|
Ccif adds the layout description of one or more
cells (symbols in CIF) to the ICD database. If a cell
description is read, ccif creates a new cell
directory, and translates the CIF cell description into an
internal database format.
The cell descriptions are read from the specified CIF
file, which must be the last argument. In most cases a file
with suffix ".cif". The input must be conform the
syntax rules of CIF. If not, a syntax error will be the
result, and an appropriate error message will be
generated.
IMPORTANT: ccif requires that cells (symbols) that
are called in the CIF file are defined earlier in the
file.
NEW: ccif can use the comment after a
'C0'-call to set the cell name, if symbol0 is
not earlier defined in the file. Thus, you can use library
cells not defined in the CIF file.
To help you in making a correct cell definition you can
invoke ccif with option -s. In that case ccif
performs only a syntax check, and two kinds of semantic
checks, i.e.:
1) a cell may not be redefined,
2) a cell called in a cell description must already exist.
With the origin mode (option -x) the lower-left corner of
the bounding box of a sub-cell is used as the translation
vector reference point. Note that, if the bounding box
information of sub-cells in the database is incorrect, this
sub-cells can be incorrectly placed.
If an error is detected the cell will not be added to the
database. This can result in subsequent error messages,
complaining about the fact that the cell does not exist.
The unit of the parameters is assumed to be in centi
microns. Since the unit of the database is lambda, each
parameter is multiplied by a resolution value:
new-parameter = parameter * resolution
resolution = (A/B) * (1/cifunit) * (1/projectlambda)
where A and B are the scaling factors of the Definition
Start statement, and cifunit is 100 (default). Thus, with a
cifunit of 100 and a projectlambda of 0.01 micron the
scaling (resolution) used is A/B.
|
