Key idea & worked example
A coordinate frame defines axes, orientation and units. A verified physical scale additionally requires evidence linking those coordinates to the physical object.
Coordinate consistency allows interchange; metrological validity requires additional evidence.See the worked example →On this page
Choose the representation before the tool
| Approach | Useful output | Question it can support |
|---|---|---|
| Photographic reconstruction | Estimated surface observations, point clouds and camera paths. | What spatial structure is supported by the available views? |
| Parametric component modeling | Geometry generated from explicit dimensions and construction rules. | What follows from this stated geometric hypothesis? |
| Passive pressure-acoustic analysis | Modes or response behavior under specified geometry and boundary assumptions. | How does this modeled acoustic domain respond under those conditions? |
| Room and propagation modeling | A response or field for a stated spatial arrangement. | How might sound reach a receiver within the supported domain? |
| Interactive assembly | A usable combination of visuals, controls, signals and modeled behavior. | How can a person explore or operate this research representation? |
Modelgenerator: preserve what the cameras support
The Modelgenerator code inspected on 9 October 2026 requests a GLB export from Depth Anything 3 and retains it as a reconstruction artifact. An optional path also preserves camera estimates for an experimental localization bundle. These are implemented adapter paths, not new measurements of reconstruction accuracy. The September portable source snapshot separately documents point clouds and camera paths.
Inspect what the file contains, rather than treating GLB as a guarantee of a complete textured surface. The dependency-free fallback writes one lightweight plane per selected image; it is an authored test representation, not recovered object geometry. Camera registration, visible surface coverage, scale and transport verification are distinct checks. Occluded interiors remain unsupported by the photographs.
Record the input photographs, processing configurationConfigurationAn identified arrangement of components and their relationships. Rebuilding or reassigning parts can change a configuration while the instrument’s broader identity persists.Work through the Cuntz example → and coordinate relationship to measurements. Photograph-based geometry and a manually supplied dimension can be combined, but the resulting scale or correction should have a derivation record. A hidden pipe mouth, internal cavity or inaccessible mechanism cannot become an observed surface merely because software produces a complete-looking shape.
Can we measure the original organ by rotating its model?
A coordinate frame defines axes, orientation and units. A verified physical scale additionally requires evidence linking those coordinates to the physical object.
- Model coordinatesScene geometryA viewer can rotate and zoom the representation.
- Physical measurementCalibrated dimensionsNeeds a documented unit, frame and calibration.
Work through the example 3 STEPS
The released model declares a right-handed, Y-up frame with metre units, while its purpose limitation says the source scale has not been verified by a physical survey.
Read the coordinate convention
The convention tells a renderer how to interpret model coordinates and orient the scene. It helps two applications agree about the same geometry.
Read the qualification
A declared metre unit does not supply a missing calibration. If the original geometry scale was not verified, a distance computed from its vertices must not be advertised as a measured organ dimension.
Use appropriate evidence for dimensions
For a documented pipe quantity, inspect the identified measurement observation and workbook source. Use the model to locate and discuss shapes within its stated limitations.
- Cuntz Positiv · VAO 0.5.0-rc.2 ↗
Published manifest: identities, measurement observations, realization digests, profiles and rights. Exact examples checked against this release.
- VAO Standard 0.5.0 ↗
The standard contract is separate from the Cuntz dataset content version 0.5.0-rc.2.
The notebook deliberately provides viewing controls without a physical measurement ruler.
The useful distinction. Coordinate consistency allows interchange; metrological validity requires additional evidence.
VAO-Pipe: make the geometric hypothesis inspectable
VAO-Pipe belongs to a different route: procedural pipe geometry and passive pressure-acoustic evaluation. Its parameters make a hypothesis explicit and allow controlled variation. That is useful for investigating how selected dimensions and boundary assumptions influence a modeled result. It should remain distinct from a complete account of self-sustained sound production, historical voicing or the actual condition of a surviving pipe.
The thesis emphasizes the danger of tuning underdetermined parameters until a desired result becomes inevitable. Preserve the source of each dimension, distinguish measured and inferred values, and identify which observations were used to adjust the model. Compare against independent evidenceEvidenceMaterial offered as grounds for a claim. Its relevance depends on the claim, context, transmission and interpretation.Work through the Cuntz example → where available. A close frequency agreement after optimization describes that model and fitting procedure; it does not automatically recover a unique historical geometry.
When does a modeled pipe become a historical hypothesis?
A reconstruction combines evidence with assumptions to produce a possible representation. A simulation computes behavior under specified model conditions.
- Observed evidenceWhat is documentedRetain the original sources and gaps.add stated assumptions
- Modeling decisionAn explicit hypothesisRecord inferred or reconstructed parts.inspect the result
- ResultA testable representationExplain which conclusions it can support.
