This section describes how PHPlot calculates the range of World Coordinate space, and the tick increment for each axis, when those values have not been set manually.
In order to plot a data set, PHPlot needs to map the data points from world coordinate space to device coordinate space. The world coordinates of the data points are generally in realworld units, such as kilometers, seconds, degrees centigrade, etc. Device coordinates are pixels in an image file, display screen, or paper hardcopy. PHPlot knows the pixel coordinates of the limits of the plot area in device space. In order to translate data points from world coordinates into pixel coordinates, it needs to know the limits of world coordinate space. These can be provided manually, or computed by PHPlot.
The limits of world coordinate space are the endpoints of the X and Y axis lines. The two endpoints of an axis are also referred to as the plot range along that axis. Each of those plot ranges is divided into uniformly sized tick intervals, which may or may not be marked by visible tick marks. The space between tick marks is called the tick increment. PHPlot will calculate a suitable tick increment if necessary.
The rest of this section describes how PHPlot calculates the limits of the plot ranges, and the tick increments.
This section does not apply to pie charts, which do not have X or Y axis lines.
If you use SetPlotAreaWorld to set both
Xmin
and Xmax
to nonNULL values,
this will fix both ends of the X axis. Similarly, setting both
Ymin
and Ymax
to nonNULL values
will fix both ends of the Y axis.
If you set both ends of an axis, PHPlot will use exactly that for the plot
range on that axis. This is recommended if your data has a range which is
known, predictable, and/or 'natural'. For example, if your Y axis represents
percentage values from 0 to 100, you might use this to fix the Y axis ends:
$plot>SetPlotAreaWorld(NULL, 0, NULL, 100);
You can use SetXTickIncrement to set a fixed X tick increment, and SetYTickIncrement to set a fixed Y tick increment. If you set a fixed tick increment along an axis, PHPlot will use exactly that value (even if it results in too many, or too few tick marks).
Another way to manually specify tick increments is with
SetNumXTicks or SetNumYTicks.
These set the desired number of tick intervals.
PHPlot will then calculate the tick increment by dividing the plot range into
exactly that many intervals.
For example, if the Y axis range is 0 to 17, and you use
$plot>SetNumYTicks(6)
to get 6 intervals, then the tick
mark spacing will be about 2.833333.
This is probably not a desirable value for the tick increment.
Using SetNumXTicks
or SetNumYTicks
to set the number of tick intervals is usually not a good idea, unless you
also set the plot range with SetPlotAreaWorld.
If you set both ends of the X or Y plot range, PHPlot will position tick marks starting from the lower end of that range, by default. For example, if you set the Y axis range from 1 to 20, with a tick increment of 5, the tick marks will be at 1, 6, 11, and 16. You can set a tick anchor to change this. See Section 4.6.9, “Tick Positions” below for more information.
If you use SetPlotAreaWorld to set either
Xmin
or Xmax
(but not both) to a
nonNULL value, or you set either
Ymin
or Ymax
(but not both) to a
nonNULL value, then you are fixing one end of the plot range.
PHPlot will use your fixed value for that end of the range, and calculate the
other end as described below.
This may make sense if your data has a fixed lower (or upper) bound. For example, if you are plotting outdoor summer temperatures in degrees Fahrenheit, you might use something like this:
$plot>SetPlotAreaWorld(NULL, 60);
which sets Ymin to 60.
Since PHPlot positions tick marks starting from the lower limit of the X
and Y plot ranges, setting Xmin
or Ymin
establishes the basis for tick marks along that axis.
For example, if you set the lower end of the X axis range to 12, with a tick
increment of 10, the tick marks along X will be at 12, 22, 32, etc.
You can set a tick anchor to change this.
See Section 4.6.9, “Tick Positions” below for more information.
Each end of the plot range (for both X and Y) which is not specified using SetPlotAreaWorld will be calculated by PHPlot. The algorithm used by PHPlot to calculate the ends of a plot range is not perfect, but is meant to create reasonable plots in a majority of cases.
This description applies starting with PHPlot6.0.0. In older versions, PHPlot used a more simplistic approach to calculating the plot range.
Here is an overview of the method, using the Y axis as an example.
DataMin
and DataMax
are the smallest and largest Y values in the data array (or the values
derived from the data array, for some plot types).
PlotMin
and PlotMax
are the
calculated limits of the plot range. Remember that PHPlot only calculates
PlotMax and PlotMin if they have not already been set using
SetPlotAreaWorld.
