Releases: n8bot/PrusaSlicer
PSN82.4.0b1 Release 2
NOTE: GRID SUPPORTS ARE DEPRECATED IN THIS VERSION! USE SNUG SUPPORTS INSTEAD!
Everything from Release 1, PLUS:
Added ability to use machine limits and time estimates for RepRapFirmware
Much more accurate print times when using RRF, with the ability to emit to G-Code or use parameters only for estimates. Uses proper RRF g-codes.
Dramatically improved perspective camera behaviour (see: prusa3d#7338):
Camera now adapts to print volume/scene size. Small printers actually see perspective. Larger printers see appropriate level of perspective. No more telephoto paparazzo perspective camera.
Notice the difference between perspective and orthographic cameras on the SL1S, patched vs unpatched:
PrusaSlicer-n8 2.4.0 Beta1
Windows and macOS
¡¡¡GRID SUPPORTS ARE DEPRECATED IN THIS VERSION. USE SNUG SUPPORTS!!!
More info:
New settings:
Feature name, description, default value,, identifier and type (c++ config variable):
Auto-adjust solid infill spacing:
Calculate a new spacing to fill width with possibly integer number of lines, the first and last line being centered at the interval ends. This function possibly increases the spacing, never decreases, and for a narrow width the increase in spacing may become severe, therefore the adjustment is limited to 20% increase.
Default value: Enabled
bool
solid_infill_adjust_spacing
Loop clipping:
Clip loop paths a bit (default PrusaSlicer behaviour).
Default value: 15% of nozzle width
float
loop_clipping_length
Little move inwards:
Some weird little move inwards.
Default value: Enabled
bool
little_move_inwards
Optimize island extrusions:
Print perimeter, infill and ironing for each island at once. If disabled, group extrusion roles across object islands.
Default value: Disabled
bool
optimize_island_extrusions
Optimize tool order:
Optimize the order that tools are printed on a layer to minimize tool changes.
Default value: Enabled
bool
optimize_tool_order
Retraction wipe speed:
The X/Y speed (feedrate) which the nozzle moves wile wiping during a retraction.
Default value: 12 mm/s
float
retract_wipe_speed
Small perimeters length:
The threshold of length for a loop to be extruded at small perimeter speed.
Default value: 40 mm
float
small_perimeter_length
Solid infill acceleration:
This is the acceleration your printer will use for solid (including top) infill. Set zero to disable acceleration control for solid infill.
Default value: 0 mm/s/s
float
solid_infill_acceleration
Support material acceleration:
This is the acceleration your printer will use for support material. Set zero to disable acceleration control for support material.
Default value: 0 mm/s/s
float
support_material_acceleration
Support material interface acceleration:
This is the acceleration your printer will use for support material interface. Set zero to disable acceleration control for support material interface.
Default value: 0 mm/s/s
float
support_material_interface_acceleration
Skirt speed:
X/Y speed (feedrate) to print the skirt.
Default value: 100 mm/s
float
skirt_speed
Additional XY separation for base/sparse support:
Additional XY separation between an object base/sparse support. Only used if dense interfaces are used. If expressed as percentage (for example 50%), it will be calculated over external perimeter width.
Default value: 50%
floatOrPercent
support_material_additional_xy_spacing
Expand or filter support structures:
Expand or filter the support structures by width. Positive values expand structures, negative values reduce them. Default of 0% ensures all requested areas are supported with minimum expansion of 1x extrusion width. Value of +50% expands support structures further to 1.5x extrusion width. Value of -100% reduces expansion of structures to 0, thereby filtering out support structures less than 1x extrusion width. Value of -150% filters out structures less than 1.5x extrusion width. Note: values below -100% filter out larger widths, without contracting the edge of structures. Expressed as a percentage of the support material extrusion width.
Default value: 0%
percent
support_material_expand_or_filter
Extra skirt base layers:
Add extra loops to inside and outside of skirt for given number of layers.
Default value: loops: 0, layers: 1
int
skirt_extra_loops
, skirt_extra_loop_layers
With sheath around the support interface:
Add a double sheath (two perimeter lines) around the support interface. This provides cleaner contact points when using rectilinear interface pattern. Concentric can leave more gaps than rectilinear, providing insufficient support.
Default value: Disabled
bool
support_material_interface_with_sheath
Prevent retraction within support islands:
Prevents retraction within support islands, which can be faster but lead to stringing/oozing.
Default value: Enabled
bool
retract_skip_support_islands
Modified behaviour:
Lower polyline/polygon simplification threshold:
Minimum for RESOLUTION was changed from 0.0125 to 0.001 mm.
