The situation for visible diode lasers is much worse. Sure, the power tends to be lower, but they're still powerful enough that looking at a diffuse reflection will result in dangerous power densities on your retina. Unfortunately, the brain is really good at hiding this sort of damage, so it's possible to not notice until it's too late.
1.064um fiber lasers are the worst of both worlds. Very high powers, invisible so you have no idea how much stray light is getting out or if you're staring at a reflection, and expensive + hard to verify safety glasses.
I like doing things with high power lasers (next up for the collection is probably a 355nm ns system?), but am glad that I had to take a lot of laser safety training before I bough my first big laser source.
The worst you'll get is 100% illumination on all pixels on the LCD inside the headset.
You can't stare at the Sun, but you can stare at a video of the Sun.
Edit: I've updated the article with a hopefully better text and a reference to your comment, if you could review it and indicate if I should make further modifications I'd be very grateful.
These weren't small errors, > 2 mm in either axis across the 61 cm x 61 cm work area of the machine. But now that it has been set up properly it is quite usable.
A 'real' machine has a calibration procedure that will allow you to correct for such errors as well as a large variety of others but these cheap machines just output stepper pulses in fire-and-forget mode without any feedback at all besides the ability to re-home in case they get lost (and the homing switches are so crummy that I'm amazed they work at all). But that usually implies a ruined work piece.
That being said, it's also great to see the beam's location, so one possible solution is to mix in low-power red/green laser into your cutting beam. It can be as simple as gluing a strand from a fiber optic cable next to the main cutting head.
Not relevant to laser cutting at those powers.
I disagree with this. Both diode and tube lasers can instantly blind you. But diode lasers can be a pain to block - because they’re in the visible spectrum you need specialized blocking materials that are opaque in that very specific part of the visible spectrum. Worse still, many have leakage into other frequencies, making them even more difficult to block. On the flip side, a co2 tube is much simpler - some common and readily available plastics transparent to visible light are opaque to the IR wavelengths emitted by co2 tubes, dramatically simplifying blocking material selection.
You should never operate any sort of cutting laser without an enclosure and safety goggles, and it’s MUCH easier to do this for co2 lasers.