July 21, 2019
Up to Crazy 8 Extreme beta 0h now. This game puts up a good fight. I don't win at all. The program doesn't cheat but it does make adjustments in play based on the number of cards other players are holding. I was going to refine that and only test the player in the direction of play but so far the computer players have no problem beating the human.
July 20, 2019
The Crazy 8 Extreme program has been updated to beta 0g. All the features I needed to add are in. For play of 3 to 5 players. It saves games and restores them. It is pretty much done. I may polish things up for the release version and remove generation a log file. That is for testing.
July 14, 2019
The Crazy 8 Extreme program has been updated to beta 0c. Programming the progressive two option was a bit difficult. But it seems to be working. It is set to automatically play the players hand if they have to play a 2. I may have to add back in some player control where the player may choose to draw cards instead of playing a 2. Thus maybe preventing a full circle of twos to come back on them..
The algorithm is really pretty good. It does beat me quite often. I wouldn't call it an A.I. But rather a formula of play. It uses a weight system to prioritize cards. Some weights have random elements so it isn't all rote.
July 12, 2019
The Crazy 8 Extreme program has been updated to beta 0b.
July 7, 2019
I decided that I could write a program that played Crazy Eights with the card rules I wrote up. This is in beta form right now. Some other features for saved games need to be added and tests for failure have to be done.
June 18, 2019
And now for something completely different. Rules for a Crazy Eight like game. [PDF]
What is amazing is that the Crazy Eight system has been incorporated in a number of commercial games over the years in a number of different countries. Uno is an American version but there are games from Germany, Israel, Czech Republic and Hungary. Looking up various derivations on Wikipedia rules sometimes enforce mostly silence to calling out cards played in silly voices to conspiring between players.
May 30, 2019
The Panzer War Japanese gun tables slightly revised. Sights went to 24% and some penetration values revised as they now use calculated values from NaaB2.
May 28, 2019
The Panzer War German, British and French gun data tables have been updated as a format problem only displayed one digit of the stun value.
May 9, 2019
I added a new page explaining the problems of the 152mm gun data. Link
May 8, 2019
Trying to produce my own firing tables for the 152mm shells I find that Wiki has a lot of wrong data. I don't know how many times I had to make a redo on my Russian tables and PW tables. Finally had to rely on CIA pdfs of Russian data.
May 7, 2019
I came across two original documents on the Russian 152mm gun/howitzer. The earlier one of 1937 report of the ML-20 updated and released in 1944. This has a firing table of the BR-540 AP shell without striking velocities but with time-of-flights which are impossible as the time for the first 500m is the same as from 1500-2000 range. In effect the shell does not lose velocity in its first 2000 meters. This report has a penetration table based on the Demarre formula of the standard K=2400 type. Solving for velocity using this table gets some incredible numbers as well.
The other document is a 1943 report of the 152mm ML-10 updated and released in 1968. This has a firing table of the BR-540 AP shell with striking velocity as well as realistic time of flight. So this appears to have good numbers. I can use the ballistics of the ML-10 BR-540 for the BR-540 of the ML-20. The velocity difference is only 40 m/s. What I may do is run both of these shells through NAAB and see what I get.
April 19, 2019
NAAB has been updated.
April 13, 2019
I updated the tank gun tables for the German, Russian, USA, and UK. This reflects accuracy calculations of guns based on optics.
April 9, 2019
My run for the March Madness this year chose Gonzanga to win but Virginia to come in second. As they say missed it by that much.