Development will be conducted in close collaboration with Snapmaker ecosystem and with Radoux/Radu, author of FullSpectrum the now official part of the Snapmaker team. So from v1.9 forward expect big things! By Neotko — inventor of Ironing/Neosanding (Ultimaker Cura, PrusaSlicer)
Check https://sebsucmor-alt.github.io/OrcaFS-NeotkoCM/ For a Basic tutorial and Introduction to the world of delicious Color Sandwiches!
Check Releases
Features conceived and designed by Neotko — inventor of Neosanding, now known as Ironing in OrcaSlicer, PrusaSlicer, Bambu Studio and Cura.
This is the Neotko FullSpectrum feature pack ported on top of the official Snapmaker Orca 2.3.4 base. It adds a set of surface-quality, colour-blending, wall-generation and workflow features. Everything here is opt-in — with the new options left at their defaults, Snapmaker Orca behaves like the stock build. This guide explains what each feature does and how to use it; no programming knowledge required.
About this build. This is a feature pack meant for pick & choose — each block (ColorStitch, Penultimate, Libre Mode, NeoArachne, NeoTower, Align & Stack) can be adopted independently. Some surfaces are marked (WIP): they are usable but still being refined, or not yet wired in this base. Review your G-code before long prints.
Everything on the colour side of this pack revolves around one idea: a Sandwich of passes you build yourself.
A normal slicer treats the top of your part as one thing — one filament, one pattern, printed once. The Sandwich breaks that open. Each surface zone (the Top layer and the Penultimate layer) becomes a small stack of passes, and you decide what each pass is:
- A plain Solid pass (a normal solid fill with its own tool, angle and Z) — stack two or three of them and you get the old MultiPass glazing/cross-hatch effect.
- A ColorStitch pass that decides the filament for each fill line — stripes, dithered blends, hard bands or custom patterns.
- A PathBlend pass — a continuous gradient that fades between filaments across the surface.
The two zones together form the Sandwich over your fill. You build it in one place — the Sandwich Editor (Quality → Surface ColorStitch → Sandwich editor…) — drag dividers to split the layer height between passes, save the result as a profile, and reuse it or paint it onto specific faces with the ColorStitch Painter (§6).
The goal is not "presets that work" — it is a playground. Mix, stack, experiment, save profiles, paint them, and see what surface comes out.
Beyond surface effects the pack also adds a new wall-generation engine — NeoArachne (§8) — a post-slice wipe-tower planner — NeoTower (§9) — and a set of professional/experimental unlocks under Libre Mode (§4), including the Align & Stack gizmo (§7).
- Surface ColorStitch — the Sandwich Editor
- Neoweaving + Monotonic Interlayer Nesting (WIP)
- Penultimate Top Layers
- Libre Mode
- 4a. Enabling Libre Mode
- 4b. Floating objects
- 4c. Assembled Boolean mode
- 4d. Per-volume XY compensation
- 4e. Copy / Paste Process Settings
- 4f. Assembled Parts — full options
- 4g. World-space import (WIP)
- 4h. Internal-bridge handling
- 4i. Realistic Shading (2.3.9) — Phong + SSAO + contact shadow in Prepare
- S3DFactory — Simplify3D project import
- Surface Effect Profiles & ColorStitch Painter
- 6a. Saving and managing profiles
- 6b. The ColorStitch Painter gizmo
- 6c. Palette groups, slot cap & Save All
- 6d. Painter mode at slice time
- 6e. Profile persistence and 3MF round-trip
- 6f. Weave preview on the painted surface
- 6g. MixedFilament Object mode (Beta)
- 6h. Highlight active colour — where is this colour applied? (2.4.0)
- 6i. Painted colours stay visible outside the painter (2.4.0)
- Align & Stack — align and stack two objects
- NeoArachne — alternative wall generator
- 8a. Turning it on
- 8b. Per-feature engine choice
- 8c. Edge Closure controls
- 8d. Preview Lab
- NeoTower — post-slice wipe tower
- Bump Mapping Editor — texture-driven wall & top-surface relief
- Precision Adaptive Layer Height — point-based layer height curve
- NeoWave Support (WIP) — Wave-Huygens roof + hollow pillar
- Painter Pro Mode — precision tools for the stock Color Painting gizmo
- NeoStitch Interlock (WIP,
⚠️ UNTESTED) — Z-axis layer interlocking - Expert G-code Reprocessor (2.3.8) — layer-ranged, per-tool G-code post-processing
- PerObject Support (2.3.9) — support that avoids the other objects on the plate
- Gravity ("True Objects") — real floor, honest bridges
- Typographic Spacing (2.3.9) — real kerning for embossed text
- Bridging infill extra expansion (2.4.0) — anchor bridges before they cross
- RealColor View — see the colour you are actually going to print
- Real prints — what this actually looks like off the bed
The Sandwich Editor is where you build the per-surface effect stack. It lives under Quality → Surface ColorStitch → Sandwich editor….
Naming. The feature family is called ColorStitch in the UI. You will still see ColorMix in a few internal places — config keys and 3MF data were not renamed, so old projects keep working. They refer to the same system.
The dialog has two columns — Top layer and Penultimate layer. Each is an independent stack of 1–3 passes, chosen with the Passes selector. The passes are stacked as thin virtual sub-layers inside the same nominal layer; you drag the dividers between them to split the layer height (each pass gets its own Z share). Per pass you can also move it up/down in the stack.
Each pass exposes:
- a kind (see §1a),
- a Z mm height box,
- an angle box (
-1 = auto; scroll the wheel over the box to rotate). For a PathBlend pass this same box is repurposed to show ramp end (the top height of the ramp, in mm) instead — PathBlend's own fill angle isn't edited from this box (see §1c), - an Advanced ⚙ button to open that pass's detailed settings (a
*marks non-default values) — for PathBlend this is where the start/end zone editor lives (§1c).
A single Perimeter override checkbox per zone clones the walls into every Solid pass — useful when you want the perimeter reprinted with each glaze.
At the bottom are the Filament & TD preview (§1e) and the ColorStitch Studio (§1g).
Each pass in a zone is one of:
| Kind | What it does |
|---|---|
| None | Empty slot — no pass here. |
| Solid | A normal solid fill pass with its own tool, angle and Z share. Stack 2–3 Solid passes and you get the classic MultiPass effect: cross-hatch (two angles, two colours), a glaze pass over a base, or an optical colour blend. |
| ColorStitch | Decides the filament line by line across the surface — stripes, dithered blends, hard bands or a custom pattern. Configured in the Edit gradient… dialog (§1b). |
| PathBlend Half | A gradient pass with no complementary cap — the ramp climbs from its floor straight up, one tool only (§1c). |
| PathBlend Full | A gradient pass with a ramp and a complementary cap in a second tool, filling the rest of the layer height (§1c). |
MultiPass = multiple Solid passes. There is no separate "MultiPass" button anymore — you simply add Solid passes and split the height between them with the dividers. Aim for the height shares to add up to the full layer for full coverage.
PathBlend is always a single full-height gradient — when you pick a PathBlend kind the stack collapses to one pass.
A ColorStitch pass decides which filament prints each fill line. Open its Edit gradient… button to configure it — the same editor also opens from the ColorStitch Painter's Pro tray (ADV… button), so whichever entry point you use, you get the identical dialog.
Which infill patterns does ColorStitch work on? (2.4.0) Since 2.4.0 it works on Monotonic, Monotonic Line, Rectilinear and Hilbert Curve out of the box — the old "ColorStitch on Monotonic (continuous)" setting is gone and its behaviour is always on. It also works on Concentric, Octogram and Archimedean: the line distribution is not always uniform there, so the dithering is less predictable, but the results can be worth having. Getting a clean dither on Hilbert Curve in particular takes some tuning and does not always come out.
Pattern style picks where the pattern comes from. Only one style is active at a time — they're mutually exclusive by design (a custom string, a MixedFilament recipe, a weave, a blend and a set of stripes can't all drive the same pass at once), and a one-line note under the selector always says what the active style does, shown in amber when it overrides everything else:
| Style | What it produces |
|---|---|
| Custom pattern | Click filament buttons to build a digit string by hand; the slicer loops it across lines (line 1 → first digit, etc.). For exact, repeating stripes you design yourself. Clear / ⌫ Undo digit / Invert live here. |
| MixedFilament recipe | Pick one of your MixedFilament combinations (e.g. F1+F2 50/50) — it becomes the whole pattern. Only shown when MixedFilament virtual digits are enabled. Switching to another style clears the active recipe. |
| Textile weave | Ready-made weave structures generated for you — see below. |
| Smooth blend — 2 colours | Two filaments distributed across the surface with a percentage split, dithered so the transition looks smooth. The most common choice. |
| Smooth blend — 3 colours | Three filaments at configurable percentages; the middle colour concentrates in the centre. |
| Stripes — manual band sizes | Explicit band counts: N lines of Colour 1, M of Colour 2, … repeating. |
Two of those styles, with the live preview of how the surface will look:
Colours used — pick Color 1–4 (each maps to a loaded filament). Only shown for the two blend styles and Stripes — Custom/MixedFilament/Weave patterns already carry their own colours in the digit string.
Preview — a horizontal strip plus a square "how it will look" sample, drawn at the pass's infill angle, always visible for whatever style is active, including Custom strings and MixedFilament recipes (previously these had no colour preview at all). Blends stretch to fill the square once, the same way the slicer spreads a gradient across the whole surface; patterns (Custom, MixedFilament, Weave, Stripes) tile as a repeating motif, matching how they actually print.
Blend controls (Smooth blend styles):
| Control | What it does |
|---|---|
| How much Color 1 / Color 2 | The percentage split. (In 3-colour, Color 3 fills the rest automatically.) |
| Color overlap (soft ← hard zones) | How much colours bleed into each other in 3-colour blends. |
| Transition shape | Even (same density everywhere) · Slow start (the default when you first pick a blend style) · Slow end · S-curve (smooth start & end) · Custom shape (set γ) · Hard step. |
| Skip tiny areas | Surfaces with fewer than N fill lines use Color 1 only. |
| Invert direction ⇆ | Reverses the per-line sequence (blend and stripe styles; Custom/Weave have their own Invert button instead). |
Textile weave — classic weave structures from real fabric construction, generated as pattern strings from two colours you pick:
| Weave | Pattern | Look |
|---|---|---|
| Plain (tafetán) | 1212 alternation |
Balanced 50/50 mix, neither colour dominates. |
| Twill 2/2 (sarga) | 1122 |
Diagonal weave — the true per-layer diagonal offset isn't implemented yet, so today it prints as a static repeat (the dialog flags this). |
| Twill 3/1 | 1112 |
One colour dominates, the other marks a diagonal (same offset caveat). |
| Satin 5 (satén) | 11112 |
One colour covers ~80% of the surface; the other appears as scattered accent points. |
| Houndstooth | 1122 top / 2211 penultimate |
The classic pattern only appears where Top and Penultimate cross at 90° — the dialog writes the correct half automatically for the surface you're editing and shows you the matching string for the other one. |
Pick Colour A / Colour B to substitute into the weave; if you set both to the same filament, Colour B is nudged to the next one automatically (a one-colour weave isn't a weave). Edit as custom pattern… copies the generated string into the Custom style for hand-tweaking.
Stripes controls — for each of Color 1–4: a swatch (following the colour picked above) and a lines: count. Bands repeat [Color 1 × lines, Color 2 × lines, …] until the surface is filled; 0 lines skips that colour.
Print options (apply to every style):
| Control | What it does |
|---|---|
| Infill angle override | -1 = Auto. Scroll the mouse wheel over the pass preview bar to rotate it live — the bar's stripes rotate with it. A fixed angle (≥ 0) is now honoured exactly in the G-code (the per-layer fill rotation is locked out for that pass), so the print keeps the angle you set. -1 = Auto lets the slicer alternate per layer (uniform finish, but the orientation won't match a static preview). |
| Gradient repetitions | 1 = single; higher repeats the pattern across the surface. |
| ColorStitch min. line length | Fill lines shorter than this (mm) are skipped, so they keep the surrounding colour and avoid toolchanges on tiny segments. Default 0 (don't skip). |
An estimate of how many lines a 60×60 mm surface would have at your filament width is shown next to the strip preview.
A PathBlend pass creates a continuous gradient across the surface: one filament dominates at one edge, another at the opposite edge (Full mode), with the two flows changing in exact proportion path-by-path in between — a real, physical Z gradient within a single layer, not a dithered pattern. Choose Half (ramp only, no cap) or Full (ramp + complementary cap) when picking the kind.
Basic controls, on the pass's own row:
- floor — the ramp's starting height (mm), at the low edge of the surface.
- ramp end — the ramp's top height (mm), at the high edge. In Half mode this is locked to the full layer height (no cap exists to fill the rest). In Full mode you can drag it all the way up to the layer height too — that leaves zero of the cap's colour in that area ("techo"), useful when a translucent (high-TD) filament needs full opaque coverage instead of a thin sliver of the wrong colour on top.
- Mode — a cycling button through Linear / Ease In / Ease Out / Ease In-Out, shaping how quickly the ramp climbs (only used by the older Sandwich-Editor-only engine path; the per-scanline staircase engine that actually prints today ramps linearly in
tbefore the start/end zone below is applied).
