[master] weird fat skin lines

[smartavionics]

Application Version

master

Platform

Linux

Printer

Kossel XL

Steps to Reproduce

Open cura, select printer, select profile, load model, slice.

Actual Results

Every other skin layer has a weird region where the lines are too fat and some lines are missing.

Expected results

Normal skin

Additional Information

The region that is bad is at the start of the skin.

[smartavionics]

Here's the project, it's not a standard printer, hope you can load it...

KXL_glasses-leg.3mf-renamed-to.zip

[diegopradogesto]

Don't worry, it's very easy to reproduce even with a Custom FFF Printer.

[diegopradogesto]

Devs.: CURA-5690

[LipuFei]

@smartavionics Also found that in your kossel_xl_extruder_0.def.json, material_diameter's default_value should actually be 1.75 instead of "1.75".

    "default_value": "1.75",   Change it to --> 1.75 without double quotes

[smartavionics]

OK, thanks a lot for spotting that.

[LipuFei]

@diegopradogesto I couldn't reproduce this with today's master. Do I need to change some settings to make this happen?

[smartavionics]

Shouldn't all the settings be in the project?

If you can't reproduce it, let's assume it was a glitch and close this issue, if I find it happening again, I will reopen.

[LipuFei]

@smartavionics Tried with your settings, too. No luck. @diegopradogesto Do you want to have one last try?

[diegopradogesto]

Try with this project file and see layer 13 for example.

CFFFP_example.zip (Rename to 3mf)

[smartavionics]

With my cura it slices but fails to display the layer view and I get these messages:

2018-09-15 11:28:40,932 - ERROR - [Thread-3] UM.Logger.logException [84]: Exception: Job <CuraEngineBackend.ProcessSlicedLayersJob.ProcessSlicedLayersJob object at 0x15aca5a8> caused an exception
2018-09-15 11:28:40,932 - ERROR - [Thread-3] UM.Logger.logException [88]: Traceback (most recent call last):
2018-09-15 11:28:40,933 - ERROR - [Thread-3] UM.Logger.logException [88]:   File "/opt/cura/lib/python3/dist-packages/UM/JobQueue.py", line 121, in run
2018-09-15 11:28:40,933 - ERROR - [Thread-3] UM.Logger.logException [88]:     job.run()
2018-09-15 11:28:40,934 - ERROR - [Thread-3] UM.Logger.logException [88]:   File "/opt/cura/lib/cura/plugins/CuraEngineBackend/ProcessSlicedLayersJob.py", line 192, in run
2018-09-15 11:28:40,934 - ERROR - [Thread-3] UM.Logger.logException [88]:     material_color_map[position, :] = color
2018-09-15 11:28:40,934 - ERROR - [Thread-3] UM.Logger.logException [88]: IndexError: index 1 is out of bounds for axis 0 with size 1

[smartavionics]

I updated my cura+uranium to the current masters and the above failure goes away. Furthermore, I am not seeing the fat skin lines.

I vote that we close this issue for now and if I see them appear again, will reopen it. Thanks for looking at it anyway!

[smartavionics]

See https://community.ultimaker.com/topic/24494-introducing-ultimaker-cura-35-beta/?tab=comments#comment-219435

[LipuFei]

@smartavionics It's caused by this commit: Ultimaker/CuraEngine@06dc32c

[smartavionics]

Hi @LipuFei , well the bug is in the 3.5 beta and is biting people!

[LipuFei]

We are on it.

[BagelOrb]

This is inherent to how the mergeInfillLines works.

I don't believe we can improve on this without making it worse in some other situations.

We should consider replacing the MergeInfill strategy all together with some new approach.

[smartavionics]

No longer a problem for me.