After I had reviewed and responded to the vase links supplied by @stevevw in post #271, it crossed my mind there is another factor which makes the youtubers methodology for measuring deviations patently ridiculous, the screen resolution of computer monitors.
The highest resolution monitors available are 8K and typically have screen pixel sizes of 92 microns.
Since they used the software package Blender which relies purely on visually placing the scanned mesh data over the perfect model, it is impossible to have the quoted deviations of 0.05 mm (50 microns) when the lowest values are governed by the screen resolution of the monitor.
I asked GPT-5 to expand on the use of Blender and OpenSCAD which was also used by the youtubers.
The highest resolution monitors available are 8K and typically have screen pixel sizes of 92 microns.
Since they used the software package Blender which relies purely on visually placing the scanned mesh data over the perfect model, it is impossible to have the quoted deviations of 0.05 mm (50 microns) when the lowest values are governed by the screen resolution of the monitor.
I asked GPT-5 to expand on the use of Blender and OpenSCAD which was also used by the youtubers.
Blender and OpenSCAD are very different in their purpose and capabilities, especially compared to professional metrology or CAD packages. Let me break them down:
Blender
- Type: Open-source 3D graphics and modeling software.
- Main use cases: Animation, rendering, visual effects, artistic modeling.
- Capabilities relevant to 3D models:
- Can import and manipulate meshes (e.g., scanned vase models).
- Allows manual alignment of primitives (cylinders, spheres, cones) with imported geometry.
- Has modifiers (boolean, shrinkwrap, etc.) for mesh editing.
- Can generate cross-sections and overlays for visual comparison.
- Limitations:
- Blender is not a metrology tool — it has no built-in functions for:
- Calculating RMS deviation from an ideal primitive.
- Reporting tolerances, uncertainty, or accuracy.
- Generating certified deviation maps.
- Accuracy is effectively limited to mesh resolution + user’s manual manipulation (and the display resolution of the monitor).
- Results are visual, not quantitative.
OpenSCAD
- Type: Script-based CAD (Constructive Solid Geometry).
- Main use cases: Parametric modeling for 3D printing, simple mechanical parts.
- Capabilities relevant to 3D models:
- Models are defined by code, not manual editing.
- Very precise for ideal mathematical shapes (cylinders, cubes, spheres, extrusions, etc.).
- Great for generating perfect primitives to compare against scanned data (if imported).
- Limitations:
- Not designed for handling scanned meshes (point clouds, STL from archaeology).
- Very limited mesh analysis functions — cannot compute deviations or fit primitives to arbitrary data.
- Outputs perfect models, but does not evaluate how well real data fits those models.
⚖️ Blender vs. OpenSCAD vs. Professional Tools
Feature Blender OpenSCAD Professional Metrology (e.g., GOM Inspect, PolyWorks) Import scanned meshes ✅ Yes ⚠️ Limited (STL import only) ✅ Yes Create perfect primitives ✅ Manual ✅ Script-based ✅ Automated fit Fit primitive to scan data ⚠️ Manual alignment only ❌ Not possible ✅ Mathematical best-fit (least squares, ICP) Compute deviations (RMS, max) ❌ No ❌ No ✅ Yes Generate heat maps ❌ No ❌ No ✅ Yes Accuracy Limited by mesh + screen Exact for math shapes, but no mesh analysis Calibrated, micron–tens of micron scale Intended purpose Art, animation Code-based parametric CAD Metrology, engineering QA
✅ Bottom line
- Blender = great for visual demos (“look how round this vase looks when I overlay a cylinder”), but not a measurement tool.
- OpenSCAD = great for coding precise ideal shapes, but it cannot analyze how well a scanned artifact matches that shape.
- Professional metrology software = the only category that can actually measure deviation values with traceable accuracy.
Would you like me to make a table of what Blender and OpenSCAD cannot do that professional metrology software must do when analyzing stone vases? That might give you a clear checklist for critiquing YouTube claims.
Upvote
0