Understand yield estimates in vFactory

Read the defect-density inputs and the three model probabilities in a yield estimate.

Defect density

In Cloud vFactory, “Geometry and yield estimate” brings together wafer geometry, defect-density assumptions and yield probabilities.

Understand yield estimates in vFactory

Read the defect-density inputs and the three model probabilities in a yield estimate.

Model assumptions

  • 1. User D0 (defects/cm²) · Assumption
  • 2. Environment D0 (defects/cm²)
  • 4. Die area (cm²)
  • 5. Clustering α

Total D0 (defects/cm²)

  • 3. Total D0 (defects/cm²)

Yield probabilities

  • 6. Poisson probability
  • 7. Murphy probability
  • 8. Negative binomial probability

Connections

  • 1. User D0 (defects/cm²) · Assumption → 3. Total D0 (defects/cm²): +
  • 2. Environment D0 (defects/cm²) → 3. Total D0 (defects/cm²): +
  • 3. Total D0 (defects/cm²) → 6. Poisson probability: D0
  • 3. Total D0 (defects/cm²) → 7. Murphy probability: D0
  • 3. Total D0 (defects/cm²) → 8. Negative binomial probability: D0
  • 4. Die area (cm²) → 6. Poisson probability: A
  • 4. Die area (cm²) → 7. Murphy probability: A
  • 4. Die area (cm²) → 8. Negative binomial probability: A
  • 5. Clustering α → 8. Negative binomial probability: α

“User D0 (defects/cm²) · Assumption” and “Environment D0 (defects/cm²)” add up to “Total D0 (defects/cm²)”. D0 is expressed in defects/cm². The environment contribution is calculated from the cleanroom assumptions.

Three models

“Poisson probability”, “Murphy probability” and “Negative binomial probability” use the same die area in cm² and total D0. The negative binomial model also uses α to control clustering. Compare the three probabilities with the assumptions in mind.

vFactory

The estimates are model outputs from stated assumptions, not measured yield, certification or a prediction for a real fab.