and Water Engineering > Modeling
Software > Slope
Stability Software > XSTABL
> Revision History
XSTABL v5 -- An Integrated Slope Stability Analysis Program
for Personal Computers
Version 5.203--December 23, 1998
- Fixed problem with display of 2-digit year (date) for beyond 1999.
- Fixed the mis-drawn profile in the top-left plot in *.GP4 output
for right-facing slopes.
Version 5.202--February 7, 1997
- Fixed program to allow use of negative values for a GLE
function during execution phase.
- Corrected the "wording" of the reported error message 51.
Version 5.201a--December 17, 1996
- Modified the scaling options by including consistent starting
values on the x and y axes.
Version 5.201--November 21, 1996
- Program automatically uses the moist unit weight if the user
fails to specify the saturated unit weight. Please note that an
error is NOT reported..... the value shows up in the output.
- There were problems with a "diminishing" plot for right-facing
slopes. After the first analysis, the minimum x-axis value on
the plots would start changing. This has been fixed, but could
be avoided by starting the profile with x-minimum of zero.
- The labels on the hard-copy plots failed to print properly if
the scale increment was 2.5. They worked fine for ALL other
- Version 5.200 allowed scale increments that were limited to
factors of 1, 2, 2.5, 5 and 10, unlike the previous versions.
This doubling was (perhaps) too fast and so three more increment
factors--3, 4 and 6--have been added to make sure that the scales
do not expand too quickly. Also, these agree with most scales
(rulers) used by the enginering profession.
Version 5.200--July 10, 1996
Version 5.107--June 10, 1996
- Fixed problem with surcharge load intensities for right-facing
slopes. Program adjusted coordinates, but incorrectly flipped
the VALUES of the load intensities. Not a problem for a single
boundary load, or for loads with the same intensity.
- Improved the scale annotation for the screen plots and hard
copies. This version uses "nice" multiples of 1, 2, 2.5, 5
and 10 for tick-mark increments. Overall this may reduce the
plot resolution, but the hardcopy plots (often included in
reports) will now have a more professional look.
- Fixed some additional minor bugs concerning screen plots and
potential scaling problems.
- Program underestimated the vertical effective stress used to
assign the "undrained" strength values in cases of high pore
Version 5.106--April 15, 1996
- For cases where a soil unit was specified as undrained AND
anisotropic, the program would use the c-value specified in
the isotropic properties table rather than the anisotropic
value. This is an error and may be serious if the isotropic
c-value is greater than the anisotropic one.
- Fixed some formatting errors in the writing of the undrained
input data. Also fixed a very minor formatting error in writing
the input anisotropic parameters.
Version 5.105a--March 15, 1996 (no change)
- In cases where XSTABL generated more than the 100 slices limit,
the program reported ERROR 50. This should have been ERROR 28.
Version 5.105a--January 23, 1996
- Minor problem with *.GP4 plot for cases where the y-coords are
much greater than the x-coords (e.g. y ranges from 5800 to 6000
and x ranges from 400 to 800). Version 5.105 plot did not
always correctly scale within the displayed axes.
Version 5.105--September 22, 1995
Fixed problem concerning the save of a *.?P4 file for cases where
the *.GLE or *.RJM file is read from disk. XSTABL would give a
warning message and allow user to continue with a viewing of the
final plot, which also could not be saved. This problem could be
bypassed if the *.GLE file was NOT read immediately before the
analysis, i.e. perform a dummy run.
Version 5.104--July 1, 1995
Modified program to accommodate negative values of an interslice
force angle function. Previously, only positive values were
allowed by the program.
Version 5.103a---June 16, 1995
Problems in correctly assigning the alternative strength types if
there are more than 9 soil units. In the interface, the new data
were incorrectly assigned to the soil unit that was next to the one
originally intended; i.e. if soil 9 was to be "undrained", the
screen would show soil 10 as being "undrained". This problem
also has occurred for the isotropic/anisotropic option. However,
please note that the resulting data file was written correctly.
Version 5.103---April 20, 1995
- Fixed potential difficulties associated with generating more than
9999 surfaces. Program will now work for up to 32,000 surfaces,
which is the restriction imposed by the adopted integer size.
- Fixed problem with the Ru value NOT being zeroed properly if a new
*.IPT file was read into the tables. This problem would result in
a lower FOS value being calculated as the "old" Ru value may be
used in error. This "bug" can be easily by-passed by explicitly
setting the Ru value in the appropriate table.
- Fixed some minor formatting problems.
Version 5.102 --- March 1, 1995
- The summary table following a "REINFORCEMENT" analysis listed the
driving moments/forces rather than the resisting moments/forces.
This has been fixed and the numeric values are now consistent with
the column label.
- Following use of Janbu's Method, the REINFORCEMENT summary had an
incorrect label for units of resisting force. This has been fixed.
- The y-coordinate for the circle center is now written to TWO decimal
places in the final summary table following a circular search.
