Uncompressed raster format — pixels stored individually, massive file size, zero quality loss. Legacy in VFX; TIF and OpenEXR are standard today.
You don't really need BMP anymore today — but to understand why we use TIF and OpenEXR today, you need to know what BMP was and why it was the standard on set and in VFX workflows for a long time.
BMP stores each pixel as raw data, without compression. This means: a 2K shot (2048 × 1080) with 8-bit RGB needs almost 6.5 megabytes per frame. At 24fps, that's 155 MB per second. Sounds normal today, but in the 90s and early 2000s, it was the only way to ensure that digital effects could work losslessly. No codec loss, no artifacts — just raw data, pixel by pixel.
In practical VFX workflows, you primarily used BMP in frame sequences: you exported your CG renders not as a video file, but as 150 individual BMP files. Each file is uncompressed, each file is identical in size and structure. This makes them stable for the compositing process. A compositor can load the sequence into the editing system, and there are no surprises — no GOP structure, no keyframe problems, no codec overhead.
The problem: storage space and network bandwidth quickly became the bottleneck. A 90-second VFX shot with multiple render layers (Diffuse, Specular, Normal, Depth, Beauty Pass) — we quickly reach terabyte amounts. That's why since the 2010s, we've switched to TIF with LZW compression or OpenEXR. Both allow lossless storage with a significantly smaller footprint.
Today, you only encounter BMP in legacy projects or in very specialized contexts — for example, when a 20-year-old compositing system only reads BMP, or when someone deliberately wants to work uncompressed and has the storage. But fundamentally: BMP was the practical reason why we started working with frame sequences instead of video files in the first place. This logic remains relevant, even if the format itself is outdated.