--------------------------------------------------------------------------------
17 Jan 1996    News for NTIA/ITS HF Propagation Models (WINDOWS version)
--------------------------------------------------------------------------------

17 Jan 1996    1. This is the first release of the Windows version of the
                  NTIA/ITS HF Propagation models.
                  It has been tested and seems to run on:
                      Windows 3.1
                      Windows 3.11
                      Windows-95.
                  Functionally, everything should work except:
                  a.) output to color printers. Black & White works.
                  b.) calculating SSN from vertical incident sounder foF2 data.

               2. For the graphics to look good on the screen, it is suggested
                  that your display support a minimum of 256 colors.  If you
                  use the standard default VGA display driver, some of the
                  colors are performed with dithering and crosshatching.
                  It is also suggested that a minimum display size of
                  800x600 pixels be used.  The standard 640x480 does work,
                  but is not as visually pleasing.  The larger your display
                  size, the better it will look.

               3. Possible printer problems may include not enough memory
                  in your printer.  I have a very basic HP IIp with 512K bytes
                  of memory.  I can only output at 150 dots/inch without a
                  border drawn around the plot.  I have used other printers
                  with more memory and can output at 300 dots/inch resolution.

31 Jan 1996    1. Further compatibility with Windows-95 has been tested.
                  Some minor problems have been solved.

               2. This software should now work for both SMALL and LARGE text
                  for screen sizes (480x640  600x800  768x1024  1024x1280).
                  If you notice any problems, let me know.

               3. Output to color printers seems to work.  If it does not
                  work, it may be due to a problem in the driver.  I have
                  problems with the HP PaintJet driver.  It displays the
                  border around each cell when color shading is plotted.
                  I do not have that problem when using a Tektronix Phaser.
                  It has not been tested on HP 500C series printers.
                  They may not have the driver problem that the PaintJet has.

--------------------------------------------------------------------------------
--------------------------------------------------------------------------------

07 Feb 1996    1. The Point-to-Point programs now produce real graphs for
                  Methods 11 (MUF-FOT-EsMUF) and 28 (MUF-FOT-LUF).
                  Previously, these were "line printer" graphs.  These are
                  accessed by setting METHOD to 11 or 28 and selecting
                  CIRCUIT (not GRAPH).  Remember, GRAPH always produces
                  contour plots of parameters.

               2. The Point-to-Point GRAPH output now has the ability to 
                  allow the user to point to a location with the mouse, and
                  the UT, FREQ, parameter value at that point will be displayed.
                  If the LEFT mouse button is used, the point location will
                  be interpolated to the exact point selected.  If the RIGHT
                  mouse button is used, the nearest grid point (integer UT
                  and FREQ) will be selected.

               3. Also, while re-writing the Point-to-Point graphing, the
                  program PERFORMW.EXE was incorporated into POINTWIN.EXE.
                  Thus the program file in ..\BIN_WIN\PERFORMW.EXE is no
                  longer necessary.  This saves about 700,000 bytes of disk.

13 Feb 1996    1. Some problems were discovered in HFANTWIN when changing
                  data parameter values.  If the program said the range was
                  [.1 to .9], and you changed the parameter to .9, you would
                  get a protection fault, or some such fatal error.
                  If you changed it to .1, it would say not in the range.
                  These problems have been corrected.

               2. Selecting Transmit Antenna in Point-to-Point models has been
                  changed so that you may choose 4 different antennas for
                  frequency ranges.  This makes it the same as the DOS version.

               3. A lot of progress has been made in the HELP/CONTENTS for
                  the Point-to-Point programs.  Take a look.  It is not yet
                  complete.  They will be added as they are created.

20 Feb 1996    1. A problem was discovered in Worldwin when plotting Lat/Lon
                  projections.  That has been fixed.

