It used to be in 'support' since it's something you may want to decide on depending on whether you have support or not. It could also go in 'Special Modes', but I think this is more specified.
When support is enabled then it doesn't matter if the user choose
whether to remove or not the empty layers, because it's done after
generating supports. In order to avoid confusion to the user, the
setting is disabled.
Instead of having a default layer height of 0.2mm, we change the default quality profile to be the one with a layer height of 0.2mm. This fixes up the quality slider.
Cross, Cross 3D, Concentric and Concentric 3D don't work well with Gradual Infill. Lines and zigzag don't work well with Gradual Support Infill (in addition to the previously mentioned patterns) since they aren't alternating in support.
Contributes to issue CURA-4095.
Conflict in ClusterControlItem.qml where there was a new element in Master that wasn't present in 3.0. That element wasn't referenced anywhere else so I just removed it.
Modified version of the Prusa i3's definition. While the i3 values work,
the print time estimations were horribly imprecise (often the print
takes 2x to 3x as long as Cura estimated). Therefore, I inspected the
problem, and found a Reddit thread [1] about it. Instead of hacking my
own installation, I just wrote a defition including the correct
acceleration/jerk values that work with my firmware (custom build of
Marlin based on https://github.com/JimBrown/MarlinTarantula/).
There's downsides however. I don't know how it performs with the stock
firmware (although I doubt there'd be many differences). Furthermore,
the `jerk_xy` is set to 4 in my definition, but that's only true for
the `x`value,the Marlin firmware has a `y` jerk value of 7. That leads
to minor differences in the printing time (e.g., just finished a print
which Cura estimated with 13h 30m, but it took 12h 50m).
Feedback from stock firmware users is therefore highly welcome. Also, if
someone could provide information about the jerk values and help
improving those inaccuracies, that'd be awesome.
[1] https://www.reddit.com/r/3Dprinting/comments/5wz0xs/
