Table of Contents
ploterps_manual - ploterps User’s Manual (ERP manuals)
Ploterps User’s Manual
Ronald Ohst
University of California, San Diego
1. INTRODUCTION
Example 1 - a simple plot
2. USAGE
2.1 Plotting Devices
2.2 Plotting Without a Command File
2.3 Command Line Options
2.4 General Command File Format
3. COMMAND FILE ARGUMENT AND PARAMETER CONVENTIONS
3.1 Attributes
3.2 Coordinate System
4. COMMANDS
4.1 Style Commands
4.3 Commands Related to ERP Data
4.4 Plotting Commands
4.4.1 Plotting Calibration Lines
5. EXAMPLES
Example 2 - quick and dirty approach (no command file)
Example 3 - timeline and calibration bar
Example 4 - several bins from one channel
Example 5 - shading between waveforms, and labeled arrows
Example 6 - standard error
Example 7 - labpos
Example 8 - drawvoltline
Example 9 - windowtext
Example 10a - linetypes (solid, dashed, ldashed) and linewidths
Example 10b - linetypes (solid, dotted, dotdash) and linewidths
Example 11 - colors
6. APPENDIX A - List Of Commands
7. APPENDIX B - List of Predefined Strings
8. APPENDIX C - Input Naming Conventions For Special Characters
ploterps is a program that is designed to allow quick and easy plotting
of ERP data. By expending
more effort, it is also possible to produce
plots in almost any format; if the output device is of
sufficiently high
quality, publishable figures can be drawn using ploterps.
ploterps provides the user with about 50 commands to control the manner
in which data are
plotted, but it is possible to plot data without using
any commands at all. A series of these
commands is placed in a "command
file" which can be generated using any text editor. With a
little practice,
it’s relatively easy to produce high quality ERP data plots.
A number of example command files are given in this manual. For each example,
the command file as shown also exists on the linux system, and the reader
is
encouraged to view the plot it generates. Simply enter the command
line given
with the example at any xterm window’s command line prompt (a
cut-n-paste will
do) and the plot will appear in a new window on the display.
To remove the plot
move the mouse cursor over top of it and hit ESC.
Example 1 - a simple plot
This simple plot is meant to give the reader a feel for what command files
generally
look like. It plots 2 channels of data in the center of the
screen (or page).
View the plot for the following command file by entering:
ploterps xwin /usr/local/erp/man/man7/ploterps_manual/ex1.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 1 - a simple plot
**
** Any line not beginning with an alphabetic character is a comment.
** Commands and their parameters are given here in lower case, but
** ploterps ignores case so the user is free to use either upper or
** lower case. Comments appear just before relevant command.
**
** Set the top and bottom margins to be .25 of page height each
topmargin .25
bottommargin .25
** Set the left and right margins to be .1 of page width each
leftmargin .1
rightmargin .1
** Set the textsize to 7.5 and the font to complex
textstyle textsize=7.5 complex
** Tell Ploterps the name of the data file
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
** Tell Ploterps what bin to get data from
bin 1
** Divide the drawing page into a 1 X 2 grid for 2 data traces
** The 7.5 means that 7.5 microvolts will correspond to the
** distance from the middle to the top of each of the
** rectangles in the 1 X 2 grid
chanlocations 1 2 7.5
** Plot channel 1 in position 0 of the grid, label it "Channel 1"
chan 1 0 "Channel 1"
** Plot channel 2 in position 1 of the grid, label it "Channel 2"
chan 2 1 "Channel 2"
********************************************************************
One invokes ploterps like this:
ploterps plot_device commandfile [options]
where plot_device is one of the plotting devices that are available
(see
Plotting Devices below); commandfile is a text file consisting of
ploterps
commands, one per line (see General Command File Format below); and
options
are the optional program parameters (see Command Line Options below).
If
commandfile is "none", no explicit commands will be needed and plots will
be generated via command line arguments alone, as discussed below.
The ploterps command may fail from time to time with a "broken pipe" message;
if it does, simply repeat the command.
Plotting devices can be
separated into two types: landscape and portrait.
A landscape plot is one where the x dimension
is greater than the y dimension.
A portrait plot is one where the y dimension is greater than the
x dimension.
All plotting devices maintain a standard 4:3 ratio.
The following table lists some typical devices (the complete list can
be found in
/usr/local/erp/lib/graphicdevs).
NAME DESCRIPTION
xwin -- xwindow (640x480 pixels, landscape)
pxwin -- portrait xwindow (480x640 pixels)
xwinb -- xwindow (640x480 pixels, landscape), black background
hxwin -- huge xwindow (1024x768 pixels, landscape)
phxwin -- portrait hugh xwindow (768x1024 pixels)
hxwinb -- huge xwindow (1024x768 pixels, landscape), black background
fortel -- HP LaserJet 5100tn laser printer, landscape mode
pfortel -- HP LaserJet 5100tn laser printer, portrait mode
hutel -- HP Color LaserJet 3800n color laser printer, landscape
mode
phutel -- HP Color LaserJet 3800n color laser printer, portrait
mode
pastel -- HP Business Inkjet 2280TN color printer, landscape mode
ppastel -- HP Business Inkjet 2280TN color printer, portrait mode
motel -- HP LaserJet 5M laser printer, landscape mode
pmotel -- HP LaserJet 5M laser printer, portait mode
pso -- Encapsulated Postscript Level 2 to stdout, landscape mode
ppso -- Encapsulated Postscript Level 2 to stdout, portrait mode
gv -- Encapsulated Postscript Level 2 to ghostview previewer,
landscape
pgv -- Encapsulated Postscript Level 2 to ghostview previewer,
portrait
psf -- Encapsulated Postscript Level 2 to file out.ps, landscape
ppsf -- Encapsulated Postscript Level 2 to file out.ps, portrait
pdff -- PDF (Portable Document Format) to file out.pdf, landscape
ppdff -- PDF (Portable Document Format) to file out.pdf, portrait
jpgf -- JPEG (Joint Photographic Experts Group) to file out.jpg,
landscape
pjpgf -- JPEG (Joint Photographic Experts Group) to file out.jpg,
portrait
save:file -- save to file (landscape), for subsequent use with multiplot
psave:file -- save to file (portrait), for subsequent use with multiplot
Command line options are provided to enable users to plot data without
using any ploterps
commands or a command file. When these command line
options are present, ploterps first plots
data as directed by these options,
and then reads the commandfile (if any, as specified above) for
more commands.
If one wishes to only plot data via the command invocation options, "none"
or
one of the default configuration files should be specified as the commandfile.
In these modes of
operation, the -files and -bins options must be specified,
and their order of appearance in the
command line option list determines
the method of plotting, as described below under the -files
option. Various
parameters controlling the format of the plot are set to default values,
unless
otherwise specified via additional command line option(s). The default
parameters for this mode
cause the plots to be drawn negative "up", with
5 uV calibration markers. The current default files
are:
Here are the currently implemented options:
-files file1 [file2 file3 ...]
This option emulates the definefiles command (below). It allows the user
to specify files by number
(the first file is number 0) in subsequent
commands, enabling one to use one command file to plot
data from many
data files by merely specifying different data file names when invoking
ploterps.
When used in conjunction with the -bins option (described next)
and "none" as the commandfile,
the -files option allows plotting data in
a standardized manner without using any commands. In this
case, if the
-files option precedes the -bins option on the command line, all the specified
bins are
overlaid on a single page (or screen) for each file; if -bins
precedes the -files option, data from
all the specified files are overlaid
on a single page (screen) for each bin specified. For these
standardized
plots, either different colors or different styles of lines (or both) are
used to
distinguish the data that are overlaid on each page, depending
on the capabilities of the specific
plot device in use. It is usually difficult
to interpret plots having more than four traces overlaid,
so this is the
recommended limit for the number of overlaid traces.
-bins #bin1 [#bin2 #bin3 ...]
This option is meaningless unless used with the -files option described
above. It specifies the bins
that should be plotted from the ERP data file(s).
The format in which the plots are drawn (all bins
overlaid for each file,
or vice versa) is determined by the order of appearance of the -bins and
-files options in the command line, as explained above.
-scale uVolts
This option determines the scaling that will be applied to the data when
they are plotted
from command line options (-files and -bins). The argument
uVolts is a floating point number
which specifies the number of microvolts
that span one half of the vertical space allotted to each
data trace. It
is the same as the argument to the position and chanlocations commands,
discussed
below. The number should be positive, but the absolute value
is used, as the polarity of the data
can be inverted via the -cal option.
The default scale is 10.0.
-horzspace value
This option determines the horizontal spacing that will appear around
each data trace.
The value argument is a floating point number that determines
the fraction of the horizontal space
allotted to each data trace that
should be used for both left and right margins preceding and
following
the actual data trace. This is the same as the optional argument to the
position and
chanlocations commands, discussed below. The default horizontal
space is 0.05.
-cal uVolts
This option specifies the size of the calibration bar that will accompany
the data that are plotted.
The argument uVolts is a floating point number
that corresponds to the amplitude in microvolts
that the bar will represent.
Normally two such bars are drawn, one on each side of the baseline
for
the data. A negative value indicates that negative voltages should be plotted
"up". The default
value is -5.0 microvolts per side.
-smooth npts
This option tells ploterps to use a smoothing algorithm on all data before
plotting it. A running
average of npts before and after each data point
will be used as the value of the data to plot.