Version 5.101a --- February 20, 1995
- Version 5.101 would display the *.GP3 plot even if there was an error
in the data file or if the analysis was terminated.
This problem is tolerable but may lead to some confusion as different
geometries may get mixed-up. Version 5.100 did not do this.
- Version 5.101a now correctly reports ERROR 50.
- Small modification made to the algorithms that control the FOS
iterations. This may eliminate cases where the iteration sequence
incorrectly leads to a very small FOS value.
Version 5.101 --- January, 25, 1995
- Program did not allow a GP3 file to be saved if GP2 file was saved
earlier in the same analysis. This condition can be bypassed by
performing two identical runs .... but was fixed.
- If user exits analysis by pressing ESCAPE before all the requested
surfaces have been analyzed, program will NOW provide a summary of
the TEN most critical surfaces. This summary was not provided in
earlier versions as the ESCAPE was construed as a request for
termination of the analysis.
Version 5.100 ---January, 1, 1995
- Up to TWENTY search boxes can now be used to generate block-shaped
surfaces. The previous limit was 10 boxes.
- Improvements made to the iteration function used for the FOS
calculations. This now gives a better "chance" for convergence,
especially for cases where NLMC soils have been specified.
- Fixed potential problem concerning files being viewed following a
directory search. Previously, the program would (possibly) have
the wrong screen information, terminated or locked-up.
- Other minor code fixes that improve on the readability and
functionality of the complete program.
Version 5.007 --- November 14, 1994
- Inconsistencies noted for NLMC data option. It is possible that
XSTABL may assign NLMC data for a soil unit, which may have
linear strength parameters. Problem does NOT occur if ALL soil
units have NLMC parameters or if the a fresh XSTABL session has
been started. The NLMC option is now correctly initialized before
the start of each analysis.
Version 5.006a --- October 28, 1994
- The convergence routine for the GLE iteration was modified with the
addition of an accelerator function that speeds up the solution.
This is based on the bisection method and is very useful for cases
where the rate of convergence was very slow, or (just) failed to
complete within the allocated 50 iteration steps.
- In a very few, unique cases, the ground surface would not be
plotted above the thrust-line and failure surface for the GLE
option. This problem has been fixed in this version.
Version 5.006 --- October 11, 1994
- The NLMC properties were not correctly assigned if there was a
mixture of conventional and NLMC soil units. Versions prior to
5.006 incorrectly assigned the first NLMC data set to soil unit-1,
the second NLMC data set to soil unit-2, and so on.
- XSTABL now checks for consistency between the defined water
surfaces and the water surface numbers assigned to each soil unit.
An appropriate WARNING message is now printed to the output file
if any defined water surface is not used by any soil unit. The
analysis is then performed by ignoring the unused water surface.
Version 5.005 --- August 10, 1994
- XSTABL did not always allow user to enter a negative value for
the upper/lower angular limits. This has been now fixed.
- The failure surface coordinate data could not be read directly
into the data table for a specified noncircular surface. This
problem was traced to a mismatch of file identifiers and has now
been fixed in this version.
Version 5.004 --- May 25, 1994
- Corrected the anisotropic option. Previous versions did not assign
the correct anisotropic strength values for any soil unit that was
numbered as unit #1. So if the subsoil model had more than one soil
unit and the anisotropic soil was NOT unit #1, the analysis would
use the assigned isotropic values.
- Problem reported with single circular surface generation for
- Clipping problems with display of slope profile for the Thrust
Line Location graph following analysis of a single surface.
- Fixed problem with display of current seismic coefficients.
Program failed to display the kh and kv values if either one
happened to be zero. Analysis was OK if the *.IPT file had
these values written to it.
- Fixed problems with the creation of a *.GP2 file if the analysis
was aborted due to errors with weight calculation or use of the
nonlinear M-C envelope.
- Some minor Network capability added for printing plots to
- The on-screen graphs displayed following the analysis of a single
surface can now be seen on monochrome displays.
Version 5.003 ---May 20, 1994
- Corrected the anisotropic option. Previous versions labeled this
data "ANISO2" in the input datafile (*.IPT) and thus it was
rejected during the analysis phase as the correct name should
have been "ANISO".
Version 5.002 ---May 8, 1994
- Changed the OLD error messages #48, #49 and #50 to WARNINGS.
This is considered to be a more appropriate report as, logically
speaking, errors should be followed by automatic termination of
Version 5.001 --- April 29, 1994
- The *.CP? graphics format has been "dropped" due to an unknown
incompatibility with WordPerfect 6.0b (for DOS).
However, you can save a graphics file using the native WordPerfect
Graphics (WPG) format with THIS version under the *.WP? file
extension offered by version 5.001.
For third-party software that cannot read the WordPerfect Graphics
format, try saving plots to the *.HP? (HPGL) format.
- This version will operate successfully on all computers that have
an EGA/VGA graphics adapter. If you want to use XSTABL with a
CGA/EGA/VGA combination, please contact Interactive Software
Designs, Inc., for a SPECIAL version.
