243synthetic DOE rowsUNKNOWNInspect metric contract
4factor classesUNKNOWNInspect metric contract
51candidate windowsUNKNOWNInspect metric contract
0live server runsUNKNOWNInspect metric contract
SIMULATION / SA-PROC-ETCH-UQ-002
Plasma etch heteroscedastic UQ replay
Etch UQ replay
A browser CSV replay that separates mean etch response, spatial uncertainty, CD-loss risk, selectivity proxy, and acquisition priority without exposing a process recipe.
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 wafer zone, pressure class, bias class, polymer balance, and mask opening to separate mean etch behavior from uncertainty behavior.
- Compare rate proxy, CD-loss proxy, selectivity proxy, sigma, profile taper, and objective score before interpreting any candidate window.
- Use acquisition priority as a teaching label for what to measure next, not as a fab recipe or equipment instruction.
- Open the model card before reusing a row in a paper note, slide, or lesson.
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.
objective scorecd loss proxy nmsigma rate nm min
Each metric is normalized independently for shape inspection. Vertical positions do not compare magnitude across metrics.
n/aobjective score
n/acd loss proxy nm
n/asigma rate nm min
Change one or more controls to find a saved row. No unrelated CSV value is substituted.
| case id | wafer zone | pressure class | bias class | polymer balance | mask opening nm | rate proxy nm min | cd loss proxy nm | sigma rate nm min | objective score | acquisition priority | label |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Loading saved replay rows... | |||||||||||
RUN CARD
Conditions before conclusions.
The browser filters saved CSV rows and redraws a local plot. It does not run a plasma model, expose gas ratios, accept uploads, or issue process settings.
Primary knobs
- wafer_zone
- pressure_class
- bias_class
- polymer_balance
- mask_opening_nm
Saved outputs
- rate_proxy_nm_min
- cd_loss_proxy_nm
- selectivity_proxy
- sigma_rate_nm_min
- profile_taper_deg
- objective_score
- acquisition_priority
- label
EXPORT
Download the public artifacts.
These exports are the supported public files for review. PDK files and restricted third-party material are not redistributed.
Interactive CSV rows
/data/plasma-etch-heteroscedastic-uq-replay-v76.csv
Open file ->Web result JSON
/data/plasma-etch-heteroscedastic-uq-replay-web-v1.json
Open file ->Model card
/data/plasma-etch-heteroscedastic-uq-replay-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.
Generate a deterministic 243-row factor-class grid over wafer zone, pressure class, bias class, polymer balance, and mask opening.
Keep this method attached when reusing the figure or metric.
Compute original proxy metrics for etch-rate response, CD-loss risk, selectivity, spatial sigma, profile taper, objective score, acquisition priority, and outcome label.
Keep this method attached when reusing the figure or metric.
Keep class labels and model-card limits visible so the replay teaches UQ structure without exposing process recipes or equipment actions.
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.
The factor classes are teaching abstractions, not recipe settings, gas ratios, tool conditions, or chamber instructions.
Carry this boundary into any derivative note, paper draft, or slide.
The rows are synthetic public teaching data, not measured wafer data, copied paper data, process qualification evidence, or yield evidence.
Carry this boundary into any derivative note, paper draft, or slide.
No live server solver execution, uploaded design ingestion, equipment control, ATE programming, or fab recommendation is enabled.
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.
Daily factory run 20260819_135955 passed with 108 candidates, 101 endpoint rows, 0 errors, and 0 warnings.
SA-PROC-ETCH-UQ-002
The public CSV and JSON are generated from an original deterministic teaching surrogate with 0 live server runs.
SA-PROC-ETCH-UQ-002
The arXiv paper remains a methodology anchor only; no paper data, plots, or text are copied into the replay.
SA-PROC-ETCH-UQ-002
Source spine
These links anchor the public teaching model. They do not convert the replay into a qualified process recipe or signoff result.
Public execution boundary
No server-side live ngspice execution, uploads, accounts, payments, comments, newsletter signup, or job/event submissions are active on this page.