FDM snug support fixes:
Bring snug supports to perfection. Support clipped corners/thin geometry. Otherwise, small pieces of the model are not accounted for in support generation, and missing layers occur. Revert interface/raft angles to 90 degrees to allow 45 degrees on all rectilinear fill. 45 degrees is best.
Better handle solid infill expansion:
Not as extreme, not as unsupported.
Remove slic3r non-rectangular-prism flow math:
Model extrusions as simple rectangular prisms.
G-Code numerical precision:
6 digits XYZ and 9 digits E, at least until extrusion rate is fixed.
Perspective camera FOV change:
Turn up the perspective to eleven.
Darker BG gradient:
I think it looks nicer.
Bottom solid infill is printed like top solid infill:
To balance the behaviour, and provide more control. Bridges are still bridges.
PrusaSlicer 2.3.0 n8 Edition
This is the first 2.3.0 release of my modded PrusaSlicer.
Windows 64 bit binaries provided, only. Sorry.
Mods are:
- Decrease the threshold for polyline simplification
This allows more faithful reproduction of the mesh shape. Can still be adjusted with the Resolution setting. - Increase precision of G-Code output
This fixes a problem of fluctuating/incorrect extrusion rates for small moves. This is due to rounding of coordinates after they are used to calculate extrusion amounts. - Fix painted supports not generating on shallow slopes
This allows supports to generate on areas where they previously wouldn't. - Disable 'little move inwards before leaving loop'
Did you know this existed? It doesn't anymore! - Fix support 'bleed' due to rotation
Couldn't rotate support structures by 45 degrees without crazy errors? Now you can. - Print all perimeters and infills at once, ignoring islands
For each instance of an object, print all perimeters together, then all infill, ignoring islands of the object. Helps reduce changes in extrusion rate. - Prevent optimizing of tool order
This helps maintain consistent cooling times for each section of the print. Small object regions might be printed too quickly otherwise, overheating them. - Remove support links
Rectilinear support structures would over-extrude by printing over themselves to link lines. Now they are just lines. Use a sheath, or honeycomb infill. - Remove non-rectangular-prism math
I will tell you the extrusion width and spacing and you will listen to me, slic3r. - Disable loop clipping
If I wanted a hole in my print I'd design it that way. - Fix rectilinear solid fill alignment
Convoluted math with my "fixes" sometimes "breaks." Fix it again, tony! - Change sheath to erSupportMaterialInterface
Because sometimes it touches and I get the no feeling. - Change skirt to print at support interface speed
Because sometimes support speed is lightning fast. Hopefully interface isn't... I tried, kinda. - Change wipe speed to external perimeter speed
Why wipe at 80% of travel speed when you can wipe at a reasonable speed instead?
Disclaimer: I am a noob and all of these modifications are cringey hacks. They seem to work, though. They've been tested pretty well. Report bugs and feedback! Thanks.
PrusaSlicer 2.3.0-alpha0 n8 Precision+ Minus Edition (infill+support fix)
The Precision Edition, with NEW! updated features! (And a few more tweaks)
This is PrusaSlicer 2.3.0-alpha0 with several key modifications:
-
RESOLUTION, controlling toolpath simplification, has been decreased to provide more accurate output. This may lead to problems if your firmware can not handle the extra resolution, so you can set the slicer setting Resolution to 0.0125 to achieve the default behaviour of PrusaSlicer and all other Slic3r variants.
-
NEW! E_NUM precision increased to allow
six9 digits of decimal precision in GCode output E values. -
NEW! XYZF_NUM precision increased to allow 6 digits of decimal precision in GCode output X/Y/Z/F values.
- This change completely alleviates inconsistent extrusion rate calculation. The inconsistency results from PrusaSlicer's calculation of extrusion amounts based on the un-rounded vector's norm, whereas the GCode was outputting rounded values. Without refactoring the extrusion amounts to be calculated based on the rounded vector's norm, outputting the vector to GCode with full-precision alleviates the majority of the inconsistency.
-
Flow math spacing and flow adjustment removed. See Slic3r flow math. (Bridge flow math unchanged, I hope. This is experimental, not thoroughly tested.)
-
Infill and Support adjustments to fix small issues. Don't worry about it, lol.
Only Windows 64 binaries are provided. Sorry.
This code is located in my n8_precision_minus_infill_support
branch.
PrusaSlicer 2.3.0-alpha0 n8 Precision+ Minus Edition
The Precision Edition, with NEW! updated features!
