Can we change raster engraving to go top-down, to avoid smoke discoloration?

haha, i dunno, maybe all you’d get is the same ui with a checkbox to reveal ‘advanced settings.’ i’ve definitely seen some cases of ‘well this was obviously designed by engineers,’ but there’s also no reason not to let people have information if they feel they’re capable. you just tell them that something isn’t supported for one reason or another.


I find typically the response to that to be:

:wine_glass::wine_glass::wine_glass: But I paid my money so I don’t have to do better. :wine_glass::wine_glass::wine_glass:

[My personal favorite is: I’m surrounded by people who can’t do what I do but know I’m doing it wrong.]


as far as cleaning, you might try dampening a paper towel in alcohol and wrapping the work in it and place it in or under plastic to let the solvent soften the deposit.


Until we get some folks chiming in with clear examples of an Epilogue cutting and engraving back to front and what post procession is needed, it’s hard to make a judgement. I do know that in cutting an object I have many designs that process from back to front and leave soot on the leading edge. But I almost always have masking. As to engraving, even going from back to front you will have soot deposited on the unengraved surfaces that you have to attend to. And I can’t imagine having to laser through the ash and smoke of natural rubber as you go down. I guess if the air assist blows the ash totally across and away, it would work. That would be an interesting test.

I can imagine that this is a feature that might have its uses. It’s good that it is in the hopper. From all that I have read on the forum and Glowforge staff discussing this, and @smcgathyfay’s input somewhere along the line with her experience, choosing to engrave bottom up was clearly intended, taking into consideration all these issues that we brought up.

I’ve thought about it many times and haven’t ever come up with a design where I said, this is exactly where back to front would make it easier. But then, I can’t think abstractly very well about vectors and motion and plot it in my mind.


Yeah. I’ll give that a try for sure. If I have time, I’ll do that tonight.


Believe me in my line of work EVERYONE knows how to do it best but weird they continue to hire me and the company I work for to keep doing it. I guess it’s like that in every job you have, someone always knows it better but weirdly hire you to do it lol


A funny response to a simple suggestion of how to make the GF better. If somebody comes up justification of why it is better the way they do it now then fair enough. If not why not change it and make it better?

It is starting to look like belief in GF as being the one true way is a religion. Anybody that suggests an aspect might be improved is a heretic to be slapped down.

I just can’t see how a simple suggestion of an improvement that is trivial to implement becomes a big controversy. Don’t we all want the GF to be the best machine possible?


I don’t think that if you offered a simple solution to make things better it would bother anyone. It’s when it’s a constant, non stop barrage of complaints and whining, that people start to get annoyed.


Why an Epilog… I have 2 perfectly good Universal Lasers… :wink:

I cant say for the GF since I obviously dont have one in front of me.
But on my Universals, the airflow goes down through the air assist right with the laser beam then gets pulled towards the back of the bed.
When doing heavy etching, the smoke and debris get pulled back over the etched portion if cutting back to front (which is default).
For light etching it doesnt matter too much, but with deep etch on wood or rubber/silicone, that leaves alot of residue, its always cleaner to cut front to back and leave the newly etched portions clear of the shoot, smoke and debris…
Unfortunatly there will still always be some debris, but its much less when I cut front to back…


I just did make a simple suggestion of an improvement. Constructive criticism. People got upset.

I think the GF is a very good machine at this point, but it isn’t perfect. For example, I would much rather have it than an FS Muse, but the muse can do cuts and engraves without out over burning the corners. So GF have that to learn from it. I have even posted demo code to do it. If this was an open source project I would submit pull requests. But it isn’t so all I can do is make suggestions.

I just can’t understand the attitude where people can’t accept that there might be better ways to do some things. I can see that the machine is good enough for a lot of people but it doesn’t meet its own specifications on the front page yet. And here is what looks like a simple oversight that should be very easy to correct. Happy to be told it is deliberate for some reason I have overlooked, or very hard to change due to some bizarre software architecture but from where I am looking it is a trivial thing to change the direction of the Y axis and reverse the loop iterating over the image.


I’d love to see this feature added for engraves. It makes a lot of sense to have it self clean as it moves down instead of dirtying up what was just engraved.

There are a lot of pieces that I’ve done that would have come out much better if not for the direction of the engrave. Some engraves I’ve done I’m not able to clean or it will damage the engrave, so I have to leave the residue there…


So I’m not sure if you’re describing the same motion. On the Glowforge the air is pushed back-to-front from the head area, then the exhaust sucks it out the back. So it kinda shoots the smoke to the front makes a u-turn to the left, and exits. Since the engrave is front-to-back, the smoke always passes across your fresh engrave.

1 Like

Thats similar to what it does on mine…on Monday I’ll do 2 vids…one etching front to back and the other back to front…
Generally the back to front is not as clean…


I think the reason it does it bottom up is that you can see the engraving emerge. If it was top down it would be mainly covered by the head and gantry until the end. So I think it was done to make videos look more impressive.

So it might need to be an option as it seems to be on other machines. A bit more work to add a tick box in the GUI, but still trivial in the GUI frameworks I have used.


Could very well be the reason. Sure hope that option makes it in and quickly.

Could also be to keep the laser and cables out of the way if your project would catch fire


Hmm, yes that could be a reason.I think there is metal between the cables and the fire though. The main things that seem to be exposed are the lenses and the head cam.

1 Like

Doesn’t this also mean that if an engraved part remains hot by continuing to blow air assist over the just engraved stuff you continue to cool it?

1 Like

I agree with most of what you said, except this part. We don’t know enough about their codebase to make any sort of judgement about how trivial or not any change would be to implement. Nor do we know anything about their software qualification or release processes.

I think your choice of language here is part of why people get riled up. Your choice of phrasing takes what is a straightforward and pretty uncontroversial observation that reversing the direction of engraving of raster images is likely to reduce the smoke contamination, and layers a value judgement about how long it should take Glowforge to implement it. Riding along with that judgement is what could easily be perceived as an implication that they’re somehow neglecting us if they don’t prioritize such a trivial change – after all it should only take a few minutes.

For the record, I don’t think that’s what you meant by your statement and I believe that you’re genuinely trying to provide context from your experience. And it’s interesting context for me! I have spent 20+ years building software but I don’t have any experience in image processing so I enjoy reading your insights. I also don’t think that you were intentionally trying to cast any aspersions on the Glowforge team’s efforts to improve the system. Still, I have to go out of my way to remind myself of my belief that you’re just trying to provide helpful context when I read language like that in order to read it in the best possible light rather than assuming that it’s a passive aggressive attack.

Thanks for your contributions to the forum, I really do appreciate them. I hope you’ll read this in the spirit of constructive feedback that I intend!


I think the Y axis advances so slowly that it would still have time to cool things.