Image File Size Calculator — Exact Bytes

Before compression bargains begin, every image has a true size: its pixels times the bytes each pixel costs. Enter the width, height and bit depth, and this calculator returns the uncompressed file size — the number that BMP and TIFF honour, that memory buffers must hold, and that JPEG’s savings should always be measured against.

Pixels Times Bytes: The Only Exact Figure in All Imaging

An uncompressed image is a ledger: width × height pixels, each pixel a fixed number of bytes — 3 for ordinary 24-bit colour (one byte each for red, green and blue), 4 with an alpha channel, 6 for 16-bit-per-channel editing files, 1 for greyscale, a single bit for line art. The multiplication admits no negotiation, which makes this page's answer the one figure in all of imaging that is exact rather than estimated — and therefore the anchor for every other size conversation: what a memory buffer must hold when the image is opened, what an uncompressed TIFF will weigh to the byte, how much bandwidth a raw transfer truly moves, and the denominator of every compression ratio ever quoted. Formats that compress (JPEG, PNG, WebP) make size depend on content — a blue sky compresses to nearly nothing, a forest floor barely at all — and their numbers are necessarily forecasts; this page computes the sky-independent truth the forecasts depart from.

Worked Example: The 24-Megapixel Photograph

The defaults price the modern camera's standard frame: 6,000 × 4,000 pixels = 24.0 megapixels, at 3 bytes per pixel = 72,000,000 bytes — 72.00 MB in the decimal units storage is sold by, 68.66 MiB in the binary units operating systems often report (and the calculator prints both, because the 5% gap between the two dialects has launched a million "missing gigabytes" complaints that are really dialect disputes). A quality JPEG of this frame typically lands near 6–9 MB — the note's eighth-to-fifteenth — while the same frame edited in 48-bit swells to 144 MB, and a hundred-frame shoot becomes 7.2 GB of raw truth before a single compressed derivative exists. Storage planning, card sizing and backup windows all begin at this multiplication; photographers who run it before the trip do not run out of card during it.

The Megapixel Mirage and the Print Connection Explained

Megapixels describe a count, not a dimension, and the calculator's inputs insist on the distinction: the same 24 megapixels can be 6,000 × 4,000 (3:2), 5,657 × 4,243 (4:3) or a panorama 12,000 × 2,000 — identical file sizes, utterly different pictures and print shapes. The print translation runs through resolution: at the gallery standard of 300 pixels per inch, the default frame prints 20 × 13.3 inches; at the billboard's forgiving 30 ppi it spans nearly seventeen feet — size is pixels divided by the ppi the viewing distance demands, which is why this hub's pixels-to-print thinking and this page's arithmetic are two ends of one ruler. What never changes across the shapes and densities is the byte ledger at the top: the file does not know what it will become, only what it costs to exist.

The Scan and the Archive: Where Raw Size Is Law

Two professions still live inside this calculator's arithmetic daily. Archivists scanning photographs at 600 ppi produce files whose sizes obey the ledger exactly — an 8×10 print at 600 ppi is 4,800 × 6,000 pixels, 28.8 MP, 86.4 MB per 24-bit scan — and national digitisation programmes plan petabytes by this multiplication, because preservation formats (uncompressed or losslessly compressed TIFF) refuse the JPEG bargain on principle: the archive's job is to keep the information, not the appearance. Print production runs the same ledger in reverse, demanding files whose pixel counts at the required ppi cover the physical page — the calculator's width and height fields are the prepress checklist's first two lines. In a web economy built on compression cleverness, the archive and the press remain the ledger's loyal jurisdictions, and their storage budgets are this page's output denominated in years.

The Screenshot and the Sensor: A Modern Postscript

Most images made today are not photographs but screen captures and renders, and their ledgers hold a surprise: a 4K screenshot (3,840 × 2,160, 24-bit) weighs 24.9 MB raw, yet its PNG is often under 2 MB — because interface graphics are oceans of identical pixels, and lossless compression feasts on repetition exactly as lossy feasts on texture-blindness. The pair of extremes teaches the calculator's final lesson: raw size is a property of the canvas, compressed size a property of the picture, and confusing the two produces both kinds of planning error — the photographer whose card dies mid-shoot (canvas arithmetic ignored) and the web team whose "tiny" hero image ships at photographic complexity to a million phones (picture arithmetic ignored). This page owns the canvas; wisdom is knowing which arithmetic your problem is running, before the card fills, the upload stalls, or the million phones render their verdict on a picture whose ledger nobody computed in advance.

