Pixel-based image format — JPG, PNG, TIFF. Loses sharpness when scaled up, but captures color depth perfectly. Standard for final delivery and archival.
On set and in the DI suite, we talk about raster graphics when we work with pixel-based images — this is the standard in digital film production. Each image is composed of a regular grid of colored pixels that together convey the visual information. In contrast to vector graphics (which mathematically defines curves and shapes), raster graphics are a sampling method: reality is broken down into discrete points, each point carrying a color value. This is not abstract — this is what we deal with daily.
The practical consequence of this pixel structure is immediately apparent when scaling. If you enlarge raster graphics beyond their native resolution, you lose sharpness — the pixels become visible, edges appear jagged. This is not a weakness of the format, but its physical reality. Therefore, in editing and VFX compositing, we work with the highest possible resolutions — 2K, 4K, often even RAW material with 16 or 32 bits of color depth per channel. This oversampling gives us room for zooms, repositioning, and color grading without immediately hitting the limits.
The strength of raster graphics lies in color reproduction. JPG, PNG, TIF — these formats store millions of color nuances per pixel, which is indispensable for photorealistic shots and complex composites. A 16-bit TIF sequence captures subtleties needed for correction or keying. At the same time, it is compressible: lossless (PNG, TIF) or with calculated quality loss (JPG), which becomes relevant for archiving and exchange.
For DCP creation and long-term archiving, raster graphics are standard — the DCP itself is a raster-based format. RAW material is stored in 12-bit DPX or comparable raster sequences. This means you cannot avoid raster graphics when making films. The question is not whether, but which resolution, which bit depth, which codec.