I'm not sure how much work is done by the CAM and how much by the controller.
I tried 3dpBurner-Image2Gcode and it works but it has a big drawback: any time it needs to change the laser intensity (basically M3 Sx command) it stops the cnc. The results it's an ugly joggling of the machine... instead as you see in the video above, the movement should be continue, just the PWM has to change for each pixel.
I answer by myself. Yes, the patyork's fork seems to do the trick! The S commands don't lead to a deceleration anymore. The movements are quite smooth, just a little vibration. Anyway it's a HUGE improvements from the official code! I hope they will merge the changes into the mainstream.
The issue you noted (114) is the most I have seen on this topic here.
Many folks are doing laser on/off laser "machining", but I understand the differences
My swag would be that integrating S planning int X-Y(-Z) motion would be a challenge.
Given that 3D printing is now the rage, I wonder, without any deep knowledge, if using a 3D printing stream with the 'material feed rate' parameter being used to modulate the laser would not be more effective, planning wise at least.
My guess is there is a lot more focus on 3d 'Gcode enhancements'
Anyway, maybe someone else will stop by here but most often items that need significant FW rework get action over at Issues.
Done. But I'm not satisfied with the results. I'm going to use GRBL if neither Synthetos G2 has the 'laser mode' feature. It's a pity because TinyG has a lot of interesting features!
Long answer: I don't think a "user over on Ox Forum" might be interested in the SPWM/Coolant swap. And this thread is about the "laser mode" for S commands.
Short answer: no, I compiled the repository for you as is, without any swapping... :)
Agreed, this is your "laser mode" for S commands thread.
I probably assumed your other thread "Alternative to PWM Spindle" was a necessary precursor to