Initialization: Start by setting the plot range to the data range: PlotMin = DataMin, and PlotMax = DataMax. The plot range will now include all the data, but just barely.
Zero Magnet: If the plot range does not currently include zero (PlotMin <= 0 and 0 <= PlotMax), PHPlot tries to extend the range so it will begin or end at zero. This is done by either by moving PlotMin down to zero (for positive data), or by moving PlotMax up to zero (for negative data). There is a limit as to how far PHPlot is willing to extend the plot range to include zero (see below).
End Adjust / Increase range: PHPlot may then adjust the ends of the plot range by a factor (adjust_amount) multiplied by the plot range. This provides some extra room above or below the data for data value labels, and prevents the top (or bottom) of the plotted data from hitting the edge of the plot area, which would detract from the appearance. Adjustment only occurs for PlotMin if it is negative, and for PlotMax if it is positive. This corresponds to the direction of the plot data 'away' from zero. For example, if 0 < PlotMin < PlotMax (all the data is positive), then PlotMax will be increased. If PlotMin < PlotMax < 0 (all the data is negative), then PlotMin will be decreased. If the range spans 0 with PlotMin < 0 < PlotMax, then both PlotMin will be decreased and PlotMax will be increased.
End Adjust / Finalize:
PHPlot may then adjust the ends of the plot range further, based on the
adjustment mode (adjust_mode
), but only if the end of
the range is not already zero.
If the adjustment mode is T
(Tick, the default mode),
then PlotMax is adjusted up to the next higher tick mark position, and
PlotMin is adjusted down to the next lower tick mark position. Note that
this happens even if tick marks are not displayed.
If a tick anchor is set, this is taken into account when adjusting the
range (see Section 4.6.9, “Tick Positions” below).
If the adjustment mode is R
(Range), there is no further
adjustment.
If the adjustment mode is I
(Integer), then PlotMax is
adjusted up to the next higher integer, and PlotMin is adjusted down to the
next lower integer.
Note that adjustment modes R
and I
are
intended for special applications where the default T
(Tick) mode produces undesirable results.
The range calculation uses parameters which can be set using TuneXAutoRange and TuneYAutoRange. The parameters are described in the next sections.
The zero_magnet
parameter sets the strength of the
zero magnet, which controls how far PHPlot is
willing to extend the plot range to include zero, if the range does not
already include zero.
This is a floating point number greater than or equal to 0, and less than
or equal to 1.0.
A value of 0 means the magnet is disabled. PHPlot will not extend the
plot range to include zero.
A value of 1.0 means the magnet strength is infinite, and PHPlot will
always extend the plot range to include zero.
For increasing values between 0.0 and 1.0, the zero magnet becomes
stronger, and PHPlot will go further in extending the plot range to include
zero.
Between 0 and 1.0, the zero magnet factor is applied as follows. Let ZF = zero_magnet / (1  zero_magnet). This maps the zero magnet domain of 0:1 into the range 0:infinity. PHPlot will extend the plot range to include 0 if that increases the range by a factor of ZF or less.
For example, the default zero magnet factor of 6/7 results in ZF=6, so PHPlot will 'pull' the bottom of the plot range down to zero if doing so will stretch the range by 600% or less. (If the data is all negative, it will apply the zero magnet to 'pushing' the top of the range to zero in the same manner.) If the data range is 500 to 600, PHPlot will adjust it to become 0 to 600, because adjusting the lower range by 500 is less than the maximum 6*100. If the data range is 900 to 1000, PHPlot will not adjust it to start at 0, because the zero magnet is only 'strong enough' to pull the range down by 6*100, not 900.
The purpose of having a variable zero magnet is to balance two conflicting goals:
Including zero in the data range improves the utility of plots, as it facilitates comparing relative changes and differences.
Increasing the data range of a plot makes it harder to see relatively small changes in the data.
The adjust_mode
parameter selects from one of three methods
PHPlot has to determine how to extend the end of the plot range.
The value is a single character, selecting one of three available modes.
Mode  Description 

T  Tick : extend the range, then to the next tick mark. 
R  Range : extend the range. 
I  Integer : extend the range, then to the next integer value. 
The default mode is T
, extend to tick mark.
Mode I
, extend to integer, is similar to the only
available adjustment mode available before PHPlot6.0.0.
