TLDR
For most hollow vs solid terrain decisions, begin with a closed shell that has intentional walls and adequate top and bottom thickness. Add internal structure only beneath playable roofs, floors, bridge decks, broad summits, fastening points, or other features that face a defined load. Reserve 100% infill for small parts that need concentrated strength or weight. Compare the variants in your slicer using the same machine, nozzle, layer height, material, orientation, and exterior-quality settings.
A ruined wall, rock outcrop, or closed building rarely becomes more useful just because every hidden cubic millimetre contains plastic. The efficient question is not whether solid terrain is stronger in the abstract. It is whether that additional material and printer time improve handling, gameplay, stability, or durability in a place that matters.
Hollow vs solid terrain is not a two-option choice
“Hollow” can describe several substantially different constructions. Treating them as interchangeable can lead to fragile prints or misleading slicer comparisons.
- Modeled cavity: The terrain file contains an intentional internal void bounded by inner and outer surfaces. Its shell thickness is partly determined by the model itself, although slicer settings still affect how those surfaces print.
- Zero sparse infill: The model is conventionally solid, but the slicer is told not to generate sparse infill. Walls and top or bottom shells remain, and geometry can still produce internal solid regions.
- Single-wall or vase-style construction: The object is produced as a very thin continuous wall with limited or no conventional top surfaces. This is a specialized approach, not a general synonym for hollow terrain.
- Locally reinforced shell: Most of the piece remains hollow or lightly filled, while selected regions receive thicker walls, denser infill, ribs, separate supports, or other reinforcement.
- Fully solid slice: The interior is filled at 100% infill, subject to how the slicer interprets the geometry and its other strength settings.
For ordinary scenery, the locally reinforced shell is usually the most useful target. It avoids filling large unseen volumes while preserving material where the model must support miniatures, resist squeezing, hold hardware, or remain stable on the table.
Start with walls, not a solid interior
Walls or perimeters form the outer structure that receives most handling contact. Prusa’s documentation identifies perimeters as a major contributor to part strength, while infill contributes to rigidity and resistance to compression; both also affect time and material use. Prusa’s infill documentation provides a useful overview of that relationship.
That distinction matters for terrain. A closed rock formation normally needs a durable exterior more than it needs maximum resistance to crushing from above. A building with a playable roof has a different requirement: the roof must resist local downward loads and transfer them into the walls. Increasing infill throughout the whole building may work, but it spends material in regions that do not contribute much to that specific task.
Additional wall loops can increase rigidity and strength, but they also increase print time. This makes walls a deliberate production setting rather than a free upgrade. Add enough exterior structure for handling, then inspect whether the geometry creates a separate need for internal support.
Choose construction by terrain feature
| Terrain feature | Efficient starting point | Where strength is needed | When to consider more structure |
|---|---|---|---|
| Ruined wall or barricade | Shell-first construction | Exposed edges, thin projections, base connection | Add local thickness around narrow uprights, joints, or frequently handled sections |
| Closed rock or boulder | Hollow or nearly hollow shell | Outer walls and contact area at the base | Reinforce a broad flat summit if it regularly carries miniatures |
| Building with no playable interior | Closed shell | Corners, door frames, thin wall sections | Support large roofs or overhangs rather than filling the entire building |
| Building with playable upper floor | Shell plus local reinforcement | Floor plate, supporting walls, stair or balcony connections | Use ribs, thicker floor shells, or localized infill beneath occupied areas |
| Bridge or gantry | Feature-specific reinforcement | Deck, joints, supports, and load path to the ground | Use internal structure where the deck spans between supports |
| Broad hill with flat summit | Hollow body with supported top | Top skin and the structure beneath its widest spans | Add internal ribs, sparse infill, or designed columns if the top cannot bridge reliably |
| Small marker or objective | Strong shell or solid print | Thin necks, attachment points, and base stability | Use solid construction when the part is tiny or intentionally needs ballast |
This table is a decision framework, not a universal recipe. A narrow bridge deck and a wide roof can behave differently even when their overall dimensions look similar. Material, nozzle, layer height, line width, orientation, layer adhesion, geometry, and expected tabletop load all change the result.
Where local reinforcement earns its filament
Playable roofs and upper floors
A roof that only closes a building needs to print cleanly and survive ordinary handling. A roof intended to hold metal miniatures, scenery pieces, or repeated hand pressure has a structural job. Reinforce the roof plate and its path into the walls instead of automatically filling the entire structure.