               2. Point-to-Point HELP should be almost complete.  Also, Area
                  Coverage HELP is 90% completed.

               3. An error was corrected in VOACAPW.EXE (the propagation
                  module for VOACAP) that corrected the last line in the
                  All Modes output (METHOD 25) that reads:
                          SPROB =    xx.x   yy.y   zz.z
                  The output for yy.y was wrong.  It represents a variable
                  named D50S which represents the Median Signal-to-Noise ratio
                  deviation of the BEST mode, with BEST defined as the mode
                  with the highest Service Probability.

--------------------------------------------------------------------------------
--------------------------------------------------------------------------------

05 Mar 1996    1. An error in ICEPAC was corrected.  Setting the SYSTEM
                  variable Multipath Power tolerance to zero (to not consider
                  multipath) actually set it to 10dB.  That has been fixed.
                  The only parameter this affects is MPATH.

               2. A correction was made to the Windows COMBINE module so that
                  the BIN operation would work correctly.  Look at the HELP
                  information under COMBINE for a rather lengthy example.
                  It demonstrates an analysis that can be performed in a matter
                  of hours now, that organizations previously did in months.

               3. The WORLDWIN plotting program was modified so that it will
                  Color BINned maps (an output from COMBINE).

               4. All on-line HELP should be "reasonably" complete.
                  Enhancements (and corrections) will always be made.

14 Mar 1996    1. In HFANT, the ability to set the DESIGN FREQUENCY for
                  Type 12 (NTIA Report 87-215 Curtain Arrays) was lost.
                  That has been repaired.

               2. You can now specify Longitude coordinates >180 degrees.
                  Thus, 340 = -20 = 20 West.
                       -340 =  20 = 20 East.

               3. REC533 (ITU Recommendation 533) is now available for Windows.
                  It is obtained from:
                     elbert.its.bldrdoc.gov/hf_prop/Windows/PKZIPed/recwind.exe
                  It does not require USERNAME and PASSWORD to execute.

               4. On certain size displays (1024x768 with LARGE text) all of
                  the data in a TRANSMIT.??? file was not visible.  That has
                  been fixed.  (Also for RECEIVE.??? files).

--------------------------------------------------------------------------------
--------------------------------------------------------------------------------

02 May 1996    1. A new implementation of ICEPAC/VOACAP has been added.
                  The ability to perform Signal-to-Interference calculations
                  in Point-to-Point mode has been added.  Essentially,
                  you specify all the inputs you do for Point-to-Point mode,
                  with the addition of specifying the interferer data
                  (location, Method, antenna, path, FPROB).
                  The HELP information will eventually contain a detailed
                  description of the S/I algorithm used.

               2. The COMBINE program has been modified to use the same
                  S/I combine algorithm as the new Point-to-Point S/I
                  algorithm.  This required the addition of some new input
                  parameters to allow S/I Probability to be calculated.
                  This also allows the output parameters SIPGN (S/I power gain),
                  SIREL (S/I reliability), and SIRxx (S/I at required 
                  probability) to be plotted.

               3. Repairs were made to HFMUFES antenna calculations.
                  There was an incompatible COMMON block in the source code
                  so that results for VOACAP calculations may have been in
                  error.

               4. Several documents have been added that may be of interest
                  to the users.  They can be found under the HELP/GENERAL
                  option.  Then select the line:
                       Helpful documents on using the propagation models.

15 May 1996    1. This represents the first official release of the Windows
                  version of the software.  All previous versions may be
                  considered to be Beta versions.  If you have installed a
                  previous version, you MUST re-install the DBOS device
                  driver.  Unfortunately, if you do not re-install it, you
                  probably will not see this message!

               2. The current version has solved some long standing problems
                  of printing color maps to print devices.  You must make sure
                  you have obtained the most recent device driver for your
                  printer.  Look under the WORLDWIN/HELP/CONTENTS for more
                  information on Printer Problems and Solutions.

               3. As funding for HF propagation studies has ended at NTIA/ITS,
                  technical assistance can only be given to those users who have
                  registered their software.  Anyone interested in contributing
                  to further enhancements, are encouraged to contact me.

