What Is JPEG? The Format Behind Almost Every Photo

If you have ever taken a photo on a phone, downloaded a picture from a website, or attached an image to an email, you have almost certainly used JPEG. It is the most common image format on Earth — the quiet default behind billions of photos a day. Yet most people have never asked what actually happens inside a .jpg file. This is a plain-English tour of how JPEG works, why it looks the way it does, and when it is still the right choice in 2026.

Quick answer

JPEG is a lossy image format designed specifically for photographs. It shrinks files by permanently throwing away detail your eyes barely notice, controlled by a quality setting from 1 to 100. It cannot store transparency or animation, but it is supported by every browser, camera, phone, and app ever made — which is exactly why it refuses to disappear.

A little history

JPEG stands for the Joint Photographic Experts Group, the standards committee that published the specification in 1992. That name describes the group of experts, not the file — but the format inherited it. The committee's goal was to compress continuous-tone photographs far more aggressively than the lossless formats of the day could, while keeping the result convincing to the human eye. More than three decades later that original design is still doing its job on modern hardware, which is a remarkable run for any piece of technology.

How JPEG compression actually works

The engine at the heart of JPEG is the discrete cosine transform (DCT). The image is split into small 8×8 pixel blocks, and each block is converted from raw pixel values into a set of frequency components — smooth, gradual changes plus fine, rapid detail. Here is the clever part: human vision is far less sensitive to fine high-frequency detail than to broad shapes and tones. JPEG exploits that by quantizing — rounding off — the high-frequency components more coarsely, discarding information you would struggle to see anyway.

The quality slider is simply how hard that rounding is applied. At quality 95 the rounding is gentle and almost nothing is lost; at quality 40 whole ranges of subtle detail are collapsed to zero, which is why low-quality JPEGs develop blocky patches and halos around edges. This is the direct trade you are making: lower quality means smaller files because more frequency data is thrown away. In practice, quality 75–85 is the sweet spot where photos stay visually clean while files stay small.

Chroma subsampling: why JPEG loves photos and hates text

JPEG separates an image into brightness (luma) and color (chroma) channels, then usually stores the color channels at half resolution — a trick called chroma subsampling. Human eyes are much sharper at detecting changes in brightness than in color, so halving the color detail saves space with little visible cost on a photograph.

But this is also why JPEG is a poor fit for screenshots, logos, and line art. Sharp, high-contrast edges — the boundary of black text on white, a crisp icon, a solid-color chart — are exactly the high-frequency, high-color-contrast content that subsampling and quantization smear. You get muddy fringes and ringing around every letter. For that kind of content a lossless format like PNG is the right tool. If you have graphics saved as JPEG that need clean edges, the fix is to keep or re-export them as PNG rather than pushing the JPEG quality higher.

No transparency, no animation

Two things JPEG simply cannot do: it has no alpha channel, so there is no transparency, and it holds a single still frame, so there is no animation. If you convert a transparent PNG to JPEG, the see-through areas get filled with a solid background color — Picovert flattens them to white by default. When you genuinely need a cut-out logo or an animated loop, JPEG is the wrong container; reach for PNG, WebP, or AVIF instead.

Baseline vs progressive JPEG

There are two ways a JPEG can be laid out on disk. Baseline JPEGs decode top to bottom, so on a slow connection they appear to load in a downward wipe. Progressive JPEGs store the image in successive passes — a blurry version of the whole frame appears first and sharpens as more data arrives. Progressive files are often slightly smaller and feel faster on the web because the viewer sees the full composition almost immediately. The pixels are identical once fully loaded; only the encoding order differs. Most modern export tools default to progressive for web images.

Generation loss: the one habit to avoid

Because JPEG is lossy, every save runs the compression again and discards a little more detail. Open a JPEG, make an edit, save it, and repeat, and the quality degrades a bit each cycle — a phenomenon called generation loss. It is the reason a meme that has been screenshotted and re-shared dozens of times ends up crunchy and full of artifacts. The defense is simple: keep an original master (ideally the untouched camera file), edit from that, and export a fresh JPEG each time instead of re-saving the same working file over and over.

.jpg vs .jpeg — a non-issue

People sometimes worry about the difference between a .jpg and a .jpeg file. There is none. They are byte-for-byte the same format. The three-letter .jpg extension survives from an era when Windows only allowed three-character extensions, while .jpeg is the fuller spelling used on systems without that limit. Rename one to the other and nothing changes.

EXIF metadata inside camera JPEGs

A JPEG straight from a camera or phone carries an EXIF block — embedded metadata such as the capture date, camera model, exposure settings, orientation, and often precise GPS coordinates. That data is genuinely useful for organizing a photo library, but it can also leak where a picture was taken when you post it publicly. If you are sharing photos online and would rather not broadcast your location, strip that metadata first with a tool like Remove EXIF Data, which cleans the file in your browser without touching the visible image.

How JPEG compares for a photo

For a photograph, JPEG is dramatically smaller than PNG at similar visible quality, which is why PNG is the wrong choice for camera images. WebP typically beats JPEG by another 25–35%, and AVIF can go further still.

Should you still choose JPEG in 2026?

Yes, in plenty of situations — even though newer formats compress better:

• Maximum compatibility. Email clients, older devices, print shops, and random apps all understand JPEG without question. Nothing else is quite as universal.
• Photographs without transparency. This is exactly what JPEG was built for, and it does the job well.
• Handing files to someone else. When you cannot control what software the recipient uses, JPEG is the safe lowest common denominator.

Reach for WebP or AVIF instead when you control the delivery — a website you build, an app you ship — and want the smallest files for the same quality. For a modern site, serving WebP with a JPEG fallback is a common and reliable pattern. You can turn a batch of photos into WebP in one click with JPG to WebP, all locally in your browser.

Working with JPEGs on Picovert

Everything above happens client-side on Picovert — your files never leave your device. If you have graphics or transparent images that need to become JPEGs, use PNG to JPG (it flattens transparency to a solid background for you). To shrink existing JPEGs by tuning that quality slider intelligently, run them through the Image Compressor, which finds a smaller file at a quality you choose. And whenever you are cleaning up camera photos for the web, strip the location metadata first — small habits that keep your images fast and your privacy intact.