EGAINT 0.94.12 EXECUTABLE DISTRIBUTION PACKAGE

This program comes with ABSOLUTELY NO WARRANTY; see the accompanying GNU
General Public License for full details.  You should have received a
copy along with this program (see the file COPYING).  If not, write to:
Free Software Foundation, Inc., 675 Massachusetts Avenue, Cambridge,
Massachusetts 02139.

See the included file MANIFEST for the contents of this distribution
package.

--
Eric Ng, 1906 Milvia Street, Berkeley, California 94704
...!ucbvax!irss.njit.edu!erc -or- erc@irss.njit.edu


WHAT'S NEW WITH THIS RELEASE

Version 0.94.12 contains an extremely inelegant implementation of
user-definable EGA/VGA palette settings.  After running egaint 0.94.12
for the first time, your configuration file should contain an entry
which looks like "palette=0,7,63,47,49,25,27,10,50,44,37,39,36,38,55,62"
or so.  Using an external editor (if you don't have one, EDLIN or even
DEBUG will work; consult your MS-DOS manual for more details), it is now
possible to have egaint use your favorite colors out of the 64-color EGA
palette instead of the above defaults.  For more information, consult
a manual for one of the following languages:

        Microsoft QuickBASIC or GW-BASIC (BASICA) 3.x, see "PALETTE"
        Microsoft C or QuickC, see "_remappalette()"
        Borland Turbo Pascal, see "SetPalette()"
        Borland Turbo C, see "setpalette()"

In addition, the source code distribution contains the source code for a
quick and dirty EGA/VGA palette editor written in Turbo Pascal.  Also,
if you don't have a color monitor, this won't apply to you.

Version 0.94.11 contains a inelegant method for saving the values of the
user-defined key bindings in the configuration file.


VERSION 0.95 NEWS

Work on version 0.95 has, unfortunately, been suspended due to time
constraints and most likely will not resume until this summer.
Therefore, the target release date has been pushed back from the
original late January to early February to late March to Fall 1990.

Just in case you're interested, below is a partial list of new or
possible features:

        * the long-awaited implementation of show shadow
        * background music and external music editor
        * optional arcade levels and external arcade level editor
        * external blockstyle editor
        * full Tetris compatibility
        * Nintendo Game Boy "Game B" mode
        * less ambiguous shape set (extended shapes) interface
        * user-definable pit dimensions
        * support for both Hercules monochrome and InColor cards (maybe)
        * support for MCGA and VGA 256-color modes (maybe)
        * backgrounds and external backgrounds editor (maybe)
        * and the ability to configure just about everything!

This new version of egaint is being completely rewritten in C (with some
hand-optimized assembly code).  Since portability is an issue (albeit a
minor one), an X rendition of egaint is a definite possibility.

Meanwhile, your suggestions and comments are always welcome.


KEYBOARD ENHANCERS

Egaint uses conventional BIOS calls for keyboard reads, meaning that it
should be compatible with TSR keyboard enhancers.  In fact, egaint
actually plays better with them:  if you have a keyboard enhancer which
adjusts the typematic rate, you will notice that egaint has a more
responsive feel, something which should, in most cases, be more to your
liking.

I personally use this archaic program (circa 1986) by Skip Gilbrech,
kbfix2.  Though it's old, it's still quite versatile:  I have yet to
find a keyboard enhancer which offers more useful features than kbfix2.
Unfortunately, I got it several years and computers ago and have since
lost the documentation, so I can offer no information on it.


DESQVIEW, OS/2, AND WINDOWS

Users of graphical interfaces will most likely encounter difficulties in
coaxing egaint to perform properly; this is due to the fact that egaint
does some "greedy" things with the EGA/VGA display (like using both of
the two available video pages and changing the palette to use better
colors than the original, mucky sixteen which users are given).

I have personally attempted to run egaint under both OS/2 and Windows:
the results, I am pleased to report, are, for lack of a better term,
amusing.

Windows/386 spit out several angry messages, hemming and hawing about
how egaint is an "exclusive" application (among other epithets).  Once
finally into egaint (with all other applications idling), I discovered
that Windows/386 had commandeered the EGA/VGA palette, producing
interesting results.  If you insist on using the Windows/386 interface,
I suggest that you investigate Klotz, a nicely-implemented Windows-based
Tetris from Germany (I liked it, but I don't run Windows much).

OS/2, on the other hand, runs egaint fine (albeit not in a little
window); I seem to recall it being called a "DOS box" or something along
those lines.  A warning:  do not exit back into OS/2 while egaint is
still running, because OS/2 will commandeer one of the video pages
(without restoring it to its previous state), producing an almost
psychedelic effect upon return to egaint (if your machine is fast
enough).

As for DESQview, for those running DESQview 2.2 or higher, it has been
reported to me that the following settings provide optimal results:
"virtualization" off, "writes directly to screen" on, "runs only in
foreground" on, "displays graphics information" on, and "uses its own
colors" on.

Later versions of egaint will probably malfunction in exactly the same
way, as I will stubbornly refuse to support graphical interfaces which I
do not possess.  In fact, X is the only interface with which I possess
any intention of supporting in the future.


CHECKSUMS AND SUCH

For those of you who are virus-weary (like myself), here are some
statistics about the executable file under various compression schemas
(nota bene all were executed with the "plain vanilla" add function and
no other options enabled):

                                       Final          Reported
Compression Schema                     Size      %    CRC
-------------------------------------- --------- ---- --------
PKWARE's                PKZIP 1.10     40,785    44   96a65ce3
Nogate Consulting's     Pak   2.10     41,227    44   7fb3
Haruyasu Yoshizaki's    LHarc 1.13c    41,233    43   7fb3
Fabrice Bellard's       LZEXE 0.91     41,629    43   n/a
Miki/Okumura/Masuyama's LArc  3.33     46,620    36   7fb3
PKWARE's                PKPAK 3.61     53,500    27   7fb3
Dean Cooper's           DWC   A5.01    53,589    27   n/a
SEA's                   ARC   6.02     55,221    24   7fb3
Rahul Dhesi's           Zoo   2.01     55,560    23   7fb3

In addition, here are the values returned by both McAfee Associates'
Validate 0.3 and Rahul Dhesi's Brik 2.0 file integrity programs for the
executable file:

Validate                          Brik -Gb
Check Method 1:  7fb3             CRC-32:  1767482140b
Check Method 2:  1236