ADV… — start/end zone editor. Opens a small cross-section graph of the layer: the horizontal axis is position across the surface, the vertical axis is real height in mm. Two draggable handles set the shape:
- the low handle — where the ramp starts to rise, and how low its floor sits;
- the high handle — where the ramp finishes rising, and how high its top reaches.
By default the ramp spans the full surface edge-to-edge (the classic behaviour). Drag the low handle right to add a flat "start zone" before the ramp begins climbing; drag the high handle left to add a flat "end zone" after it's done. This is the same editor, same model, in both the Sandwich Editor's ADV…/Advanced button and the ColorStitch Painter's Pro tray — whichever one you use, they write the same pass data.
(2.4.0) Fixed — a flat PathBlend on some surfaces. The staircase gives one height per fill line, which quietly assumed the surface pattern hands it one line at a time.
Monotonic linedoes; plainMonotonic(and other patterns that chain their lines into a long zigzag) does not — the whole surface arrived as a single line, got a single height, and the gradient came out flat with no Z change at all. It was most visible on bottom surfaces, whose default pattern isMonotonic, while tops set toMonotonic linelooked fine — but the same top would have broken with the same setting. PathBlend now asks for unchained lines itself, so the gradient no longer depends on the surface pattern you chose.
The gradient runs across the build-plate Y axis — rotate the object to change direction. PathBlend works best on surfaces with many fill lines; on small surfaces the gradient is coarse. It shares the Line distribution mode (§1f) — if a gradient looks broken across holes, try LaneQuant or DirCluster. In the ColorStitch Painter specifically, PathBlend also exposes its own fill angle field (-1 = auto, or a fixed 0–359° override) next to the Mode button — the Sandwich Editor doesn't have a separate control for this and leaves it on auto.
⚠️ PathBlend is the most fragile part of the engine. Its per-scanline staircase model — one physical print-height step per fill line, each step resting on the layer below and its neighbours — is validated and must not be disturbed by unrelated changes. The start/end zone and floor/ramp-end controls above are an intentionally safe extension of that model: left at their defaults they reproduce the exact same G-code as before. A future, more ambitious idea — letting the ramp rise and fall within one pass instead of always climbing — was considered and set aside for now, because the current staircase can't do that without risking unsupported overhangs at the print head; it would need a different, multi-pass engine (closer to the Bump Mapping Editor's approach, §10) to do safely.
These apply to whichever passes are active on each zone.
Zone — All surfaces vs. Topmost only. On many models the "top surface" appears on every horizontal face. All surfaces applies the effect everywhere; Topmost only restricts it to the single highest horizontal surface. Available independently for Top and Penultimate — use Topmost only on stepped objects to colour just the very top.
Filament filter (0–16). Apply the effect only to regions assigned to a given filament number. 0 = no filter. Example: red body (filament 1) + white logo (filament 2) → set 1 to leave the logo untouched.
The Filament & TD panel visualizes how passes combine optically. Each filament has a Transmission Density (TD) value — low TD = opaque, high TD = translucent:
| TD range | Type |
|---|---|
| 0.1 – 0.5 | Highly opaque — 1–2 passes fully cover the lower colour |
| 0.5 – 3.0 | Opaque-translucent — some lower layer shows through |
| 3.0 – 7.0 | Translucent — needs several passes to block |
| 7.0 – 10+ | Highly translucent — lower colour almost always visible |
Four TD sliders (one per filament) are saved per machine (neotko_td_1..4) — they describe your actual filaments, not the print profile. The preview shows the blended Top result, the Penu result, and the final on-print result (penu showing through the top by opacity), plus a transmit= readout.
The inverse colour-match ("find the recipe closest to a target colour") lives in ColorStitch Studio → Target + Match ▸ (§1g), using ΔE2000.
This controls how the slicer maps colour assignments (ColorStitch slots / PathBlend positions) to the physical fill lines of a surface. It does not change the pattern — only how slots find which lines belong to which spatial "lane." It lives in Quality → Surface ColorStitch → Line distribution mode (directly below Minimum line length) and affects both ColorStitch and PathBlend.
| Mode | Best for |
|---|---|
| Default | Raw print order. Simple rectangular surfaces, Custom-text patterns. |
| GeoSort | Print order scrambled but the spatial direction is clean. |
| LaneQuant | Surfaces with holes, concavities or disconnected sub-regions — fragmented stripes stay the same colour. Recommended for complex tops. |
| DirCluster | The fill engine rotated direction per sub-region — each region keeps its own coherent gradient. |
Quick rule: if the gradient looks wrong, move one mode up and re-slice.
The Studio (bottom of the Sandwich Editor) generates a strip of colour swatches from your loaded filaments + TD, each one a complete pass recipe with its predicted colour. Click a swatch to load it into the live editor. The Mode dropdown:
| Mode | What it generates |
|---|---|
| Gradient ramp | A manual top-only ramp between two tools (A visible side, B contrast). Set Steps and the Split (thinnest → thickest top pass, mm; floor 0.04). |
| Flat color (predict) | The gamut reachable by stacking solid passes — robust, predictable colours. |
| Mixed approximation (predict) | An extended gamut: a dithered ColorStitch base plus a translucent solid on top — colours no single filament can make. |
Target + Match ▸ — pick a MixedColor target and press Match; the Studio runs an inverse search (minimising ΔE2000) and loads the closest achievable recipe, showing the resulting ΔE.
Name + Export — turns swatches into saved Surface Effect Profiles (§6). The strips react live to the TD sliders.
(WIP — not wired in this build.) Neoweaving and its companion, Monotonic Interlayer Nesting, are not yet ported to this Snapmaker 2.3.4 base and will arrive in a later release. The description below documents the intended behaviour.
Neoweaving alternates the Z height of successive fill lines on each layer: odd lines at the nominal height, even lines slightly higher (by an amplitude). The next layer inverts the pattern, so the elevated lines nestle into the recesses below — mechanical interlocking between layers, like puzzle pieces. It improves inter-layer adhesion and vibration damping without changing external dimensions. This is a structural technique, not a visual one.
Monotonic Interlayer Nesting is the companion that makes Neoweaving clean and controllable: it shifts the monotonic fill reference by half the line spacing on alternate layers, so the lines of layer N sit over the gaps of layer N−1. That precise registration is what lets Neoweaving's raised/recessed lines lock together layer to layer — so the two ship together.
The layer(s) just below the top surface — the penultimate layer — normally behave like ordinary solid infill. This feature classifies them as their own zone so they can carry their own density and their own ColorStitch/PathBlend/Solid passes.
Set it in Strength → Top/bottom shells:
- Penultimate top layers — number of solid layers below the top treated as penultimate. Default 0 (feature off); set 1 or 2 to enable (max 20). These use reduced density for a smoother transition and faster printing.
- Penultimate solid infill density — density of that layer (default 100%).
When painting (§6), penu autonomy auto-forces 2 penultimate layers for an object whose painted recipe declares penu activity, so the effect has a surface to print on.
Libre Mode unlocks Snapmaker Orca for workflows where the normal constraints get in the way — multi-part assemblies, professional workflows, experimental printing.
Libre Mode uses a two-key gate so it never interferes with normal use:
- Master switch — Preferences → "Enable Neotko LibreMode (requires restart)" (off by default). Until this is on, no Libre Mode UI exists at all (Snapmaker sees a stock build).
- Active toggle — once the master switch is on (and after the restart), a side button labelled "Neotko LM: Off / On" appears in the toolbar. Click it to toggle the active state.
⚠️ Because the master switch builds UI at start-up, enabling it requires a restart. Some panels (e.g. full Assembled-parts options, §4f) only appear after that.
While active, Libre Mode also signals the slicer: an object with no first layer becomes a warning instead of a hard error, and split/assemble operations are allowed across Z=0.
⚠️ Moved. Floating/anchoring is no longer part of Libre Mode — it is its own independent toggle, "True Objects" (Gravity), covered in full in §17. This section stays as a pointer so old links keep working. Libre Mode still opens the door to it (the button only exists once the Libre Mode master switch is on), but turning Libre Mode itself on/off no longer affects whether objects float — only True Objects does.
With True Objects active, objects can sit at any Z height — floating above the bed or partly below it — instead of being snapped to the plate. The slicer still generates G-code and warns (instead of erroring) when an object has no initial layer. Use it for assemblies whose parts print at specific heights, or parts that clip into a structure already on the bed.
The floating Z is preserved across object operations — copy/paste, Paste Process Settings, reload-from-disk, replace-STL, boolean, mesh simplify, move/rotate/scale/mirror and face the camera no longer snap a floating object back to Z=0. To drop a floating object to the bed on purpose, use the sinking column in the object list (that path is left intact).
Real "is it floating?" detection. The stock warning used a blind heuristic — empty first layer = floating — which is wrong the moment an object rests on top of another object (empty first layer of its own, but not floating). This build measures it instead: for an object whose lowest geometry starts above the bed, each instance is checked against the bed and against the top surface of every other object's instances — real Z gap, real XY footprint, per-instance. An object stacked on another no longer gets a bogus warning, and a genuinely floating island still surfaces one (the old blanket suppression that could hide real floaters is gone). The tree-support sharp-tail seed uses the same check. This is also the foundation of PerObject Support (§16) and True Objects / Gravity (§17), which goes further than "can it float" into "what does the slicer actually do with a stacked piece" (honest bridges instead of guesswork).
When you combine parts into an Assembled object, the stock slicer performs a boolean union. Assembled Boolean mode (right-click an object → toggle; neotko_assemble_boolean, default on) lets you turn that off per object: the parts are sliced and merged without a union, so overlapping multi-material geometry (inserts, interlocking colour regions) is kept as separate co-existing meshes instead of being collapsed. This mode is print-verified.
Normally XY contour/hole compensation is per-object. With Libre Mode and an Assembled object, each volume (individual mesh / part) can have its own XY compensation; the slicer applies the delta between the part's value and the object's to that part before merging. Use it for multi-material assemblies whose materials shrink differently (e.g. a PETG insert in a PLA shell). The controls appear in the part settings.
Libre Mode adds Copy Process Settings (a submenu with Speed / Quality / Strength and All) and Paste Process Settings to the object context menu. Copy one object's process settings and paste a chosen block onto another:
- All → replaces the target's process settings.
- A single category (Speed / Quality / Strength) → merges into the target, leaving the others untouched.
Use it to propagate a tuned block across many parts without overwriting their other settings.
In stock OrcaSlicer the parts inside an Assembled object expose only a limited subset of settings. With Libre Mode active each part's settings tab exposes the full option set (the combined Print-object + Print-region keys).
⚠️ The parts tab is built once at start-up — toggling Libre Mode may need a restart for the full set to appear.
(WIP — basic functionality works.) With Libre Mode active, objects can be imported at their source-file world coordinates instead of being re-centred on the plate, preserving relative positions across an assembly.
Recommended workflow in this build: rather than relying on world-space import alone, import as an Assembled object and then split in Libre Mode. That is the reliable route while the world-space path is being finished.
On floating objects and unusual geometries, internal-bridge detection can misfire and apply bridging where it isn't wanted. Note that the stock 2.3.4 default internal_bridge_density is 25% — if a top surface looks unexpectedly filled or empty, that stock setting is usually the cause, separate from Libre Mode. (The older fork's automatic "disable internal bridges" behaviour is deliberately not carried as-is here; it caused a layer-count bug and will return later as an explicit opt-in.)
With Libre Mode active, the Prepare tab's 3D objects render with the same Phong + fresnel + screen-space ambient occlusion + projected contact shadow shading that RealColor already used for shells in the Preview tab. Turn Libre Mode off and objects go back to the stock flat/Gouraud look — nothing changes for a normal build.
- Applies to the normal opaque object view only; the Assemble tab and the sinking/transparent pass are untouched.
- The contact shadow appears on the bed under each object, same as it already did in Preview.
- No new toggle to learn — it's automatic once Libre Mode is on, and falls back silently to the stock shader if a shader/framebuffer isn't available on your GPU.
This pack can open Simplify3D .factory project files — a complete project with multiple objects, positions and extruder assignments. A .factory always loads as a single Assembled object (its parts share one world-space layout — base, texts, etc.). Use File → Import → Import 3D model (or drag & drop) and pick the .factory file.
Recommended workflow in this build: after import, split in Libre Mode to recover the individual parts in place. The importer is functional this way; a little positioning fine-tuning is still unfinished, so the assembled→split path is the reliable route.
Save Sandwich configurations as named profiles, then paint them onto specific surfaces of your model with a brush gizmo — different parts of one object can carry different Sandwiches. Profiles and painted areas are saved inside the 3MF, so they travel with the print.
At slice time, when an object has any painted facets it switches to painter mode: the preset Sandwich settings are ignored for that object and each painted area uses its own profile. This is the cleanest way to apply several different surface effects to one object without splitting the mesh.
In the Sandwich Editor, Save as profile… captures the current pass stacks as a named profile (a popup reports how many keys were captured). Manage Sandwich Profiles lets you Load into dialog, Update from current, Rename and Delete.