Top and bottom shells are best considered in terms of physical thickness as well as layer count. The same number of layers produces a different physical shell when layer height changes. OrcaSlicer documents controls for top and bottom shell thickness, making physical thickness the more transferable way to reason about the feature.
Bridges, gantries, balconies, and cantilevers
These features combine gameplay loads with unsupported spans. A hollow deck may be adequate if it is short, well supported, and sufficiently thick. A long span may need ribs, infill, columns, or a redesigned cross-section. Bridging settings affect how well a printer closes unsupported internal spans, so a hollow design can fail because of printability before it fails under tabletop use.
Fasteners, magnets, and assembly points
Screw locations, magnet pockets, pegs, clips, and glued joints concentrate force into small areas. Give those regions surrounding material. A local boss, thicker modeled section, or modifier-controlled reinforcement is generally more efficient than making a large building solid merely to strengthen one attachment point.
Weight and stability
If a tall, narrow terrain piece needs ballast, place weight where it improves stability: low and near the base. Printing the entire model solid adds mass at every height, including places that do less to resist tipping. A weighted base, captured washer, enclosed ballast compartment, or locally dense base can be more efficient, provided the design safely retains the added material.
Why zero infill may save less than expected
Setting sparse infill to zero does not remove every internal extrusion. Thin features may already consist mostly of walls. Broad horizontal surfaces still need top and bottom shells. Small gaps can become solid because there is insufficient room for a separate infill region. Sloped geometry can also receive added internal material as the slicer maintains vertical shell thickness. Prusa documents this vertical-shell behavior in its layer and perimeter settings.
Supports can also dominate the job. A hollow building with deep overhangs, decorative arches, or a difficult roof may consume substantial time and filament in support structures. Likewise, reducing infill will not necessarily deliver a large time reduction when exterior walls, fine details, acceleration limits, cooling constraints, or minimum layer times control most of the print.
The practical lesson is simple: do not estimate savings from infill percentage alone. Slice the exact model and inspect where the toolpaths actually go.
A fair slicer comparison workflow
Compare construction strategies without changing unrelated variables. If one version uses a different orientation, layer height, or quality preset, you are no longer isolating the hollow-versus-solid decision.
- Select the actual printer profile, nozzle, filament profile, and layer height you plan to use.
- Load the model at its intended tabletop scale and choose the production orientation you would genuinely print.
- Keep exterior quality settings, speeds, supports, seam strategy, and adhesion settings constant across the comparison.
- Create a shell-first version with no sparse infill, while retaining intentional walls and top or bottom shells.
- Create a locally reinforced version using modeled ribs, separate internal supports, or slicer modifiers where your software and model permit them.
- Create a higher-infill or fully solid version only if it represents a plausible production choice.
- Inspect the layer preview, not just the infill field. Look for unsupported roofs, long bridges, tiny solid regions, internal supports, gaps, and features that receive more material than expected.
- Record the slicer’s estimated time and filament for every version. Treat both as configuration-dependent estimates rather than universal performance figures.
- Choose the lightest version that meets the piece’s handling, load, stability, and print-reliability requirements. Test the exact geometry before producing a large batch.
For a useful comparison, label the configuration with the result: printer, nozzle diameter, layer height, material, orientation, support strategy, and quality target. “Hollow was faster” tells another maker very little. “The shell-first version reduced the estimate under this fixed configuration” is a reproducible and appropriately limited conclusion.
When 100% infill is justified
Full infill is reasonable when the whole part is small, when internal volume is already limited, or when the object must resist concentrated compression throughout its body. It can also make sense for compact feet, pins, mounting blocks, or small markers where creating a separate local strategy would add complexity without meaningful savings.
It is less compelling as a default for large rocks, closed buildings, hills, and other volume-heavy scenery. In those cases, ask what failure the solid interior prevents. If the answer is vague, preview a reinforced shell first. If the terrain must reliably carry substantial loads, accept fasteners, survive transport abuse, or meet a club’s repeatable production standard, print a representative test piece before committing a batch.
The practical answer
Use a shell-first approach for ordinary non-load-bearing terrain, then reinforce the features that do real work. Walls protect the exterior. Top and bottom shells close surfaces. Ribs, infill, columns, or thicker modeled sections support roofs, decks, attachments, and broad spans. Full infill belongs where the entire volume contributes useful strength or deliberate weight.
Your next step is to slice one representative model three ways: shell-first, locally reinforced, and solid. Hold the printer, nozzle, layer height, material, orientation, supports, and quality target constant. Inspect the layer paths, compare estimated time and grams, and choose the least material-intensive version that still performs its tabletop job.