This is PrusaSlicer 2.3.0-alpha0 as at f202d11
on Aug 3 2020, with several key modifications:
-
RESOLUTION, controlling toolpath simplification, has been decreased to provide more accurate output. This may lead to problems if your firmware can not handle the extra resolution, so you can set the slicer setting Resolution to 0.0125 to achieve the default behaviour of PrusaSlicer and all other Slic3r variants.
-
NEW! E_NUM precision increased to allow
six9 digits of decimal precision in GCode output E values. -
NEW! XYZF_NUM precision increased to allow 6 digits of decimal precision in GCode output X/Y/Z/F values.
- This change
completelyalleviates inconsistent extrusion rate calculation. [Edit: not completely. Some infills are still problematic. I will investigate this and sort it out for a future release.] The inconsistency results from Slic3r's calculation of extrusion amounts based on the un-rounded vector's norm, whereas the GCode was outputting rounded values. Without refactoring the extrusion amounts to be calculated based on the rounded vector's norm, outputting the vector to GCode with full-precision alleviates the majority of the inconsistency.
- This change
-
Flow math spacing and flow adjustment removed. See Slic3r flow math. (Bridge flow math unchanged, I hope. This is experimental, not thoroughly tested.)
Only Windows 64 binaries are provided. Sorry.
This code is located in my default n8_precision_minus
branch.
PrusaSlicer 2.3.0-alpha0 n8 Precision+ Minus Plus Klipper Edition
The Precision (plus Klipper!) Edition, with NEW! updated features!
This is PrusaSlicer 2.3.0-alpha0 as at f202d11
on Aug 3 2020, with several key modifications:
-
RESOLUTION, controlling toolpath simplification, has been decreased to provide more accurate output. This may lead to problems if your firmware can not handle the extra resolution, so you can set the slicer setting Resolution to 0.0125 to achieve the default behaviour of PrusaSlicer and all other Slic3r variants.
-
NEW! E_NUM precision increased to allow
six9 digits of decimal precision in GCode output E values. -
NEW! XYZF_NUM precision increased to allow 6 digits of decimal precision in GCode output X/Y/Z/F values.
- This change
completelyalleviates inconsistent extrusion rate calculation. [Edit: not completely. Some infills are still problematic. I will investigate this and sort it out for a future release.] The inconsistency results from Slic3r's calculation of extrusion amounts based on the un-rounded vector's norm, whereas the GCode was outputting rounded values. Without refactoring the extrusion amounts to be calculated based on the rounded vector's norm, outputting the vector to GCode with full-precision alleviates the majority of the inconsistency.
- This change
-
Flow math spacing and flow adjustment removed. See Slic3r flow math. (Bridge flow math unchanged, I hope. This is experimental, not thoroughly tested.)
-
Klipper GCode flavor cherry-picked from SuperSlicer.
Only Windows 64 binaries are provided. Sorry.
This code is located in my n8_precision_minus_plus_klipper
branch.
PrusaSlicer 2.3.0-alpha0 n8 Precision Minus Plus Klipper Edition
This is PrusaSlicer 2.3.0-alpha0 as at Jul 22 2020, with several key modifications:
-
RESOLUTION, controlling toolpath simplification, has been decreased to provide more accurate output. This may lead to problems if your firmware can not handle the extra resolution, so you can set the slicer setting Resolution to 0.0125 to achieve the default behaviour of PrusaSlicer and all other Slic3r variants.
-
E_NUM precision increased to allow 6 digits of decimal precision in GCode output E values.
-
Flow math spacing and flow adjustment removed. See Slic3r flow math. (Bridge flow math unchanged, I hope. This is experimental, not thoroughly tested.)
-
Klipper GCode flavor cherry-picked from SuperSlicer.
Only Windows 64 binaries are provided. Sorry.
This code is located in my n8_precision_minus_plus_klipper
branch.
PrusaSlicer 2.3.0-alpha0 n8 Precision Minus Edition
This is PrusaSlicer 2.3.0-alpha0 as on Jul 22 2020, with several key modifications:
-
RESOLUTION, controlling toolpath simplification, has been decreased to provide more accurate output. This may lead to problems if your firmware can not handle the extra resolution, so you can set the slicer setting Resolution to 0.0125 to achieve the default behaviour of PrusaSlicer and all other Slic3r variants.
-
E_NUM precision increased to allow 6 digits of decimal precision in GCode output E values.
-
Flow math spacing and flow adjustment removed. See Slic3r flow math. (Bridge flow math unchanged, I hope. This is experimental, not thoroughly tested.)
Only Windows 64 binaries are provided. Sorry.
This code is located in my n8_precision_minus
branch.