Note that smoothing can
be turned on and off in the command file with the smooth command.
-filter filter_type cutoff [cutoff]
This option tells ploterps to filter the data before plotting it. The
filter_type parameter must be
one of "lp" for lowpass, "hp" for highpass,
or "bp" for bandpass. The cutoff parameter is the cutoff
frequency to
use for the filtering. If using a bandpass filter, both a high and low
cutoff frequency
must be specified. Note that filtering can be turned
on and off in the command file with the filter
command.
-ticks ms_per_tick
This option specifies the interval in milliseconds (ms_per_tick) between
ticks on the abscissa. It
can be a floating point number or an integer.
-pause
This enables the pause command, such that ploterps will pause until the
user presses the enter key
whenever the pause command is encountered in
a command file. If plotting is being performed
using invocation line options
alone (-files and -bins), the program will pause between pages as if
a pause
command had been encountered.
ploterps provides about 50 commands. Each command should appear together
with its associated
arguments on a single line of the command file. Commands
and arguments are separated by white
space (spaces, tabs, or newlines
immediately preceded by a ’\’). Any line in a command file that
begins with
anything other than an alphabetic character is treated as comment line,
and its content
ignored (the line count is incremented, however). Unless
otherwise noted, the case of either
commands or their alphabetic parameters
and arguments is immaterial; they are all converted to
lower case before
interpretation.
If ploterps detects an error either in syntax or during the execution
of a command, it will print an
error message containing the line number
at which the error was detected, and continue processing
the input, if
possible.
Note that several commands will affect the global format of the plots
(e.g. where things are plotted
on the page), and should thus appear first.
For example, if one wishes to adjust the top margin,
it should be done
before any plotting commands are given. As another example, it is necessary
to
plot the "header" (via the plotheader command) before determining the
locations where data are
to be plotted with the chanlocations or position
commands, so that the space used by the header
will diminish the available
drawing area allocated by these positioning commands. If this is not
done,
the data will be drawn over the textual information in the "header".
Many of the commands for ploterps employ arguments (often optional) that
specify certain
attributes for text, data, or other graphic elements in
the plot. An "attribute" is simply some
property, such as the thickness
of the lines used to draw the data traces, that you can designate.
Other
attributes, for example, are the typestyle (font) used when plotting text,
the color for data
traces, and the type of line (e.g. dotted, dashed, dotdash,
etc.) used when plotting data traces.
The values and conventions for these
attributes are collected here, since they apply to more than
one command.
In addition, the user may need to specify a location on the drawing surface.
How
one does this is also described in this section. Generally, both the
attribute and its value are
specified as ASCII keywords, such as "dotted",
or "thin". In some cases numeric values are required;
these are sometimes
of the form "attribute=nnn", where "attribute" is an ASCII mnemonic for
the
particular parameter being specified, and "nnn" is a numeric value
appropriate for the mnemonic,
for example, color=12. No spaces are allowed
between the mnemonic, the equal sign, or the value,
and the quotation
marks are not part of the actual command text. Finally and again, the case
of
the parameters and/or keywords is immaterial. Here are the attributes:
font
Currently, four Hershey vector fonts are available for use in plotting
text and special symbols; when
a font argument is to be specified, the
user supplies one of the following keywords: simplex (single
stroke, plain),
duplex (double stroke), complex (multiple stroke, nice), or italic (much
like complex).
The default font is simplex.
textsize
textsize is analogous to the point size of printed text. This parameter
is specified by the string
textsize=XXX, where XXX represents a positive
number, possibly containing a decimal point and
fractional values. Again,
no spaces or other characters are allowed between the "textsize", "=",
and
"XXX" in the specification. A textsize of 2.0 is about as small as you
can go, and this is only useful
on devices with high resolution. A character
plotted at textsize 5.0 will be twice as tall and twice
as wide as one
plotted at 2.5.
color
The color attribute is specified in a manner similar to textsize, with
a string of the form:
color=NNN, where NNN is a positive integer. Again,
no intervening spaces are allowed between
elements. On pen plotters, this
attribute selects the pen (and hence perhaps the width of the lines).
The
number of colors or pens available is device dependent. In general, for
black and white
devices, a color of 0 means no color at all (black), and
anything else means the device’s one color
(white). For pen plotters,
color 0 means no color, color 1 is the first pen, color 2 is the second
pen, etc. It’s permissible to use unavailable colors, since most monochrome
devices map these into
color "1", but beware of other color devices and
particularly pen plotters, which require more user
attention than raster
devices. The default color is 1.
linewidth
When a linewidth argument is needed, the user should supply one of the
following (hopefully self-
explanatory) keywords: verythin, thin, medium,
thick, or verythick. The linewidth attribute interacts
with the linetype
attribute, in that thicker lines have different spacings for elements of
a linestyle
(i.e. dots and dashes). For text, solid lines of varying widths
are very useful for drawing bold
characters. On raster devices, the line
width is mapped into an appropriate number of pixels for
the width of
lines that are generated. For pen devices, it only alters the spacings for
elements of
a linetype; the user must alter the color attribute to select
various (pre-loaded) pens with different
tip widths. The verythin line width
is drawn fastest, and is the default.
linetype
linetype specifies the style or type of lines to draw. Four types are available:
solid, dotted, dashed,
ldashed and dotdash. Dashed and ldashed are both
for dashed lines, the difference between them
is that ldashed has longer
dashed segments than dashed. The user specifies the linetype attribute
by supplying one of these keywords. The default line type is solid.
labelorigin
The labelorigin attribute is associated with plotting text, and specifies
the positioning of text with
respect to a location on the drawing surface.
The labelorigin attribute is specified by a string of
the form laborigin=CCC,
where CCC represents a code of from one to five ASCII characters. This
encoding is a bit tricky, since there are 34 possibilities to be encoded,
but is intended to be
mnemonic, and works like this: Imagine that a rectangle
encloses some text. The label origin
specifies some point on, in, or outside
of the rectangle and serves as the reference location for the
positioning
of the text. When one moves to a particular location on the drawing surface
and then
issues a text command, the reference point becomes the current
location, and text is drawn relative
to that point. The 9 basic locations
for the reference point are depicted schematically thus:
LT----------------MT----------------RT
| | |
| | |
| | |
LM-----------------M----------------RM
| | |
| | |
| | |
LB----------------MB----------------RB
Note that these 9 label origins are encoded with a combination of the
letters L (left), M (middle),
R (right), T (top), and B (bottom). This
encoding allows the letters to occur in any order and be
of either upper
or lower case. As an example, to center some text immediately above a particular
coordinate, one would specify the labelorigin attribute as laborigin=MB,
since the reference point
would be in the middle of the bottom of the
rectangle.
Two other letters are combined with these 5 to complete the full set of
label origins. They are: V,
which causes the text to be plotted vertically,
one character on top of another, and E, which adds
an additional half-character-width
of extra space around the text in the imaginary rectangle (for
example,
laborigin=LME would place the reference point one half of a character space
to the left
of the beginning of the text). The letters that encode these
modifiers of the 9 basic positions can
also appear in any order, even
before and intermingled with the basic five letters, and can be either
upper or lower case.
Some commands allow arbitrary placement of text, lines, plotted data,
etc, on the drawing surface.
Whenever a coordinate appears as an argument
to a command, no prepended mnemonic should
appear and the numbers should
start with a digit or a ’.’. A floating point coordinate system
(different
from that used by barplot and linegraf) is employed. The x direction always
goes from
0.0 to 1.0 for any device, whether it is used in landscape or
portrait mode. An x coordinate of 1.0
corresponds to the point immediately
beyond the rightmost point on the drawing surface, while 0.0
corresponds
to the leftmost point on the drawing surface. The bottom-most point on the
screen
corresponds to 0.0 in the y direction, while the topmost point depends
on the specific device in
use. For most devices, which have a 4 to 3 aspect
ratio (x to y available surface ratio), this will be
0.75 in landscape
mode, and 1.33 in portrait mode.
Commands are grouped into four different categories: those that affect
the style and drawing
attributes of the plot (e.g. color, font, linewidth,
linestyle), those that are used for positioning plot
elements on the drawing
surface, those that are related to ERP data, and those that actually cause
plotting to take place. In many cases, arguments are optional, and are so
denoted by enclosure in
"[" and "]". The possible types for these optional
arguments are listed in the square brackets; the
user should substitute
the desired keyword and/or value as discussed in
COMMAND FILE ARGUMENT
AND PARAMETER CONVENTIONS, above. Usually any number and/or combination
of the different attribute types can
appear as optional arguments for
a command. Following each command is an alternate,
abbreviated version
of the command name in parentheses, which can be used in place of the full
command name.
ploterps keeps track of the drawing attributes for text, lines (including
data traces), and the
baseline grids separately (remember that attributes
are things like line thickness and color. Refer
to section 3.1 for a description
of the drawing attributes). The idea is that you might, for
example,
want to use thin lines for drawing data traces and thick lines for drawing
text (to give a
bold effect). You use the linestyle command to set the
attributes used for drawing lines, the
textstyle command to set the attributes
for text, and the gridstyle command to set the attributes used
for drawing
baseline grids. The commands that tell ploterps to actually plot something
then refer
the attributes that you have set with the style commands to
decide how to do the plotting. Also,
each plotting command lets you temporarily
override the current settings for the attributes. For
example, you could
use the textstyle command described in this section to set the color for
text to
be color # 1. All commands that plot text would then use color
# 1 unless you use an optional
command parameter that says to override
the current text color and use the one you specify
instead.
