What is a sprite sheet? A clear explainer on how game characters move
Once you start drawing characters in pixel art, sooner or later you run into the term "sprite sheet." I am sharing a beginner-friendly explainer that Keitaro wrote on note, because it walks through how the system works step by step and I found it useful.
A sprite is a 2D image shown on the game screen. Characters, enemies, items, bullets and effects all count, and a sprite sheet is what you get when you gather them into one large image. For a walk cycle, you show a frame with the right foot forward, then a frame with both feet together, then a frame with the left foot forward, switching quickly so it looks like walking. In other words, motion comes from swapping still pictures, not from video.
The game does not display the big image as a whole. It cuts out only the frame it needs by splitting the image into coordinates. For example, if each frame is 128×128px and eight sit side by side, the display area moves through 0–127px, then 128–255px, and so on. The article also gives an example of arranging the horizontal axis as 8 directions and the vertical axis as actions (walking, attacking, being hit, guarding), so you manage 8 directions × 7 states, or 56 frames.
When you actually use them, the article stresses that lining up the reference position in every frame matters most. If the feet drift by a few pixels from frame to frame, the character appears to jitter up and down during the animation. Keeping the background transparent is also essential.
Finally, it explains the difference between a sprite sheet, which is simply related pictures arranged together, and a sprite atlas, which packs several images into one texture for drawing efficiency. My personal takeaway from the article is its summary that making the images is less difficult than shaping them into a form the game can use correctly. I think it will be useful for pixel artists as well, especially at the export stage.
Source: note (Keitaro)
A sprite is a 2D image shown on the game screen. Characters, enemies, items, bullets and effects all count, and a sprite sheet is what you get when you gather them into one large image. For a walk cycle, you show a frame with the right foot forward, then a frame with both feet together, then a frame with the left foot forward, switching quickly so it looks like walking. In other words, motion comes from swapping still pictures, not from video.
The game does not display the big image as a whole. It cuts out only the frame it needs by splitting the image into coordinates. For example, if each frame is 128×128px and eight sit side by side, the display area moves through 0–127px, then 128–255px, and so on. The article also gives an example of arranging the horizontal axis as 8 directions and the vertical axis as actions (walking, attacking, being hit, guarding), so you manage 8 directions × 7 states, or 56 frames.
When you actually use them, the article stresses that lining up the reference position in every frame matters most. If the feet drift by a few pixels from frame to frame, the character appears to jitter up and down during the animation. Keeping the background transparent is also essential.
Finally, it explains the difference between a sprite sheet, which is simply related pictures arranged together, and a sprite atlas, which packs several images into one texture for drawing efficiency. My personal takeaway from the article is its summary that making the images is less difficult than shaping them into a form the game can use correctly. I think it will be useful for pixel artists as well, especially at the export stage.
Source: note (Keitaro)
Comments 0
Leave the first comment, in any language.
Log in to comment