Overlap setting does nothing with Modifier Meshes [Arachne]

[Theguffy1990]

Application Version

Arachne

Platform

Windows 10

Printer

Custom

Reproduction steps

Brought in a model, printed, had issues with top layer not connecting so went to increase the overlap percentage beyond 10% with a testing process involving sequential printing, and support blocker modification. When this didn't work, I verified that it also did not work with just a single model in "all at once" printing.

Actual results

Nothing. It's almost as though the default of 10% is not even being followed and it's stuck at 0%. Attempted infill mesh and cutting mesh.

Expected results

Expected to increase overlap, tested to extremes from 50, 100 and even 300 percent.

Checklist of files to include

• Log file
• Project file

Additional information & file uploads

Program crash on save project, Log file included (I never save projects, this is a non-issue for me).
cura.log

[nallath]

I've just tried this with arachne. I can't seem to reproduce your issue.

Can you check if this project reproduces what you mean?

OverlapTestProject.zip

[Theguffy1990]

I've just tried this with arachne. I can't seem to reproduce your issue.

-snip-

Ahh, apologies for incorrect terminology on the title there @nallath.

It would seem that this is definitely a "me" issue, and I am really unsure what the problem is resulting from. Your Project file did not work (it did not crash at any point either). Would there be any changes if I were to open the project fresh and save it again providing it doesn't crash? It almost feels as though something is conflicting with it, but this was tried with my version that has a moderate amount of usability plugins, but also with a clean install.

Again, sorry for the incorrect title, and the new knowledge of what to refer to them as.

[nallath]

The project should be the same if you open & save it again. One main difference between our setups is that I was using the nightly build of Arachne. We just completed a full sprint of fixing things in Arachne, so it could be that we already fixed the issue that you're encountering.

[Ghostkeeper]

One way of disambiguating that would be if Theguffy1990 could also save his project, to allow us to reproduce his case. It could be that Nallath's test case was missing some override that Theguffy doesn't know could influence this result. If Theguffy would go to File -> Save project... and share that file here, we should be able to reproduce their case precisely.

And if it doesn't reproduce then, we can (more) confidently say that the issue has been fixed in the past half year.

[Theguffy1990]

Apologies, I hadn't noticed an update to this.

I stated in the Additional Information that when I try to save that specific kind of project, it crashes outright. 32GB of ram (24GB free usually when trying to replicate), I7-8700K, trying to save to an NVME, SSD, SSHD and HDD, all fail to save and crash. That's why I included my most recent log file with the original post. There was not much information I could extrapolate from it, likely because I did not know what I should be looking for, so left it to the more experienced to either tell me I had screwed up somewhere, or that there was a genuine bug there.

I doubt I have any right to ask, but curiosity can't be helped; when would the latest version of Arachne be expected? I use it exclusively now, and would be interested in knowing if/when my workflow will be altered.

Thank you again for keeping me in the loop,
Guffy.

[Ghostkeeper]

I doubt I have any right to ask, but curiosity can't be helped; when would the latest version of Arachne be expected? I use it exclusively now, and would be interested in knowing if/when my workflow will be altered.

Asking is fine, but I can't really give an answer to this yet. It came up again recently and we plan to bring out another beta soonish, when we've ironed out the issues people encountered when creating profiles. The stable release is certainly going to take a handful of months.

[Ghostkeeper]

In that log file it indeed seems to stop logging abruptly. There is no mention of even an attempt to save project files, but it could be that the program crashed then before it had the chance to flush the log file. There is not much we can deduce from that.

It's worrisome that saving the project file would cause Cura to crash. I don't know of any other bug reports about that, so it's probably quite rare.

[Theguffy1990]

Asking is fine, but I can't really give an answer to this yet.

Of course, I won't press on the matter, but I look forward to it!

It's worrisome that saving the project file would cause Cura to crash. I don't know of any other bug reports about that, so it's probably quite rare.

Almost definitely. I suspect that it's likely a memory leak issue, as the program doesn't run very well to begin with from when I first found it, to the newest variant I have which is the same version, but much more recently installed.

I have the issue that if I go down to a low layer height, say, 0.01mm (up to around 0.03mm), it just will not slice. It gets maybe a third of the way, but then just stops. The bar goes back to the start from the position it was in, and doesn't do anything. Comparatively, mainline Cura does this successfully. Watching the memory usage, it goes up to around maybe 10GB then the bar resets, then goes up to around 12GB and stops there. It doesn't increase, the program doesn't stop responding, it doesn't drop down after the loading bar drops. It's loaded something, then experienced an overflow of some description, then just hits a break almost.

Obviously, I don't expect a beta branch to be optimised, but it would be nice if it were able to use all the resources available to it, even if it does so inefficiently. Working poorly is always better than not at all.

[Theguffy1990]

Apologies for the amendment, but here is an updated log, along with an exact process for how to achieve:

1. Open Cura Arachne
2. Drag & Drop xyzCubeTop.stl into workspace
3. Right Click model > Multiply > 8
4. Reposition all models in a grid, specifically for me 75,75, 0,75, -75,75, 75,0, 0,0, -75,0, 75,-75, 0,-75, -75,-75
5. Place Modify Mesh cubes on each space, reposition all to same coords
6. File > Save
7. Choose name and location
8. Save
9. Crash

cura.log

I will also attach a log for when it does not slice a model, along with a screenshot:

(as you can see, this time it did seem to dump memory, only got to 4GB max)
cura.log
I have tried different models, including just simple cubes, with a "fresh" computer, as well as a "stale" computer, in a VM, running as administrator, the works. The failure mode seems to be consistent, though the effects afterwards seem to have some variability.

[Ghostkeeper]

Thanks for those reproduction steps. They don't cause Cura to crash for me though. I've been loading this cube which might not be identical though, and with "modify mesh cubes" I'm assuming you meant support blockers changed to be a modifier mesh.

It saved the project file fine:
CE3PRO_xyzCalibration_cube.zip

[GregValiant]

Is this still an issue in current versions of Cura (5.8.0 and up)? Can this be closed?

[github-actions]

This issue has been automatically closed because there has been no response to our request for more information from the original author. With only the information that is currently in the issue, we don't have enough information to take action. Please reach out if you have or find the answers we need so that we can investigate further.