- linestyle (ls)
[color linewidth linetype]
-
The linestyle command sets the current line color, linewidth, and/or linetype
attributes that
are used when any line is drawn (such as during the plotting
of data) up until the next
linestyle command. It does not affect the style
used for plotting text, or the style used for
plotting baseline grids.
One can supply one, two, or all three types of attributes; unspecified
attributes retain their existing values. However, if no arguments at all
are specified, all
linestyle attributes are reset to their default values.
- gridstyle (gs) [color linewidth linetype]
-
This is very similar to the linestyle command, except that it only applies
to the plotting of
the "grid" on which the data is drawn. The grid includes
the timebase, baseline, tick marks,
and calibration bar. The gridstyle
is kept as a separate class of style attributes so that, for
example,
one can plot data with dashed lines and the baseline with solid lines. As
with the
linestyle command, any or all of the arguments can appear; however,
if no arguments are
supplied, all gridstyle attributes revert to their
default values.
- textstyle (ts) [font textsize labelorigin color linewidth linetype]
-
The textstyle command sets the current font, textsize, color, linewidth,
and linetype that will
be used for all text plotting up until the next
textstyle command. Any combination of
arguments can be supplied; if none
appear, all textstyle attributes are set to their default
values.
- labelprec (lp)
precision
-
Set the precision for tick labels. Precision is the number of digits that
are to appear after
the decimal point. If precision is 0, no decimal point
is printed. Note that label precision
applies only to ticklabels. Precision
defaults to 0.
- vlabelprec (vlp) precision
-
Similar to labelprec. Sets the number of decimal places that should appear
in the
labels used for the voltline. See the Plotting Calibration Lines
section for info
on the drawvoltline command.
Positioning commands specify the locations, sizes, and margins for elements
of the plot. All these
commands require that the numeric arguments appear
in the proper order. If no arguments are
supplied for any one of these
commands, the value is set to the corresponding default(s) for the
command.
Usually one will want to place these in the command file prior to actual
plotting
commands.
- bottommargin (bm) value
-
Set the amount of space to be reserved for the bottom margin of the drawing
surface as
a fraction of the screen height. For example, if one were using
an 11 inch high piece of
paper (portrait mode) and wanted a 1 inch bottom
margin, the command would be:
bottommargin 0.0909
- topmargin (tm) value
-
This command is analogous to the bottommargin command. It reserves space
for the top
margin on the drawing surface as a fraction of the screen
height.
- leftmargin (lm)
value
-
Sets the left margin as a fraction of the screen width.
- rightmargin (rm) value
-
Sets the right margin as a fraction of the screen width.
- chanlocations (cl) rows cols [scalefactor horzspace show]
-
Later commands are going to require a positioning argument that selects
a position on the
drawing surface. That position and its associated position
number is defined using this
command, or by the position command below.
The position command allows one to specify
exactly where to plot data
using the coordinate system described in COMMAND FILE ARGUMENT
AND PARAMETER
CONVENTIONS above. The chanlocations command allows one to automatically
define an
array of positions, one for each channel of data, laid out as
follows. The available drawing
surface is divided into "rows" horizontal
strips and "cols" vertical strips, the combination of
which creates an
imaginary array of rectangles. These rectangles, numbered as discussed
below, define the space that can be used to plot a particular channel of
data. Screen
coordinates are calculated and saved, but the user denotes
positions by an integer, counting
positions from the top left using row-major
order. For example, if a chanlocations command
is given with the rows and
cols arguments being 2 and 2, the upper left area would then
correspond
to a position number of 0, the upper right to 1, the lower left to 2, and
the
lower right to 3:
--------
| 0 1 |
| 2 3 |
--------
The optional scalefactor argument determines the scaling of the data (and
thus its apparent size) in
the rectangle by specifying the number of microvolts
(as a floating point number)
corresponding to one half of the height of
the rectangle (one side of the baseline). The
size of the actual calibration
marker is determined by the calibration command below, its
presence or
absence by the calmarks command, also below.
The optional horzspace
argument represents a fraction of the width of
each rectangle that should be used for both
the left and right margins
preceding and following the actual data trace. For example, if one
wanted
traces separated by one quarter the width of the traces, one would supply
0.10 for
this argument. In this case the space between traces would be .25
of the width of the traces,
but the left and right margins would be increased
by only one eighth of the width of a
trace. Be sure to set the margins
before executing a chanlocations command.
The optional show parameter indicates to display the grids.
If only one optional parameter is given, and it is not show, then it is
assumed to be scalefactor. If there are two optional parameters, the first
must be the scalefactor, and the second can be either show or horzspace.
- channelorder (co) chan [chan ...]
-
Define which channels of data should be plotted in which locations (as
defined by the
chanlocations command). The first argument is the channel
number of the data channel
that should be plotted in location 0, the second
argument is the channel number of data
channel that should be plotted
in location 1, and so on. If no data should be plotted in
a particular
position, use a -1 for that location. The channelorder command works only
in
conjunction with the plotchans command (below). Once a list of channels
has been defined
with the channelorder command, the plotchans command
is used to plot all of the channels
in the list. This is merely a shortcut
to get around plotting each channel one at a time with
the chan command
(below).
- position (po)
posnumber xl xr yt yb scalefactor [horzspace]
-
This command is used to define the drawing rectangle (as described above
for the
chanlocations command) for a single a channel of data. It is useful
if one wishes to place
data at arbitrary locations on the drawing surface,
rather than just on a rectangular matrix,
or when one wants the scaling
to differ at different positions. Posnumber represents an
integer, and
specifies the position being defined. It will be referred to in subsequent
chan
commands. The xl, xr, yt, and yb arguments represent floating point
numbers using the
coordinates system defined above in COMMAND FILE ARGUMENT
AND PARAMETER CONVENTIONS
corresponding to the left,
right, top, and
bottom edges of the rectangle. The scalefactor argument is the same as in
the chanlocations command (above) as is the optional horzspace argument.
- scalefactor (sf) scalefactor posnumber
-
This command is used to redefine the scalefactor of a drawing rectangle
previously defined
by the chanlocations or position command (described
above). It is useful, for example, if
one wants to use the chanlocations
command to define the locations on the drawing surface
for plotting the
data, but does not want to use the same scalefactor for each position. The
scalefactor argument is the same as in the chanlocations and position commands
(above).
These commands are used to select the particular ERP data that are to
be plotted, and various
miscellaneous aspects of the drawing that depend
on the particular data being plotted.
- definefiles (df) file0 [file1 file2 ...]
-
Defines a list of files that can later be referenced by numbers. Any files
defined by a
previous definefiles command will be closed and their names
and indices forgotten. This
command can be emulated by the -files command
line option, allowing one to use the same
command file for different data
sets. In this case, however, no definefiles commands should
appear, as
they will override the command line option.
- file (f) file_spec
-
Select the current file from which to retrieve data. File_spec can either
be a file name, or
an integer specifying one of the files defined in the
last definefiles command (or from the
-files command line option).
- bin (bn) bin_number
-
Select the current bin from which to retrieve data as an integer. Remember
that bin
numbers start at 0.
- procfunc (pf)
pfnumber1 [ +/- pfnumber2 ]
-
Select the current processing function from which to receive data to be
pfnumber, an
integer. Processing functions numbers start at 0. If the optional
arguments are present,
pfnumber2 is added to (or subtracted from) pfnumber1
point by point. This would be
useful if one wanted to plot a simple average
with the standard error processing function
added to or subtracted from
(or probably both) the original simple average. Note that the
spaces around
the "+" or "-" must be present. Read the manual on ERP data files for
more
information on processing functions.
- ticks (tk) mspertick
-
Specify how far apart ticks (if drawn) should be spaced in milliseconds,
as a floating point
number. Ticks are turned on or off with the tickmarks
command, below. If no argument
is present, the tick interval is set to
the default value of 100 milliseconds per tick.
- calibration (cal) uVolts
-
Specify how many microvolts via uVolts (as a floating point number) are
represented by the
calibration bar. A separate bar is drawn on each side
of the baseline, so this is half the
number for the combined size of both
bars spanning the baseline. If uVolts is negative, the
data are plotted
with negative voltages "up". If no argument are present, the calibration
size
is set to -5.0, the default value.
- smooth (sm) npts(on, yes)/off(no)
-
This is very similar to the -smooth command line option. The data is "smoothed"
before
plotting by computing a running average of npts before and after
each data point. If yes
or on is used instead of a number, smoothing is
done by using a weighted average: 2 times
each data point value plus the
point before it plus the point after it, divided by 4. If no
or off is
used as the argument, smoothing will not be done for any plotting that
follows the
smooth command.
- filter (fi) off/no
- filter (fi) filter_type cutoff [cutoff]
-
This is very similar to the -filter command line option. The data is filtered
prior to
plotting. filter_type specifies the type of filtering to perform.
It must be one of: "lp" (for
lowpass), "hp" (for highpass), or "bp" (for
bandpass). cutoff is the cutoff frequency. Note
that for bandpass filtering
both a high and low cutoff frequency must be specified. If the
only parameter
specified is off (or no), no filtering is done for subsequent plotting.
- timewindow (tw) start_time end_time
-
Specify the time window of the data that you wish to plot. The start_time
and stop_time
are in milliseconds, as floating point numbers. If no arguments
are present, the default (the
entire epoch) is set.
