Advanced

Settings and Logs

Interface preferences, performance limits, log visualization, and pre-run warnings

This chapter covers three groups of functionality: the Settings page for interface behavior and frontend protection limits; the Logs page for tracking calculation, sweep, and optimization execution; and the pre-run warning dialogs that intercept invalid configurations or overloaded tasks when clicking Run / Sweep / Optimize.

Settings Page

The Settings page is divided into interface preferences, performance limits, and reset. All settings are automatically persisted to browser localStorage.

Settings page

Interface Settings

Item	Options	Default	Description
Language	`English` / `中文`	`English`	Switch UI copy; also affects documentation entry language
Appearance	`System` / `Light` / `Dark`	`System`	`System` follows the operating system dark mode state
Tooltips	Toggle	Enabled	Control whether field descriptions appear on hover
Local Auto Save	Toggle (`Beta`)	Enabled	Automatically save the current model to browser local storage; disabling requires confirmation and clears the saved local draft

Local Auto Save is marked Beta. It is useful for loss protection, but should not replace explicit case export and version tracking.

Performance Settings

Performance settings

Item	UI Range	Default	Direct Effect
Concurrent Requests	`1 ~ 32`	`4`	Parallel request count for Sweep and Optimizer
Max Sweep Combinations	`1,000 ~ 10,000`	`4,000`	Blocks sweep submission when exceeded
Max Calculation Data Points	`100,000 ~ 20,000,000`	`2,000,000`	Limits total result data points for one run (all sweep requests combined)
Maximum Data Points per Request	`100,000 ~ 1,000,000`	`1,000,000`	Limits result data points for each individual request in a sweep run. Hard cap is 1,000,000

The performance settings exist to prevent the frontend from becoming overloaded by large sweeps or oversized result sets:

They control what the frontend is allowed to submit and render.
They do not modify the governing equations of a single Run.
If a task is blocked, reduce parameter range, increase wavelength step, or reduce depth sampling before raising the limits.

Maximum Data Points per Request has a hard cap of 1,000,000 defined by the system constant MAX_SINGLE_REQUEST_DATA_POINTS and cannot be exceeded in Settings. Max Calculation Data Points has a hard cap of 20,000,000.

Reset

Click the Reset button to restore all settings to their defaults. A confirmation dialog is required; the action cannot be undone.

Logs Page

Logs is an execution monitor for tracking the run process, diagnosing failures, and assessing task scale.

Logs page

Element	Function
Log count badge	Displays the total number of log entries
Latest status indicator	Shows the status icon and text of the most recent log
Timeline / Matrix toggle	Select visualization mode; deselect both to collapse the visualization panel
Top / Bottom buttons	Scroll to the top or bottom of the log list
Export dropdown	Export logs as JSON or CSV
Clear button	Clear all logs (requires confirmation)

Timeline Visualization

Timeline visualization

The Timeline view displays the execution process of the latest task on a time axis.

Component	Description
Time axis	Top row with time grid markers; time range is computed automatically
Prep row	Shows a summary bar for preparation phases (RI Check, Structure Prep, RI Inject, Dispatch)
Core lanes	One row per concurrent core, showing request execution bars
Status dots	Green (success), red (error), orange (terminated), blue blinking (running), gray (idle)

For optimization tasks, the Timeline uses fallback bars instead of per-request display. Each lane represents a Grid row or a Seed row.

Matrix Visualization

Matrix visualization

The Matrix view displays request status distribution as a grid.

Standard mode (Run / Sweep)

Component	Description
Summary bar	Cloud Compute Cores, Active / Idle Cores, Completed / Total Requests, Data Points
Prep row	RI Check, Structure Prep, RI Inject, Dispatch phase cells
Request grid	One cell per request, filled row by row; cell size shrinks automatically with total request count

Optimization mode

Component	Description
Summary bar	Completed Grid Points / Total Grid Points, Completed Seeds / Total Seeds
Prep row	Same as standard mode
Grid row	Shows grid search progress (when enabled)
Seed rows	One row per seed, showing local optimization progress

Cell colors

Color	Status
Blue (pulse animation)	Running / pending
Green	Success
Red	Error
Orange	Terminated
Gray	Idle / not started

Log Entries

Each log entry records one calculation, sweep, or optimization task.

