SVG vs JPG: what's the difference and why does it matter for makers?

Someone in a Facebook group posted a photo last week. They'd tried to cut a logo for a customer, loaded the file into Lightburn, and the edges came out like a staircase. Blocky, jagged, not even close to what they wanted. Three people replied with the same answer: that's a JPG. You need an SVG.

It's one of those things that nobody really explains properly when you're starting out, and it costs people a lot of wasted material before they figure it out.

Pixels are the problem

A JPG is made of pixels. Little squares of colour arranged in a grid. Your phone camera captures millions of them every time you take a photo, and that grid is the entire file. There's nothing else in there.

The issue is that pixels don't stretch. If your image is 500 pixels wide and you try to cut it at 500mm, the software has to invent information that was never there. It fills in the gaps as best it can and the result is blurry, soft edges. Or if it doesn't soften them, you get the staircase effect — the actual pixel grid showing through. Neither is what you want on a finished cut.

PNGs have the same problem by the way. And BMPs. Any format that stores an image as a grid of pixels will behave the same way when you try to scale it up.

SVGs are instructions, not pictures

An SVG doesn't store pixels at all. It stores instructions. Instead of a grid of coloured squares it contains things like: here's a circle, this radius, this position. Here's a curve, starting at this point, ending here, with this handle. It's maths describing shapes, not a photograph of them.

That's why they scale perfectly. Whether you're cutting something at 30mm or 300mm, the machine reads the same instructions and draws the same shapes. The edges are always sharp because the file is telling the machine exactly where to go, not asking it to guess from a blurry pixel grid.

You'll also see DXF mentioned a lot in laser and plasma cutting. Same idea — vector paths, not pixels — just a different format that's more common in CAD tools. The plasma cutting files here are DXF for that reason.

What your cutter is actually doing

Laser cutters and Cricut machines don't look at your file like you look at a picture. They read paths. Lines and curves that tell the head where to move. An SVG hands them those paths directly. Load it into Lightburn or Design Space and the software knows exactly what to do with it.

Load a JPG and the software has to trace it first — guess where the edges are by looking at colour changes in the pixel grid. It can do a reasonable job on something simple with high contrast. On anything detailed, or anything with soft edges or gradients, it falls apart pretty quickly.

Photo engraving is different

Worth saying: if you're burning a photograph into wood or leather, a raster file is actually what you want. You're reproducing shading dot by dot and the pixel grid works in your favour. A high resolution JPG or PNG is fine for that. It's cutting and line engraving where SVG makes the difference.

What to do if you only have a JPG

You can convert it. The process is called vectorisation — tracing the pixel image and generating SVG paths from it. You can do it manually in Inkscape if you have the patience, or use an auto-tracer.

The vectoriser on this site does it automatically. Drop in your PNG or JPG, adjust how much detail gets captured, and download an SVG. It works well on logos, silhouettes, anything with clear edges and decent contrast. A photograph of a face is harder. A black logo on a white background is easy.

Every file on Bonny Creations is SVG or DXF. The monogram builder outputs a custom SVG sized for your cutter. The jig files are SVG templates. If you've been struggling with blurry cuts or jagged edges, that's the first thing to check.