- baseline (bl) start_time end_time
-
Specify the time window which should be used to calculate the baseline
amplitude, which
is subtracted from all the points of the channel of data
prior to plotting it. The start_time
and stop_time are in milliseconds,
as floating point values. If no arguments are present, the
default interval
is set, which is the entire presampling period.
- ticklength (tl) value
-
Determines the size of tick marks on the grid abscissa, as a floating
point number
representing a fraction of the entire width of the drawing
surface. If no argument is
supplied, the ticklength is reset to the default
value.
- calmarks (cm) on(yes, y, 1)/off(no, n, 0)
-
Specify whether or not you want a calmark on the baseline grids. The argument
may be:
on, yes, y, or 1 for drawing the calmarks; or off, no, n, or 0,
for not drawing the calmarks.
If no argument is supplied, they are turned
on.
- tickmarks (tkm) on(yes, y, 1)/off(no, n, 0)
-
Specify whether or not you want tick marks on the baselines. Argument
is the same as for
calmarks command. If no argument is supplied, tickmarks
will be drawn.
- grids (g) on(yes, y, 1)/off(no, n, 0)
-
If the argument is off (no, n, 0), neither the baseline grid nor the
channel label are plotted
when data are plotted. This is nice when you
are overlaying several channels at the same
position and don’t want to
redraw the baseline grid or channel label each time. If no
argument is
supplied, grids and channel labels will be drawn.
- polindicators (pi)
on(yes, y, 1)/off(no, n, 0)
-
Specify whether or not to put polarity indicators on the calibration bars
when they are
plotted. This does not affect the polarity indicators plotted
by the drawcal command
(below). If no argument is supplied, indicators
will be drawn.
- ticklabels (tlb) on(yes, y, 1)/off(no, n, 0)
-
Specify whether or not to label the tickmarks on the baseline grids. If
they are labeled,
the textsize used will be 1.0 less than the current textsize.
If no argument is supplied,
ticklabels will not be drawn.
- utick (ut) time [+/-] [label] [font labelorigin textsize color linewidth
linestyle]
-
Specify a time (in milliseconds) to locate a tick mark on the baselines.
If "+" is supplied
as an optional argument, only the top half of the
tick is drawn. If "-" is present, only the
bottom half (below the baseline)
is drawn. If the user wants to label the tick mark, he may
provide a label
in the optional label argument (if it contains spaces, enclose it in quotes).
These two optional arguments should precede the following optional attribute
arguments
on the command line. The current font, textsize, color, linewidth,
and linetype from the last
textstyle command will be used unless the user
overrides them with the optional attribute
arguments.
- clearuticks (ct)
-
Requests that any user requested tickmarks and labels (defined via the
utick command) be
forgotten.
- plotheader (ph)
[font textsize color linewidth linestyle]
-
Plot a legend, or "header" at the bottom of the page, describing the data
from the current
bin, as well as various plotting s. The s plotted includes:
sums, calibration, time window
being plotted, milliseconds per tick mark,
onset delay, subject description, experiment
description, condition description,
and bin description. ploterps assumes that all information
other than the
sums, bin description, and experiment description will be the same for
each
bin, and reserves space for this information for four overlaid traces.
These fields will be
arranged in some default fashion at the bottom of
the page. See example 2 for the default
header layout. The current font,
textsize, color, linewidth, and linestyle will be used, unless
overridden
with the optional arguments. This command should appear before any space
allocating (such as chanlocations or position) or plotting commands, and
after specifying
bin numbers.
- chan (ch) chan_num posnumber [labpos] [label] [font textsize col linewidth
linetype]
-
The chan command is used to plot a particular channel of data at a particular
position.
Chan_num specifies the channel number from the current file,
bin, and processing function.
Posnumber selects a position as defined in
a previous chanlocations or position command,
and is an integer.
The optional labpos argument specifies where the label for the channel
(from the channel
description field in the data file bin header) will
be plotted with respect to the baseline. It
is similar to the argument
for the labelorigin attribute, but uses a different code, and is
introduced
by the sequence labpos=CC, where CC is the code, as follows. The single
letter
"N" specifies that no channel description should be drawn, whereas
a combination of the
letters "T" (top), "L" (left), and "R" (right) specify
a location around the data trace. If you
want the label just to the left
of the baseline, the argument is the single letter "L". If you
want the
label just to the right of the baseline, the labpos argument is the single
letter "R".
If you want the label centered over the top of the baseline
(just above the top of where
the calibration bar would be), the argument
is the single letter "T". The combination of
letters "T" and "L" says
to put the label on the top left side, and the combination "TR"
says to
put the label on the top right side. Order of the letters is not important,
and upper
or lower case may be used.
The optional label argument for the chan command is for labeling a channel
with one’s own
label. The program will normally get the channel description
out of the bin header and use
that for the label, but if the user specifies
another label by supplying this label argument,
it will be used instead.
Note that the optional labpos and label arguments must appear
before any
of the optional over-riding attribute arguments on the command line. Either
or
both of these optional arguments can appear in either order, as long
as they occur before
the attribute arguments, if present.
For example, let’s say you wanted to plot channels 0, 1, and 2 down the
center of the page.
You could first use the chanlocations command with
arguments of 3 (rows) and 1 (cols),
and an appropriate scaling factor.
Then, the posnumber argument for channel 0 would
be 0 (the top position
of the 3 X 1 matrix), the posnumber argument for channel 1 would
be 1,
and the posnumber argument for channel 2 would be 2. If this is still confusing,
please
look at the examples before giving up. Interaction with a human
expert might also be
helpful. In any case, the automatic, easy to use chanlocations
facility should be sufficient
for most applications.
- chanlabel (chl) chan_num posnumber [labpos] [label] [font textsize col
linewidth linetype]
-
This command merely puts the label for chan number chan_num at the specified
posnumber. The syntax is exactly the same as for the chan command described
above.
Refer to the description of the chan command for details.
- plotchans (pc)
[yes(on 1)/no(off 0)]
-
Plots the channels as defined by a previous channelorder command. If no,
off or 0 is
specified, doesn’t plot channel labels. The default is to plot
the channel labels.
- plotallchans (pa)
[rows cols scalefactor [horzspace]]
-
This command plots all channels in the current bin. The channels of data
are plotted in the
order they are found in the average file (use the plotchans
command to plot several
channels in some special order). If none of the
optional arguments are supplied, ploterps
figures out some reasonable
way to position the channels, and automatically plots them,
labeling them
with the default labels from the header. If the optional arguments are
supplied, the specified rows, cols, scalefactor (and optionally horzspace)
are used to position
and scale the data for all of the channels in the
current bin. Note that rows, cols, and
scalefactor must all appear if
the user wishes to specify how to position the data traces, and
that they
must appear in the given order. These optional arguments are the same as
the
arguments to the chanlocations command (above).
- drawbaseline (db) posnumber [color linewidth linetype]
-
The drawbaseline command plots a baseline without plotting data on it.
The posnumber
argument is exactly the same as the posnumber argument for
the chan command. The
color, linewidth, and linetype from the last gridstyle
command (or the defaults if there was
no previous gridstyle command) will
be used, unless they are overridden with the optional
attribute arguments.
- drawcal (dc) posnumber calorigin [label] [font textsize color linewidth
linetype]
-
(NOTE: the drawcal command is from the original version of ploterps. A
newer, easier
technique is available with the newer versions of ploterps.
See the Plotting Calibration
Lines section below.)
The drawcal command is for plotting a calibration bar by itself at an
arbitrary location.
Because there are so many ways one might want to plot
a separate calibration marker, the
command is quite involved.
The posnumber argument specifies one of the positions previously defined
by either the
chanlocations or position command (above) to be used as
a reference point for the
calibration marker. One must also specify where
the calibration marker will be located with
respect to that reference
point using the calorigin argument. In other commands, the
position defined
by a posnumber argument specifies a rectangular area on the drawing
surface
in which data will be plotted. Usually a baseline is plotted, centered
in this
rectangle. The imaginary place where this baseline would be drawn
is used a reference point
for drawing the calibration marker. There are
several possibilities, all of which are encoded
in the calorigin argument
with a combination of letters, in much the same fashion as the
argument
for the labelorigin attribute. The calorigin argument also specifies whether
or not
and where to put tick marks on the calibration marker, and which
side of the calibration
marker to draw the positive and negative signs.
The calorigin argument has the form
calorigin=CCC, where a combination
of one or more letters constitute the CCC code, as
follows. The letters
that can be combined are: "N", "T", "M", "B", "L", "R", "E", "O", and "+".
Their meanings are as follows:
N: do not put ticks or positive and negative indicators on the calibration
marker.
T: place the calibration marker above (on top of) the location specified
by the
posnumber argument.
M: center the calibration marker (middle) vertically on the location
specified by the
posnumber argument.
B: place the calibration marker below the location specified by the posnumber
argument.
L: put tick marks on the left side of the calibration marker only.
R: put tick marks on the right side of the calibration marker only.
E: when placing the calibration marker below or above the location specified
by the
posnumber argument, use extra space.
O: ploterps will label the calibration marker with the value (in microvolts)
that it
represents (or the optional label string, if present), unless
this letter appears. The
side of the calibration marker on which the label
is placed is determined by the
current label origin associated with the
linestyle attributes, or by the optional, over-
riding labelorigin attribute
argument. Normally the positive and negative signs will
be placed on the
marker, on the same side as the calibration label. This letter
means to
put the positive and negative signs on the "O"pposite side from the label
(picky yes, but somebody wanted to be able to do this).