Orphan warning: deleting a profile that has painted areas emits a non-critical slicing warning ("…painted regions referencing deleted Surface Effect Profile(s)…"); the slice continues and those areas just get no effect. Re-paint or re-create the profile to fix.
The Painter lives in the left-side gizmo toolbar of the 3D view.
Tools (top row)
| Tool | What it does |
|---|---|
| Select | Click objects in the scene to choose which ones you can paint. Shift-click removes one. |
| Paint (smart fill) | Click a flat face → flood-fills the coplanar region with the active profile. |
| Eraser | Smart-fill removes paint under the cursor. |
| Pick (eyedropper) | Reads the recipe under the cursor and loads it as the active colour. |
| Sticker | Places the loaded SVG on a flat top face (§ Stickers, in Palette). |
Mouse rules — left-click paints/erases/picks with the active tool; right-click is camera only; Shift + left-click is a one-shot erase.
(2.4.0) You can paint several objects. Open the Painter with nothing selected and it starts in Select so you can click the objects you want; open it with a selection and those objects are already picked. While painting, moving the cursor onto another chosen object switches to it automatically, and clicking an object that isn't in the set adds it (that first click adopts it — the one after paints).
(2.4.0) The active colour tells you its state. Next to the swatch at the top of the panel you will see one of:
| Badge | Meaning |
|---|---|
| slot N (green) | The colour already owns a paint slot on this object — painting applies it right away. |
| ready (grey) | A colour is chosen; it takes a slot the first time you actually paint with it. |
| no colour (amber) | Nothing is selected — clicking the model will not paint (and will not erase either). |
This replaces a long-standing trap where the swatch kept showing a colour that had quietly stopped being paintable after a slice or a tab change, so clicks did nothing until you re-picked it from the palette.
The panel
- Palette strips — collapsible Gradient ramp and Flat color sections, scrollable strips of swatches generated from your filaments + TD (same engine as the Studio, §1g). They regenerate when colours/TD change.
- Pro — the composer, and the Pro panel IS the active colour: build Top / Penultimate / Bottom passes (Solid / ColorStitch / PathBlend Half|Full) with a per-pass Z box and a Perimeter override checkbox. If the active colour is linked to a saved profile, editing it here rewrites that profile in place. (2.4.0) The three zones are edited one after another in a single panel, top to bottom in printing order — the old Top Surface / Bottom Surface switch is gone. The Recipe | Result preview sits at the top of the panel, and the (TD) grid has moved out to the Object & TD department. (2.4.0) Right-click any pass's preview bar for Duplicate pass, Move up / Move down and Delete pass — the clone splits the original's thickness in half, so nothing else in the stack moves. Under each zone a copy to: row copies the whole zone onto another one (Top → Penultimate / Bottom, and back). The ColorStitch pattern is translated to the destination zone's keys on the way, and a stack landing on Bottom is normalised to the Bottom rules (max 2 Solid, max 1 ColorStitch, PathBlend forced to Full).
- Save — promote the active recipe into the saved Profiles library.
- Duplicate (2.4.0) — make an independent copy of the active colour and open it in Pro. Because editing in Pro rewrites the linked profile in place (everywhere it is already painted), this is how you make a variant without touching the original.
- Profiles — saved palettes; click one to load and paint with it. Right-click a swatch for Duplicate / Save to palette / Delete.
- In use on this object (2.4.0) — every paint slot this object is spending: its colour, its name, how many facets it covers, and two actions — Use (make it the active colour) and Free (erase that colour from this object and release the slot).
- Brush & view (2.4.0) — Smart fill angle and the section-view clipping slider. These live outside the department tabs now, because the brush keeps working whichever tab is open; before, you had to go back to Palette to adjust them.
- Erase all painting — the coral button in the tool row. It now asks for confirmation and tells you how many objects it will clear (it wipes every chosen object, not just the active one).
How to paint: pick a swatch or compose one in Pro mode (it becomes the active colour) → click the surface to paint. Use Pick to grab a colour already on the model.
(2.4.0) Fluidity. Dragging a divider or a number in Pro used to re-schedule a slice on every frame of the drag. Now the heavy work is committed once, when you let go — the same rule the (TD) sliders already followed. The Recipe/Result previews still update live while you drag.
The four departments
Walkthrough — painting a multi-pass Sandwich, step by step
The panel chrome in these seven shots predates the 2.4.0 reorganisation (they still show the Top Surface / Bottom Surface switch and (TD) inside Pro). The workflow is unchanged — for the current layout see the four department shots just above.
Layout and pass rows (2.4.0). The tool row (Select · Paint · Eraser · Eyedropper · Sticker · ? · Erase all) sits on its own line under the active-colour header, instead of sharing it — the panel had grown too wide. Each zone is titled by a coloured chip (green Top, darker green Penultimate, orange Bottom — the same codes the 3D highlight uses, see §6h), because three plain labels did not read as three different things. In every pass row:
- The thickness bar on the left is twice as wide, so its millimetre value no longer collides with the pass number, and its drag handles between passes are easier to grab (they light up teal under the cursor and stay clear of each other on thin bands).
^vreorder andxremove close the row on the right, with thexset apart: reordering passes (raise the Solid, lower the ColorStitch, swap them) is a core move when composing a recipe and it used to be hidden in a right-click menu, while thexsat next to the thickness control — one adjusts, the other destroys.- A
!CSor!PBbutton appears on a Solid pass that still carries a leftover ColorStitch or PathBlend payload — a pass degraded by an older build. The engine slices it as Solid (the kind wins) but the recipe looks like an effect, so the preview came out flat. Click it to restore the pass. New ones cannot be created: switching a pass to Solid now clears its payload.
Working vs saved. Colours you paint are working colours — created on demand, deduplicated, and garbage-collected when no face uses them (shown with an amber border while occupying a slot). Browsing palettes does not consume slots. Saved palettes are deliberate, named, and travel in the 3MF.
Palette groups — saved palettes are organised into groups (up to 10, global). Use + New group to add one and the Group selector to switch; deleting a group moves its colours to Group 1.
Slot cap — up to 254 painted slots per object (slot 0 = unpainted).
Save all — promotes every unsaved working colour into the active palette group at once, so a later Erase all leaves nothing dangling.
Painting and the slot→profile mapping are recorded for undo/redo within the session.
Save keeps the Bottom zone (2.4.0, bug fix). A colour's recipe is carried as Top + Penultimate, and Save built the saved palette from those two only — a recipe with a Bottom zone was saved without it, silently, in the very gesture meant to preserve your work. The Bottom now travels with the colour, and it also counts when Save looks for an identical existing palette: two recipes that match on top and differ underneath are different colours, and collapsing them was how twin profiles appeared, one of them carrying the Bottom and the other not.
The list no longer reshuffles under the cursor (2.4.0, bug fix). Working colours are listed while they occupy a slot, and that was checked against the active object — which changes on simple hover in Paint/Eyedropper mode. The grid reordered itself as you moved the mouse across the plate, and with it whatever Save appeared to do. It now counts the active object and every chosen one, so the list stays put until you change your selection. And if Save promotes a colour that lives in another group, the view now jumps to that group instead of leaving you looking at an unchanged grid — the colour is filed where it belongs, not moved behind your back.
| Situation | What applies |
|---|---|
| Object with zero painted facets | Preset mode — Sandwich Editor values apply |
| Painted object — painted area | The painted profile applies; preset ignored for that area |
| Painted object — unpainted area | No effect (preset suppressed for the whole object) |
This "all or nothing" rule prevents mixing preset and painted effects. For each top/penu fill at each layer the slicer uses the dominant painted slot in that Z range, so different regions at the same height each get their own effect. The wipe-tower planner uses the same lookup as the slice, so plan and G-code stay in sync.
Everything is saved inside the 3MF: the profile library (project-level base64 JSON), the per-volume slot tables (slot → profile id), and the per-triangle paint (mirroring MMU painting). Opening a 3MF restores all three. Profiles live inside the 3MF only — there is no cross-project library; save a template 3MF to reuse a set.
Known limitation — PathBlend on Penu. The penultimate PathBlend has a gradient-direction bug on multi-stair objects, so the PathBlend pass is restricted to the Top zone for now. The engine supports penu PathBlend; it will be re-enabled when the bug is fixed.
Painted top surfaces show the ColorStitch weave directly on the model — the per-line tool stripes (or dither / gradient / hard bands) instead of a flat swatch colour. The preview is built from the same per-line sequence the slicer produces (build_dithered_tools_* / build_custom_bands / pattern), so the filament colours, density and pattern match the G-code. The same sequence builder also drives the small pass strip in the Pro tray and the Sandwich editor, so the strips and the 3D view stay identical. (The preview is always on now; the old Preview weave toggle was retired.)
Scale fits the painted area at the real line width. The stripe pitch comes from the resolved top line width (config, no slice needed), and the gradient/pattern is scaled to the painted region's own extent — computed per island: each flat zone (e.g. a stair step) is detected as a connected component (edge-adjacency, so zones that only touch at a corner stay separate) and gets its own gradient ramp, just like the slice. Tiled patterns repeat at the real line width (shader wrap), so the stripe width matches the print regardless of zone size.
Orientation matches the slice for a fixed angle. The stripes run at the pass's fixed angle, and that angle is now honoured exactly in the G-code: for a fixed ColorStitch angle the slicer's per-layer fill-angle rotation is locked out (internally via the template-angle flag), so every layer keeps the painted orientation. Set the angle by scrolling the mouse wheel over the pass preview bar (Pro tray and Sandwich editor) — the bar, the 3D model and the print all rotate together in real time.
Auto angle (
-1). With auto angle the slicer alternates the fill direction every layer (this is what gives a uniform finish), so a static preview cannot match the print. An amber "auto angle" tag appears next to ADV in that case — set a fixed angle (wheel over the bar) to lock the orientation.
The Bottom zone is previewed too (2.4.0). Until now every on-model preview read the Top recipe only, so a colour whose Bottom zone carried its own ColorStitch or PathBlend showed up flat — or worse, wearing the Top colour. Now each painted slot is previewed per zone: upward-facing facets show the Top recipe, downward-facing facets show the Bottom one, each with its own islands and its own scale. A Bottom made only of Solid passes has no weave to draw, so it gets its own composed colour instead of borrowing the Top's.
Where you can paint. Painting follows the zones the colour actually uses: a recipe with Top/Penultimate content keeps upward-facing facets, one with Bottom content keeps downward-facing ones, and a recipe with both keeps both. Side walls are never painted — that is where the effect would not print anyway.
The preview no longer flickers when you change object (2.4.0). Chosen objects that are not the active one are drawn with their own weave now; previously they fell back to a flat composed colour, so sweeping the cursor across a plate made patterns blink in and out.
Remaining limitations (this version). The stripe scale uses the painted-area projected extent, not the exact line count after perimeters/gap-fill are subtracted, so it can differ by a line or two. Islands wider than ~64 lines coarsen in the preview (64-entry shader LUT) — gradients just lower resolution; patterns still tile at real width.
A MixedFilament (Filament Settings → the MixedFilament rows built from two of your loaded filaments) can be assigned to a whole object as its extruder, the same way you'd assign any normal filament. MixedFilament Object mode is a one-click way to make that object's top surface, penultimate layer and bottom surface actually look like that MixedFilament's colour, instead of printing with whatever the default treatment would be.
2.4.0 — the Bottom zone is new. Before, the mode replaced Top and Penultimate but left Bottom resolving to whatever painted recipe was underneath, so an object could print under two different recipes at once while the interface said it was fully governed. Fixed.
How to use it: open the ColorStitch Painter gizmo (§6b) on an object whose extruder is a MixedFilament. A new checkbox — "MixedFilament Object" — appears above the palette strips, with a small colour swatch next to it showing the approximated result.
- If the object's extruder is not a MixedFilament, the checkbox is greyed out with a tooltip telling you to assign one first.
- Turning it on:
- Auto-generates a small sandwich (up to 3 solid passes) that approximates the MixedFilament's colour using your other loaded filaments and their TD values (§1e) — the same colour-matching math the ColorStitch Studio uses.
- Turns Perimeter override on automatically, so the walls get reprinted to match too.
- Locks out manual painting/patterns for that object (the palette strips, zone editors and the Perimeter override checkbox grey out) — the object is either "painted by hand" or "driven by its MixedFilament," not both at once. (2.4.0) The lock now covers the brush in the 3D view as well, not just the panel. Before, the controls greyed out but clicking the model still painted: slots were spent and re-slices scheduled for paint the engine then ignored — work lost with no warning. Placing a sticker on such an object is blocked for the same reason. Select and the eyedropper keep working: changing object or reading a recipe are still useful.
- Turning it off restores whatever was painted before (nothing is lost).
Beta. This feature is functional and print-verified in principle, but still young — report anything that looks off. One known rough edge: the swatch shows the colour only, not a preview of the pattern/passes that will actually print.
The panel always knew which colour was active; the 3D view never said where that colour is already applied. With two similar colours on one plate that question had no answer other than squinting at the mesh. Highlight active colour (checkbox at the bottom of the painter, next to the brush and section-view controls) answers it.
What it draws. The outline of the painted region for one slot — the boundary edges only, so it frames the area without covering the weave preview you are looking at — plus a faint box around each painted island for reading at a distance, and a badge carrying the slot number. The outline is drawn twice: solid on the surface, and as a ghost through the object, so a zone facing away from you still shows without orbiting blind. A slow pulse keeps it apart from the paint itself.
