.de Sq
.ie n\\h'-\\w' 'u'^2
.el \\h'-\\w' 'u'\\u2\\d
..
.TH NETCHANGE 5 "" "" "IRSIM User's Manual"
.UC
.SH NAME
netchange \- format of netchange files read by irsim.

.SH DESCRIPTION
A \fBnetchange\fP file consists of a series of lines, each of which begins
with a key letter.
The key letter beginning a line determines how the remainder
of the line is interpreted.
The following are the list of key letters understood.
.TP
.B "| \fIany text\fR"
Lines beginning with a vertical bar are treated as comments and
ignored by the program.
.TP
.B "a\fPdd\fI type gate source drain length width [area]\fR"
Add a new transistor of
.I type
to the network.
Currently, \fItype\fP may be:
.RS
.PD 0
.RS
.IP \fBn\fP 4
n-channel enhancement transistor.
.IP \fBp\fP 4
p-channel enhancement transistor.
.IP \fBd\fP 4
depletion transistor (for NMOS).
.RE
.PD 1
.PP
The names of the nodes to which the gate, source, and drain of the
transistor are connected are given by
.I gate,
.I source,
and
.I drain
respectively.
The length and width of the transistor are given by
.I length
and
.I width
respectively.
The
.I area
parameter, if given, will use that number as the
area for calculating the gate capacitance.
.I Length
and
.I width
should be given in lambda units,
.I area
should be in lambda
.Sq 
units, these
will internally be multiplied by the LAMBDA factor from the configuration
(.prm) file.
.RE
.TP
.B "d\fPelete\fI type gate source drain length width [area]\fR"
Delete an existing transistor from the net.
All the parameters have the same meaning as for the \fBa\fPdd command.
.TP
.B "m\fPove \fItype gate source drain length width [area] g s d\fR"
Move an existing transistor to a new location the net.
.I type, gate, source, drain, length, width, and area
have the same meaning as for the \fBa\fPdd command.
.I g, s,
and
.I d
are the names of nodes to which the gate, source and drain should be
connected.  If a particular terminal(s) is not to be re-connected, the
name can be specified using an "*".  Any or all of
.I g, s,
and
.I d
may be "*".  For example, to move the gate of an n-channel transistor from
node
.I old
to
.I new
the following command would be used:
.HP
		\fBm n old src_node drn_node 4 2.2 new * *\fP
.IP
Note that the
.I drain
and
.I source
terminals, and the
.I g
and
.I s
terminals are interchangeable; the simulator will know if these are
swapped.  So the last example could also have been written:
.HP
		\fBm n old drn_node src_node 4 2.2 new * *\fP
.TP
.B "c\fPapacitance \fInode value\fR"
Change the capacitance of a node by
.I value
picofarads.  
.I Value
may be negative, thereby
decreasing the node's capacitance.
.I Node
is the node name.
.TP
.B "N \fInode metal-area poly-area diff-area diff-perimeter\fR"
Change the capacitance of
.I node
using the area and perimeter information of the metal, polysilicon,
and diffusion layers.
All the parameters should be in lambda (or lambda
.Sq
for areas) units, they will internally
be converted to the appropriate capacitance as defined in the configuration
file.  The values can be negative to decrease the capacitance.
.TP
.B "M \fInode M2A M2P MA MP PA PP DA DP PDA PDP\fR"
Change the capacitance of
.I node,
using the following geometrical information:
.PP
.RS
.PD 0
.RS
.IP \fIM2A\fP 6
area of 2nd-level metal
.IP \fIM2P\fP 6
perimeter of 2nd-level metal
.IP \fIMA\fP 6
area of 1st-level metal
.IP \fIMP\fP 6
perimeter of 1st-level metal
.IP \fIPA\fP 6
area of polysilicon
.IP \fIPP\fP 6
perimeter of polysilicon
.IP \fIDA\fP 6
area of n-diffusion
.IP \fIDP\fP 6
perimeter of n-diffusion
.IP \fIPDA\fP 6
area of p-diffusion
.IP \fIPDP\fP 6
perimeter of p-diffusion
.RE
.PD 1
.PP
All perimeter values should be in lambda units, area values should
be in lambda
.Sq
units.
The perimeter measures are half of the actual total perimeter (i.e.,
they are the sum of the lengths of the top and one side).
Again, the values may be negative to decrease the capacitance of
the node.
.RE
.TP
.B "t\fPhreshold \fInode low high\fR"
Change the threhsold voltages of
.I node.
.I Low
and
.I high
should be in normalized voltage units
(i.e. floating-point numbers in the range 0.0 to 1.0).
.TP
.B "D\fPelay \fInode tplh tphl\fR"
Change the delays for
.I node
to be
.I tplh
nanoseconds for low-to-high transitions, and
.I tphl
ns. for high-to-low transitions.  These should be absolute numbers,
not relative increments/dercrements.
.IP \fBNOTE:\fP 6
For all commands, only the first letter is significant, the rest
of the string will be ignored.  They are only shown here for clarity.

.SH BUGS
This is an experimental interface for the incremental simulator and
is very likely to change in the future.

.SH "SEE ALSO"
irsim(5) sim(5) presim(1)