+: put calibration marker on the left side of the location specified
by the posnumber
argument.
As usual, the default label (the size of the marker in microvolts, to
one decimal place) can
be overridden by supplying an optional label argument.
The current font, textsize, etc can
also be overridden if the optional
arguments are present, but the label argument must
precede them.
Note that the optional attribute arguments following the optional label
affect only the label,
but not the calibration bar line itself. This is
controlled by the gridstyle command.
- line (l)
startx starty endx endy [color linewidth linetype]
-
The line command simply draws a line from (startx, starty) to (endx,
endy). The current
color, linewidth, and linestyle (from the last linestyle
command) will be used, unless
overridden with the optional attribute arguments.
- lineto (lt)
x y [color linewidth linetype]
-
This command is similar to the line command above, except the line is
drawn from the end
of the last line, lineto, or moveto command to the
location specified in x and y, as drawing
surface coordinates. The current
color, linewidth, and linestyle (from the last linestyle
command) will
be used, unless overridden with the optional attribute arguments.
- moveto (mt) x y
-
Simply move to the specified location on the drawing surface. This command
is used mainly
with lineto command, to reduce the typing of coordinates.
- text (tx) xcrd ycrd string [string args] [font labelorigin textsize color
linewidth linestyle]
-
Plot the text string at the coordinate (xcrd, ycrd) using the current
label origin, textsize,
font, linewidth, color, and linetype attributes
as determined by the last textstyle command,
unless overridden by optional
attribute arguments supplied for the command. If the string
argument contains
white space, enclose it in double quotes. Users have a number of
predefined
strings at their disposal. The syntax for accessing these predefined strings
(mostly from the bin headers of the data file) is similar to that of the
printf function of the
C language. At the point where a predefined string
is desired, use the "control" sequence
"%v". Then, after the string (in
the [string args] position), put name of the predefined
string desired.
The available predefined strings are described in APPENDIX B. One
complication
is that some predefined strings need an additional "index" argument. Such
is
the case for the channel descriptions, which require the channel number.
For example,
if you wanted to plot the experiment description for the
current bin on the center of the
screen, the command would be:
text .5 .5 %v expdesc
If you wanted to print the channel description in the bin header for channel
number 5, the
command would look something like this:
text .5 .5 "Channel 5 is %v" chandesc 5
Note that any string arguments must precede any optional text style arguments.
Several
special characters (mostly greek alphabet and mathematical symbols)
are available. These
are accessed in the same manner employed by the
text processing program nroff.
APPENDIX C lists the special characters
that are available.
- arrow (ar) location time volt arrowcode label [textstyle options]
-
Draw an arrow with a label. Location is one of the channel locations specified
with either
the chanlocations or the position command. Time is the time
in milliseconds that
corresponds to the horizontal position of the arrowhead.
Volt is the voltage in microvolts
that corresponds to the vertical position
of the arrowhead. Arrowcode is a combination of
the letters "U" (up),
"D" (down), "L" (left), and "R" (right) used to define one of 8 possible
orientations for the arrow. If you want the arrow pointing up, the arrowcode
is "U", if you
want the arrow pointing down and to the left, the arrowcode
is "DL", etc. Label is a text
string that will be placed near the tail
of the arrow. If the label has white space in it,
enclose it in quotes.
The exact position of the label in relation to the arrow is determined
by the label origin as described above in the Attributes section. The label
origin reference
location is the tip of the tail of the arrow. The size
of the arrow is determined by the
textsize.
- shade (sh) location start end file1 bin1 chan1 file2 bin2 chan2 [greyscale]
-
Shade the region between two erp waveforms. The location argument is the
location,
defined previously by either the chanlabel or the position command,
where the two erps are
plotted. The start and end arguments specify the
times in milliseconds where shading should
start and end. The next six
arguments specify where the two waveforms came from. File1,
bin1, and
chan1 define the file, bin, and channel number of the first waveform. File1
may
be either a data file name, or a file number as defined by the definefiles
command or the
-files command line option. Similarly, file2, bin2, and
chan2 define the file, bin, and channel
number for the second waveform.
The optional greyscale parameter is used to specify lighter
or darker
shading. greyscale is a number in the range 1-10, 1 being the lightest and
10
being the darkest.
- rect (r) ULx ULy LRx LRy [fill] [color linewidth linestyle]
-
Draw a rectangle with the upper left corner coordinate specified by the
first two
arguments, and the lower right corner specified by the second
two arguments, as drawing
surface coordinates. The current attributes specified
by the last linestyle command are
employed when the rectangle is outlined,
unless overridden by the optional attribute
arguments. If the optional
fill argument is specified, it must appear before any of the
attribute
arguments, and selects the pattern used to fill the rectangle. The fill
argument is
an integer specifying one of a number of fill patterns available.
If not present, the
rectangle will only be outlined.
- newpage (np)
-
Depending on the device, the current page is ejected and a new one loaded,
or the screen
is erased. All subsequent plotting will be done on a "new
page".
- readfile (rf) filename xtrans ytrans
-
Read commands from the file filename, and then resume reading commands
from the
current command file. While the file is being read, all coordinates
are translated by adding
the xtrans and ytrans arguments to those in the
file, thus effecting a translation of the sub-
image. Hence, the coordinates
in a command file that is to be read in this manner can be
considered
relative, if the translations are non-zero, or absolute, if the translations
are
both zero. Actually, a stack of offsets is maintained to allow the
possibility that one of the
commands in filename is itself another readfile
command. The actual location of coordinates
in a nested command file is
translated by the sum of all the xtrans and ytrans s in the
"reading"
files. The limit on the number of nested files is 10. The readfile command
is
quite useful for things like drawing a head or other figures at various
locations on the
drawing surface. One command file could contain a long
sequence of line, lineto, and
moveto commands for the head, and another
command file could use the readfile command
to read the specifications
for drawing the head, so that the drawing commands need not
be replicated
repeatedly.
Note: only actual plotting performed in the readfile is translated; commands
that don’t do
plotting (e.g. position) won’t be affected by the xtrans and
ytrans coordinates.
- pause (pz)
-
The pause command only works if -pause is specified on the command line.
When a pause
command is encountered, the program waits for the user to
type a carriage return on the
terminal keyboard.
- windowtext (wt) posnumber position text [font labelorigin textsize color
linewidth linestyle]
-
The windowtext command displays text relative to a specified window. This
is
useful if the user changes the order of the plots with the channelorder
command.
Instead of recalculating the x and y coordinates for the plot’s
label, the user only
has to change the posnumber to the corresponding
posnumber of the plot that has
changed to display the text. The first
argument for the windowtext command is a
posnumber, the posnumbers are
initially allocated with the chanlocations command,
which specifies which
window in the array of windows to display the text. The
second argument
specifies which of the nine preset locations the text should be
displayed
in the selected window. The nine preset locations are as follows:
TL/LT-------------TM/MT-------------TR/RT
| | |
| | |
ML/LM---------------M---------------MR/RM
| | |
| | |
BL/LB-------------BM/MB-------------BR/RB
The last argument is the actual text to be diplayed.
Plotting a calibration line is somewhat of a special operation, and as
such, gets a section all to
itself in this manual. By calibration line,
we are referring to the calibration and timeline that are
often put as
a legend at the bottom of an erp plot. In previous versions of the ploterps
program,
this task was performed with the drawcal and drawbaseline commands.
While these commands are
still available, it is suggested that the new
commands described in this section be used, as the older
commands have
proven to be rather cumbersome to use.
The task of drawing a calibration line can be divided into 4 subtasks:
(1) drawing the horizontal
time line at some predefined channel location,
(2) putting time ticks on and labeling them, (3)
drawing a vertical voltage
line at some location on the time line and labeling it, and (4)
labeling
the voltage line with a polarity label. Commands are provided for performing
each of these
subtasks.
- drawtimeline (dt) location [linestyle options]
-
Simply draw a horizontal line in the position specified by the location
argument. The
location argument refers to one of the locations defined
by either a previous chanlocations
or position command. If no linestyle
options are given, the linestyle attributes from the last
gridstyle command
are used.
- labelticks (lbt) size time [lab="my label"] [time [lab="mylabel"]...] [text
options]
-
Put labeled tickmarks on the timeline drawn with the last drawtimeline
command (note that
drawtimeline must be used before the labelticks command
can be used). The size
argument determines the size of the tickmarks as
a floating point number representing a
fraction of the entire width of
the drawing surface. Each time argument specifies a time in
milliseconds
at which a tickmark should be placed on the timeline. Normally ploterps
will
label the tickmark with the time you specify, however, if you wish
to override the default
label for a particular tickmark, you may supply
your own label by following the time
argument for the desired tickmark
with a lab= argument. You are allowed to specify up
to 20 tickmark times
and labels. The text attributes of the tickmark labels may be specified
with the usual optional text options. If no text options are specified,
the text attributes are
determined by the last textstyle command. The
labels are normally centered underneath the
tickmarks; this can be overridden
by specifying a laborigin= text attribute as one of the text
options,
making it possible to right or left justify a tick label with a tickmark.
- nolabelticks (nlt) size time [time ...]
-
Put tickmarks on the timeline drawn with the last drawtimeline command,
but don’t label
them. This functions similarly to the labelticks command
(above), except that no labels are
drawn with the tickmarks.
