dieopt

dieopt is a Python package for optimal die placement on circular wafers, supporting semiconductor layout and process engineering. It provides fast, flexible algorithms to maximize die-per-wafer (DPW) yield, visualize placements, and compare strategies.

---

Features

• Three fixed-offset strategies: Center, half-offset, and full-offset placement modes.
• Optimal DPW calculation: Quickly compute the best die arrangement for given wafer and die dimensions.
• Visualization: Easily plot wafer layouts and compare placement strategies.
• Simple API: Functional interface for both quick overviews, and GDS-II files.

---

Installation

pip install dieopt

---

Quickstart

from dieopt import get_solution, show_solution

# Visualize all three placement strategies
show_solution(
    wafer_diameter=50.8,
    edge_exclusion=2.0,
    width=1.0,
    height=1.0,
    scribe=7.0,
    solution="comparison"
)

# Get coordinates for the optimal placement
coords = get_solution(
    wafer_diameter=50.8,
    edge_exclusion=2.0,
    width=1.0,
    height=1.0,
    scribe=7.0,
    solution="optimal" # or "center", "half_offset", "full_offset"
)
print(coords)

>>> [(-12.0, -16.0), (-4.0, -16.0), (4.0, -16.0), (12.0, -16.0), (-20.0, -8.0), (-12.0, -8.0), (-4.0, -8.0), (4.0, -8.0), (12.0, -8.0), (20.0, -8.0), (-20.0, 0.0), (-12.0, 0.0), (-4.0, 0.0), (4.0, 0.0), (12.0, 0.0), (20.0, 0.0), (-20.0, 8.0), (-12.0, 8.0), (-4.0, 8.0), (4.0, 8.0), (12.0, 8.0), (20.0, 8.0), (-12.0, 16.0), (-4.0, 16.0), (4.0, 16.0), (12.0, 16.0)]

---

API Reference

get_solution(...)

Compute die placements for a wafer.

Parameters:

• wafer_diameter (float): Wafer diameter in mm.
• edge_exclusion (float): Edge exclusion in mm.
• width (float): Die width in mm.
• height (float): Die height in mm.
• scribe (float): Scribe street width in mm.
• solution (str): "center", "half_offset", "full_offset", or "optimal".

Returns:
List[Tuple[float, float]] — Die center coordinates.

---

show_solution(...)

Visualize die placements on a wafer.

Parameters:

• Same as get_solution
• solution (str): "center", "half_offset", "full_offset", "comparison", or "optimal"

Returns:
Shows a matplotlib plot of the wafer and die placements.

---

Placement Modes

• center: Dies placed with grid centered on wafer.
• half_offset: Grid offset by half a die.
• full_offset: Grid offset by a full die.
• optimal: Automatically selects the best DPW.
• comparison: Visualizes all three strategies side-by-side.

---

License

Mozilla License

---

Authors

• Bjorn Funch Schrøder

---

Acknowledgements

Adapted from my Masters Thesis: "Site controlled epitaxy of quantum dots for nano- and quantum photonic applications in telecom wavelength range".