Range end adjustment works as follows. First, PlotMin is extended by the
adjust_amount
multiplied by the current plot range
(PlotMax  PlotMin), but only if PlotMin is negative.
Similarly, PlotMax is extended by the same factor, but only if it is positive.
This corresponds to where extra spaces is needed for data value labels, for
example: above positive values, and below negative values.
For adjustment mode R
, that completes the range adjustment.
For example, in this mode, with the adjustment amount 0.025 and data range
10 to 10, the resulting plot range will be 10.5 to 10.5 (having been
increased at each end by 2.5% times the range of 20).
For adjustment mode I
, after extending the range,
PHPlot adjusts PlotMin to the next lowest integer, and PlotMax to the
next highest integer.
Using the same example, with the data range 10 to 10 and adjustment amount
0.025, the range is now 11 to 11 using this mode.
Adjustment mode T
, the default mode, extends the range
end to coincide with a tick mark (or where a tick mark would be, if tick
marks were displayed).
In this mode, the adjust_amount
can be thought
of as controlling the minimum amount of space between the top of the data and
the next tick mark. If the adjust_amount is 0, and the maximum data value
DataMax falls exactly on a tick mark, PHPlot will end the range at that
tick mark. But if adjust_amount is greater than zero, PHPlot will extend
the range by least one additional tick mark.
As an example of T
adjustment mode, consider a Y tick
increment of 10, and adjustment amount 0.03 (3%). With the Y data range of
0 to 95, PHPlot will set PlotMax to 100  extending the range to the very
next tick mark. But with a Y data range of 0 to 98, the space between
DataMax and the tick mark at 100 is not enough (less than 3% of the range),
so the range is extended to the next tick mark after that, at 110.
The adjust_amount
parameter determines how much the end of
the plot range is adjusted (extended) to leave room for labels within the plot
area, or to keep the plotted data from pushing against the edge of the plot.
This is a nonnegative floating point value.
See the previous section on how the adjust_mode and adjust_amount parameters are used.
The default adjustment amount is either 0 (0%) or 0.03 (3%), depending on the plot type, and on whether the adjustment is for the independent or dependent variable. For most plot types, the default is 0 for the independent variable axis (X for vertical plots, Y for horizontal plots), and 3% for the dependent variable axis. Some plot types, such as bubbles, benefit from adjustment on both axes, so the default value for adjust_amount is 3% for both X and Y. Other plot types, such as candlesticks, do not need any adjustment on either axis (because there are no data value labels, and the upper wick may touch the edge of the plot without harm). For these plot types, the default adjust_amount is 0% for both X and Y.
Here are 3 examples showing the automatic range calculation. Each figure contains 4 plots showing the cumulative effect of the numbered steps which are detailed above: (1) initial range, (2) zero magnet, (3) end adjust / increase range, and (4) end adjust / finalize ('Tick' adjust mode).
In the first example, the data is all positive (DataMin > 0).
Note: You can see in the figure about that the small adjustment to the top
of the Y range made at step (3) is completely absorbed in the adjustment to
reach the tick mark at step (4). This is not always the case. If the data
happens to end exactly at a tick mark, and the adjust_amount
parameter is greater than zero, PHPlot will extend the range to the next tick
mark after that. In a more general sense, when using 'Tick' adjustment mode,
the adjustment amount represents the minimum amount of space to allow between
the data maximum and the plot maximum.
In the second example, the data is both negative and positive. Note that in this case the zero magnet (step 2) has no effect, and the end adjustment (step 3) and adjust to tick (step 4) change both ends of the range.
The third example has all negative data (DataMax < 0). Zero magnet applies to the top of the range, and end adjustment applies to the bottom of the range in this case.
There are two implied independent variable cases:
When using the textdata data type, you are drawing a vertical plot without specifying the values of the independent variable X. Your data array will contain only labels and Y values; the X values are implied.
When using the textdatayx data type, you are drawing a horizontal plot without specifying the values of the independent variable Y. Your data array will contain only labels and X values; the Y values are implied.
If you don't use SetPlotAreaWorld to set an explicit plot range for the independent variable axis in these cases, PHPlot sets the plot range for that axis to start at 0 and end at N, where N is the number of rows in the data array. The automatic range calculation is disabled for that axis, and there is no adjustment.
For both the X axis and Y axis, PHPlot calculates a tick increment if it was not set as described above. (Even if your plot has no visible tick marks along an axis, PHPlot will calculate a tick increment for that axis, since it might be used in adjusting the plot range.)