- drawvoltline (dv) time tickmarkpos labelpos [text options]
-
Draw a vertical voltage line on the timeline drawn by the last drawtimeline
command. The
time argument specifies the time in milliseconds on the timeline
at which to draw the
voltline. The tickmarkpos specifies the orientation
of the tickmark that appears at the top
of the voltline. It must be one
of "right", "left", or "center". The labelpos argument
encodes one of
6 possible positions for the microvolt label to be placed with respect
to the
voltline. It is constructed of one or a combination of two of the
letters "M", "T", "R", "L",
and "N". The first four are the usual middle,
top, right, and left specifiers. Using "N" for
the labelpos argument specifies
that no microvolt label should be drawn for the voltline.
Examples of
all 6 positions are shown in the Examples section below.
- labelpol (lbp) neg [pos]
-
Put a negative or positive (or both) polarity sign by the voltline drawn
with the last
drawvoltline command. The polarity indicator is drawn on
the side opposite from the side
that the tickmark was drawn on the voltline,
or to the left if the tickmark was centered on
the voltline.
Perhaps the easiest way to learn about ploterps is to study the examples
given here. Each example
is followed by the actual command file used to
produce it, except for Example 2 which generates a plot
without using
a command file. Keep in mind that there are often several ways to produce
the same output. For most of these examples, we use a grand average (i.e.
many subjects) erp
datafile called g28.gnd.
This example demonstrates the "quick and dirty" approach to plotting data.
This is probably the
best way to take a quick look at some data without
having to mess with a command file. The plot
is produced using "none"
for the commandfile argument, and instead uses command line
options to
select the data of interest.
View the plot by entering the following (enter the command all on one
line,
here it is broken into multiple lines for readability):
ploterps pxwin none
-files /usr/local/erp/man/man7/ploterps_manual/g28.gnd
-bins 1 8 11 -scale 7.5
For each channel in the file g28.gnd the bins 1, 8, and 11 are
overlaid
on the plot, using the default calibration of 5 microvolts per side.
The
-scale option scales the data up from the default of 10.0
microvolts for
each half of the rectangle allotted to each data trace.
In this example, the use of a command file is demonstrated. Keep in mind
that users that don’t
like to type a lot can use the abbreviated command
names, however I’ve chosen to use the full
command names to make things
more readable. I’ve overlaid data from channels 3 through 13 of
bins 21
and 17 from the data file g28.gnd, using the chanlocations and channelorder
command to
determine positioning. This is also a good example of how to
draw a timeline and a calibration
bar for the legend.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex3.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 3 - timeline and calibration bar
**
** comments appear just before relevant command
**
polindicators no
** put time ticks every 100 msec, set length of ticks to .02
ticks 100
ticklength .02
** default timewindow is entire epoch (-100 to 924 in this case)
timewindow -100 800
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
topmargin .08
bottommargin .02
rightmargin .10
leftmargin .10
** linewidth for grids
gridstyle medium
calibration -2.0
textstyle complex
** next command puts a title at specified x and y locations, it will be
** centered around the x-coordinate because of the M (for middle) in
** the "labelorigin" parameter. Font type is "complex", line thickness for
** letters is "medium", size of letters is "5".
text .50 1.293 "Congruent Associated" textsize=5 medium laborigin=MTE
** next command divides page up into sections for the different channels,
** 7 rows, 3 columns. 3.0 determines how much vertical space will be
** allowed between channels. .04 sets how much horizontal space between
** channels. Nothing will be drawn in these sections until we request it.
chanlocations 7 3 3.0 .04
** bin to be plotted
bin 21
** line thickness for the ERP and color of line
linestyle thick color=1 solid
** define which channels go in which locations. -1 means put no channel
** in this location. We have channel 4 in location 0, nothing in location
** 1, channel 5 in location 2, etc. This will stay in effect until we use
** the channelorder command again.
channelorder 4 -1 5 6 2 7 8 3 9 10 -1 11 12 -1 13
** set the textsize to 3, the text color to 1 and the text thickness to
** thin. This will affect the channel labels for the data we are about
** to plot.
textstyle thin textsize=3 color=1
** Plot the data for the channels and locations we defined above with
** the channelorder command.
plotchans
** we’re going to plot another bin on top of the first,
** no need to redraw the grid.
grids off
** plot this bin with a dashed line, color 3 (color command will
** be ignored by laser)
linestyle color=2 dashed thick
bin 17
plotchans no
** done plotting data, now we will worry about plotting the calibration
** info, next command sets text characteristics for the msec info, and
** the uV info.
textstyle complex textsize=4.4 color=1 medium
** First draw the timeline only in position 17.
drawtimeline 17
** Now put some labeled ticks on the timeline we just drew. Tick size
** is .04. We want labeled ticks at 0, 400, and 800 milliseconds. For
** the tick at 800 we don’t use the default label. The spaces to the
** left of 800 in " 800 ms" will just about center the 800 under the
** tick mark.
labelticks .04 0 400 800 lab=" 800 ms"
** Now put some ticks without labels on the same timeline. Tick size
** is .02. Put ticks at 200 and 600 milliseconds.
nolabelticks .02 200 600
** Now draw the voltline. Draw it at 0 milliseconds. Put the tickmark
** on the right size of the voltline only. Put the micrvolt label in the
** the middle and to the left of the voltline.
drawvoltline 0 right ML
** Finally draw the negative polarity indicator.
labelpol neg
** write "First word" at specified x and y coordinates, color 3
text .26 .28 "First word" color=2
** draw a dashed line from x-coord .14 to x-coord .24, vertical (y-coord)
** location is .27. Make the line thick.
line .14 .27 .24 .27 color=2 thick dashed
** write "Second word" at specified x and y coordinates, color 1.
text .26 .24 "Second word" color=1
** draw a solid line from x-coord .14 to x-coord .24, vertical (y-coord)
** location is .23. Make the line medium thickness, color 1.
line .14 .23 .24 .23 color=1 medium solid
** Center a title at the bottom of the page.
text .50 .10 "Ploterps Manual, Example 3" color=1 textsize=5 laborigin=MTE
********************************************************************
This example is very similar to the last. I have plotted channel 2 from
several
different bins. Notice that we could not use the channelorder and
plotchans commands this time since we use only one channel from each bin.
Also notice the special handling of the calibration bar label.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex4.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 4 - several bins from one channel
**
** comments appear just before relevant command
**
timewindow -100 799
topmargin .10
bottommargin .20
** In setting the left hand margin, I left plenty of space to put some
** labels there.
rightmargin .09
leftmargin .28
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
gridstyle medium
calibration -2.0
ticklength .02
textstyle complex textsize=5 thin laborigin=LTE
text .31 1.273 "Congruent"
text .63 1.273 "Anomalous"
** The following labels appear on the left side of the page, so I
** wanted them left-justified and used "lte" (left top, extra space)
** for the "laborigin". Notice that I set laborigin to "lte" in the
** above textstyle command.
textstyle textsize=4
text .15 .517 "Final"
text .15 .477 "word"
text .15 .65 "Words"
text .15 .61 "13 on"
text .15 .785 "Words"
text .15 .74 "10-12"
text .15 .92 "Words"
text .15 .88 "7-9"
text .15 1.055 "Words"
text .15 1.015 "4-6"
text .15 1.19 "Words"
text .15 1.15 "2&3"
** Specify a 7 row by 2 column grid for channel locations. The
** 4.5 is for the vertical spacing. It means that there is room
** for 4.5 microvolts on each side of the baseline. Another way to
** look at it is that each baseline is 9 microvolts apart from the
** baseline above or below it.
chanlocations 7 2 4.5
** Ticks every 100 milliseconds.
ticks 100
** No polarity indicators.
polindicators no
** I don’t wan’t to the channel label (Cz) to show up, so I set the
** text color to 0.
textstyle color=0
** Draw several bins with a medium-width, solid line, color 1.
linestyle medium color=1 solid
bin 1
chan 2 0
bin 3
chan 2 2
bin 5
chan 2 4
bin 7
chan 2 6
bin 9
chan 2 8
bin 12
chan 2 10
** I’m going to draw in the same locations again, so no need to redraw
** the baseline grids and calmarks. Turn grid drawing off.
grids off
** Now draw several bins with thick-width, dashed lines, color 2.
** Note that color command does nothing for laserjet.
linestyle thick color=2 dashed
bin 2
chan 2 0
bin 4
chan 2 2
bin 6
chan 2 4
bin 8
chan 2 6
bin 10
chan 2 8
bin 11
chan 2 10
** Now I’m going to plot data in locations that I haven’t used yet, so
** I turn grid drawing back on.
grids on
** Plot several more bins of data with medium solid lines, color 1.
linestyle medium color=1 solid
bin 13
chan 2 1
bin 15
chan 2 3
bin 17
chan 2 5
bin 19
chan 2 7
bin 21
chan 2 9
bin 24
chan 2 11
** Turn the grid drawing off again.
grids off
** Plot several more bins of data with thick-width, dashed lines, color
2.
linestyle thick color=2 dashed
bin 14
chan 2 1
bin 16
chan 2 3
bin 18
chan 2 5
bin 20
chan 2 7
bin 22
chan 2 9
bin 23
chan 2 11
** Prepare to draw the baseline/calibration stuff at the bottom.
textstyle complex textsize=4.5 color=1 thin
** Put a timeline only at location 13
drawtimeline 13
** Put labeled ticks at 0 and 400 milliseconds. The ticklength
** is .04. Override the default label for 400, the extra spaces
** will center the tickmark over the "400".
labelticks .04 0 400 lab=" 400 ms"
** Put ticks without labels at 200 and 600 milliseconds. The .02
** is the ticksize.
nolabelticks .02 200 600
** The voltline and microvolt label is a bit tricky here. When I
** did it the first time, I noticed that everything was crammed
** together because the voltline is only 2.0 microvolts long.
** So, in order to spread it out, I draw the voltline without the
** normal microvolt label (the N).
drawvoltline 0 right N
** Now put the microvolt label in using the chanlabel command. This
** puts my label to the left of location 13. Since I am supplying a
** label, the channel number argument (0) is ignored. Also notice
** the nroff-like way of specifying the Greek micro sign. Available
** Greek and special characters are in the appendix.
chanlabel 0 13 labpos=L "2.0 \(*mV"
** Now put the polarity indicator up.
labelpol neg
** Finally some more explanatory text.
textstyle textsize=4
text .34 .28 "high freq"
line .19 .26 .31 .26 medium color=1 solid
text .34 .24 "low freq"
line .19 .22 .31 .22 thick color=2 dashed
text .50 .10 "Ploterps Manual, Example 4" textsize=5 laborigin=MTE
********************************************************************
This example demonstrates shading between waveforms, and labeled arrows.