This change has been implemented as the support for the Color
Graphics Adapter (CGA) tends to slow the graphic screen displays.
Version 4.111 --- October 21, 1993
- User is now prompted for decision on option
to handle negative normal stresses for cases
where slope has nonlinear and linear M-C strength
specified. Otherwise program would exit with
Version 4.110 --- August 27, 1993
- If soil at base of slice was affected by
piezometric surface, program would neglect
any influence of submerged conditions above
Version 4.109 --- August 1993
- Minor cosmetic fix to plotting routine with availability of SciPlot 5.0
Version 4.108 --- May 13, 1993
- Fixed problem with FOS calculation that suggested lack of convergence "too-early"
Version 4.107 --- May 13, 1993
- Improved chances of convergence for cases where the FOS may be less than 1.0
Version 4.106 --- April 16, 1993
- Program did not include surcharge loads for right-facing slopes
Version 4.106 --- April 23, 1993
- This version release fixes a problem where a right facing slope
is being analyzed with a surcharge load. For such cases, prior
XSTABL versions did not correctly place the surcharge within the
assumed geometry. The user would see the surcharge located
correctly on the screen and hardcopy plots, but the location of
the surcharged zone was incorrectly established for the analysis.
Typically, the analysis would have neglected the surcharge loads
or applied them to wrong slices.
- If such an analysis was performed, the reported factor of safety
values would be incorrect. The magnitude of this error would be
a function of the size, direction and position of the applied
surcharge. Typically, the reported FOS will have been (slightly)
Version 4.103 --- October 5, 1992
- version 4.103 of the XSTABL program,
- a prototype version (2.01) of Spencer analysis program XSP, and
all of the example input files from the XSTABL technical reference manual.
Version 4.100 --- August 1, 1992
- Changed consequences of ERROR 48. Now the user has three
options available to control the handling of negative
effective stresses. See "Slope Analysis with XSTABL" for
- The current unit-system (English or Metric) is indicated in
the upper right hand corner of the screen menus.
- User can switch between English and metric unit systems
without exiting program. The F4 key, as shown in lower
right corner of all menu screens, is the selected toggle
- All data files generated by XSTABL are stamped with the
unit-system that was being used at the time the file was
saved as a *.IPT file. Upon reading this file again, XSTABL
will automatically switch to the appropriate unit-system.
- The saved *.IPT files have the filename, and the time,
written on the first line. This allows printed data files
to be quickly associated with the correct output files.
- The option to vary the number of plots shown on-screen and
written to the output files is now accessible from within
the program. Previously the user had to exit the program,
modify file: SETUP1.STB, and then restart the program. This
option is available via the F5 key from the screen where the
user elects to proceed with the analysis.
- Inconsistencies between the number of decimal places
displayed on-screen in the data tables and the data written
to the *.IPT file have been eliminated. Now, the on-screen,
*.IPT and *.OPT files comply with the same format.
- Negative effective stresses calculated as a consequence of
excessive pore water pressures are now handled internally.
If the specified pore water pressure is greater than the
calculated total normal stress, XSTABL reduces the magnitude
of the pore water pressure such that the shear strength
(c' + [sigma]' tan[phi]) is always greater than or equal to zero.
A message is still printed to the output file, but only once!
Previous versions used the negative effective stress, but
reported a misleading factor of safety along with an error
- If the user elects to use option 3 for controlling negative
effective stresses, an abbreviated message is now printed to
the *.OPT file in an attempt to reduce the size of the file.
Previously, about 15 lines were printed for each anomalous
- At the request of many users, the beep that accompanied the
calculation of a "misleading" factor of safety has been
eliminated, even though such errors are less likely with
- If the user specifies a vertical surface boundary segment,
ERROR 51 will now be reported to the user and the execution
terminated. The current version would "crash" with a
reported floating point error.
- The algorithms, for the FOS calculation, have been modified
considerably to implement the various options now available.
The theory along with the select limit equilibrium equations
is now published in a supporting technical manual.
- Units for all data are now reported in the *.OPT output
file. This should prevent any further confusion regarding
the slice-data printed following the analysis of a single
- The program will only acknowledge the existence of a maximum
of 200 files in any subdirectory. This limitation was not
always known and thus an added emphasis is indicated if more
than 200 XSTABL files are ever found to reside in a single
- The plot for the reinforced soil analysis now indicates the
projected FOS value along with the required force.
- Plots that currently show a FOS = 500.000 (i.e. a misleading
FOS) will now show "*******" which is probably a more
realistic label for such cases.
- This version assumes that the connected printer is switched
on and ready to print a file. This eliminates the request
for a user-response following the selection of the print
- A separate demonstration program for calculating the FOS
according to Spencer's Method is now available. This program
uses an XSTABL output file (*.OPT) generated following the
analysis of a single surface.
It is expected that a new version of XSTABL with an integrated
Spencer's method of analysis will be available in 1993.
Version 3.22 --- July 30, 1990
Version 1.03 --- April 1989