Exposure Bracketing / HDR Stops Planner mechanism

Model scope

The core exposure-duration formula is mathematically correct, but the instrument cannot be verified as fully specified until the frame-indexing convention, duration-versus-reciprocal input representation, frame-count constraints, and camera rounding behavior are defined.

• The multiplier formula base*2^(n*stopStep) is correct for exposure duration in seconds: each positive stop doubles the duration and each negative stop halves it.
• The specification does not define how n is assigned from number of frames. To produce brackets around the base exposure, n must be explicitly centered, such as -(frames-1)/2 through +(frames-1)/2.
• Even frame counts require fractional n values under centered indexing; this is valid mathematically but must be specified.
• The formula assumes shutter speed means exposure duration. If shutter speed is represented as the denominator in notation such as 1/125, applying the formula directly reverses the intended direction.
• The specification does not state whether the base exposure must be included, whether frames must be a positive integer, or whether one-frame and zero-stop-step requests are allowed.

Units and conventions

• base is an exposure duration, not a reciprocal shutter-speed denominator; all shutter durations must use one consistent unit.
• The test vectors use seconds or milliseconds consistently within each vector; multiplying by a dimensionless stop factor preserves that unit.
• A one-stop increase means multiplying exposure duration by 2; a one-stop decrease means multiplying by 1/2.
• The vectors assume centered n values ordered from darkest/shortest exposure to brightest/longest exposure.

Inspectable vector

One of 4 source-bound review vectors:

{
  "inputs": {
    "base_seconds": 0.008,
    "number_of_frames": 3,
    "stop_step": 1,
    "n_indices": [
      -1,
      0,
      1
    ]
  },
  "expected_outputs": {
    "shutter_seconds": [
      0.004,
      0.008,
      0.016
    ]
  },
  "working": "For n=-1: 0.008*2^(-1*1)=0.008*(1/2)=0.004 s. For n=0: 0.008*2^(0*1)=0.008*1=0.008 s. For n=1: 0.008*2^(1*1)=0.008*2=0.016 s."
}

Failure boundary

BREAK_CLAIMED · SELF_FLAGGED

The extreme valid-number input produces a non-finite Infinity output, and the specification does not define validation or rejection behavior. The unspecified frame-index and shutter-direction conventions can also produce reversed or differently centered bracket lists.

Other records

Sources and evidence

These pointers bind the page to evidence. They do not grant authority or collapse verification into deployment.

• VERIFIED_LEDGER.json
• evidence-tranche4/tranche4-results.json
• BOUNDARY_HAZARD_CATALOG.json