Field	Meaning
Status icon and text	`running` / `success` / `error` / `terminated`
Timestamp	Task creation time
Data Points tag	Total result data points
Total Requests tag	Total request count
Cloud Compute Cores tag	Concurrent core count
Sweep Items / Sweep Parameters	Sweep metadata (combination count, parameter labels)
Duration	Task runtime
Inline phase events	Real-time Prep and Compute phase events shown while running
Expandable details	Phase events and error details available after completion

Log entries have a colored left border: blue (running), green (success), red (error), orange (terminated).

Log Export

Format	Content	Best Use Case
JSON	Full structured log payload	Preserve complete context for debugging
CSV	Flattened log table	Count runs, durations, and status distribution

Log export captures the execution process, not the optical result data.

Pre-Run Warnings

When clicking Run / Sweep / Optimize, the system performs pre-checks. The following dialogs may appear depending on the results.

Validation Dialog

When the model has validation errors, a Cannot Run dialog appears.

Element	Description
Title	Context-specific message ("Fix the issues below before running / starting Sweep / starting the optimizer")
Issue list	Table listing all issues: location (Structure / Optics / Sweep / Optimizer), key, and error message
Click on entry	Navigates to the page where the issue is located
Go to First Issue button	Jumps to the page of the first issue

For optimization tasks, additional checks verify that at least one objective is enabled and at least one variable is checked.

Sweep Combination Limit Exceeded

When sweep parameter combinations exceed the Max Sweep Combinations setting, a Too Many Sweep Combinations dialog appears.

Element	Description
Current combinations	The computed combination count
Settings limit	The configured upper limit from Settings
Go to Sweep button	Navigate to the Sweep page to reduce combinations

Compute Load Warning

Appears when the TMM solve count for a single request exceeds the solver core capacity (MAX_SINGLE_REQUEST_TMM_SOLVE_COUNT = 50,000).

Element	Description
Capacity exceeded banner	Red message indicating the computation load exceeds TMM solver core capacity
TMM Solve Count per Request panel	Shows Current value and Limit value
Main Contributors	Lists primary factors (e.g., "wavelength 101 x polarization 2 x cone-angle rays 7 = 1,414")
Suggestions	Recommended actions: reduce wavelength range, increase wavelength step, reduce cone-angle sampling, lower dispersion quality, disable unneeded depth detectors

The TMM solve count limit is a system safety guard and is not user-adjustable in Settings. It targets actual computation load rather than frontend rendering volume. For example, enabling cone angle significantly multiplies the solve count for intensity detectors, even when the returned result point count is modest.

Result Data Volume Warning

Appears when total data points or per-request data points exceed limits.

Element	Description
Warning banner	Yellow warning for soft limit, red error for hard limit
Request Count	Total request count (shown for sweep runs)
Total Result Data Points panel	Shows Current, Settings Limit, and Hard Limit values
Result Data Points per Request panel	Shows Current, Settings Limit, and Hard Limit values
Peak request summary	Identifies the sweep parameter combination with the highest data volume
Suggestions	Recommended actions to reduce data volume
Open Settings button	Navigate to Settings to adjust limits

Soft limit vs. hard limit

Category	Condition	Behavior
Soft limit (Settings limit)	Exceeds the Settings threshold but stays under the hard cap	Continue button is enabled; the user can choose to proceed
Hard limit	Exceeds the system hard cap (per-request 1,000,000 / total 20,000,000)	Continue button is disabled; parameters must be reduced first

Troubleshooting Sequence

When a run does not produce the expected result, follow this order:

Check the latest Logs entry status and message to confirm the task type.
Review Data Points / Total Requests / Cloud Compute Cores to determine whether the failure is scale-related.
If the status is error, expand details and read the failure message.
If a pre-run warning dialog appeared, use the dialog content to locate the problem source.
Return to Structure, Optics, Sweep, or Optimizer to correct the model or parameters.

Data Import and Export

Material files, spectra, and result export

MCP

What MCP is, how it helps AI understand Dreapex TMM, and how to add the Dreapex TMM MCP in mainstream AI tools.