Zone colours, the same ones as the panel. The zone chips in the Pro tray and the highlight share one palette: green = Top (the Penultimate is a darker green — it is the layer under the same top surface, not a separate thing), orange = Bottom. So a recipe that paints both zones is obvious at a glance: green outline above, orange below. Each badge shows the slot number, a disc in the colour of the slot (what colour it is), and a ring plus a wedge in the zone colour (where it is applied) — the wedge points up for a top island, down for a bottom one, and bottom badges sit under their island.
Which slot is highlighted. The active colour by default, so while you paint you always see where that colour already is. Hovering a swatch in the palette grid, or a row of In use on this object, highlights that colour instead — "show me where this one is". Move the cursor off the panel and it returns to the active colour.
The counter next to the checkbox reads sN — 137 (slot and facet count) or sN — not painted here. That distinction matters: "the colour is active and I see nothing" has two very different
causes — it is not painted on this object, or it is painted on a face pointing away from you.
It is an aid, not a preview of the result — turn it off to check the clean weave preview.
Notes. On a multi-instance object only the first instance is highlighted. A bottom outline can be hidden by the build plate, since the plate is drawn after the objects — that is what the badge under the island is for.
Assembled objects now show their paint (2.4.0, bug fix). On an object made of several parts (what Assemble produces) painting a colour that already had a slot recorded the facets but not the colour itself on the other parts: the part sliced correctly while the painter showed it grey, the eyedropper read no recipe there, and the highlight had nothing to light up. Paint slots are per part, and the profile is what identifies a colour across parts — that is now respected everywhere (painting, eyedropper, highlight, preview). Objects already in this state are repaired when the object is opened in the painter: orphan paint recovers the colour from the sibling part that still had it.
Until now a painted Sandwich only existed while the ColorStitch Painter was open. Close the gizmo and the object went back to one flat colour: nothing on the plate told you which parts were painted, with what, or how the effect would land — you had to reopen the painter, object by object, to find out. Plate thumbnails had the same blind spot.
Painted objects are now drawn painted in the normal 3D view, gizmo closed, with the full weave — the same per-line stripes, dithers, hard bands and PathBlend gradients you see inside the painter, per island and at the real line width, composed against the object's actual base colour and its TD. It is the same code doing the drawing in both places, so there are no two versions of the truth to drift apart.
What is identical to the painter, and what is not. The colours and the pattern are the same calculation. Two differences worth knowing:
- Lighting is not the same. Inside the gizmo the model is drawn by the painter's own shader; outside it is the normal one — or Libre Mode's realistic shading with its ambient occlusion and shadows. Same bands, different light on top of them.
- Small leftover facets print flat here. Inside the painter, faces too small to form an island fall back to a whole-object weave. Outside they take the slot's flat composed colour. In practice this is stray fragments between zones.
- With the object selected, the unpainted part carries the selection tint and the painted zones do not — the same behaviour MMU painting has always had.
Turning it off. A checkbox, Keep paint visible outside this gizmo, sits with the other view aids at the bottom of the painter (next to Highlight active colour). It is on by default and applies to the whole project.
With MMU painting on the same object (2.4.0). Both are drawn, each on its own faces: the Sandwich first, the MMU on top of the faces it owns. That is now exactly what the slicer does — where you painted MMU, MMU rules; everywhere else the Sandwich applies. What you see is what prints.
In the MMU area you get no Sandwich effect: that surface prints plain, in its own filament. Both painters tell you how much of your paint is affected, in amber, when it actually happens.
The MMU painter draws your Sandwich painting too, with the full weave — the same way the normal 3D view does — so you can decide where MMU paint goes without working blind. Where both meet, MMU is drawn on top, matching the slicer.
A checkbox, Show Sandwich effects, appears in the MMU panel on objects that carry Sandwich painting. On by default, project-wide.
Known limitation. Where a surface is split — by MMU paint, or anything else — each piece restarts its gradient instead of continuing it. The line spacing stays continuous across the boundary; the colours restart. Cross a ColorStitch or PathBlend zone with MMU paint and the pattern begins again on the far side.
Load an SVG in the Palette tab and the Sticker tool places it on a flat top face. It is not geometry: the shape becomes a 2D mask carrying a Sandwich recipe, resolved at slice time. Stack several and the topmost one wins where they overlap — it occludes, it does not blend.
It works and it slices. But it is the least finished part of the painter, and you should know what you are getting before you build anything around it:
- You cannot see a sticker anywhere. Not in the normal 3D view, not inside the MMU painter, and not even inside the Sandwich painter itself unless you are actively editing that sticker. You place it, it disappears, and you find out what it did in the preview or the G-code.
- A sticker overrides everything under it, silently — hand-painted zones and MMU paint. It is applied last, over whatever survived, so it takes the surface back from both. Nothing warns you.
- Top surfaces only. Bottom stickers were never implemented.
- The wipe tower does not currently switch between a painted tower and a sticker one.
All of this is being addressed together in a Mask-Painting section planned for a future release, where stickers and a drawing tool become one thing. Until then, treat stickers as experimental.
Align & Stack is a gizmo (left-side gizmo toolbar, "Align & Stack") for placing one object against another — #1, the object clicked first, is the anchor; #2 is the one that moves. Click two objects in the scene to set them (click a third to swap out #2 — the gizmo only ever relates two objects, so there's no chain to manage); click a chip or Reset to clear.
Two modes:
| Mode | What it does |
|---|---|
| Place against | #2 comes to rest touching the chosen face of #1. This is the stacking mode. |
| Align flush | #2's same-side face becomes coplanar with #1's (Illustrator-style alignment). |
A row of face / centre buttons picks which face or centre axis to align or stack against (the tooltip changes with the mode). Z gap (mm) sets a controllable gap between stacked objects, and Drop to bed (Z = 0) drops every ordered object back onto the plate.
Viewport AABB aid — see the landing spot before you click it. With #1 picked, its bounding box is drawn as a wireframe "zone" directly in the 3D view. Once #2 is picked too, every possible placement (all 5 face ops + the 3 centering ops) is previewed live as a translucent ghost wireframe of #2 at exactly the position it would land — computed with the same math the click would run, so the preview never lies. Each ghost carries its own big, semi-transparent mini-cube icon floating right at the seam between #1 and the ghost: hover it for a tooltip, click it to run that placement immediately, no need to go back to the side panel or decode which abstract X‑/X+ button means "behind". Ghost previews only show for a #1+#2 pair currently on-screen; from a camera angle where a given ghost isn't really visible, its icon simply doesn't appear rather than showing up somewhere meaningless.
Works together with Libre Mode (§4) for floating/assembled workflows — align or stack the parts, then slice with the arrangement you need. Pair it with True Objects / Gravity (§17) when stacking separate (non-Assembled) objects — Align & Stack places the pieces exactly, Gravity is what makes the slicer treat the touching face honestly instead of as a false bridge.
NeoArachne is a wall-generation engine that sits beside the stock Classic and Arachne engines and lets you choose, per feature (outer wall, inner walls, gap fill), which underlying engine prints it — plus extra controls targeting Arachne's known failure modes (width breathing, blobs at thin transitions). The default recipe — Neotko Hybrid v2 — uses Classic for the outer wall and stock Arachne for the inner walls. It ships a Preview Lab (§8d) that renders wall paths before slicing.
NeoArachne is exposed only under Libre Mode (§4): the NeoArachne entry in the wall-generator dropdown, and its controls, appear only when Libre Mode is active. It is opt-in — existing presets are unaffected.
- Enable Libre Mode (§4a).
- Quality → Wall generator (Advanced) now offers Classic / Arachne / NeoArachne.
- Select NeoArachne — its controls appear inline below the dropdown. Leave them at defaults for a first slice (the Neotko Hybrid v2 recipe).
Three source dropdowns route each wall kind through an engine:
| Setting | Options |
|---|---|
| NA — outer wall source | Classic / Arachne (stock) / Arachne (NeotkoEdge) |
| NA — inner walls source | Classic / Arachne (stock) / Arachne (NeotkoEdge) |
| NA — gap-fill source | Off / Classic / Arachne (stock) / Arachne (NeotkoEdge) |
- Classic — constant width, cleanest visible surface, no breathing; can't adapt to thin features without a gap-fill pass.
- Arachne (stock) — variable-width beading with integrated gap fill. Best for inner walls.
- Arachne (NeotkoEdge) — Arachne with the NeotkoEdge bead-count stabiliser. (Its extra tuning knobs from the older fork are not exposed in this build yet.)
The default Neotko Hybrid v2 = outer Classic, inner Arachne (stock), gap fill Off.
| Control | Range / default | What it does |
|---|---|---|
| NA — allowed perimeter overlap | 0–100%, default 0% | How much the first inner Arachne bead may overlap the Classic outer. Raise to 5–15% only if you see a seam; high values blob. |
| Min Line Width | 5–100% of nozzle, default 40% | Minimum bead width Arachne emits. Higher widens thin features (blob risk). 30–50% recommended. |
| Max Line Width | 100–200% of nozzle, default 200% | Ceiling on bead width. Lower (130–150%) for more, narrower beads. |
| Min Feature Threshold | 1–100% of nozzle, default 10% | Geometry thinner than this is discarded. Keep low; must stay ≤ Min Line Width. |
| Preserve Thin Edges | on/off, default on | Keeps short closure tails near the outer perimeter for cleaner seams. |
A panel inside the NeoArachne section that renders planned wall paths before slicing: outer + inner paths in distinct colours, the execution (chain) order, seam dots, travel moves, and a head animation. Controls include a layer slider, animation speed, a ghost/printed build mode, zoom, "use selection," and a Dump button that exports the full plan as JSON for off-line diagnosis.
Known limitation: a small visual divergence vs the real slice on the second-to-last inner wall in narrow "waist" regions of some geometries — use Dump to share the JSON (and the 3MF) for diagnosis.
NeoTower is an alternative wipe-tower planner. The stock planner decides geometry up front; NeoTower runs after slicing, when every toolchange is known — including the sub-layer primes that Sandwiches and Solid-pass stacks insert inside a layer. Seeing the full toolchange list first lets it build a fixed, predictable footprint that stays in sync with the real G-code and understands variable layer heights. Options live in Quality → Prime tower.
You usually don't have to choose it. Any scene that uses a Sandwich or a multi-pass stack auto-promotes to NeoTower regardless of the setting below. NeoTower requires a prime tower to be active.
| Setting | Behaviour |
|---|---|
| Classic (default) | The standard WipeTower2 planner. |
| NeoTower | The post-slice planner — fixed footprint, delta-Z aware, with the options below. |
Default: on. Limits how fast the footprint may shrink between consecutive real layers (one perimeter width per side) so every wall ring rests on the ring below (wall-on-wall). Disable to save material and time, at the cost of rings only partially supported by the sparse interior grid.
Sandwich / multi-pass purges run at microscopic heights and would inflate the tower. Purge compaction is a flow-boost cap that compacts those thin purges into a narrower band (the surplus hangs into the hollow interior), reducing the footprint.
- 1.0 = off. Higher = more aggressive. Default 1.7. Range 1.0–5.0. Confirm with a test print.
The volume (mm³) purged before each Sandwich sub-layer toolchange (Solid / ColorStitch / PathBlend, Top + Penultimate). The config key is multipass_prime_volume, labelled "Sandwich wipe reserve".
- Default 10 mm³. Lower = thinner/shorter tower; higher = better purge. Set 0 to disable. Requires a wipe tower active.
Because NeoTower plans from the real post-slice toolchange list and is delta-Z aware, it is the mechanism that lets adaptive / variable layer height + multiple tools + a Sandwich coexist on one coherent tower. Stock Orca refuses to slice such scenes; NeoTower can purge each toolchange at the real per-layer height.
A new option Variable layer height (Experimental) sits under Tower type and is exposed only with Tower type = NeoTower and Libre Mode active (it is visible but greyed-out otherwise). Default: off. When on, the slicer stops blocking:
- scenes that mix objects with different layer heights, and
- adaptive / variable layer height combined with more than one filament.
(Experimental.) The wipe-tower issue that previously left empty/short tower layers (missing "drawers" on real layers, including the "empty first layer" abort) is fixed in 2.3.1 — the tower now stays coherent under variable layer height. The capability is proven and was solid on the 2.2 line (older FS099 fork), and is now consistent on this 2.3.4 base too. It is still flagged Experimental: review G-code before long multi-tool runs. The option only takes effect with Tower type = NeoTower.
The Bump Mapping Editor is one gizmo (GLGizmoTextureBump) with three modes, switched via a
bar at the top of its panel: All, Painter, and Top. All three turn a grayscale (or
any) PNG into physical Z relief at slice time — the difference is where the relief goes and
how it's scoped.
