When we started looking into how to make normals for a 2D object with no high poly model, we did come across applications with functionality to do this to some degree. What are the advantages of hand-drawing normal maps this way, over using off-the-shelf tools like Sprite DLight? We use Nvidia’s TextureTools for Photoshop to correct/normalize vectors after finishing painting the normalmap. It meant that the shapes and details we had to mimic in the normal were already very obvious and clear.ĭependent on the nature of the sprite being normalled, we use a mix of painting normal angle colors from a ‘normal map dome’ (basically a sphere which acts as a palette of all the possible angles you can paint a tangent normal map with) and painting in the individual RGB grayscale channels, with other applications such as NDo, Crazybump or AwesomeBump helping on some shapes, and blending multiple parts of normals together correctly. We actually found this approach to work really well. This meant that when it came to draw the normals for the sprites and tiles, the shape, surface detail and materials were already decided. In our case, this master light is top down with a right side bias for most sprites and foreground tiles, with background tiles having a more generic ‘light from the camera’ to simulate dark rooms and not distract the player from the foreground. Generally, pixel art is drawn with a ‘master’ light direction in mind. Sculpting shapes with normal mapsĬan you briefly walk me through your process of developing an asset, particularly how it differs from how you'd otherwise make a similar simple 2D bitmap asset? This time, the normal map not for small surface detail such as you might in 3D art, but to use the normal map to ‘sculpt’ the large obvious shapes of the pixel art instead. When we saw Motion Twin’s Gwenael Massé’s art on Dead Cells, we saw normalled pixel art done very well and decided to have a second try at it. Our first attempt at hand painting normals on pixel art led to a very detailed normal-as that’s what its main purpose is in 3D art-but as each tile in our game is only 32x32 pixels, there’s not a lot of pixels to sell the surface information and so the end result was just a noisy mess. We were initially put off using the effect, as many examples we’ve seen had a habit of destroying the charm and aesthetic that pixel art has, giving it too much of a shiny metallic look. We’re certainly not the first studio to use them in modern pixel art titles, but the results vary. What initially attracted you to this method of using normals for lighting in 2D? Where did you learn about it, or first see it done well?Īpplying normals to 2D artwork was a natural extension from their use in 3D, though obviously it wasn’t possible on the older systems that inspire the aesthetic. If you’re going for bright simple colors, or a more true retro look, then flat lighting works well. We want the player to feel more immersed in the environment, so that the parkour has a sense of depth and physicality, rather than that paper cut-out feeling some pixel games can have. Our use of normal mapping is mostly for visual benefit, rather than as a functional gameplay choice. Used correctly, normal maps can add a sense of depth and weight to objects in a scene, in a way that flat 2D bitmaps can't quite match. The normals help us achieve that, giving a sense of shape and solidity to our world and the things within it. We wanted our art to be rich, deep, and really ‘lift off the page’. Keith Duke-Cox: Art direction is all about individual style and the objective of the delivery. What advantages does using normal mapping offer over traditional, flat lighting in a 2D game? Questions and answers have been edited for brevity and clarity. Keith Duke-Cox, one of the art directors at Cardboard Sword, recently took the time to speak with Gamasutra about how they arrived on this particular technique, and why it works in their game-and how you can adopt these techniques for use in your own 2D games. One of the chief technical differences distinguishing The Siege and the Sandfox is their use of hand-painted normal-mapped tilesets and sprites, allowing them to use 3D lighting methods in their otherwise fairly conventionally-design 2D pixel art game. UK-based developer Cardboard Sword is developing The Siege and the Sandfox, a 2D stealth-based platformer in Unreal Engine 4 using Paper2D, and they've been posting a series of development diaries introducing other would-be developers to using these tools to make their own games.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |