Lossy vs Lossless Compression Explained
Every time you save a photo smaller, you are using compression. But there are two very different kinds, and picking the wrong one is why some images look terrible at a small size while others stay crisp. The two kinds are lossy and lossless, and understanding the difference makes you far better at hitting a KB limit without ruining your picture.
No computer science degree required. Here is how they work in plain language.
What compression actually is
Compression is a way of storing the same image using fewer bytes. Think of it like packing a suitcase more cleverly so it weighs less. The question is whether you get everything back exactly when you unpack it, or whether you left a few things behind to save weight. That is the whole difference between lossless and lossy.
Lossless compression
Lossless compression makes a file smaller without throwing anything away. Unpack it and you get back every single pixel exactly as it was.
How it works
It looks for patterns and repetition and stores them more efficiently. If a picture has a big block of identical white pixels, instead of storing "white, white, white" ten thousand times, it stores "white, ten thousand times." Nothing is lost, but there is a limit to how much you can save this way.
Where you see it
PNG is the common lossless image format. GIF is another. They are excellent for logos, screenshots, diagrams and line art — anything with flat colours and sharp edges. The trade-off is that photographs, which have millions of subtly different colours and few repeating patterns, do not shrink much when stored losslessly. A photo saved as PNG is often several times larger than the same photo as a JPG.
Lossy compression
Lossy compression makes a file much smaller by permanently discarding information — specifically, the details your eye is least likely to notice. You cannot get that information back, but for photographs the savings are enormous and the loss is usually invisible.
How JPEG works, roughly
JPEG breaks the image into small blocks, usually 8 x 8 pixels, and simplifies the detail inside each block. It leans on a fact about human vision: we notice broad shapes and brightness far more than tiny colour variations. So JPEG keeps the important structure and rounds off the fine details, which slashes the file size.
What gets discarded
Mostly subtle colour differences in areas that look uniform to you — a smooth cheek, a plain wall, a patch of sky. At high quality you would need to zoom in hard to see any difference at all. That is why a 3MB camera photo can become a 40KB JPG and still look perfect at passport size.
Compression artifacts: what they look like
Push lossy compression too far and the shortcuts start to show. These flaws are called artifacts, and they come in a few flavours:
- Blockiness — the 8 x 8 blocks become visible as little squares, especially in smooth areas.
- Ringing or halos — faint ghosting around sharp edges and text.
- Colour banding — smooth gradients like a sky break into visible stripes.
Artifacts are the price of going too small. The skill is stopping just before they appear, which a good target-size tool does automatically by choosing the highest quality that still fits your KB target.
Which should you use for forms?
For form photos and signatures, the answer is almost always lossy JPEG. Forms enforce tight KB limits, and only lossy compression can get a photo that small while still looking clean. PNG's lossless quality is wasted on a small form photo and just makes the file too big to fit.
The exception is content that is not a photograph — a screenshot of text or a simple logo — where lossless PNG can look sharper. But for the ID photos and signatures these forms ask for, JPEG wins.
The generation-loss trap
Here is a mistake that quietly ruins images. Every time you open a JPEG, edit it, and save it again as JPEG, it is compressed again, losing a little more each time. Do this five or six times and you will see real damage — this is called generation loss.
The fix: keep your original photo, and export to JPEG once at the final size you need. Do not repeatedly save-edit-save the same JPEG. If you must make several edits, do them in one session and export at the end.
Quality settings explained
When a tool asks for a "quality" from 0 to 100, it is telling the lossy compressor how aggressive to be. High numbers keep more detail and produce a bigger file; low numbers throw away more and produce a smaller file. There is no single "right" number, because the same setting yields a different KB on every image. That is precisely why targeting the file size directly beats guessing a quality percentage.
A quick format cheat sheet
- JPG/JPEG — lossy, best for photos and form uploads.
- PNG — lossless, best for logos, screenshots and transparency.
- WebP — can do both lossy and lossless, smaller than JPG, but poorly accepted by forms.
- GIF — lossless but limited colours, mainly for simple animations.
When lossless actually matters
Lossless is the right call when every pixel counts: medical or archival images, graphics you will edit repeatedly, artwork with hard edges, or anything with transparency. For a one-off photo upload to a form, none of that applies, so lossy is the practical choice.
An everyday analogy: MP3 versus WAV
If images feel abstract, music makes it click. A WAV file is like a lossless image — it stores the full audio in perfect detail and takes up a lot of space. An MP3 is like a lossy JPEG — it throws away the parts of the sound most people cannot hear, so the file is a fraction of the size, and at a decent bitrate you cannot tell the difference.
The same trade-off runs through both. Lossless keeps everything and stays large; lossy discards the imperceptible bits and shrinks dramatically. And just like a heavily compressed MP3 starts to sound tinny, a heavily compressed JPEG starts to look blocky. The art is stopping before that point.
How much smaller does lossy actually go?
The savings are hard to overstate. A photograph that is 3MB as an uncompressed or lossless file often lands around 200 to 400KB as a high-quality JPEG, and can go under 50KB at form quality — all while looking fine at the sizes forms use. That is a reduction of ten to fifty times or more. This is the entire reason the web runs on JPEG for photos: without lossy compression, image-heavy pages would be enormous and slow.
WebP: the modern hybrid
WebP is worth a mention because it blurs the line. It can operate in either a lossy or a lossless mode, and in both it usually beats the older formats — a lossy WebP is typically smaller than the equivalent JPEG at the same quality, and a lossless WebP is smaller than PNG. On paper it is the best of both worlds.
