|
Shannon V. OKeets -> RE: Production Planning (4/2/2013 8:36:54 PM)
|
April 1, 2013 Status Report for Matrix Games’ MWIF Forum
Accomplishments of March 2013
This report is a day late because I had a meeting with our architect yesterday morning and a dental appointment in the afternoon. Together they subtracted 5 hours from my normal workday. The dentist was just replacing a couple of small, decides-old, amalgam fillings with resin fillings (black transformed to white).
With the architect we signed the final contract for her to begin work on the renovations to our kitchen and adjoining guest bathroom. Both are pretty much going to be gutted: new ceramic tile for the floors and shower surround, new tub, toilet, sinks, cabinets (23 by my last count), furring out 17 feet of walls by 4 inches, relocating the range, dishwasher, refrigerator, clothes washer/dryer, and the kitchen sink. Moving things around means rerouting the electric and plumbing lines. We are keeping the same appliances because they were purchased new over the past couple of years. There will be new countertops, including a built in table for breakfast/lunch for two. As part of all this, we are widening several doorways to ADA (Americans with Disabilities) specifications. The work is scheduled to start mid-June (ye gods, city permitting takes a long time) and the project will last 8 weeks! Eight weeks without a kitchen, the refrigerator in the living room, microwave meals on paper plates, the only running water in the master bathroom. It will be just delightful I’m sure. And the cost is 50% more that the price we paid in 1978 for a 4 story town house in Philadelphia. Everybody should by 3 or 4 copies of MWIF so they can give away copies to friends and relatives.[:)]
Hardware and Software
The open items for Theme Engine remain unchanged: (1) minimizing the game generates a Mad Except error, and (2) so does trying to display detailed listings of file directories (i.e., the dates and stuff) when opening or saving a file.
Beta Testing
In March I released 6 new versions to the beta testers: 10.02.02 (27 fixes), 10.02.03 (10 fixes), 10.02.04 (15 fixes), 10.02.05 (10 fixes), 10.02.06 (13 fixes) and 10.02.07 (7 fixes). That’s 6 new versions and 82 fixes, which is well below my average of 116 fixes/month. But on the other hand I did beat both production planning and supply into submission while concurrently making substantial progress on NetPlay.
Below is the summary of my Master Task List (MTL) as of April 1st. My task list count is 70, down from 78 at the start of the month. The NetPlay list is very volatile , since as I fix NetPlay bugs the beta testers reach additional sections of the code to test. Presently I am more concerned about NetPlay than the other bugs, which reached a new low of 54. A dozen of those bugs are “not quite fixed” items I mostly corrected last month. I just need to touch up a few lines of code (where the paint hasn’t dried). Another half dozen have been reported in the last couple of days, so I haven’t really looked at them. I expect there to be ~16 that I can knock off at the rate of 8/day (based on my experience over the past couple months).
NetPlay [14] 1589, 1594, 1619, 1785, 1826, 1827, 1859, 1880, 1881, 1882, 1883, 1884, 1885, 1886
Sequence of Play [42]
Supply [5]: 191, 1070, 1036, 1081, 1707
Setup Phases [1]: 1877
DOW [1]: 1866
Air Missions [1]: 1611
Naval Movement [2]: 1813, 1816
Naval Combat [10]: {874, 1531}, 1566, 1599, 1701, 1724, 1815, 1847, 1868, 1869, 1872
Land Movement [1]: 276, 1878
Land Combat Resolution [1]: 1873
Reorganization [2]: 1855, 1856
Production Planning [10]: 1107, {847, 961, 1347}, 326, 1644, 1671, 1825, 1862, 1863, 1864, 1871
Search Seizure [1]: 409
Reform Units [2]: 1851, 1879
Conquest [1]: 1876
Vichy [2]: 1803, 1811
Liberation [1]: 891
Final Reorganization [1]: 1733
Non-sequence of Play [12]
Detailed Map [5]: 1188, 142, 769,140, 1501
Game Save/Restore [5]: 695, 517, 110, 118, 1778
Theme Engine [2]: {1050, 568}, {1513, 1467, 966, 1455, 1573, 1655}
Saved Games
Done, except for 5 bugs.