Hidden behind a double gate in this build — expert-only, on purpose. The gizmo only appears in the toolbar with both (1) Libre Mode active (§4a: master switch in Preferences + restart, then the toolbar toggle) and (2) the debug env var
ORCA_DEBUG_TEXTUREBUMPset before launch; Top mode additionally needsORCA_DEBUG_ZBUMP(orORCA_DEBUG_ALL=1) for ZBump to actually apply when slicing. This isn't a placeholder gate — some option combinations here are genuinely capable of producing bad extrusion (see the Classic-wall-generator note in §10a and the per-mode limitations below), so it's kept off unless you've deliberately opted in and are prepared to read the resulting G-code. Full per-OS activation steps and the safety notes that go with them are inNEOTKOCM_RELEASE_2_35.md.
Applies one texture to every wall of the selected object, projected via Planar / Cylindrical / Spherical / Cubic (Quality → wall texture settings, or the panel's Source/Relief/Transform sections). Three on-canvas 3D handles adjust it without leaving the viewport:
| Handle | Colour | Drag controls |
|---|---|---|
| V | blue | Scale (tile size, mm) |
| U | orange | Repeat count |
| Yaw ring | purple | Rotation around the pivot |
| Pivot | pink | Pan — moves the tile origin |
A live textured overlay shows the pattern on the object before slicing, so you can line it up by eye. The real object dims while the gizmo is open so the projection isn't hidden behind it.
⚠️ Wall generator: Arachne only. Wall-texture bump (All + Painter modes) requires the Arachne wall generator. It's deliberately disabled on Classic — a real test print showed Classic silently over-extrudes when adjacent walls carry different bump amounts, and the preview hides it while the G-code doesn't. NeoArachne isn't wired up for it yet either. Even on Arachne, adjacent walls with very different bump amounts can still show a visible gap between them (nothing adjusts line width to compensate, only the centreline moves) — this is a known, open limitation, not something the gate gets you past.
Paint zones on the model (same brush/erase interaction as the other painter gizmos) and assign each zone its own texture, projection mode, scale, and thickness — independent of the object's base (All-mode) settings and of every other zone. Each zone gets its own overlay preview and a thumbnail in the zone list so you can tell them apart at a glance.
A different feature entirely (no shared engine code with All/Painter's wall texture) that modulates the Z height of the top-surface fill, not wall XY. A grayscale image becomes a literal height field: brighter pixels raise the top surface, sampled point-by-point along every top-fill line for a real 2D relief (not one flat Z per line).
| Control | What it does |
|---|---|
| Height (mm) | Max Z displacement at full-white texture value. Shown in red past a safe-height estimate (0.8× nozzle − layer height) — a warning, not a hard cap; you can go past it and judge from the resulting G-code. |
| Reinforcement passes | Splits the total height across multiple stacked passes when it exceeds what one pass can safely reach on its own. |
| Scale / Repeat | Physical size of one image tile, and how many times it repeats. |
| Pan X / Y | Shifts the tiling phase — drag the on-canvas green handle, or type exact values. |
| Edge ramp (mm) | Smoothstep margin from the top-fill's own contour, so the wall itself stays flat. |
The green pan handle rests at the object's own center plus the current offset, and a live textured overlay (cropped to roughly where top-fill actually starts, inset from the object's outer edge by its perimeter walls) previews the pattern before slicing — drag the handle or the numeric fields, both update the same preview live.
Preview vs. slice. The overlay is a GUI-only approximation (perimeter inset is estimated, not the engine's exact fill boundary) — always meant as a placement guide, not a pixel-perfect match. It's cross-checked against the real slicing math via a calibration log (
ORCA_DEBUG_ZBUMP,/tmp/neotko_zbump.log) that logs identical probe points from both the GUI and the engine side for direct comparison.
Known limitation: the on-canvas scale/rotation handles (in any mode of this gizmo) can disappear when the print bed renders behind them, depending on camera angle — visible fine from below. Long-standing, not specific to any one mode; not yet root-caused.
A gizmo (left-side toolbar) that replaces the stock "Layers editing" brush with an exact, point-based curve editor for variable layer height. Instead of clicking and dragging to add/remove detail (imprecise — you can't say "I want exactly 0.28 mm at Z=14.2 mm"), you place control points at an exact Z and height, drag them, and set a tension per segment to shape the curve between them.
Like Align & Stack (§7), this gizmo is gated behind Libre Mode (§4) — its icon is always visible in the toolbar but stays disabled (greyed out) until Libre Mode is active. It never affects the stock Variable Layer Height dialog/brush, which remains untouched and fully usable with Libre Mode off.
How to use it
- Enable Libre Mode (§4a), select a single object, open the gizmo.
- The panel shows a graph: height (mm) across, Z across the object's height vertically.
- The bottom point (grey) is the object's fixed first layer — it can't be moved or deleted.
- The top point (amber) can be dragged in height only.
- Click anywhere else in the graph to add a new point; drag any non-bottom point to move it; right-click a point to delete it.
- With 3+ points, a tension slider appears per segment below the graph:
0= straight line (identical to the engine's default linear interpolation),1= a smooth curve through the segment (monotone — it never overshoots past the two points' heights, so it can't spike outsidemin/max layer height). - While dragging or hovering a point, a translucent band lights up on the object itself, showing exactly which Z-slice you're affecting — teal on hover, amber while dragging. A small label over the point shows the exact Z and layer height live.
- Every discrete edit (point add/move/delete, tension change, Reset) is a normal undo step and triggers a re-slice, same as the stock brush.
Current limitations (first version)
- Min/max layer height are read-only (from the printer/nozzle) — no per-object override yet.
- Reopening the gizmo on an object that already has a very dense profile (e.g. one painted with the old stock brush) falls back to a flat 2-point start rather than importing hundreds of points as control points.
- No result preview from the classic Adaptive/Smooth buttons — this gizmo is a separate, precise path, not a replacement for those.
Layer height and MixedFilament color are not independent: the height you slice at changes how the color reads on the part. Turn on the Adapt to Color checkbox and the height editor shades the ranges where the object's color setup actually works:
- Red on the right — pattern-resolution ceiling. Above the mix band (follows the "Dithering cadence" upper bound in your print settings), a Cycle/gradient pattern gets too coarse to read as a blend. Applies object-wide to any object with a mixed pattern, paint or Sandwich.
- Red on the left — color-fidelity floor (Sandwich zones only). Where the mesh has a plausible Sandwich top surface, the filaments' TD (translucency, from the ColorStitch settings) sets a minimum thickness: thinner than this and a translucent pass washes out over what's below. This never applies outside real Sandwich zones — a plain Cycle object shows only the ceiling.
- Green line — optimal height per Z, and a Snap to optimal button that rewrites every editable point of your curve to it in one click.
Dragging stays completely free within the nozzle limits — the guidance is visual (a point sitting in a red zone turns orange), and the emitted profile is hard-clamped to the safe range only at commit, so what reaches the slicer always respects the color. The info panel shows the color-safe range for whichever point you hover or drag.
The problem (no stock slicer addresses this): on a sloped surface, each layer's contour steps
inward, and that staircase ledge exposes the interior perimeter rings to view. Those rings
print in whatever the layer's color happens to be, so a clean MixedFilament banding degrades into
noise exactly where the model curves. The wider the step (d = layer height × tan(slope)), the
more rings show — at 0.2 mm on a steep slope you're already looking at one or two interior rings.
What it does: with the Slope recolor checkbox on, the gizmo scans the mesh for slope bands, computes at your committed layer heights how many interior rings each band exposes, and stores a per-object recolor plan (in the project 3mf, undoable, erased when you untick). At slice time the engine applies it:
- The external perimeter is never touched — its per-layer alternation IS the pattern's visible rhythm and it keeps printing exactly as your recipe dictates.
- The exposed interior rings print a side-by-side combination of the recipe's own components, chosen by ΔE2000 color distance so the ledge's blended appearance matches the recipe's intended mix color — the step fills with the mix instead of a random solid.
Works with Cycle, gradients and manual patterns. The violet shading in the height editor shows which heights expose rings at each Z (drag the curve below the violet edge and thin layers cover the slope instead — the two strategies are complementary), and the info panel reports the exposed ring count plus the suggested ring colors for the focused point.
Interaction with adaptive height: you now have both answers to the same geometry problem — fine layers shrink the ledge until nothing extra shows (slow, maximum quality), Slope recolor accepts tall layers and colors what shows (fast, great finish at 0.12–0.2 mm). Use the violet shading to choose per zone.
Current limitations (WIP — needs broad print testing)
- Best on surfaces with one dominant slope. Two very different slopes sharing the same height range currently share one plan (the steeper wins) — refinement planned.
- The plan is per height-band, not per-region: on a model that is sloped on one side and vertical on the other at the same Z, the vertical side's interior rings recolor too (invisible there, but it costs tool changes).
- A "Sandwich + slope" zone (top surface on a slope) still prioritizes the thick-top Sandwich — spreading the recipe across several thin top layers ("Sandwich 2.0") is a planned future system; the panel warns when you're in one of these zones.
(WIP, paused mid-implementation — Libre Mode only.) The support roof and hollow pillar described below are implemented and print-validated. The second mechanism — a contact-layer toggle (§12a) that ripples the object's own bridge fill directly above the roof — is now shipped and slice-verified, but has not yet been through a real print (its first print is still ahead). Both the support engine and the contact layer are experimental; expect this section to expand in a later release.
NeoWave is a support type that replaces the interface/roof fill with a wave-front pattern — long, continuous paths that diffract around concavities (ported from a published wave-overhang algorithm), instead of straight parallel lines. It optionally hollows out the support body itself (perimeter-only, no infill) underneath that roof, trading material and print time for a support that still closes cleanly on top.
Exposed only under Libre Mode (§4): the NeoWave entry in Support type, and the Wave roof controls below, appear only when Libre Mode is active.
Turning it on
- Enable Libre Mode (§4a).
- Support → Support type: select NeoWave. Selecting NeoWave now locks the support to its tested shape automatically — Base pattern → Hollow and Interface pattern → Wave (NeoWave roof) are set for you and greyed out, since the other base/interface patterns don't apply to NeoWave and only added clutter. Switch back to a normal support type and the full choices return. (Previously you had to set both by hand.)
Wave roof controls (appear once Interface pattern = Wave)
| Control | Options / default | What it does |
|---|---|---|
| Wave roof shape | Concentric / Wave | Concentric collapses nested rings inward from the roof boundary. Wave sweeps open arcs across the region, diffracting around concavities. Print-tested: Wave is the one that works — Concentric fails to close cleanly over a hollow pillar (see below). |
| Wave roof order | Smart / ZigZag / Monotonic | Print order of the same fronts — the drawn shape doesn't change, only how it's traversed. Smart hooks each front onto whatever is already printed (fewest long travels); ZigZag chains fronts into one continuous path; Monotonic keeps every front separate. |
| Reverse wave roof order | on/off, default off | Flips which edge (or, for Concentric, which ring) the fill starts from. |
| Hollow pillar walls | 0–10, default 0 | Number of perimeter walls for the support body, printed with no infill — a hollow pillar. 0 keeps the normal solid body. The first layer always stays solid for bed adhesion. |
🖨️ Print result so far: a hollow pillar (Hollow pillar walls ≥ 1) capped with a Wave roof of 2 interface layers printed cleanly. The same pillar capped with a Concentric roof did not close properly over the hollow — use Wave, which is already the default shape.
A separate, independent toggle that ripples the Z of the object's own bridge fill wherever it lands directly above a support roof — valleys touch the roof, crests stay in the air, leaving intentional microscopic contact gaps meant to reduce bonding force so the support separates more easily. Unlike the original idea (a paintable Sandwich effect), this shipped as a plain on/off toggle under Support, independent of the Sandwich/ColorStitch system entirely: it works whether or not the object has any painting on it, and it never touches colour/pattern — only Z.
| Control | Options / default | What it does |
|---|---|---|
| Support neoweave contact | on/off, default off | Enables the wave ripple on bridge fill directly above a support roof. |
| Contact wave amplitude | mm, default 0.1 | How far the wave deflects (upward only — it never digs into the roof below, by construction). |
| Contact wave period | mm, default 0.6 | Distance between successive wave peaks along the fill line. |
⚠️ Slice-verified, print-pending. The toggle is confirmed to fire correctly wherever bridge fill sits over a support roof, and the resulting G-code looks correct — but this specific mechanism has never been through a real print. Orca's G-code preview also doesn't render the Z variation this produces (same known limitation as Neoweave top/penu and ZBump) — the G-code is right even though the on-screen preview looks flat.
Known gaps (this build)
- No dedicated key yet for roof layer count — it uses the existing Interface top layers setting.
- No dedicated UI panel/gizmo yet; all controls above live as plain fields under Support.
Not the same gizmo as §6. This section is about Orca's own stock multi-material painter — the Color Painting tool in the left-side gizmo toolbar (needs 2+ filaments configured; it's what you'd use to hand-assign filaments to triangles in any Orca build). It is a different tool from this pack's own ColorStitch Painter (§6), which paints Sandwich effect profiles, not raw filament assignment. Pro Mode is a collapsible section added to the bottom of the stock Color Painting panel with four precision add-ons on top of the regular brush. It is always available — no Libre Mode needed.
The stock brush paints by hand with a circle/sphere cursor, which is naturally imprecise on small or fine details — a click can bleed well past where you meant to paint. Pro Mode's tools attack that problem from different angles: finer brush subdivision, limiting paint to a thin perimeter ring instead of filling solid, two "mask" tools that paint an exact area in one shot instead of brushing it by hand, and (2.3.8) Surface depth — projecting a painted top/bottom design into the object as solid material, with optional per-color control.
