showing, I'd also brush different colors in this step as well.
For example red for knees/elbows/blushing, or in this specific example, an orangish tint on the wood.


7. Post Processing

We are almost done now, in most cases I do not want an outline for the map as it is an element that always exists beneath the sprites that in contrast do have an outline.
I could color the outline in order to make it stand out less (I actually do that for CGs since I want an outline there, but sometimes pure black just doesn't look good or stands out too much), but it is much quicker to simply set the outline layer to soft light.
In some cases I also decrease the opacity, I rarely if ever simply delete it.

Once that has been taken care of, I postprocess it futher with a few simple effects, in some CG I use much heavier effects such as some of Hestia's CGs in the recent update.
Last but not least, I finally downscale it to an useable format (1920x1080), or in case of sprites or other elements that exist to be overlayed on top of others, in a scale of 4.


That's all!
I hope it was an interesting read, but remember this is not a tutorial!
It is in no way the correct way to create images, simply my way to handle the need for images as a non-artist.

Comments

No comments found for this post.