Related Tools

More from the Other Calculators hub: the Camera Field of View Calculator for the geometry that filled these pixels, and the Upload Time Calculator — the natural sequel to knowing a file's size.

Frequently Asked Questions

Why is my JPEG so much smaller than this figure?

Because JPEG discards information the eye is bad at missing: it converts colour into brightness-plus-colour-difference channels, subsamples the colour, transforms blocks into frequencies and quantises the high frequencies coarsely — throwing away perhaps 90% of the ledger while keeping nearly all of the appearance. The ratio is content-dependent (flat areas compress almost free; fine texture and noise are expensive, which is why a grainy high-ISO shot's JPEG can be three times a clean shot's at identical dimensions and settings). The calculator's figure is the pre-bargain truth; the JPEG is the negotiated settlement, and knowing the truth is how you judge whether the settlement was fair.

MB or MiB — which is right, then?

Both, in their dialects: storage manufacturers and this page's first figure use decimal megabytes (10⁶ bytes); many operating systems report mebibytes (2²⁰ = 1,048,576 bytes) while labelling them MB, which is why a 72,000,000-byte file can appear as "68.7 MB" in a file manager and "72 MB" on the box the drive came in. The calculator prints both to end the argument locally. The same dual bookkeeping applies all the way up (GB/GiB), with the dialect gap growing to about 7% at terabyte scale — a discrepancy entire support forums have mistaken for fraud.

How big is an uncompressed video, then?

The image figure times the frame rate, with consequences that explain the entire compression industry: the note's example, 4K (3,840 × 2,160 = 8.3 MP) at 24-bit and 30 frames per second, is about 24.9 MB per frame × 30 ≈ 746 MB per second — 45 GB per minute — which is why raw video lives on specialised media and every delivered video is a compression artifact by necessity, not choice. Audio engineers meet the same wall smaller (CD audio's 10 MB per minute is why MP3s happened); the multiplication on this page is the seed of all of it.

Does bit depth really change quality that much?

Bit depth buys editing headroom, not visible colours: a 24-bit image already encodes 16.7 million colours, more than the eye resolves in a finished picture — but when you stretch tones in editing (lifting shadows, grading skies), 8 bits per channel band into visible steps where 16 bits per channel (the 48-bit setting, 281 trillion colours) keeps gradients smooth. Hence the professional pattern: capture and edit deep (48-bit ledger: double the 24-bit file size), deliver shallow (24-bit). The calculator prices both ends of the pattern; the workflow exists because the ledger and the eye disagree about what matters.

What size is a phone photo?

Run your own sensor's numbers: a 12-megapixel phone frame (4,032 × 3,024) at 24-bit is 36.6 MB uncompressed — yet the HEIC or JPEG in your gallery is typically 2–4 MB, a ratio near a tenth, achieved by the same discarding bargain as the JPEG FAQ, refined by two decades of codec research. The screen-sized truth this reveals is worth sitting with: nearly everything visual on the internet is a small, clever fraction of its ledger size, and this page computes the ledgers — the sizes things would be if information were free to keep.

How many photos fit on a card or drive?

Divide the capacity by the size you actually store: a 128 GB card at the worked example's 72 MB raw holds about 1,780 frames; at a typical 8 MB JPEG, about 16,000. The honest planning figure sits between (raw plus JPEG pairs, bursts, the video clips nobody admits to planning) and the calculator's role is to make the raw end of the range exact, because it is the end that ambushes: cards fill at the ledger's pace whenever the photographer switches to raw mid-trip, and the multiplication takes ten seconds here against a ruined evening in the field, which is the standing exchange rate between arithmetic done early and regret experienced at leisure, in imaging as in everything else this hub computes.