The shade command is used to shade the area between two erps. The arrow
command is used to point out an N400.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex5.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 5 - shading between waveforms, and labeled arrows
**
** comments appear just before relevant command
**
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
topmargin .09
calibration -5.0
ticklength .02
bin 2
text .5 1.333 "Ploterps Manual, Example 5" italic textsize=6 medium laborigin=MTE
chanlocations 5 2 8
channelorder 0 1 2 3 4 5 6 7
linestyle thin
textstyle complex textsize=5 thin
gridstyle thin
ticks 200
polindicators no
plotchans
linestyle dotted medium
grids off
bin 3
plotchans
drawtimeline 9
labelticks .04 0 400 800
nolabelticks .02 200 600
drawvoltline -100 right mr
labelpol neg
text .75 0.01 (msec) italic textsize=6.5 laborigin=BME
line .45 .16 .31 .16 solid
textstyle laborigin=lm complex textsize=6.5
text 0 0.16 "Congruent"
line .45 .09 .31 .09 dotted
text 0 0.09 "Anomalous"
** shade the area between the two waveforms in location 0.
** Start at 175 milliseconds, end at 425. First waveform
** is from file g28.gnd, bin 2, chan 0. Second waveform
** is from file g28.gnd, bin 3, chan 0.
shade 0 175 425 /usr/local/erp/man/man7/ploterps_manual/g28.gnd 2 0 /usr/local/erp/man/man7/ploterps_manual/g28.gnd
3 0
** Similar shading for locations 2 and 3.
shade 2 350 760 /usr/local/erp/man/man7/ploterps_manual/g28.gnd 2 2 /usr/local/erp/man/man7/ploterps_manual/g28.gnd
3 2
shade 3 300 800 /usr/local/erp/man/man7/ploterps_manual/g28.gnd 2 3 /usr/local/erp/man/man7/ploterps_manual/g28.gnd
3 3
** put an arrow in location 3 with the arrowhead at 400 milliseconds
** and -4.5 microvolts. Use an empty string for the arrow label.
arrow 3 400 -4.5 DL N400 laborigin=LME textsize=4.5
********************************************************************
This example shows how to use the procfunc command to add or subtract
the data that may be contained in a second dataset, or processing
function,
in the same data file. The purpose of providing this feature
is to facilitate
the plotting of standard error around erps. In this
example I have used
the procfunc command to plot the average, the
average plus standard error,
and the average minus standard error from a
data file called flex.avg,
which has standard error as its second
processing function. I have also
demonstrated the use of predefined strings
in the text command.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex6.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 6 - standard error
**
** comments appear just before relevant command
**
file /usr/local/erp/man/man7/ploterps_manual/flex.avg
** Put some margins on the plot. This must go before any
** chanlocations commands.
rightmargin .10
leftmargin .10
topmargin .10
** Set the current font to be the complex font since we will
** use this font for the entire plot.
textstyle complex
** Center a title at the top of the page.
text .5 1.33 "Standard Error" textsize=6 laborigin=tme
timewindow -100 800
bin 61
ls thin
** Don’t draw the positive and negative signs on all the
** cal bars.
polindicators no
calibration -7.5
chanlocations 6 3 10
** Use channelorder command so we can plot the first 15 channels
** with one plotchans command.
channelorder 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
plotchans
** No need to redraw the grids again.
grids off
** We will draw the standard error with dotted, medium width lines,
** using color 2.
ls color=2 dotted medium
** The standard error is in processing function 1 of the data file.
** We want to plot the erp +/- the standard error. Do the "+" part
** first.
procfunc 0 + 1
plotchans
** Now do the erp minus the standard error.
pf 0 - 1
plotchans
** Draw a timeline and a calibration bar at the bottom.
drawtimeline 17
drawvoltline 0 right M
labelticks .04 0 400 800
nolabelticks .02 200 600
labelpol neg
textstyle textsize=4
** We can get text out of the data file by using "%v" in the
** textstring and supplying one of the predefined string names.
** (Available predefined strings are listed in Appendix B).
text .12 .1 "%v" expdesc laborigin=ble
text .12 .1 "Condition: %v" condesc laborigin=tle
********************************************************************
This example merely shows all of the possible values for the labpos argument
for
the chan and chanlabel commands. Notice that the chanlabel command
can be used to label the electrode as well as for displaying explanatory
text. In
certain situations, it is easier to use the chanlabel command
than it is to
use the text command when displaying arbitrary text.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex7.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 7 - labpos
**
** comments appear just before relevant command
**
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
bin 2
topmargin .15
text .5 1.33 "Chanlab labpos Locations" laborigin=tme textsize=7 medium Complex
ts textsize=6.0 italics
chanlocations 3 3 7.5
db 1
chanlabel 0 1 labpos=L "labpos=L"
chanlabel 0 1 labpos=R "labpos=R"
chl 0 1 labpos=T "labpos=T"
db 3
chl 0 3 labpos=TL "labpos=TL"
db 5
chl 0 5 labpos=TR "labpos=TR"
*db 7
chan 3 7 labpos=N
ts complex textsize=5.35
chl 0 7 labpos=T "Wernicke’s "
chl 0 7 labpos=L "Low Frequency"
ts textsize=2.9
** When putting double quotes within a quoted string, precede the
** double quotes with the backslash () character
arrow 7 250 5.5 UL "You can Use
arrow 7 250 5.5 UL "label electrode and for explanatory text" laborigin=tle
********************************************************************
This example shows all of the possible locations for the microvolt label
when using the drawvoltline command.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex8.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 8 - drawvoltline
**
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
bin 1
tm .15
rm .10
gs thin
ticklength .02
ts complex thin textsize=5.5
tx .5 1.33 "Drawvoltline \(*mV-Label Positions" laborigin=mte textsize=7
cl 6 3 6
cal -7.5
vlabelprec 1
dt 2
dv 0 right tl textsize=4
labelpol pos
chl 0 0 labpos=TL "Label Position is TL"
dt 5
dv 0 right t textsize=4
labelpol pos
chl 0 3 labpos=TL "Label Position is T"
dt 8
dv 0 right tr textsize=4
labelpol neg
chl 0 6 labpos=TL "Label Position is TR"
dt 11
dv 0 right ml textsize=4
labelpol neg
chl 0 9 labpos=TL "Label Position is ML"
dt 14
dv 0 right m textsize=4
labelpol neg
chl 0 12 labpos=TL "Label Position is M"
dt 17
dv 0 right mr textsize=4
labelpol neg
chl 0 15 labpos=TL "Label Position is MR"
********************************************************************
This example just shows all the possible values for the windowtext
command’s
position argument.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex9.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 9 - windowtext
**
topmargin .05
bottommargin .35
leftmargin .1
rightmargin .1
textstyle textsize=7.5 complex
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
bin 1
** I am dividing the window into a 2 x 2 array
** so therefore there are 4 subwindows numbered 0 through 4
chanlocations 2 2 10 show
**
** Using the windowtext command to show where the
** 9 preset locations appear in the subwindow
** All the text will be displayed in window number 3
**
windowtext 3 TR "TR" textsize=5
windowtext 3 TM "TM" textsize=5
windowtext 3 TL "TL" textsize=5
windowtext 3 MR "MR" textsize=5
windowtext 3 M "M" textsize=5
windowtext 3 ML "ML" textsize=5
windowtext 3 BR "BR" textsize=5
windowtext 3 BM "BM" textsize=5
windowtext 3 BL "BL" textsize=5
********************************************************************
This example shows the solid, dashed, and ldashed linetypes in the
varying
line widths.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex10a.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 10a - linetypes (solid, dashed, ldashed) and linewidths
**
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
bin 4
topmargin .1
leftmargin .2
textstyle complex textsize=4.5 thin
text .5 1.33 "Linetype and Linewidth Examples" laborigin=tm textsize=6.0 medium
cl 5 3 7.5
linestyle solid verythin
chan 13 0 labpos=t " solid "
ls dashed
chan 13 1 labpos=t " dashed "
ls ldashed
chan 13 2 labpos=t " ldashed "
ls thin solid
chan 13 3 labpos=n
ls dashed
chan 13 4 labpos=n
ls ldashed
chan 13 5 labpos=n
ls medium solid
chan 13 6 labpos=n
ls dashed
chan 13 7 labpos=n
ls ldashed
chan 13 8 labpos=n
ls thick solid
chan 13 9 labpos=n
ls dashed
chan 13 10 labpos=n
ls ldashed
chan 13 11 labpos=n
ls verythick solid
chan 13 12 labpos=n
ls dashed
chan 13 13 labpos=n
ls ldashed
chan 13 14 labpos=n
textstyle laborigin=m
text .110 1.081 verythin verythin
text .110 .842 thin thin
text .110 .600 medium medium
text .110 .360 thick thick
text .110 .121 verythick verythick
********************************************************************
This example shows the solid, dotted, and dotdash linetypes in the
varying
line widths.
