Shrinking geometries, new manufacturing paradigms, exploding file sizes…
It’s time to rethink everything in dummy filling.
XYALIS increases the productivity, reliability, and capability of dummy fill at the reticle level with GOTfiller, a redesigned fill engine for CMP (chemical mechanical planarization/polishing).
GOTfiller is a grid based dummy fill engine optimized to increase manufacturing yield of multi-chip assemblies or a matrix of dies.
GOTfiller inserts user defined dummy tiles or full layers in the empty areas of the assembly or a matrix of dies. Also, optionally leaves sawing lines free of fill. It does not insert dummy tiles inside the chips instantiated in the multi-chip assembly or a matrix of dies.
GOTfiller takes into account all objects at the top level of the multi-chip assembly hierarchy. It identifies empty areas in the assembly before inserting dummy tiles or full layers.
GOTfiller benefits from powerful features of GOTstyle such as:
- non orthogonal grids
- 3D dummies interconnection capabilities
- dummies merging
- multiple pass for dummies mixing
- Scripting capabilities thanks to an embedded Python interpreter environment. Dedicated classes and methods allow to easily customize a dummy map and instantiate various dummy cells.
The user provides the rules to GOTfiller to perform dummy tiles or full layer generation. He provides these rules through command line options and parameters at launch time, or through a rule file.
GOTfiller does not modify the input assembly file. It generates a GDSII or OASIS ® layout file. This file only contains the inserted fill, and which needs to be merged later with the layout of the input assembly.
GOTfiller dummy filling workflow
GOTfiller generates a GDSII or OASIS ® output file which only contains the instantiations of the dummy tile or full layers. If needed, GOTfiller generates gzipped (.gz) outputs.
Dummy filling in MPW project
GOTfiller is fully integrated in XYALIS Mask Data Preparation solution and handles standard layout formats: GDSII and OASIS ®.
- Maximize CMP yield with highly accurate tile insertion,
- Easy setup.
- Insert any Type of Dummies
- High Performance and Low Memory Usage
- Dummy tile insertion or full layer generation
- Scripting capabilities thanks to an embedded Python interpreter environment