               4. The software has been tested on a variety of display sizes:
                     640x 480  small text
                     800x 600  small text
                     800x 600  large text
                    1024x 768  small text
                    1024x 768  large text
                    1280x1024  small text
                    1280x1024  large text
                  If you encounter problems of text not being visible, let me
                  know what display size you are using, and where in the program
                  the problem occurs.  The software detects your display size
                  and some of the displays will be optimized for the higher
                  resolution displays.

21 May 1996    1. VOACAP METHOD 24 did not work.  That has been fixed.

               2. Selection of Transmitter/Receiver locations has been
                  slightly changed so that now to accept the city selected,
                  you either select the ACCEPT button, or DPUBLE CLICK on
                  the city name.  You can also now use the PAGE UP, PAGE_DOWN,
                  HOME, END, and arrow keys.  Also, typing in the first letter
                  will position you at the first city that begins with that
                  letter.

28 May 1996    1. A slight problem with Windows-95 systems on specifying
                  contour levels in AREA Coverage was observed.
                  This should now be corrected.

--------------------------------------------------------------------------------
--------------------------------------------------------------------------------

06 Jun 1996    1. In HFANT, some of the HFMUFES antennas would create math
                  errors that would cause the program to about.  That has
                  been fixed.

               2. The ITSA-1 IONCAP antenna types 23 & 24 had continuity
                  problems at high elevation angles.  At 90 degress elevation
                  angle, you would get different gain at different azimuthal
                  angles.  There appears to be a problem in the equations
                  that were developed 20 years ago.  Continuity has been forced
                  with an interpolation procedure.  Use these antennas at your
                  own discretion.

               3. A problem with the Rec 705 Crossed Dipole antenna (Type 9)
                  was also fixed.  The design frequency was extracted wrong
                  from the data file and the calculations were wrong.

               4. Antenna Types 5 & 6 (Horizontal & Vertical Log-periodic)
                  allowed the number of elements to be 1.  That caused math
                  errors.  The number of elements MUST now be 2 to 30.

               5. Because of the above problems, the new version will replace
                  data files in the ..\ANTENNAS\SAMPLES directory.  If you
                  changed any of these files (which you should not change
                  because they are MY samples), they will be changed.

               6. The Sanson-Flamsteed projection to view the full antenna
                  pattern now works in HFANT.

               7. A problem running REC533 Point-to-Point BATCH was fixed.

20 Jun 1996    1. When using the Lat/Long grid type (not Great Circle)
                  in Area Coverage, the RED HELP information was incorrect.
                  It was still referring to the Great Circle inputs.  Also,
                  if you just switched from Great Circle to Lat/Lon, the
                  Xmin/Xmax, Ymin/Ymax values from the Great Circle were
                  still there, even though they were really invalid for the
                  Lat/Lon grid type (e.g. Xmin=-4000 when for Lat/Lon Xmin
                  is the Longitude of the West plot area boundary).  This
                  has been fixed.
               2. The HELP file for AREA has been updated to reflect the
                  above correction.
               3. The WORLD map plotting did not work when plots covered
                  the Longitude 180 line.

--------------------------------------------------------------------------------
--------------------------------------------------------------------------------

03 July 1996   1. Minor changes occurred with this version that affected almost
                  all of the executable programs.  Some were required to solve
                  some problems for Windows-95 systems.  Most were done to
                  incorporate a new option requested by a sponsor and will be
                  unnoticed by other users.

               2. A feature implemented that will be of benefit to all users
                  is in the Area Coverage plot program (WorldWin).  After the
                  plot is displayed, you may now point to a location with the
                  mouse and retrieve information about the location selected.
                  Information includes: 
                       (x,y) relative to plot center,
                       (x,y) in Latitude/Longitude
                       Parameter value at point
                       Distance from point to plot center
                       Azimuth from point to plot center (and reverse)
                  Using the LEFT mouse button, will return the interpolated
                  value at the exact location selected.
                  Using the RIGHT mouse button, will return the value at the
                  nearest calculated grid point, and the mouse cursor will be
                  placed at that location.
 