Work through the example 3 STEPS
Suppose a researcher uses the Cuntz GesL observation to help constrain a pipe model. This is an illustrative reconstruction task; the single value is not a complete geometric description.
List what is supplied
The observation identifies one property, one pipe, one result and one source. It does not provide every interior dimension, material parameter or boundary condition needed by an acoustic model.
Label what is assumed
Record the additional dimensions and physical assumptions with their origin. Parameters chosen for convenience should not appear as measured values.
Keep fitting and testing separate
If parameters are adjusted to match a known frequency, that same match is not an independent validation. Compare with suitable independent evidence and retain the model’s domain of use.
- Cuntz Positiv · VAO 0.5.0-rc.2 ↗
Published manifest: identities, measurement observations, realization digests, profiles and rights. Exact examples checked against this release.
- Dominik Ukolov · Musikinstrumente im virtuellen Raum (2026)
Submitted dissertation, 11 September 2026; §§8.2.2.1 and 8.4.1, printed pp. 264–265 and 275–276. The historical description and digital inventory require a resolved mapping; the manuscript is not redistributed here.
No new simulation, measured frequency or complete pipe reconstruction is supplied by this example.
The useful distinction. A plausible reconstruction is a documented hypothesis constrained by evidence, not a recovered certainty.
What does a finite-element pipe model actually calculate?
Finite-element analysis divides a modeled domain into small connected elements and approximates the governing equations over that mesh. Its result depends on geometry, material or medium properties, and boundary conditions.
- Geometry & materialModel assumptionsExplain physical and reconstructed inputs.define the problem
- Numerical problemMesh & conditionsState boundary conditions and solver setup.solve and assess
- SolutionA computed resultReview sensitivity and compare with evidence.
Work through the example 3 STEPS
Imagine using a documented Cuntz pipe dimension as one input to a passive acoustic model. The 1281 mm observation alone does not describe the entire cavity or its excitation.
Build an explicit domain
Describe the modeled interior geometry and distinguish measured dimensions from inferred ones. A display mesh of the organ case is not automatically a suitable acoustic domain.
Specify the physical question
A passive eigenmode calculation asks which patterns and frequencies the specified domain can support under its boundary assumptions. It is not a full simulation of air flow, pipe speech and self-sustained sound production.
Evaluate the approximation
Check numerical resolution and compare with suitable independent measurements. Refining a mesh can address numerical approximation while leaving uncertain geometry or inappropriate boundary assumptions unresolved.
- Cuntz Positiv · VAO 0.5.0-rc.2 ↗
Published manifest: identities, measurement observations, realization digests, profiles and rights. Exact examples checked against this release.
- Dominik Ukolov · Musikinstrumente im virtuellen Raum (2026)
Submitted dissertation, 11 September 2026; §§8.2.2.1 and 8.4.1, printed pp. 264–265 and 275–276. The historical description and digital inventory require a resolved mapping; the manuscript is not redistributed here.
- VAO Standard 0.5.0 ↗
The standard contract is separate from the Cuntz dataset content version 0.5.0-rc.2.
This is a method explanation, not a new Cuntz finite-element result or a claim that the web model is acoustically validated.
The useful distinction. A numerical solution answers the equations and conditions supplied to it; interpretation requires checking whether those represent the research question.
Simplification can remove the feature that matters
A low polygon count is attractive for interactive playback, but visual economy and acoustic relevance do not always align. Section 8.5.1 reports that unweighted simplification below 50,000 polygons damaged features including labia, core gaps and cut-up edges in the investigated setting. A protected weighting of 100 belongs to that configuration; it is not a universal setting for every organ or reconstruction pipeline.
Retain a detailed research realizationRealizationOne exact byte sequence representing an asset. Different encodings or derivatives have different exact identities.Work through the Cuntz example → and derive a lighter presentation realization with a documented purpose. Check the features relevant to the question after simplification, not only the final file size. The two realizations can share an asset role while preserving their different bytes, derivations and intended uses. This makes a mobile visualization usable without silently replacing the geometric evidence.
PAMT: a proposed architecture for spatial acoustics
PAMT is presented as a conceptual architecture, not a field-validated room-acoustic product. The thesis discusses coupling geometric and wave-based methods because geometric propagation alone cannot represent all low-frequency modal behavior. The transition needs to depend on the room and frequency regime, with reference to the room’s Schroeder frequency, rather than a universal cutoff applied everywhere.
No in-situ sound-pressure measurement campaign is reported for PAMT in the dissertation. Its proposed field validation and uncertainty budgets remain prospective. This matters when selecting a tool: the architecture can guide research questions and implementation, but it does not supply measured accuracy for a visitor’s room. A published conceptual method, a working prototype and a calibrated acoustic model are different deliverables.