JPEG uses lossy compression optimized for photographs, balancing accuracy and file size.
Common Uses
Digital photography
Web images and thumbnails
Email photo attachments
JPEG Optimization FAQ
What is a JPEG file?
+
JPEG (Joint Photographic Experts Group) is an image compression standard for digital photographs.
Does re-saving a JPEG make it worse every time?
+
Yes, and that is the one thing worth knowing about the format. Each save re-runs the lossy encoder over an image that has already been through it, so artefacts compound — visible as blocking around text and haloes on hard edges after a handful of generations. Edit from a lossless master and export to JPEG once, at the end, rather than repeatedly opening and re-saving the same file. The bitstream is identical to a .jpg: DCT-based lossy compression, usually with 4:2:0 chroma subsampling, wrapped in a JFIF or Exif header.
Why does every camera and phone hand me a JPEG?
+
Because it is the format the whole photographic chain settled on decades ago, and nothing since has been worth the compatibility risk to a camera maker. The sensor data is processed and encoded in the camera itself, at a quality setting buried in a menu, and what lands on the card is already a finished picture. That is why the first conversion of a photograph is never the lossy step people worry about — the lossy step already happened, inside the camera. JPEG is the same standard as JPG and JFIF — the four-letter extension is the original, and the three-letter one exists only because of the DOS 8.3 filename limit.
Is JPEG the same format as its other extension?
+
Yes. This extension and the longer or shorter name you see alongside it describe one underlying format, and anything that reads either reads both. The two names are historical: an eight-character filename limit from the DOS era, a vendor naming its application rather than its file format, or two programs that simply defaulted differently. Converting either produces the same result, so use whichever extension the software at the other end is expecting.
What quality setting should I use when converting to JPEG?
+
Somewhere between 80 and 90 for anything a person will look at. Below about 75 the artefacts start to show on skin, sky and any large flat area; above 95 the file grows quickly while almost nobody can see the difference. The setting is not a property of the image, it is a decision about this particular export — so choose it for where the file is going, and keep the original for the next time you need a different answer.
Why did my transparent background turn white in the JPEG?
+
Because JPEG has no alpha channel to keep it in. Every pixel has to be a solid colour, so transparency is composited against a background — white by default — the moment the file is written. Nothing is recoverable afterwards. If the image needs to sit on a coloured page or over a photograph, convert to a format that stores transparency instead. JPEG shares every JPG limitation — no alpha, generational loss on re-save — with the added nuisance of software that treats the two extensions differently.
Can JPEG hold CMYK colour for print?
+
It can, which is why files from print workflows sometimes look strange on screen: the colours are specified in ink percentages for a press rather than in light for a monitor. Most web-facing formats are RGB only, so a conversion has to translate, and the translation is approximate — saturated cyan and orange shift most. For anything heading to a printer, keep the original alongside whatever you convert.
Does it matter that JPEG is over thirty years old?
+
Age itself is not the problem — several formats from that era are still perfectly good, precisely because everything has had decades to implement them. What matters is what the era assumed: small files, modest processors, no network, and features nobody needed yet. Where those assumptions still fit, the format is fine. Where they do not, that is the reason to convert.
Is camera and location metadata kept when I convert a JPEG?
+
It is carried across where the target format supports it — camera model, exposure, capture time, and GPS coordinates if the device recorded them. That last one is worth a thought before publishing: a photo straight off a phone can carry the exact location it was taken. Strip the metadata during conversion when the image is going somewhere public.
Can I use JPEG directly on a web page?
+
Yes — every current browser renders it natively, so it needs no fallback copy and no conversion at publish time. The remaining question is weight rather than support: the same picture in a modern web format is frequently a third of the size at the same visible quality, which is worth doing on anything that appears above the fold. JPEG is the extension most cameras, scanners and Unix-origin software prefer, and it is universally recognised.
Can I just rename the extension instead of converting JPEG?
+
Almost never, and it is worth understanding why: the extension is a label, while the bytes inside are structured for one specific format. Renaming produces a file whose name promises something its contents do not deliver, and whatever opens it either refuses the file or misreads it. The one narrow exception is two formats that are genuinely the same container under different names — and even then a proper rewrap is safer than a rename.
Can I convert a whole folder of JPEG files in one go?
+
With an account, yes — drop them in together and they are queued as one job rather than handled one at a time, with the same settings applied across the batch and the results returned together. Free conversions run a single file at a time (merge tools take two); a paid plan is what unlocks batch processing and the larger per-file limit, which is what makes converting a whole archive practical.