--------------------------------------------------------------------------------
--------------------------------------------------------------------------------

07 Aug  1996   1. A new feature has been added to the ICEPAC Point-to-Point
                  model.  Under Run, there is now a Distance option.  This will
                  cause the path from the transmitter to the receiver to be
                  broken into 51 segments.  The propagation parameters are
                  calculated at these segments and may be plotted.  The 
                  advantage to this type of plot is that you may observe how
                  the most reliable mode changes with distance along the path.
                  If your type of operation allow the transmitter or receiver
                  to move, this will give you additional useful information.

               2. Some minor changes have been made to solve some more 
                  Windows-95 problems that have been identified.

19 Aug  1996   3. The BATCH calculation in Area Coverage has been fixed.
                  A previous change caused the user to not be able to select
                  any files.

26 Aug  1996   4. Another Windows-95 incompatibility was found and corrected
                  in the HFANT program.  Sorry about that.

27 Aug  1996   5. In AREA COVERAGE, if you plotted MAP ONLY and then used the
                  mouse to obtain coordinates from the plot, a divide by zero
                  would cause the program to abort.  That has been fixed.

               6. In AREA COVERAGE, if the maximum calculated output variable
                  was below what was considered unusable, a divide by zero
                  could occur.  For example, if the maximum calculated DBU for
                  a map was -10, the automatic scaling will not plot DBU below
                  zero (this is set in ..\DATABASE\???AREA.GPH).  This would
                  cause a problem during plotting.  This has been corrected.

28 Aug  1996   7. A new feature has been added to the ICEPAC Point-to-Point
                  model.  Under Run, there is now a Time option.  This will
                  allow any of the output parameters to be plotted versus
                  UT time for any of the frequencies specified.  The
                  advantage to this type of plot is that you may observe how
                  the most reliable mode changes with time of day.

--------------------------------------------------------------------------------
--------------------------------------------------------------------------------

25 Sept 1996   1. A rare division by zero was discovered (and fixed) in ICEPAC.
                  This would occur when a control point was very close to the
                  North or South pole.  This would manifest itself with
                  a COPROCESSOR FAULT.

               2. Version 4.12 of WDBOS.386 is supplied.  The previous version
                  was 4.09, and you should not notice any difference.

--------------------------------------------------------------------------------
--------------------------------------------------------------------------------

04 Dec  1996   1. A detailed HELP page was added to the Point-to-Point
                  HELP/CONTENTS to explain the LUF problems in VOACAP.
                  This is relevant to VOACAP users and does not affect ICEPAC.
                  Any value of LUF from VOACAP should NOT be trusted.

               2. The first beta version of the propagation programs that can
                  be run on Windows-NT systems is now available.  Its version
                  is 96.11W32.  This stands for Nov 1996 Windows-32-bit version.
                  It should be valid for Windows-NT and Windows-95, but for now,
                  I recommend it only be used for Windows-NT.  Everything seems
                  to work EXCEPT ICEPAC Area Coverage calculations.  I hope to
                  have that problem solved by January 1997.
                  It is available from:
             ftp://elbert.its.bldrdoc.gov/hf_prop/Windows/PKZIPed/hfwin32.exe
                  or via my HF home page at:
             http://elbert.its.bldrdoc.gov/hf.html

06 Dec  1996   3. A minor problem was corrected for ICEPAC plots versus Distance
                  and Time.  If you printed the results and selected a scale
                  factor < 1.0, shaded polygons would not have lined up.

10 Dec  1996   4. An error in ICEPAC for Effective SSN was corrected.
                  Only the Win32 version noticed this problem, although it
                  was also in error in the 16-bit version, but did not
                  cause any problems.

--------------------------------------------------------------------------------
--------------------------------- End of 1996 ----------------------------------
--------------------------------------------------------------------------------
