By Ahsan Khowaja · Updated 27 October 2026
How to use the image size calculator
- Choose File size of an image from its pixels or New pixel size when scaling an image.
- Enter the width and height in pixels. On Windows they are in the file's Properties › Details tab; on a Mac, in Preview's inspector; on a phone, in the photo's info panel. The megapixel calculator lists common camera and screen sizes if you only know the megapixels.
- For a file size, pick the colour depth: 24-bit (3 bytes per pixel) for JPEGs and ordinary PNGs, 32-bit (4 bytes) for PNGs with transparency, 8-bit for greyscale, and 48- or 64-bit for 16-bit-per-channel files.
- For scaling, enter a percentage: 50% halves both sides, 200% doubles them.
File sizes are in decimal units (1 MB = 1,000,000 bytes). The "as a computer may show it" row gives the binary figure (1 MiB = 1,048,576 bytes) that Windows labels "MB", so both readings are covered.
How to calculate image size from pixels
An uncompressed image stores a fixed number of bytes for every pixel, so its size is simply:
Uncompressed size (bytes) = width × height × bytes per pixel
The bytes per pixel come from the colour depth: 24-bit colour means 8 bits (1 byte) for each of red, green and blue, so 3 bytes per pixel; 32-bit adds an 8-bit alpha (transparency) channel, so 4 bytes. To calculate image size in megabytes, divide by 1,000,000.
| Colour depth | Channels | Bytes per pixel | Where you meet it |
|---|---|---|---|
| 1-bit | black or white | 1/8 | Fax, scanned text |
| 8-bit | greyscale, or 256 indexed colours | 1 | GIF, PNG-8, greyscale scans |
| 16-bit | greyscale | 2 | Medical and scientific images |
| 24-bit | RGB, 8 bits each | 3 | JPEG, PNG-24, most photos |
| 32-bit | RGB + alpha | 4 | PNG with transparency, app icons, screen buffers |
| 48-bit | RGB, 16 bits each | 6 | 16-bit TIFF, high-end editing |
| 64-bit | RGBA, 16 bits each | 8 | 16-bit layered editing files |
Worked examples
- 1920 × 1080 (Full HD) at 24-bit: 2,073,600 pixels × 3 = 6,220,800 bytes, 6.22 MB or 5.93 MiB.
- 4032 × 3024 (a 12 MP phone photo) at 24-bit: 12,192,768 pixels × 3 = 36,578,304 bytes, 36.58 MB.
- 1000 × 1000 at 32-bit (a square PNG with transparency): 1,000,000 × 4 = 4 MB.
- 6000 × 4000 (24 MP) at 48-bit: 24,000,000 × 6 = 144,000,000 bytes, 144 MB, which is why 16-bit TIFFs from a camera are so large.
- 800 × 600 greyscale at 8-bit: 480,000 × 1 = 480 kB.
This uncompressed figure is what the image takes in memory when it is open, and what a BMP or uncompressed TIFF file would be on disk. It is also the figure Photoshop reports: the Image › Image Size dialog's "Image Size" line for a flattened 8-bit RGB image is width × height × 3 bytes, shown in binary megabytes, so a 1920 × 1080 document reads 5.93M. If you came here looking for an image size calculator Photoshop would agree with, that is the row to compare.
Image file size calculator: compressed estimates
Almost every image you store is compressed, so the file on disk is much smaller than the uncompressed size. The calculator gives two rough estimates, each with its assumption in the label:
- JPEG, assumes 10:1. JPEG throws away detail the eye doesn't notice, and photographs usually shrink 10:1 or more. Our own sample files bear this out: a 3024 × 2016 portrait (18.29 MB uncompressed) is 1,645,283 bytes as a JPEG, about 11:1; a 3000 × 2000 flower photo (18 MB uncompressed) is 984,536 bytes, about 18:1; and a 4032 × 2623 landscape with large smooth areas (31.73 MB uncompressed) is 389,561 bytes, more than 80:1. So 10:1 is a conservative, upper-end estimate for a photo at a high quality setting.
- PNG, assumes 2:1. PNG is lossless, so it can only remove repetition. Our 1366 × 900 screenshot (3.69 MB uncompressed) is 2,375,634 bytes as a PNG, about 1.55:1. Photos saved as PNG barely shrink at all, while flat graphics with few colours, such as logos and diagrams, can shrink 10:1 or better.
Treat these as order-of-magnitude answers. The real size depends on the scene, the quality setting and the encoder. For the actual number, compress the real file: our image compressor shows the before and after size in the browser, and the reduce image size in KB tool hits a specific target like 100 KB or 200 KB. Our guide to how image compression works explains where the savings come from.
Image size calculator in pixels: scaling an image
The second mode is an image dimensions calculator for resizing by percentage:
New width = width × scale ÷ 100 and New height = height × scale ÷ 100, both rounded to whole pixels.
The pixel count changes by the square of the scale, which surprises people: scaling to 50% leaves a quarter of the pixels, and 25% leaves one sixteenth.
- A 4032 × 3024 photo at 25% becomes 1008 × 756 px, 6.25% of the original pixels, 2.29 MB uncompressed instead of 36.58 MB.
- A 1920 × 1080 frame at 50% becomes 960 × 540 px, 25% of the pixels.
- A 1200 × 800 image at 150% becomes 1800 × 1200 px; enlarging adds pixels but not detail, so it will look softer.
For a specific target width or height rather than a percentage, the aspect ratio calculator works as an image ratio calculator and finds the matching other side. To actually resize the file, our image resizer does it in the browser; the guide to image resolution explains what the pixel count means for sharpness.
Pixel size calculator: from pixels to physical size
Pixels have no fixed physical size until you print or display them. The same 1920 × 1080 image is 6.4 × 3.6 inches at 300 DPI and 26.67 × 15 inches at 72 DPI. A pixel size calculator is really a division: inches = pixels ÷ DPI. The pixels to inches converter and the DPI calculator do that, and the print size calculator tells you how large a print your pixel count supports.
Why the stored file is smaller than width × height × 3
- Compression: described above. A 12 MP JPEG is typically 2–5 MB, not 36 MB.
- Chroma subsampling: most JPEGs store the two colour channels at quarter resolution (4:2:0), which halves the raw data before compression even starts.
- Metadata adds a little: EXIF data, embedded thumbnails and colour profiles can add tens of kilobytes. Our EXIF remover strips them.
And why it is sometimes larger: a 16-bit TIFF stores 6 bytes per pixel, a layered editing file stores every layer, and a RAW file stores the sensor data at 12 or 14 bits per pixel before any colour processing. The photo storage calculator uses typical stored sizes to work out how many photos fit on a card.
