Skip to content
Server-sideDeleted in 30 minutes

Convert PNG to BMP

This is the only conversion here that trades a good format for an obsolete one, and there is exactly one reason to do it: something downstream reads nothing else. Embedded displays, laboratory instruments, industrial control panels, software from 2003. For them, BMP's simplicity is the feature.

Use this without the search next time. Prathom Workbench puts Prathom's tools in your toolbar.

Add to Chrome — free
PNGBMP

Drop your PNG file here, or click to browse

Up to 50 MB. Deleted automatically after 30 minutes.

What it does

  • Uncompressed 24-bit output that decodes with no library
  • Identical pixels — nothing is lost, only stored inefficiently
  • Transparency is flattened, because BMP has none in practice
  • No watermark and no sign-up

How to use PNG to BMP

  1. 1

    Add the PNG

    Splash screens for a device, icons for legacy software, test patterns for an instrument, or whatever a piece of hardware documentation is demanding as a .bmp.

  2. 2

    Convert

    The image is decoded and written out as an uncompressed bitmap on our server. Conversion is instant; the download is the slow part, because the file is large by design.

  3. 3

    Check the dimensions against the device

    Embedded targets are usually strict about size, and often about bit depth too. Resize the PNG to the exact panel resolution before converting — a device expecting 320x240 will not scale for you, it will fail.

How it works

The PNG is decoded, any transparency is composited against white, and the pixels are written as an uncompressed 24-bit BMP.

There is no quality decision to make, because there is no compression to tune. BMP stores what it is given.

Where this is genuinely needed

Embedded displays. Small LCD and e-paper panels driven by microcontrollers. The firmware often has no image library at all — the display routine expects a pixel array and a BMP is one with a header attached.

Laboratory and industrial instruments. Oscilloscopes, spectrometers, machine-vision controllers and CNC panels frequently read and write BMP because the format was settled before the instrument was designed and there has been no reason to change.

Legacy Windows software. Applications with resource files, custom skins or icon sets from the era when BMP was the default.

Bootloaders and firmware splash screens. Anything that draws an image before a real operating system is running, where there is no decompression library because there is barely a system.

In every case the requirement comes from the hardware or the software, not from anyone's preference. Nobody chooses BMP.

What to check before converting

Exact dimensions. Embedded targets rarely scale. If the panel is 320x240, the bitmap must be 320x240.

Bit depth. Some devices want 24-bit, some 16-bit, some 8-bit indexed or 1-bit monochrome. This produces 24-bit, which is the most common requirement and not a universal one — check the documentation.

Row order. BMP stores rows bottom-up by convention, and a handful of naive readers assume top-down and display the image upside down. If that happens, the reader is at fault and flipping the source before converting is the practical workaround.

Color order. BMP stores blue, green, red rather than red, green, blue. Any correct reader handles it; a hand-written one sometimes does not, and the symptom is an image with the reds and blues swapped.

If nothing requires it

Do not convert. PNG is better in every measurable way — same pixels, real transparency, a fraction of the size — and BMP offers nothing in return except compatibility with software that has not been updated in twenty years. The reverse conversion is the one most people actually want.

Examples

A splash screen for an embedded display

splash.png — 320x240, 8 KB
splash.bmp — 320x240, 230 KB

Nearly thirty times larger, and correct. The device's firmware reads the pixel array straight into its framebuffer with no decoder at all, which is the only reason this format is still specified in 2026.

A logo with a transparent background

logo.png — 400x400, transparent
logo.bmp — 400x400, transparency flattened to white

The one thing that actually changes. BMP's alpha support exists on paper and is ignored by most software that reads BMP, so transparent pixels are filled rather than left to chance. If the background needs to be another color, composite it in an editor first.

Frequently asked questions

Why would anyone want BMP?

Because it needs no decoder. A 24-bit BMP is a short header followed by raw pixel values in rows, which means a microcontroller with a few hundred bytes of RAM can display one by copying bytes. PNG requires a Deflate decompressor and a filtering pass, which is a real cost on hardware that small. Simplicity is the entire feature.

Does the image lose quality?

No. Both formats store exact pixel values, so the bitmap contains precisely the image the PNG did. The only change is transparency, which is flattened. Everything else — colors, dimensions, sharpness — is bit-for-bit identical, stored far less efficiently.

What happens to transparency?

It is flattened against white. BMP does have a 32-bit variant with an alpha channel, but support for it is inconsistent enough that writing one is a gamble — much of the software that still reads BMP predates the variant and will show garbage or ignore the channel. Flattening produces a file that works everywhere BMP works.

How large will the file be?

Width times height times three bytes, plus a small header, regardless of what the image contains. A 1920x1080 bitmap is always about 6 MB whether it shows a photograph or a solid color. That predictability is exactly why embedded systems like it — the memory requirement is known before the file is read.