This description applies starting with PHPlot6.0.0. In older versions, PHPlot simply divided the plot range by 10 to calculate the tick increment.
PHPlot uses one of three methods to calculate a tick increment for a given plot
range (along the X or Y axis). The methods are called 'decimal', 'binary',
and 'date', and correspond to the tick_mode
parameter
described below. By default, PHPlot uses the 'decimal' method, unless the axis
labels have been set up to use date/time labels, in which case it uses the
'date' mode.
You can also directly control which method is used.
Using the decimal method, PHPlot picks the largest tick
increment which is 1, 2, or 5 times a power of 10, and which divides the plot
range into no fewer than the acceptable minimum number of tick intervals
(see the min_ticks
parameter below).
The power of 10 can be negative, zero, or positive.
Valid automaticallycalculated tick increments for nonnegative powers of 10
include 1, 2, 5, 10, 20, 50, 100, 200, 500, etc.
Valid automaticallycalculated tick increments for negative powers of 10
include 0.5, 0.2, 0.1, 0.05, 0.02, 0.01, etc.
For example, given the range 0 to 135, and the default min_ticks of 8, PHPlot will calculate a tick increment of 10, giving 14 intervals. A tick increment of 20 is not valid because it would result in only 7 intervals, fewer than the minimum of 8. A tick increment of 5 would result in 28 intervals; while there is no explicit maximum, PHPlot chooses the largest valid increment that produces 8 or more intervals.
When using the date method, PHPlot selects from a
predefined list of valid tick increments.
The list covers a range from 1 second up to 7 days: 1, 2, 5, 10, 15, and 30
seconds; 1, 2, 5, 10, 15, and 30 minutes, then 1, 2, 4, 8, 12, 24, 48, 96,
and 168 hours.
PHPlot will pick the largest increment in the list which will result in no
fewer than the acceptable minimum number of tick intervals
(min_ticks
).
If the plot area range exceeds 7 days, PHPlot switches to the 'decimal' method
instead, with units of 24 hours.
With the 'date' method, PHPlot never chooses an increment less than 1 second.
For example, given a date/time range which spans 1020 seconds (17 minutes), and the default min_ticks of 8, PHPlot will calculate a tick increment of 120 seconds (2 minutes), which results in 9 intervals. This is the largest increment in the list above which results in 8 or more intervals.
When using the binary method, PHPlot picks the largest
tick increment which is a power of 2, and which divides the plot range into no
fewer than the acceptable minimum number of tick intervals
(see the min_ticks
parameter below).
The power of 2 can be negative, zero, or positive.
Valid automaticallycalculated tick increments for nonnegative powers of 2
include 1, 2, 4, 8, 16, 32, etc.
Valid automaticallycalculated tick increments for negative powers of 2
include 1/2, 1/4, 1/8, 1/16, 1/32, etc. (Note that PHPlot will label these
ticks as decimal values 0.5, 0.25, 0.125 etc., not fractions.)
The tick increment calculation uses parameters which can be set using TuneXAutoTicks and TuneYAutoTicks. The parameters are described in the next sections.
The min_ticks
parameter sets the minimum number of tick
intervals along the axis. The default is 8.
The maximum number of tick intervals will be about 2.5 times the minimum.
(The value 2.5 comes from the largest ratio between adjacent acceptable tick
increment values. This is true for the 'decimal' method choices of 1, 2, 5,
10, 20, etc., the 'binary' method, and also for the predefined list of values
used with the 'date' method.)
Therefore, by default, a plot range will have between 8 and 20 tick intervals
inclusive (before possible additional tick intervals resulting from
adjustment to the range, as described above).
The tick_mode
parameter selects one of the three available
methods to calculate the tick increment:
Mode  Description 

decimal  Use a power of 10 times 1, 2, or 5 
binary  Use a power of 2 
date  Use a date/time increment value 
These methods are described in more detail above.
By default, PHPlot will automatically decide which method to use.
It will use the 'date'
method if the corresponding axis
tick labels use date/time formatting (as set with
SetXLabelType('time')
or
SetYLabelType('time')
).
Otherwise is will use the 'decimal'
method.
Use this parameter to override the automatic tick increment method selection.
