10W Laser Camera Resolution is terrible

I’m running the 10W laser with current Luban software and current firmware. The resolution on the camera capture is absolutely terrible. Is this really the best image the camera can produce? It’s very difficult to see anything well enough to align a drawing to a workpiece properly.

Even then, the camera alignment does not match the real world alignment. It’s often off by 1-4 mm. The camera calibration, such as it is, is very poor. I don’t know if that would be improved if the calibration image quality wasn’t so terrible and you could actually see lines instead of blurry pixelated streaks, but attempting to align anything with the camera to 1-2 mm resolution is effectively impossible. And the amount of wrongness isn’t consistent either, it varies with distance from the center of the image.

I specifically bought a snapmaker because the camera alignment was supposed to be integrated into the hardware and software, and its almost unusable for anything requiring precise location on a workpiece.

Is this ever going to get better, or should I look for a better tool?

1 Like

Hi, I have the same issue. but I have up to 12 mm shift, so it’s very hard to get a print at the right position. The camera is not suitable to this, i have to try manually i suppose.
I hope an update will increase accuracy.

I noticed this too. Especially going from the 1.6w laser where the camera quality is far superior. Hopefully this is a software thing and will get fixed soon.

I thought about tearing apart the 1.6W and stealing the camera out of it, but that would void any warranty I’m sure…

Respect is earned, Honesty is appreciated, Trust is gained, Loyalty is returned. You are what you do, not what you say you’ll do.

Same issue. The 10W camera is really of poor quality. Did anyone transplant their 1600mW camera? I suspect it’s the same camera. Some hen you stitch 3x3 grid vs 1 single, you’re getting 9x the pixels in the 1600mW camera preview vs the 10W.

I think they have to improve the software part, accuracy is not the same at the center or at the border.
Meanwhile, I do it manually when small parts are involved and I use a 1% laser power test file to see where it goes. Boundaries test is using a simple ractangle and is sometime not enough to check. It’s really time consuming.

I’ve found the camera (in my case) is off by about 1.5-2mm in both axis at the corner. I would never rely on the camera for any type of positioning. What I do is attach a piece of wood or acrylic in the lower left using the bed bolt holes. Then cut a notch at 10,10. Then use that corner to position my work. 100% repeatable results using that method.

I made This Guide specifically to avoid using the camera, plus repeatable setups are always best and more accurate.

Strangely 0,0 for the 10W is not the same as 0,0 for the 1600mW or the CNC.