View the plot for the following command file by entering:
ploterps pxwin /usr/local/erp/man/man7/ploterps_manual/ex10b.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 10b - linetypes (solid, dotted, dotdash) and linewidths
**
file /usr/local/erp/man/man7/ploterps_manual/g28.gnd
bin 4
topmargin .1
leftmargin .2
textstyle complex textsize=4.5 thin
text .5 1.33 "Linetype and Linewidth Examples" laborigin=tm textsize=6.0 medium
cl 5 3 7.5
linestyle solid verythin
chan 13 0 labpos=t " solid "
ls dotted
chan 13 1 labpos=t " dotted "
ls dotdash
chan 13 2 labpos=t " dotdash "
ls thin solid
chan 13 3 labpos=n
ls dotted
chan 13 4 labpos=n
ls dotdash
chan 13 5 labpos=n
ls medium solid
chan 13 6 labpos=n
ls dotted
chan 13 7 labpos=n
ls dotdash
chan 13 8 labpos=n
ls thick solid
chan 13 9 labpos=n
ls dotted
chan 13 10 labpos=n
ls dotdash
chan 13 11 labpos=n
ls verythick solid
chan 13 12 labpos=n
ls dotted
chan 13 13 labpos=n
ls dotdash
chan 13 14 labpos=n
textstyle laborigin=m
text .110 1.081 verythin verythin
text .110 .842 thin thin
text .110 .600 medium medium
text .110 .360 thick thick
text .110 .121 verythick verythick
********************************************************************
This example shows the basic colors available in ploterps.
View the plot for the following command file by entering:
ploterps xwin /usr/local/erp/man/man7/ploterps_manual/ex11.plt
********************************************************************
**
** Ploterps User’s Manual
** Example 11 - colors
**
textstyle laborigin=m textsize=8 complex thick
text .500 .690 "Plotting Colors" color=1
line .200 .640 .800 .640 color=1
textstyle laborigin=lm textsize=8 complex thick
text .155 .580 "(no color = 0)" color=1 textsize=4 thin
text .160 .516 "color=1" color=1
text .160 .446 "color=2" color=2
text .160 .377 "color=3" color=3
text .160 .308 "color=4" color=4
text .160 .238 "color=5" color=5
text .160 .169 "color=6" color=6
text .160 .100 "color=7" color=7
text .540 .585 "color=8" color=8
text .540 .516 "color=9" color=9
text .540 .446 "color=10" color=10
text .540 .377 "color=11" color=11
text .540 .308 "color=12" color=12
text .540 .238 "color=13" color=13
text .540 .169 "color=14" color=14
text .540 .100 "color=15" color=15
********************************************************************
linestyle (ls) [color linewidth linetype]
gridstyle (gs) [color linewidth linetype]
textstyle (ts) [font textsize labelorigin color linewidth linetype]
labelprec (lp) precision
vlabelprec (vlp) precision
bottommargin (bm) value
topmargin (tm) value
leftmargin (lm) value
rightmargin (rm) value
chanlocations (cl) rows cols [scalefactor horzspace show]
channelorder (co) chan [chan ...]
position (po) posnumber xl xr yt yb scalefactor [horzspace]
scalefactor (sf) scalefactor posnumber
definefiles (df) file0 [file1 file2 ...]
file (f) file_spec
bin (bn) bin_number
procfunc (pf) pfnumber1 [ +/- pfnumber2 ]
ticks (tk) mspertick
calibration (cal) uVolts
smooth (sm) npts(on, yes)/off(no)
filter (fi) off/no
filter (fi) filter_type cutoff [cutoff]
timewindow (tw) start_time end_time
baseline (bl) start_time end_time
ticklength (tl) value
calmarks (cm) on(yes, y, 1)/off(no, n, 0)
tickmarks (tkm) on(yes, y, 1)/off(no, n, 0)
grids (g) on(yes, y, 1)/off(no, n, 0)
polindicators (pi) on(yes, y, 1)/off(no, n, 0)
ticklabels (tlb) on(yes, y, 1)/off(no, n, 0)
utick (ut) time [+/-] [label] [font labelorigin textsize color linewidth
linestyle]
clearuticks (ct)
plotheader (ph) [font textsize color linewidth linestyle]
chan (ch) chan_num posnumber [labpos] [label] [font textsize col linewidth
linetype]
chanlabel (chl) chan_num posnumber [labpos] [label] [font textsize col
linewidth linetype]
plotchans (pc) [yes(on 1)/no(off 0)]
plotallchans (pa) [rows cols scalefactor [horzspace]]
drawbaseline (db) posnumber [color linewidth linetype]
drawcal (dc) posnumber calorigin [label] [font textsize color linewidth
linetype]
line (l) startx starty endx endy [color linewidth linetype]
lineto (lt) x y [color linewidth linetype]
moveto (mt) x y
text (tx) xcrd ycrd string [string args] [font labelorigin textsize color
linewidth linestyle]
arrow (ar) location time volt arrowcode label [textstyle options]
shade (sh) location start end file1 bin1 chan1 file2 bin2 chan2 [greyscale]
rect (r) ULx ULy LRx LRy [fill] [color linewidth linestyle]
newpage (np)
readfile (rf) filename xtrans ytrans
pause (pz)
windowtext (wt) posnumber position text [font labelorigin textsize color
linewidth linestyle]
drawtimeline (dt) location [linestyle options]
labelticks (lbt) size time [lab="mylabel"] [time [lab="mylabel"]...][text
options]
nolabelticks (nlt) size time [time ...]
drawvoltline (dv) time tickmarkpos labelpos [text options]
labelpol (lbp) neg [pos]
7. APPENDIX B - List of Predefined StringsNote: those entries marked with
a "*" require an integer index to
specify the element of an array.
"bindesc". . . . . . . . bin description
"binnumber". . . . . . . bin number
"chans". . . . . . . . . number of channels
"chndesc". . . . . . . . channel descriptions *
"clktick". . . . . . . . 10s of microseconds in a single clock tick
"condcode" . . . . . . . condition code
"condesc". . . . . . . . condition description
"cprecis". . . . . . . . channel precision in # 256 point blocks
"epoch". . . . . . . . . epoch length in milliseconds
"event". . . . . . . . . raw event number
"expdesc". . . . . . . . experiment description
"filename" . . . . . . . current data file name
"itemnumber" . . . . . . sequential item number
"odelay" . . . . . . . . milliseconds from trigger to stimulus
"pp10uv" . . . . . . . . points per 10 microvolts
"presampling". . . . . . pre-event time in milliseconds
"prfdesc". . . . . . . . processing function descriptions *
"rejcounts". . . . . . . individual rejection counts *
"rejdesc". . . . . . . . descriptions for rejection functions *
"subdesc". . . . . . . . subject description
"sums" . . . . . . . . . number of sums in average
"totevnt". . . . . . . . total log events
"totrawrecs" . . . . . . total raw records including rejects
"totrejects" . . . . . . total raw rejects
"tpfuncs". . . . . . . . total number of processing functions
"trfuncs". . . . . . . . total number of rejection functions
Input Character Input Character
Name Name Name Name
\(de degree \(dg dagger
\(** math star \(sc section
\(*a alpha \(*b beta
\(*g gamma \(*d delta
\(*e epsilon \(*z zeta
\(*y eta \(*h theta
\(*i iota \(*k kappa
\(*l lambda \(*m mu
\(*n nu \(*c xi
\(*o omicron \(*p pi
\(*r rho \(*s sigma
\(*t tau \(*u upsilon
\(*f phi \(*x chi
\(*q psi \(*w omega
\(*A ALPHA \(*B BETA
\(*G GAMMA \(*D DELTA
\(*E EPSILON \(*Z ZETA
\(*Y ETA \(*H THETA
\(*I IOTA \(*K KAPPA
\(*L LAMBDA \(*M MU
\(*N NU \(*C XI
\(*O OMICRON \(*P PI
\(*R RHO \(*S SIGMA
\(*T TAU \(*U UPSILON
\(*F PHI \(*X CHI
\(*Q PSI \(*W OMEGA
\(sr square root \(>= >=
\(<= <= \(-> right arrow
\(mu multiply \(di divide
\(+- plus minus \(-+ minus plus
\(if infinity \(pd partial derivative
\(gr gradient \(is integral sign
\(es empty set \(mo member of
\(li line integral sign \(cp continued product sign
\(dp dot product \(cs continued sum
\(t2 alternate theta \(<< large open angular bracket
\(>> large close angular bracket
Table of Contents