A Precision slider (1×–8×) in Pro Mode subdivides the mesh more finely right where you're painting, so the edge of a brush stroke follows the surface more closely instead of stair-stepping at low mesh resolution. Higher values cost more memory/CPU on dense meshes. Default 1× = identical to a build without Pro Mode.
Two checkboxes/fields that change where a painted colour actually gets used, without touching the brush itself:
| Control | What it does |
|---|---|
| Paint perimeters only | Limits the painted colour to a ring near the painted region's contour (about wall-count × line-width wide) instead of filling the whole painted area solid. The interior reverts to the object's base colour. Reduces wasted filament and colour changes buried in solid infill where nobody will ever see them. |
| Extra walls | Adds this many extra perimeter walls to the painted region only, on top of the object's normal wall count — for a painted logo or trim that needs a bit more wall depth than the rest of the print. 0 = disabled. |
Turning on Extra walls automatically widens the Paint perimeters only ring to make room for them — if you raise Extra walls after already enabling the ring, the ring re-widens to match.
By design: Extra walls has no effect where the painted colour matches the object's own filament. If you paint a region with the same colour the object already prints in, there is no visible colour change there anyway — so Extra Walls is skipped for that region automatically, rather than silently doubling up walls in the same physical spot for no visible benefit. This only matters if you're using the painter to add wall thickness rather than colour — paint with a genuinely different colour (even one you don't intend to keep) to force the extra walls, or use a plain modifier mesh instead for pure geometry changes.
Two alternative "paint" tools for when hand-brushing an exact shape is fiddly — you outline an area on screen and it gets painted in one shot with whichever colour is currently active, instead of brushing it by hand.
| Tool | How to use it |
|---|---|
| Rectangle mask | Enable the checkbox, then click-drag a box anywhere on screen; release to paint everything under it. |
| Polygon mask | Enable the checkbox, then click to place vertices one at a time. Click near the first vertex again (with at least 3 placed) to close the shape and paint it. Click-drag an existing vertex to reposition it before closing. Right-click cancels and clears the in-progress shape. |
Both tools only paint front-facing triangles — the side of the mesh actually facing the camera — so a mask never bleeds through to the back of the object the way a naive screen-space fill would. The two checkboxes are mutually exclusive (turning one on turns the other off), the same way Vertical/Horizontal work elsewhere in this panel. Works with either Classic or Arachne as the wall generator.
Escape doesn't cancel a polygon in progress — right-click does. Keep that in mind if you're used to Escape backing out of in-progress tools elsewhere in Orca.
Paint a design on the top (or bottom) of an object — a logo, a letter, a mark — and Surface depth extends it into the object as solid infill of the painted color, following the painted silhouette exactly, for as many layers as you choose (0–20, 0 = off).
- The painted shape projects straight down (or straight up from a bottom surface), layer after layer, keeping its size — only trimmed where the object's real geometry changes. It is not a cosmetic reclassification: those layers genuinely print as solid walls-to-walls material of the painted color, surrounded by whatever sparse infill the rest of the layer uses.
- Depth is counted in extra layers past the painted surface (the painted surface itself is already solid). Where the projection overlaps areas that were already solid (your normal top shell layers, vertical shells, Sandwich internals), nothing double-counts — the projection only converts sparse infill, and never touches Sandwich's penultimate layers.
- Works symmetrically for bottom-painted surfaces, projecting upward.
- Deep projections can add tool changes on layers that previously had none — same as if you had painted deeper by hand. The wipe tower handles it with its normal machinery.
The Per color checkbox switches "Extra walls" and "Surface depth" from one global value to a per-color table: one row per filament, showing its color swatch plus a Walls (0–8) and a Depth (0–20) field. A value of 0 in the table means "use the global value" — so you can, say, give a silver logo 8 extra walls and 5 layers of depth while a red mark next to it gets 2 and 20, without touching each other.
- Clicking a color swatch in the table also selects that color for painting — same colors, same selection highlight as the filament strip at the top of the panel.
- With the checkbox off, the two global fields behave exactly as before.
- Mixed-filament note: Depth distinguishes mixed slots individually; Extra walls for mixed slots falls back to the global value (mixed paint resolves to its physical components at wall-generation time).
Known edge case (not a Surface depth bug): a MixedFilament blend that ends in the same color as the object's own filament will not generate its lower blend layers either — the whole blend chain is skipped, not just the (correctly redundant) top layer. Suspected stock-pipeline gap. Workaround: don't end the blend on the object's base color.
⚠️ Brand new, work-in-progress, and genuinely UNTESTED — preview-only, never printed. Everything below is confirmed working in the 3D preview only. Treat this section as a curiosity/early-look, not a feature to rely on. One of its own controls (fill speed, see table below) is a confirmed no-op bug — turning it doesn't currently change anything in the resulting G-code.
NeoStitch interlocks consecutive printed layers of a chosen wall without ever moving Z. Along the wall's own path, each layer alternates short notch segments (the wall deflects inward, leaving a gap at the nominal position) and fill segments (the wall over-extrudes there instead to fill the gap left by the layer above/below). The pattern's phase flips on alternating layers, so a fill segment always lands over the notch of the layer beneath it — a vertical "stitch" between layers, mechanically different from a Z-brick-layer interlock (Z itself never moves; only XY position and extrusion width change).
Where to find it: Strength → NeoStitch Interlock (per-region setting, same place/pattern as Fuzzy Skin).
| Control | Default | What it does |
|---|---|---|
| NeoStitch Interlock | Disabled | Which wall gets the interlock: Outermost / Second / Third / Innermost wall. |
| Stitch depth | 0 mm (auto = the wall's own line width) | How far the notch deflects inward. An explicit value overrides the auto. |
| Stitch length | 3.0 mm | The flat plateau length of each notch/fill segment. |
| Ramp length | 1.0 mm | The transition ramp in/out of each segment (also smooths the pressure/flow change). |
| Stitch period | 10.0 mm | Distance between successive notch/fill events along the wall. |
| Stitch flow | 100% | Scales the fill segment's extra extrusion; 100% = automatic volume conservation for the notch depth being filled. |
| Skip bottom layers | 3 | Extra layers to skip above the object's first layer before the interlock starts. |
| Fill speed | 75% | |
| Fill margin | 1.0 mm | Shrinks the fill segment shorter than its matching notch by this much per side, so the over-extrusion lands inside the gap instead of bridging across it. Confirmed working. |
Why "notch"/"fill" instead of moving Z: the whole point is to interlock layers without changing layer height or printing outside the normal Z steps — it's a pure XY-deflection + flow-modulation effect, always fully backward compatible with normal slicing when turned off (Disabled by default, per-region, exactly like Fuzzy Skin).
Known limitations / open items (this build):
- Fill speed control doesn't work yet (see table above) — everything else in this section is otherwise implemented.
- Only verified visually in the 3D preview — the alternating notch/fill pattern is visible and registers correctly from layer to layer, but no real print has been made with it yet.
- Only Arachne wall generation is supported (v1); Classic walls are unaffected by this setting.
- Interacts with Fuzzy Skin / Bump Mapping by mutual exclusion, not composition — if either of those is active on the same wall, NeoStitch skips that wall rather than stacking effects.
- A cascade to neighbouring walls (so an interior wall "follows" the perturbed contour of the wall NeoStitch targets, instead of keeping its original offset) is designed but not implemented.
🧪 Expert-only, Libre Mode only. Edits real G-code. A warning shows once, when you turn the panel on. Print-verified (2.3.8): a real multi-tool print ran clean with by-tool speed/flow/ fan/Z-offset rules active, and the exported G-code was checked afterward line-by-line. "Avoid Wipetower" is separately G-code-verified (2.3.8): checked directly against an exported file to confirm a rule with it enabled never touches wipe-tower purge G-code.
Where to find it: G-code Preview → the view-type dropdown (the same one RealColor lives in) → "Gcode Reprocessor", when Libre Mode is on. It renders inside the same legend panel as every other view mode — no separate floating window.
Master "Reprocessor enabled" checkbox at the top gates the whole panel — when off, none of the
rules below do anything, however many you have. Below it, a GLOBAL / BY TOOL toggle switches
the chart between rules that apply everywhere and rules scoped to one tool; each tool gets its own
column (T0, T1, T2, ... — 0-based, matching the real T<n> G-code command).
The chart is the only editor — there's no separate list of fields to fill in:
- Right-click empty space in the chart to add a rule at that layer (and tool, in BY TOOL view) — a small menu offers the four rule types, color-coded.
- Drag either endpoint of a rule's bar to change its layer range — dragging the top endpoint all the way up snaps it to "to the end of the file."
- Click a rule's colored value badge (in the text summary below the chart) to type its exact number — percent for speed/flow, raw PWM for fan, mm for Z-offset.
- Right-click an existing point for a menu with "Skip WT" / "Don't Skip WT" (see below) and "Delete this rule."
A rule whose layer range no longer exists (e.g. the object got shorter after the rule was created) shows its dot pinned to the chart's edge in gray instead of disappearing off-screen — still fully draggable and deletable from there.
- Speed override rules —
M220 S<1-300%>from layer X to Y (or a tool's active stretch). - Flow override rules (2.3.8, new) —
M221 S<20-200%>, same behaviour as speed. - Fan override rules —
M106 S<0-255>(raw PWM, not percent) from layer X to Y. - Z-offset rules (2.3.8, new) —
SET_GCODE_OFFSET Z=<value>, clamped to ±0.3mm, 0.01mm steps. By-tool Z-offset rules restore right before a toolchange and re-apply right after — timed around the physical tool swap on purpose.
Global rules (the default) affect the whole ranged layers; by-tool rules only affect that range while the picked tool is actually active — the effect automatically reverts when the print switches to another tool, and re-applies when it comes back, all without touching the wipe tower's own toolchange G-code. Mode is per rule, so a global fan rule and a by-tool speed rule can both be active together. No "Apply" button — every edit saves immediately.
"Avoid Wipetower" (2.3.8, new) — any rule can independently opt out of applying inside wipe tower G-code, shown as a permanent gold glow on its chart bar. On machines where the wipe tower fires between layers for drip control, not only at toolchanges, a rule's range can otherwise land inside that purge G-code rather than only "real" printing — this excludes every wipe-tower segment from the rule's active range, splitting the range around a purge that falls in the middle rather than skipping the whole thing.
Also fixed in 2.3.7 alongside the original version of this panel: Snapmaker U1/Klipper
toolchanges were forcing M220 S100 on every color change (wiping any manual speed override) and
emitting M220 B/M220 R — both leftover from older Marlin-based Snapmaker machines and
meaningless on the U1's Klipper firmware. Both are gone now; toolchange G-code is simpler and no
longer fights a manual speed setting.
Stock slicers generate each object's support as if it were alone on the plate. Two separate (non-Assembled) objects close enough that their supports share space produce a collision — each support grows through the other object and through the other's support. The only stock workaround is to merge everything into one Assembled object, which changes how the parts slice and defeats the point when they're meant to be separate.
PerObject Support (checkbox directly under Support → Enable support, support_cross_object_avoidance,
default off) makes an object's support treat every other object on the plate — its body and
its already-generated support — as collision geometry to route around, keeping the normal
support/object XY distance from them. No Assemble, no boolean union: the objects stay independent and
the support simply stops colliding.
Turning it on
- Support → Enable support.
- Tick PerObject Support on each object that should avoid the others.
- Slice. It only takes effect when the plate prints all objects at once (by layer) — in sequential by-object printing the neighbors aren't on the bed yet, so avoidance is inert.
What it covers
| Support kind | Cross-object avoidance |
|---|---|
| Tree — Default / Slim / Strong / Hybrid | ✅ (hybrid tree engine) |
| Tree — Organic | ✅ (separate organic engine) |
| Normal / Grid | ✅ (classic engine) |
| NeoWave (§12) | ✅ (built on the classic engine) |
- Support vs. support — not just bodies. When two objects' supports would tangle, the one generated second routes around the first's finished support, not only around its body.
- Move-aware — nudging one object regenerates every nearby object's support against the new position. (Stock Orca never invalidated an object's own support on a move, because support was a per-object silo; PerObject Support adds that dependency.)
Trade-off (deliberate). The wide first-layer base/brim that support engines grow for bed adhesion is a free outward offset stock code never clips against anything (exaggerate it and it collides even with itself). Under PerObject Support that first-layer expansion is dropped to keep bases from spilling across objects — supports grab the bed a little less at the very first layer in exchange for never colliding. Only applies while the toggle is on.
Built on the real cross-instance contact detector (§4b): support avoidance and floating-object detection share the same geometry. Stored per-object in the project (3mf). First appearance of cross-object support avoidance in this slicer family.
Stock slicing assumes a few things that are only true when an object sits alone on the bed: "my layer 0 is the bed", "below me is only my own previous layer", "support only ever grows from the bed". Those assumptions break the moment two separate (non-Assembled) objects are placed one on top of the other — a face resting on another object gets misclassified as a bridge over thin air, even though there's solid material right underneath it.
