4096evidence-ladder rowsUNKNOWNInspect metric contract
10link-only source anchorsUNKNOWNInspect metric contract
0real measurement rowsUNKNOWNInspect metric contract
SIMULATION / SA-EDA-VSLICE-004
Open-silicon model-to-measurement evidence ladder
Model-to-measurement ladder
A browser replay that separates model assumptions, cell characterization, implementation evidence, test-structure intent, measurement availability, discrepancy review, provenance, uncertainty, and claim maturity without borrowing an external project's data.
EXPLORE
Start with the observable.
Use this page when the simulation is the primary artifact. The linked lesson explains the concept; the linked lab keeps the visual replay surface available.
- Filter model basis, characterization state, implementation checkpoint, test structure, measurement state, and review role before reading readiness or maturity.
- Separate missing evidence, empty owned replacement slots, pinned external author-claim metadata, and explicit synthetic teaching pairs.
- Use discrepancy, provenance, traceability, uncertainty, contradiction, and decision-gate fields to name the next safe evidence check.
- Keep the boundary visible: no row is a reproduced repository result, measured-silicon claim, signoff result, or endorsement.
BROWSER INTERACTIVE LAB
Filter the saved CSV. Inspect matching replay rows.
No server solver is running here. The browser filters a precomputed table and redraws the plot locally.
Loading replay rowsReading the saved CSV in this browser.
evidence completeness scoreprovenance strength scoretraceability scorecondition alignment scoreuncertainty pressureclaim maturity scorecross stage contradiction count
Each metric is normalized independently for shape inspection. Vertical positions do not compare magnitude across metrics.
n/aevidence completeness score
n/aprovenance strength score
n/atraceability score
n/acondition alignment score
n/auncertainty pressure
n/aclaim maturity score
n/across stage contradiction count
Change one or more controls to find a saved row. No unrelated CSV value is substituted.
| case id | model basis label | characterization state label | implementation checkpoint label | test structure label | measurement state label | review role label | predicted delay proxy ps | synthetic observed delay proxy ps | relative discrepancy pct | evidence completeness score | provenance strength score | condition alignment score | uncertainty pressure | comparison readiness score | claim maturity score | discrepancy class | decision gate | next safe check |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Loading saved replay rows... | ||||||||||||||||||
RUN CARD
Conditions before conclusions.
The browser filters a saved CSV. It does not run or accept ngspice, Yosys, OpenROAD/OpenLane/LibreLane, OpenSTA, OpenRCX, PDKs, designs, reports, test logs, or measurements.
Primary knobs
- model basis
- characterization state
- implementation checkpoint
- test structure
- measurement state
- review role
Saved outputs
- synthetic predicted delay and ring-frequency proxies
- synthetic observed comparison fields where explicitly enabled
- model, arc, STA, route, PV, extraction, test, and measurement coverage
- provenance, traceability, condition alignment, and uncertainty pressure
- cross-stage contradictions and discrepancy class
- comparison readiness and bounded claim maturity
- decision gate, evidence state, role readout, and next safe check
EXPORT
Download the public artifacts.
These exports are the supported public files for review. PDK files and restricted third-party material are not redistributed.
Model-to-measurement evidence CSV
/data/open-silicon-model-to-measurement-evidence-ladder-v93.csv
Open file ->Typed web JSON
/data/open-silicon-model-to-measurement-evidence-ladder-web-v1.json
Open file ->Model card
/data/open-silicon-model-to-measurement-evidence-ladder-model-card-v1.md
Open file ->METHODS
How the number was made.
Methods are written for citation discipline: the extraction rule matters as much as the plotted value.
Pin the public repository observation to one commit and preserve every repository statement as an author claim, not as a SemiAgora result.
Keep this method attached when reusing the figure or metric.
Generate a deterministic 4 x 4 x 4 x 4 x 4 x 4 synthetic evidence-state sweep with zero copied external numbers and zero real measurements.
Keep this method attached when reusing the figure or metric.