For example, if you are using date/time values along the X axis,
but have a custom label formatting function (rather than selecting
time
format type), you will probably want to tell PHPlot
to pick a date/time tick increment for the X axis:
// Pick a date/time tick increment on X: $plot>TuneXAutoTicks(NULL, 'date');
Another example is if you are using time value label types along X, but you want tick increments to follow the 1, 2, 5 times power of 10 rule, rather than choosing a date/time value. You can use:
// Format X labels as time values: $plot>SetXLabelType('time', '%H:%M:%S'); // Override default and use the decimal method to pick a tick increment: $plot>TuneXAutoTicks(NULL, 'decimal');
In this third example, your Y axis values represent something like memory sizes which are normally measured in powers of two. You can have PHPlot pick a power of two as a tick increment with:
// Use a power of 2 for the Y axis tick increment: $plot>TuneYAutoTicks(NULL, 'binary');
If the plot range is small enough, PHPlot may select a tick increment which
is less than one  a negative power of two.
This results in a tick increment of 1/2, 1/4, 1/8, 1/16, or smaller.
If you want to use the binary tick method, but do not want fractional tick
increments, set tick_inc_integer
(see next section):
// Valid tick increments are powers of 2 >= 1 $plot>TuneXAutoTicks(NULL, 'binary', TRUE);
The tick_inc_integer
parameter is a boolean flag which
can be set to prevent fractional tick increments.
If the flag is TRUE, PHPlot will not pick a tick increment less than 1.
This may result in fewer than min_ticks
tick intervals.
If the flag is FALSE, PHPlot may pick any tick increment.
The default is FALSE.
The 'decimal' and 'binary' tick increment selection methods (see above)
can select tick increments less than 1, using negative powers of 10 or 2
respectively.
This will happen if the plot range is small enough (and/or
min_ticks
is high enough).
Specifically, a fractional tick increment will result from having a plot
range which is less than the minimum number of ticks.
Set tick_inc_integer
to TRUE if you do not want
fractional tick increments.
This parameter does not apply to the 'date' tick increment selection method,
which does not support tick increments less than 1 second.
In 'date' mode, tick_inc_integer
is assumed TRUE.
PHPlot never picks a tick increment which is greater than one and not a whole number. Selected tick increments are either integers >= 1, or fractions between 0 and 1.
For example, if your Y data range is 0 to 1.8, and you set the
tick_inc_integer
parameter for Y to TRUE,
then PHPlot will use a tick increment of 1, and you will get only
2 tick intervals. This is less than the default minimum of 8
tick intervals.
With tick_inc_integer
FALSE, PHPlot would use a tick
increment of 0.2 in 'decimal' mode, or 0.125 in 'binary' mode.
This section describes how PHPlot positions the tick marks along the X and Y axis (if tick marks are enabled), and the tick label values which result. Several factors are involved:
The plot range minimum (PlotMin), which is the world coordinates of the left end of the X axis, or the world coordinates of the bottom of the Y axis.
Whether or not a tick anchor was set. You use a tick anchor to tell PHPlot that you want tick marks based on a particular world coordinate value.
The plot range adjustment mode adjust_mode
(see above),
if PHPlot calculated the plot range minimum PlotMin automatically.
PHPlot generally places tick marks along each axis starting from the minimum value of the plot range (PlotMin), which is the leftmost position on the X axis, and the bottommost position on the Y axis. Therefore, the value of PlotMin affects all the tick values. For example, if the tick increment is 10, and PlotMin is 12, ticks will be at 12, 22, 32, 42.
A tick anchor can be set with SetXTickAnchor or SetYTickAnchor to change tick mark positions. In the above example, with PlotMin 12, if you set a tick anchor at 10, you will get tick positions 20, 30, 40, etc. Setting the tick anchor shifts all the ticks, so one of them falls at the anchor (or would, if the axis was extended). Thus, a tick anchor can be used to 'round off' the tick values to desired numbers, when the plot range minimum (PlotMin) is not a multiple of the tick increment.
If you set the plot range minimum (PlotMin) to a fixed value with SetPlotAreaWorld, consider that this value will be used as the basis for tick marks along that axis, unless you also set a tick anchor.
If PHPlot calculates the plot range minimum using the default adjustment
mode T
(adjust to tick, see above), then PHPlot will
adjust the range to start at a tick mark. That is, PHPlot acts as if a
tick anchor was set at 0. Your tick marks will be at whole multiples
of the tick increment, unless you set a tick anchor to move them somewhere
else.
In the other adjustments modes  I
(integer) or
R
(range)  PHPlot makes no attempt to correlate the
plot range minimum with the tick increment. Your tick marks will probably
not be located at multiples of the tick increment.