The catch is support. Browsers handle WebP well, but many government and exam portals still do not accept it for uploads. So for websites WebP is excellent, but for form uploads you should stick with JPG. Knowing WebP exists helps explain why the same photo can come out smaller on some tools than others.
How compression tools decide the quality
When you set a target file size instead of a quality percentage, the tool works backwards. It compresses the image at a quality level, measures the result, and adjusts up or down until it lands just under your target — a bit like a thermostat homing in on a temperature. This is why target-size compression is so reliable: instead of you guessing a percentage that means something different on every image, the tool tries several and keeps the best one that fits. imageto10kb does exactly this in your browser, searching for the highest quality that stays under the KB you asked for.
Matching the format to the content
The real skill is not memorising rules but matching the format to what the image contains:
- Photographs — smooth tones, no hard edges: use lossy JPG.
- Screenshots and text — sharp edges, flat colours: lossless PNG looks crisper, if the form allows it.
- Logos and line art — flat colour, maybe transparency: PNG.
- Anything going to a form — the form's rule wins, and that rule is almost always JPG.
When content and format match, you get the smallest file that still looks right. When they clash — a photo forced into PNG, or a screenshot crushed into low-quality JPG — you get either a bloated file or an ugly one.
Can you reverse JPEG compression?
No — and this is the crucial thing to remember about lossy formats. Once JPEG has discarded detail to shrink the file, that detail is gone for good. Saving the JPEG back to PNG afterwards does not restore anything; you just wrap the already-damaged image in a lossless container, getting a bigger file with no better quality. This is why you should always keep your original, high-quality photo. Compress a copy to the size you need, and if you ever require a different size, go back to the original rather than re-compressing the small version.
Progressive versus baseline JPEG
You may have seen a photo load top-to-bottom on a slow connection, versus one that appears blurry and then sharpens. That is the difference between baseline and progressive JPEG. A baseline JPEG loads in order, line by line. A progressive JPEG loads the whole image at low quality first, then refines it. Both hold the same picture and both are accepted by forms; progressive can even be slightly smaller. It is a behind-the-scenes detail you do not need to choose manually, but it explains the two loading styles.
Chroma subsampling, in plain terms
Here is one more trick JPEG uses, explained simply. Human eyes are much more sensitive to brightness than to fine colour detail. JPEG exploits this by storing brightness at full detail while storing colour at lower detail — a process called chroma subsampling. You do not notice, because your eyes were not paying close attention to that colour detail anyway. It is one of the clever shortcuts that lets a photo shrink so far without looking wrong, and it is why JPEG is so good for photographs specifically.
Why JPEG became the standard
JPEG has been around since the early 1990s, and it won for a simple reason: it made photographs small enough to share and store back when storage and internet speeds were tiny. Decades of universal support followed, so today every device, browser and government portal reads it without question. Newer formats compress better, but none has JPEG's universal acceptance. For anything that has to "just work" everywhere — like an exam form upload — that reliability beats a slightly smaller file.
Troubleshooting checklist
- Is your form photo a JPG (lossy)? It should be.
- Do you see blocky squares or halos? You compressed too far — raise the target size.
- Are you re-saving the same JPEG over and over? Start from the original instead.
- Is a screenshot looking fuzzy as JPG? Try PNG if the form allows it.
- Is a photo saved as PNG too big? Convert it to JPG.
The bottom line for form uploads
Strip away the theory and it comes down to one habit: for a photograph going to a form, use lossy JPG, and stop compressing the moment before artifacts appear. That single rule handles almost every situation you will meet. Reserve lossless PNG for the times you are dealing with a screenshot, a logo, or something that needs transparency — and even then, only if the form allows it.
A myth worth dropping: bigger does not mean better
It is easy to assume a larger file is a higher-quality one, but that is not reliably true. A bloated PNG of a photo is not "better" than a well-made JPG — it is just heavier, with no visible advantage at the sizes forms use. Likewise, a 200KB photo is not automatically clearer than a 40KB one if both are sharp at their dimensions. What matters is that the image looks right and fits the rules, not that the number is big. Chasing a larger file wastes space and gets your upload rejected for exceeding the limit.
Frequently asked questions
Is JPG lossy or lossless?
JPG is lossy — it discards some detail to shrink the file. For photographs at normal quality the loss is invisible, which is why it is the standard for form uploads. If you need a truly lossless copy, save as PNG instead, but expect a much larger file that will struggle to fit a form's KB limit.
Does compressing an image always reduce quality?
Lossless compression (like PNG) does not reduce quality at all. Lossy compression (like JPEG) does, but at sensible quality levels the change is invisible. Quality only becomes a visible problem when you over-compress and artifacts appear.
Why does my image look blocky after saving?
Those blocks are JPEG artifacts from compressing too aggressively. Save at a higher quality or a larger target size, and if the image is small in pixels, consider keeping slightly larger dimensions.
Should I use PNG or JPG for a form?
JPG, in nearly all cases. Forms want small files, and lossy JPG hits tight KB limits while looking clean. Reserve PNG for graphics or when a form specifically allows or requests it.
Does converting PNG to JPG lose quality?
The conversion applies lossy compression once, which is a small, usually invisible change for a photo. It is worth it because the JPG is dramatically smaller and accepted by forms. Just avoid converting back and forth repeatedly, because each round of lossy compression stacks up and the damage eventually shows. Convert a single time from your best original, keep that original safe, and you will never notice the loss on a form-sized photo.
Quick tip: for form photos use lossy JPG and stop compressing just before artifacts appear; save lossless PNG for logos, screenshots and transparency.