Compute bounded evidence coverage, provenance, traceability, condition alignment, uncertainty, discrepancy, contradiction, readiness, maturity, gate, and next-check fields.
Keep this method attached when reusing the figure or metric.
Hash the local daily triage, structured backlog, factory proof, and resource-candidate files so the source-review decision survives cross-device handoff.
Keep this method attached when reusing the figure or metric.
LIMITS
What this page does not claim.
The MVP is useful because the limits are visible. These statements prevent a teaching simulation from being cited as silicon evidence.
No repository mirroring, source-code reuse, GDS, Liberty, PDK, model deck, report body, screenshot, figure, or external measurement-table copying.
Carry this boundary into any derivative note, paper draft, or slide.
Repository statements are preserved only as pinned author-claim metadata and are not independently reproduced or endorsed by SemiAgora.
Carry this boundary into any derivative note, paper draft, or slide.
No live ngspice, Yosys, OpenROAD, OpenLane, LibreLane, OpenSTA, OpenRCX, Magic, Netgen, KLayout, or Tiny Tapeout flow execution.
Carry this boundary into any derivative note, paper draft, or slide.
No uploaded design, model, library, PDK, netlist, GDS, DEF, LEF, SDC, SPEF, Liberty, report, bring-up log, or measurement parsing.
Carry this boundary into any derivative note, paper draft, or slide.
All numeric prediction and comparison fields are original synthetic teaching proxies; the replay contains zero real measurement rows and zero external numeric results.
Carry this boundary into any derivative note, paper draft, or slide.
No tapeout or fabrication service, signoff certification, measurement claim, silicon-performance claim, product claim, foundry claim, or project endorsement.
Carry this boundary into any derivative note, paper draft, or slide.
EVIDENCE
Trace the claim back to artifacts.
The evidence route remains the catalog-level browser; this page is the simulation-level reading card.
4096 original synthetic browser rows with explicit zero-real-measurement and zero-external-number contracts.
SA-EDA-VSLICE-004
4 hashed local source-review records plus a pinned external repository observation at commit 318eac27ebedc990af5680c8a969164709545eca.
SA-EDA-VSLICE-004
Typed JSON, model card, canonical experiment folder, source observation, source-review lineage, and link-only source spine.
SA-EDA-VSLICE-004
No external repository file, numeric result, design collateral, report body, PDK, library, screenshot, figure, or measurement table is copied.
SA-EDA-VSLICE-004
Source spine
These links anchor the public teaching model. They do not convert the replay into a qualified process recipe or signoff result.
Vertical-slice repository snapshotPinned primary repository snapshot used only as a link-first evidence-chain benchmark. Repository statements remain author claims until independently reproduced.Vertical-slice Apache-2.0 licenseLicense observation for the pinned repository snapshot. SemiAgora still mirrors no repository code, GDS, Liberty, reports, PDK data, or measurements.ngspice official documentationOfficial simulator documentation anchor for model, netlist, transient, and characterization vocabulary; no deck or output is copied.Yosys official documentationOfficial RTL synthesis and technology-mapping vocabulary anchor; no design, script, netlist, or library is ingested.OpenROAD Flow Scripts tutorialOfficial RTL-to-GDS checkpoint and report-directory vocabulary anchor; report values and examples are not copied.OpenSTA official documentationOfficial STA input, timing-check, and report vocabulary anchor; no Liberty, SDC, SPEF, netlist, or timing report is copied.OpenRCX official documentationOfficial parasitic-extraction vocabulary anchor; no extraction rules, SPEF, design database, or measured RC is copied.Tiny Tapeout official guidesOfficial public guide index for hardening, bring-up, documentation, and measurement context; no project asset or test result is copied.SkyWater SKY130 PDK documentationOfficial open-PDK status and boundary anchor. SemiAgora redistributes no PDK, model, rule deck, cell library, or process content.NIST measurement uncertaintyAuthoritative measurement-model and uncertainty context for comparison readiness; no external measurement table is copied.
Public execution boundary
No server-side live ngspice execution, uploads, accounts, payments, comments, newsletter signup, or job/event submissions are active on this page.