Gravity measures what's really underneath every surface — the bed, another object, or genuine air — and slices accordingly, by area, not by object: the part of a face that rests on something solid prints as a normal contact surface; the part that's genuinely unsupported still prints as a real bridge, in the same layer if that's how the geometry actually sits.
Turning it on
The toggle is called "True Objects: On/Off", a toolbar side button next to the Libre Mode button — it appears once the Libre Mode master switch is on (Preferences → Enable Neotko LibreMode), but it is its own independent axis: turning Libre Mode's own active state on/off does not turn True Objects on/off, and vice versa. Think of it as: Libre Mode opens the door to the fork's pro features; True Objects decides whether things fall.
Upgrading from an older build that already had Libre Mode's floating active? True Objects is seeded to match it automatically the first time, so existing floating/stacked projects keep working exactly as before — nothing falls to the bed on upgrade.
What changes with True Objects on
| Before (stock) | With True Objects |
|---|---|
| A face resting on another object → false bridge (bridge speed/fan, wrong angle) | Same face → normal solid contact surface, correct fill angle |
| An object's own floating first layer → always solid, even over open air | The part genuinely over air → a real bridge |
| Perimeter overhang measured only against this object's own layer below | Measured against the real floor — a wall resting on a neighbor isn't flagged overhang |
| Support requested under a face that's actually resting on another object | No support requested there — the neighbor's top counts as ground |
| Elephant-foot compensation applied to any first layer, even a stacked one | Never applied to a face that isn't touching the bed — stacked contact faces keep their true size |
| PerObject Support (§16) is opt-in per object | Forced on for every object while True Objects is active (nothing is overwritten — turn True Objects off and each object's own PerObject Support setting is exactly as you left it) |
Where it shows up: two objects stacked exactly on top of each other (pair this with Align & Stack, §7, to place them precisely) — the touching face prints solid instead of showing up as a bridge in the preview. An object resting partly on another and partly hanging over open air shows both in the same layer: solid where it's supported, bridge where it truly isn't — that split is the clearest way to see the feature working.
A note on Assembled objects. If your stacked pieces are already combined into one Assembled
object (one ModelObject, multiple parts), this was never broken — the slicer already sees the
whole stack as one body. Gravity/True Objects is specifically for pieces that stay separate
objects on the plate.
Limits (v1) — things Gravity does not do yet:
- Support still only lands on the bed or on the object's own body — it does not yet land on the top of another object (that's a future extension).
- Auto-arrange/auto-orient can still scatter a hand-placed stack across the plate — it has no concept of "these objects are meant to stay stacked". Don't run auto-arrange after stacking by hand.
- Only takes effect in by-layer printing (the whole plate rises together); in sequential by-object printing a neighbor object may not exist yet at a given height, so nothing is treated as floor there.
With True Objects on, dragging an object in the viewport can rest it on whatever it is really above, instead of leaving it floating wherever you dropped it. Enable it per-session from the object's right-click menu → Snap & Drag (greyed out and unavailable while True Objects itself is off — it's a sub-behaviour of True Objects, not a separate axis).
How it decides where to land. Detection is by 2D footprint overlap, not a raycast under the cursor — a corner that only barely overlaps a pillar is not treated as resting on it (the overlap must clear a threshold before it engages, and a slightly lower threshold to stay engaged once it has — this hysteresis is what stops the object flickering up/down when you drag near a pillar's edge). Once a candidate qualifies, its landing height is sampled with a handful of real raycasts against the candidate's actual mesh, not its flat bounding-box top — so a hollow box (tall rim, low interior floor) resolves correctly depending on exactly where the overlap lands, instead of always reporting rim height. If several candidates qualify, the object rests on the highest real surface under it; if an object has more than one instance and they'd land on pillars of different heights, the whole object uses the lowest of those targets, so one instance landing on something tall never silently drags the others up with it.
Does the plate count? (2.4.0) By default yes: right-click → Snap & Drag: Allow Bed is on, and an object dragged over empty space lands on the build plate, the way placement behaves in any ordinary slicer. Switch it off and only other objects can catch you — an object with nothing underneath keeps floating exactly where you dropped it, which is what you want while assembling something in mid-air. The plate never competes with a real object: it is simply a floor of height zero, and the highest surface under your footprint always wins. The option lives directly under Snap & Drag in the same menu and is greyed out unless Snap & Drag is on.
With Allow Bed off and nothing qualifying underneath, the object is left exactly where it is — Snap & Drag then only ever pulls something down onto a floor it actually finds, in keeping with True Objects' own "nothing auto-drops" promise.
Landing indicator (2.4.0). While a drag is engaged the object hovers a little above its landing spot rather than sitting flat on it, so what is underneath stays visible, and that gap is filled with the evidence behind the decision:
- the corner marks of the box where the object will come to rest;
- the recognised zone — the exact patch being read as the height — highlighted on the surface it was found on;
- a translucent column standing between that zone and the object's underside;
- the sample points the height was actually taken from, as fat dots at their real hit heights;
- colour: cyan when another object caught you, amber when the plate did.
Underneath it all is the older soft contact shadow. None of this takes part in the calculation — but it is what makes aiming possible: when a large part refuses to catch a thin rim, the dots show you that the samples went through the opening instead of onto the rim.
Dragging several objects at once (2.4.0). A multi-object selection is resolved by stacks. Anything standing on another member of the same selection travels with it and keeps its exact relative height; anything with nothing of its own underneath resolves its own floor and falls independently. Pick up a stack of three plus a loose box, move them together, and the stack lands intact while the loose box drops to the plate — in the same drag. Two objects count as stacked when one overlaps the other's footprint and sits within about a millimetre of its top, so hand-built stacks that were never seated perfectly still hold together.
Limits: vertical (-Z) detection only — it does not help with side-by-side mating inside Assemble View, which has no single "down" direction. No chaining outside the selection: moving an object that something else is resting on does not drag that object along, only things picked up together move together.
Stock Orca doesn't compose text, it drops glyphs: every letter is placed at a fixed advance and that is the whole of it. The Char gap control it offers is tracking — one shift applied identically between every pair of characters. That is not kerning.
Kerning is the correction the type designer builds into the font for specific pairs, so that
AV, To or Wa close the diagonal gap that plain advances leave gaping. Orca never read it.
Left-side gizmo toolbar → Text (Emboss) → Advanced → Font kerning.
Tick it and the embossed text uses the kerning pairs stored in the font. Char gap is unchanged
and the two are independent: tracking shifts everything uniformly, kerning fixes individual pairs.
The setting is stored per style in the project (.3mf).
With Font kerning off, embossed text is identical to what earlier versions produced — this is a pure opt-in.
Not every font ships kerning data. When the selected font carries none, the checkbox is disabled and labelled "Font has no kerning data" instead of silently doing nothing.
Typical fonts without kerning data: monospaced faces (Courier, Andale Mono, SF Mono), symbol and Braille fonts, CJK fallbacks, and a number of display faces (Copperplate, Big Caslon, Bodoni 72). Across a typical macOS font library of ~900 styles, roughly three quarters do carry usable kerning.
The font library Orca is built on reads the Microsoft kern table and OpenType GPOS, but
silently ignores the Apple kern version 1.0 table. That is precisely the format the macOS
system fonts use — Helvetica included. Without a fix, "Font kerning" would appear to work for some
fonts (Helvetica Neue, most Google/Microsoft fonts) and do nothing at all for others, with no
explanation.
This fork adds a reader for the Apple format, so both layouts work.
A handful of fonts use Apple's state-machine kerning subtables (formats 2/3 — Geeza Pro, Apple Chancery); those are still reported as having no usable kerning data.
Unrelated to kerning but in the same gizmo: typing in the font selector used to make the entire list disappear instead of narrowing it. The search was implemented correctly — but the cached font list, which is what you get from your second launch onward, populated only the names drawn on screen and left the list the search matches against empty, so any keystroke matched zero fonts. Only a first launch on a clean profile ever worked.
Fixed: the font selector filters as you type. The bug is present in upstream Orca as well.
The glyph advance used to be baked into the shape cache, which is keyed by character alone — so a
per-pair value could not be expressed at any price. Text composition now happens in one place,
Emboss::layout_text(), which the geometry goes through and which future typographic controls
(manual pair kerning, baseline shift, per-range scaling) will hook into.
Quality → Bridging → Bridging infill extra expansion (mm, default 0).
Orca already grows each bridge region a little into the area around it, so the strand has somewhere to
land instead of starting in mid-air. Two things are wrong with how it does that: the amount is
hard-wired, and it is derived from your wall count — raise wall_loops for stiffness and you
silently change the anchoring of every bridge in the part. This setting adds millimetres on top of the
automatic amount and decouples the two.
What it buys you: the bridge takes over the neighbouring region of the same part, so the nozzle is already extruding over solid, supported material before it reaches open air. A strand that starts anchored behaves very differently from one that starts unattached — each pass has something to grab, and the filament tensions across the gap instead of simply hanging.
There is no cap. Turning it up until the bridge claims the entire surrounding region is a normal way to use this — it removes the seam artefact where the bridged area meets the supported one, and gives the layer a single continuous direction. Values in the hundreds are accepted.
- Additive. At 0 the slicer behaves exactly as before, byte for byte.
- Applies to external bridges only.
- The custom bridge angle is honoured across the expanded region.
- The expansion grows into solid infill, sparse infill, top surfaces and supported bottom surfaces. That last one is the important case: a beam resting on two blocks ("dolmen") has its bridge sitting right next to the supported area, and that supported area is the best possible landing ground — solid, with material directly underneath.
- Costs a little extra material and some bridge-speed travel over a region that was already solid.
ORCA_DEBUG_BOTTOM=1 writes BRIDGE_EXPAND / BRIDGE_EXPAND_DONE lines to
/tmp/neotko_bottom.log, one pair per layer that contains a bridge:
extra_cfg— the value that reached the engine. If this is 0 when you set something else, the problem is config invalidation, not the expansion.exp_solid_mm/exp_sparse_mm— how far the bridge may grow into each kind of neighbour.zone_solid_mm2/zone_sparse_mm2/zone_top_mm2/zone_bottom_mm2— how much area of each neighbour actually exists next to the bridge. All zeros means there is nowhere to grow, which is the usual reason for "it does nothing".zone_bottom_mm2reads-1when the setting is 0, since that zone is only created when you ask for extra expansion.delta_mm2— how much bridge area was actually gained.
Where to find it: G-code Preview → the view-type dropdown at the top of the legend panel → RealColor.
Every slicer shows you multi-material G-code the same way: each extrusion drawn in its filament's raw colour. That tells you which tool prints where, which is the right answer for a tool-change preview and the wrong one for this pack. A Sandwich surface is not "some blue lines and some pink lines" — it is a stack whose colours mix optically as light passes through the upper passes and bounces back off the ones underneath. Read line by line, that surface looks like stripes. Printed, it is one blended colour.
RealColor renders the second thing. It composites each surface the way the eye will resolve it, using each filament's TD (§1e) to decide how much of what is underneath shows through.
Same G-code in both shots. Only the interpretation changes.
- TD is what makes it accurate. RealColor is only as good as the transmission-distance values you gave each filament. With TD roughly right the match to the printed part is genuinely close; with TD left at defaults on filaments that differ a lot in opacity, expect the preview to drift. Tuning TD (§1e, and the Object & TD department of the painter, §6b) pays off here more than anywhere else.
- It is a simulation, not a promise. The panel says so itself: "Approximated optical simulation — not a guarantee of the final print colour." Surface finish, lighting and the printer's own colour-blending behaviour all move the result.
- Changes nothing about the G-code. It is a rendering mode, and switching back to Filament gives you the standard view instantly.
- The RealColor legend shows Filament Usage and Time Estimation only — the per-tool tower/cost breakdown and the Travel/Retract/Seams toggles belong to the Filament view.
Everything above is previews and dialogs. This section is the printed result, so you can judge for yourself how far the simulation is from the plastic.
This pair is the clearest argument for why the G-code view alone is misleading for this pack, and why RealColor (§20) exists.
The BIGTEST.3mf project used for this board is in the fork's GitHub repository, so you can slice and
print the same reference yourself.
