Blender

1. Overview

3D creation suite with a built-in Python interface and a binary project format.

Category: Graphics & 3D.

2. Version and scope assessed

Version assessed: Source commit ded7fde71c04f034d6c065e8d0a75d9f548003a0; published SDNA layout and Pack All coverage.

4. Principle profile

5. Fundamental requirements

6. Evidence and verification

• E1: Blender saves projects as local .blend files and documents Save Copy, recovery, and packed data operations. (source), confidence medium
• E2: Pack All packs supported file/tiled images, eligible external fonts, sounds and volumes, and handles geometry-node bakes on non-linked objects. Movie/image-sequence image sources are skipped, with a warning on the verbose path; linked-object bake processing is skipped. The command is not proof that every external dependency has become self-contained. (source), confidence medium
• E3: The source documents and interprets the SDNA sections NAME, TYPE, TLEN and STRC, exposing member names, types, sizes and structure records. The repository also retains a historical blend-format explanation and reader tools targeting Blender 2.48/2.5-era files. Format knowledge is available, but neither those historical tools nor this inspection establish reconstruction of every current blend file or scene semantics. (source), confidence medium
• E4: Project creation, editing, saving, and export are local operations and do not inherently require a vendor account. (source), confidence medium
• E5: The application documents recovery and undo-related workflows, but destructive operations, scripts, and external dependencies require hands-on review. (source), confidence medium
• sovereignty: A positive Sovereignty finding requires established E1, E2 and E4 in the declared scope. The existing access/exit/custody record for Blender does not establish all three minimums. The prior positive finding is withdrawn; recorded storage and portability strengths remain facts, not a verified failure. (source), confidence medium
• longevity: E3 is not established for Blender's existing essential-record scope. The previous Longevity finding is replaced by Not assessed, not a verified failure or a positive placeholder. (source), confidence medium
• locality: Core modeling and rendering workflows operate locally; network features are not required for ordinary project editing. (source), confidence medium
• transparency: The positive control observations above do not yet establish whole-scope E5. The methodology requires that minimum before a positive Transparency and Control finding; the previous positive finding is therefore withdrawn to Not assessed, not changed to Does not meet. (source), confidence medium
• E3: Published SDNA definitions explain binary structure layout, while Pack All deliberately skips some external image/movie/linked bake dependencies. Neither a current scene was independently reconstructed nor its complete assets, scripts, node semantics and compatible runtime retained and compared. Available format knowledge is not complete scene survival. (source), confidence medium
• E5: Undo/recovery documentation and pack warnings do not establish effective Python/driver/plugin authorization, save/export overwrite control or correction across asset packing and destructive scene changes. Known pack exclusions are dependency facts, not proof of governable execution. (source), confidence medium

7. Migration and recovery

Native storage format: Binary .blend files, with Python scripting and open export formats.

Synchronization or runtime model: Local desktop application.

Example export or inspection command (not verified against any particular version or data set): blender -b scene.blend --python-expr "import bpy; print([o.name for o in bpy.data.objects])"

8. Architectural strengths

Established strengths and their scope are recorded in the evidence and assessment findings on this page.

9. Architectural limitations

• Published SDNA layout does not by itself recover geometry, dependencies or editable scene meaning.
• Pack All has explicit exclusions and is not a complete dependency inventory.

10. Assessment findings and remaining work

Primary investigation: Long-term preservation of a complex creative project and its dependencies.

Investigation priority: Sovereignty standard, Longevity high, Locality standard, Transparency and Control high. This plans the test effort and is not a finding.

Synthetic dataset: A scene with meshes, procedural modifiers, textures, materials, animation, linked libraries, external media, and rendered output.

Required tests

1. Save the native .blend project and inventory every external dependency.
2. Pack supported resources and identify anything that remains outside the project.
3. Transfer the project to a clean offline machine.
4. Export common interchange formats and measure loss of modifiers, rigs, material behavior, or animation.
5. Open or convert project data using a second compatible tool or documented procedure.
6. Test missing linked assets and library recovery.
7. Assess whether consequential destructive operations have understandable undo, recovery, or versioning paths.

Critical evidence: Dependency inventory, project reopen results, format-loss comparison, recovered scene structure, and rendered-output comparison.

Documentation to verify: Blender’s documentation is said to warn that not all external resources can be packed into a .blend file. A project may look self-contained and still depend on external assets.

Evidence update: 23 October 2026

• Resolved: The source documents and interprets the SDNA sections NAME, TYPE, TLEN and STRC, exposing member names, types, sizes and structure records. The repository also retains a historical blend-format explanation and reader tools targeting Blender 2.48/2.5-era files. Format knowledge is available, but neither those historical tools nor this inspection establish reconstruction of every current blend file or scene semantics. Source.
• Resolved: Pack All packs supported file/tiled images, eligible external fonts, sounds and volumes, and handles geometry-node bakes on non-linked objects. Movie/image-sequence image sources are skipped, with a warning on the verbose path; linked-object bake processing is skipped. The command is not proof that every external dependency has become self-contained. Source.
• Open: An actual current-version blend-file interpretation, complete offline scene/dependency rebuild, pack/move comparison, editable export losses, version compatibility and script/undo safeguards. E5 needs effective script/driver/plugin trust, overwrite and destructive-edit controls plus recoverable scene state.

Assessment follow-up: These findings now appear in the application’s evidence and limitations. At the user’s request, existing eligibility and principle ratings are unchanged; E3/E5, principle consistency and blank-status fixes will be reconciled in the next pass. Reproduction details and retained inputs are in evidence/third-pass/batch-5/.

Human-readable criteria: Can a creator preserve editable source work and linked assets, reopen it after software changes, and export a useful representation to independent tools?

Rating reconciliation: 24 October 2026

E3 and E5 remain Not Verified for the existing declared scope. Longevity and Transparency and Control are Not assessed pending their required minimums. Sovereignty is also Not assessed because the required access/exit/custody minimums are not all established. This is a targeted evidence decision, not a finding that the product fails. Existing E1, E4 and Locality findings were not re-certified.

Remaining: An actual current-version blend-file interpretation, complete offline scene/dependency rebuild, pack/move comparison, editable export losses, version compatibility and script/undo safeguards. E5 needs effective script/driver/plugin trust, overwrite and destructive-edit controls plus recoverable scene state.

Next without installing locally: Obtain a complete synthetic scene/dependency manifest, native source and interchange/reference outputs with expected object/node/animation semantics; inspect preserved files and reuse version-matched script/undo/save observations.

The individual decision and provenance are retained in evidence/rating-reconciliation/batch-6/. Overall: Insufficient evidence.

Individually reconciled in rating batch 6 on 24 October 2026. E3/E5 remain Not Verified for explicit missing coverage; dependent unsupported findings corrected. Previously resolved questions are retained, not counted again. E1/E4/Locality were not re-certified by this targeted review. The assessment procedure is described in the methodology.