In mode R
(range), your tick marks will likely not even be
whole numbers. As in all other cases, you can set a tick anchor to position
the tick marks at the values you prefer.
Here are some examples of PHPlotcalculated plot ranges and tick positions, when no tick anchor has been set.
Y data min  Y data max  Adjustment Mode  Resulting Plot Range  Resulting Tick Positions  Notes 

7  123  T (tick)  0 to 130  0, 10, 20, ... 130  The bottom of the Y range is extended to zero, and the tick marks are at multiples of 10. The top of the Y range is extended to the next tick mark. 
7  123  R (range)  0 to 126.69  0, 10, 20, ... 120 
With 'R' adjustment mode, the top of the Y range is
extended by 3% of the range. The bottom of the Y range was moved to zero
(due to the zero magnet), so the tick marks are based
at 0 and positioned at whole multiples of 10. The resulting tick marks
are the same as 'T' mode, except the Y axis does not end
at a tick mark.

7  123  I (integer)  0 to 127  0, 10, 20, ... 120 
Here 'I' adjustment mode is almost the the same as the
previous case, except the top of the range is adjusted to the next integer.
Since there is no tick mark there, the difference would not be visible.

17  33  T (tick)  20 to 35  20, 15, 10, 5, 0, 5, 10, ... 35  Both ends of plot range are adjusted to a tick position, with an implied tick anchor at 0. The resulting tick marks are at whole multiples of the tick step 5. 
17  33  R (range)  18.5 to 34.5  18.5, 13.5, 8.5, 1.5, 6.5, ... 31.5  In this mode, the range is simply extended by 3% (by default) at each end, where 3% of (33+17) = 1.5. The resulting tick mark positions are probably not ideal. 
17  33  I (integer)  19 to 35  19, 14, 9, 4, 1, 6, ... 31  In this mode, the range is extended by 3%, and then the ends are extended to the next integer. The tick mark positions are whole numbers now, but not anchored at 0. 
To summarize, setting a tick anchor to force the tick positions to align one tick at a particular value, usually 0, can be useful in the situations including:
You are letting PHPlot calculate the range, but with adjustment mode
R
(range) or I
(integer).
You are manually setting the minimum of the plot range with SetPlotAreaWorld to a value other than zero or a whole multiple of the tick increment.
You are using date/time values along the axis, and you need the ticks to start at a specific date/time value (such as the start of an experiment).
You are using a version before PHPlot6.0.0, which only supports an
adjustment mode similar to I
(integer).
Given incomplete or contradictory data, PHPlot will always produce a positive
plot range (PlotMin < PlotMax), although the range may be somewhat arbitrary.
Here are some of the regressive cases, and how PHPlot produces an automatic
range for them. DataMin
and DataMax
mean the limits of the data from the data array; PlotMin
and PlotMax
are the limits of the plot range.
In all cases, any values specified in SetPlotAreaWorld will
not be altered by PHPlot, even if it results in a plot with no visible data
points.
For positive data, PHPlot uses the range 0 to K, where K is the larger of DataMin and 10. For example, if all the Y values are 5, PHPlot uses 0 to 10 for the Y range; if all the Y values are 150, PHPlot uses 0 to 150. For negative data, PHPlot uses a similar method, setting PlotMax=0.
PHPlot uses the arbitrary range 0 to 10.
If you set PlotMax to a value, but all your data is above that (DataMin > PlotMax), you will get an empty plot. PHPlot will use a range of 0 to PlotMax if your PlotMax is positive, else a range of PlotMax10 to PlotMax. All your data points will be outside the plot area.
If you set PlotMin to a value, but all your data is below that (DataMax < PlotMin), you will get an empty plot. PHPlot will use a range of PlotMin to 0 if your PlotMin is negative, else a range of PlotMin to PlotMin+10. All your data points will be outside the plot area.
If you try to use SetPlotAreaWorld to set a negative range (PlotMin > PlotMax) or zero range (PlotMin == PlotMax), PHPlot will report an error and not produce a plot.
Note that the above range choices happen as part of the "initialization" step of automatic range calculation (see Section 4.6.3, “Automatic Range Calculation” above). That is, the selected range is then subject to the same adjustments as in other cases.
This version of the manual was produced for the
PHPlot Sourceforge project web service site, which requires the logo on each
page.
To download a logofree copy of the manual, see the
PHPlot project downloads
area.