With TD values set correctly for the filaments in use, the match is close. That is the whole point of tuning TD (§1e) rather than leaving it at defaults.
| Feature | Location in UI |
|---|---|
| Sandwich Editor (pass stack, ColorStitch, PathBlend) | Quality → Surface ColorStitch → Sandwich editor… |
| ColorStitch pass pattern | Sandwich Editor → a pass set to ColorStitch → Edit gradient… |
| ColorStitch Studio | Sandwich Editor → ColorStitch Studio panel (bottom) |
| Colour match (inverse ΔE2000) | ColorStitch Studio → Target + Match ▸ |
| Filament & TD preview | Sandwich Editor → Filament & TD panel |
| Line distribution mode | Quality → Surface ColorStitch → Line distribution mode (below Minimum line length) |
| Top surface fill pattern (needed for ColorStitch) | Quality → Top surface pattern → Monotonic Line |
| Penultimate layers / density | Strength → Top/bottom shells |
| Neoweaving (WIP) | Not wired in this build |
| Libre Mode master switch | Preferences → Enable Neotko LibreMode (requires restart) |
| Libre Mode toggle | Toolbar side button "Neotko LM: On/Off" |
| Assembled Boolean mode | Right-click an object (Libre Mode) |
| Per-volume XY compensation | Part settings inside an Assembled object (Libre Mode) |
| Copy / Paste Process Settings | Right-click object → Copy/Paste Process Settings (Libre Mode) |
| Assembled Parts full options | Part settings tab (Libre Mode) |
| Remove Slice Cache (force a full re-slice) | Right-click object(s) → Remove Slice Cache (always available) |
| PerObject Support (§16) | Support → Enable support → PerObject Support checkbox, per object |
| True Objects / Gravity (§17) | Toolbar side button "True Objects: On/Off" (independent from Libre Mode) |
| Snap & Drag (§17a) | Right-click an object → Snap & Drag (requires True Objects on) |
| Snap & Drag: Allow Bed (§17a) | Right-click an object → Snap & Drag: Allow Bed (requires Snap & Drag on) |
| World-space import (WIP) | Import with Libre Mode active (recommend assembled → split) |
| S3DFactory import | File → Import → Import 3D model → .factory (loads assembled; split in Libre Mode) |
| Save / Manage profiles | Sandwich Editor → Save as profile… / Manage Sandwich Profiles |
| ColorStitch Painter | Left-side gizmo toolbar |
| Painter tools (Paint / Eraser / Pick) | Painter panel top row |
| Palette groups / Save all / Pin to palette | Painter panel |
| MixedFilament Object mode (Beta) | Painter panel → Pro mode → "MixedFilament Object" checkbox |
| Align & Stack | Left-side gizmo toolbar → Align & Stack |
| NeoArachne (enable) | Libre Mode → Quality → Wall generator → NeoArachne |
| NeoArachne sources / line widths / Preview Lab | Quality → NeoArachne section |
| NeoTower (tower type) | Quality → Prime tower → Tower type |
| Zigurat / Sandwich purge compaction / Sandwich wipe reserve | Quality → Prime tower |
| Variable layer height (Experimental) | Quality → Prime tower → Tower type = NeoTower (greyed unless Libre Mode) |
| PathBlend start/end zone + techo editor | PathBlend pass → ADV… (Painter Pro tray or Sandwich Editor's Advanced button) |
| Bump Mapping Editor (All / Painter / Top) | Left-side gizmo toolbar (expert gate: Libre Mode active + ORCA_DEBUG_TEXTUREBUMP set, + ORCA_DEBUG_ZBUMP for Top/ZBump — see NEOTKOCM_RELEASE_2_35.md) |
| Precision Adaptive Layer Height | Left-side gizmo toolbar (icon always visible, needs Libre Mode active to use — §11) |
| NeoWave Support (enable) | Libre Mode → Support → Support type → NeoWave |
| Wave roof (shape/order/reverse/hollow pillar) | Support → Interface pattern → Wave to reveal the controls (§12) |
| NeoWave contact layer (WIP, print-pending) | Support → Advanced → Support neoweave contact toggle (§12a) |
| Painter Pro Mode (Precision / Paint perimeters only / Extra walls / Rectangle & Polygon masks) | Left-side gizmo toolbar → Color Painting → Pro Mode section (always available, no Libre Mode needed) |
| NeoStitch Interlock (WIP, |
Strength → NeoStitch Interlock (§14) |
| Font kerning / Typographic Spacing (§18) | Left-side gizmo toolbar → Text → Advanced → Font kerning |
Q: ColorStitch doesn't apply on some layers — why?
Confirm the top surface fill pattern is Monotonic Line (§1b) — the only pattern that works correctly on complex objects. Then check ColorStitch min. line length (short lines are skipped) and the Zone setting (Topmost only applies to just the top of the object).
Q: Where did "MultiPass" go?
It's now built from the pass stack: add two or three Solid passes to a zone and split the layer height between them with the dividers. That gives the cross-hatch / glaze / optical-blend effects the old MultiPass produced. See §1a.
Q: My ColorStitch stripes look fragmented across holes.
Change Line distribution mode (Quality, Advanced) to LaneQuant or DirCluster. See §1f.
Q: How do I turn on Libre Mode? I don't see the button.
Enable the master switch in Preferences ("Enable Neotko LibreMode") and restart; the "Neotko LM" side button then appears. See §4a.
Q: Do I need Libre Mode to print normally?
No. It's off by default and the build behaves like stock. It only unlocks the §4 workflows and exposes NeoArachne (§8).
Q: My .factory (or world-space) import put parts in the wrong place.
A .factory always loads as one Assembled object; split it in Libre Mode to recover the parts in place. Same recommendation for world-space import while it's being finished. See §4g and §5.
Q: In the Painter, why don't picked colours show in the Profiles list?
By design — colours you paint are working colours (created on demand, cleaned up when unused). Only palettes you deliberately Pin to palette appear in the saved list; use Save all to promote every unsaved working colour at once. See §6c.
Q: Can I use adaptive layer height with multiple tools and a Sandwich?
Yes, as of 2.3.1 — enable Libre Mode, set Tower type = NeoTower, and turn on Variable layer height (Experimental) (Quality → Prime tower). The wipe-tower "missing drawers" issue that made this rough is now fixed (§9e). It is still flagged Experimental, so review G-code before long runs.
Q: My wipe tower uses more purge than expected.
Check Sandwich wipe reserve (default 10 mm³) and Sandwich purge compaction (default 1.7) in Quality → Prime tower (§9). Lower the reserve for a thinner tower, or set compaction to 1.0 to disable it.
Q: The "MixedFilament Object" checkbox is greyed out — why?
The object's extruder isn't a MixedFilament. Assign one of your MixedFilament rows as the object's extruder (same as assigning any normal filament), then reopen the Painter — the checkbox and its colour swatch become active. See §6g.
Q: Can I make a PathBlend gradient start or end somewhere other than the surface edges — or make it fully opaque at one end?
Yes — open ADV… on the PathBlend pass (Painter Pro tray or the Sandwich Editor's Advanced button) and drag the two handles in the graph that opens. The low handle sets where the ramp starts rising (and its floor height); the high handle sets where it finishes (and its top height). Push the high handle all the way to the top for a full "techo" — zero of the cap colour in that zone. See §1c.
Q: I want to try NeoArachne safely.
Enable Libre Mode, set Quality → Wall generator → NeoArachne, leave the new controls at defaults (Neotko Hybrid v2), and slice. Switch back to Arachne/Classic anytime — the NeoArachne controls don't touch existing settings.
Q: Is the "Color Painting" gizmo's Pro Mode the same thing as the ColorStitch Painter?
No — different tools. Color Painting (§13) is Orca's own stock multi-material painter, and Pro Mode adds precision brushing/masking on top of it. The ColorStitch Painter (§6) is this pack's own gizmo for painting Sandwich effect profiles (pass stacks, gradients, patterns). Both live in the left-side toolbar as separate icons.
Q: I turned on "Extra walls" but the painted area doesn't seem to have more walls.
Check whether the colour you painted with is the same as the object's own filament — Extra Walls is skipped there by design, since it wouldn't be visible anyway (§13b). Paint with a genuinely different colour to see the extra walls take effect.
Q: My Rectangle/Polygon mask painted something on the far side of the object too.
It shouldn't — both tools only paint faces pointing toward the camera. Rotate the view to confirm what you're actually seeing is the front-facing paint, not a leak through the back.
All of this work is open and free. Fork it, improve it, credit it.
Now all the info from the Original SnapMaker 2.3.4 Readme
Snapmaker Orca is an open source slicer for FDM printers based on OrcaSlicer.
📥 **Download the Latest Stable Release Visit our GitHub Releases page for the latest stable version of Snapmaker Slicer, recommended for most users.
Windows:
- Download the installer for your preferred version from the releases page.
- For convenience there is also a portable build available.
- If you have troubles to run the build, you might need to install following runtimes:
- MicrosoftEdgeWebView2RuntimeInstallerX64
- vcredist2019_x64
- Alternative Download Link Hosted by Microsoft
- This file may already be available on your computer if you've installed visual studio. Check the following location:
%VCINSTALLDIR%Redist\MSVC\v142
Mac:
-
Download the DMG for your computer:
arm64version for Apple Silicon andx86_64for Intel CPU. -
Drag Snapmaker_Orca.app to Application folder.
-
If you want to run a build from a PR, you also need to follow the instructions below:
Details
- Option 1 (You only need to do this once. After that the app can be opened normally.): - Step 1: Hold _cmd_ and right click the app, from the context menu choose **Open**. - Step 2: A warning window will pop up, click _Open_- Option 2:
Execute this command in terminal:xattr -dr com.apple.quarantine /Applications/Snapmaker_Orca.appsoftfever@mac:~$ xattr -dr com.apple.quarantine /Applications/Snapmaker_Orca.app - Option 3:
- Step 1: open the app, a warning window will pop up
- Step 2: in
System Settings->Privacy & Security, clickOpen Anyway:
- Step 1: open the app, a warning window will pop up
- Option 2:
Linux (Ubuntu):
- If you run into trouble executing it, try this command in the terminal:
chmod +x /path_to_appimage/Snapmaker_Orca_Linux.AppImage
-
Windows 64-bit
- Tools needed: Visual Studio 2019, Cmake, git, git-lfs, Strawberry Perl.
- You will require cmake version 3.14 or later, which is available on their website.
- Strawberry Perl is available on their GitHub repository.
- Run
build_release.batinx64 Native Tools Command Prompt for VS 2019 - Note: Don't forget to run
git lfs pullafter cloning the repository to download tools on Windows
- Tools needed: Visual Studio 2019, Cmake, git, git-lfs, Strawberry Perl.
-
Mac 64-bit
- Tools needed: Xcode, Cmake, git, gettext, libtool, automake, autoconf, texinfo
- You can install most of them by running
brew install cmake gettext libtool automake autoconf texinfo
- You can install most of them by running
- run
build_release_macos.sh - To build and debug in Xcode:
- run
Xcode.app - open
build_`arch`/Snapmaker_Orca.Xcodeproj - menu bar: Product => Scheme => Snapmaker_Orca
- menu bar: Product => Scheme => Edit Scheme...
- Run => Info tab => Build Configuration:
RelWithDebInfo - Run => Options tab => Document Versions: uncheck
Allow debugging when browsing versions
- Run => Info tab => Build Configuration:
- menu bar: Product => Run
- run
- Tools needed: Xcode, Cmake, git, gettext, libtool, automake, autoconf, texinfo
-
Ubuntu
- Dependencies Will be auto-installed with the shell script:
libmspack-dev libgstreamerd-3-dev libsecret-1-dev libwebkit2gtk-4.0-dev libosmesa6-dev libssl-dev libcurl4-openssl-dev eglexternalplatform-dev libudev-dev libdbus-1-dev extra-cmake-modules libgtk2.0-dev libglew-dev libudev-dev libdbus-1-dev cmake git texinfo - run 'sudo ./BuildLinux.sh -u'
- run './BuildLinux.sh -dsir'
- Dependencies Will be auto-installed with the shell script:
If you're running Klipper, it's recommended to add the following configuration to your printer.cfg file.
# Enable object exclusion
[exclude_object]
# Enable arcs support
[gcode_arcs]
resolution: 0.1
Snapmaker Orca is originally forked from Snapmaker_Orca.
Snapmaker_Orca is originally forked from Bambu Studio, it was previously known as BambuStudio-SoftFever. Bambu Studio is forked from PrusaSlicer by Prusa Research, which is from Slic3r by Alessandro Ranellucci and the RepRap community. Orca Slicer incorporates a lot of features from SuperSlicer by @supermerill Orca Slicer's logo is designed by community member Justin Levine(@freejstnalxndr)
Snapmaker Orca is licensed under the GNU Affero General Public License, version 3. Orca Slicer is based on Snapmaker_Orca by SoftFever
Orca Slicer is licensed under the GNU Affero General Public License, version 3. Orca Slicer is based on Bambu Studio by BambuLab.
Bambu Studio is licensed under the GNU Affero General Public License, version 3. Bambu Studio is based on PrusaSlicer by PrusaResearch.
PrusaSlicer is licensed under the GNU Affero General Public License, version 3. PrusaSlicer is owned by Prusa Research. PrusaSlicer is originally based on Slic3r by Alessandro Ranellucci.
Slic3r is licensed under the GNU Affero General Public License, version 3. Slic3r was created by Alessandro Ranellucci with the help of many other contributors.
The GNU Affero General Public License, version 3 ensures that if you use any part of this software in any way (even behind a web server), your software must be released under the same license.
Orca Slicer includes a pressure advance calibration pattern test adapted from Andrew Ellis' generator, which is licensed under GNU General Public License, version 3. Ellis' generator is itself adapted from a generator developed by Sineos for Marlin, which is licensed under GNU General Public License, version 3.
The Bambu networking plugin is based on non-free libraries from BambuLab. It is optional to the Orca Slicer and provides extended functionalities for Bambulab printer users.
We greatly value feedback and contributions from our users. Your feedback will help us to further develop Snapmaker Orca for our community.
- To submit a bug or feature request, file an issue in GitHub Issues or email us at support@snapmaker.com.
- To contribute some code, make sure you have read and followed our guidelines for contributing.