Map, Units, and Scenarios
This just needs the final naval unit writeups from Warspite, who sent me writeups for the Polish land units last month.
Optional Rules
Nothing new other than fixing a couple of bugs in last month’s new code for Unlimited Breakdown.
Game Engine
I fixed most of the bugs in Production Planning this month, although the beta testers then identified a bunch of other stuff on that form that could be improved. I had cut the bugs in that phase from 26 down to 5, but they drifted back up to 10 while I was working on supply.
I still have some things to do to finish my new supply routines, but my big worry about elapsed time is gone. I reduced the time to calculate supply from scratch, when the map is full of units scattered all over the world late in the war, down to 8 seconds. It had been 45 seconds. For the small scenarios, calculating supply from scratch happens in the blink of an eye. Since the only time supply has to be calculated for all units from scratch is when a game is restored, players shouldn’t notice the additional 8 second delay when restoring a saved game.
When starting a new game, supply is calculated instantaneously, as each unit is placed on the map. Once supply has been calculated once, it’s only a matter of validating whether the previous determination for all the units’ supply is still good. For most units that status won’t change. Weather and major disruptions in overseas supply can affect a lot of units, but even then a full recalculation for all units on the map won’t be necessary. Still left on my task list are:
1. Write a routine to determine if a supply path that was previously valid is still valid, and
2. Decide precisely when during game play supply needs to be recalculated.
Following this report is one of my ‘dailies’ from working on a supply bug (master task list #1081). There are also four screenshots of the resultant Supply Sources and Paths form, with commentary.
Player Interface
This is done except for 5 bugs related to maintaining the Detailed Map display in pristine condition.
Internet - NetPlay
I continue to spent a lot of time on NetPlay. Right now there are 14 bugs on my task list concerning NetPlay, half of which Rob W. sent me over the past week while I was working on Production Planning and Supply. Some of the 14 bugs are preventing the beta testers from doing additional testing.
Getting the scenarios that begin late in the war to start correctly for NetPlay took several days of my time. Both sides can decide about lend leasing units before anyone places units on the map. That makes Lend Lease the first Setup subphase and it is executed by both sides simultaneously. Whichever side finishes making those decisions first then waits for the other side. Setup also has a Scrap Units subphase for each major power, which gets done before units are randomly selected for the Setup Tray. All of that code now seems to work correctly. But there is at least one bug in the last subphase of Setup, when partisan units are placed on the board behind enemy lines. For instance, the Russians, Chinese, and French get to do that in some of the scenarios where the Axis starts the scenario having already occupied large portions of Allied home countries.
I haven’t gotten Antiaircraft Combat to work. But then I haven’t had time to look at that yet. Air-to-air Combat took up another couple of days of effort and while I made a lot of progress there, it has a ton of possible outcomes with different players making decisions. Each of AX, DX, AC, DC, AA, and DA has its own code mass that needs to run “just so” when there are two computers involved.
The technical aspects of playing over the internet seem solid except for the automatic disconnect (i.e., crash) when both players leave the game unattended for 10 minutes. I’m not sure what to do about that. If I can get the time interval up to 30 minutes before it crashes, I think I’ll ignore the problem. 10 minutes is too short for my taste, but if you walk away from the internet for 30 minutes, you should expect to have the connection dropped.
PBEM
Nothing new.
Artificial Intelligence (AI)
Nothing new.
Player’s Manual and Rules as Coded (RAC)
Some issues arose last month about numbering the sections in RAC, since the Player’s Manual has a massive amount of cross references to RAC. These and other minor details mainly concern the Table of Contents and Index.
Tutorials and Training Videos
The Tutorials are done.
The training videos remain roughly 2/3rds done. I need to re-record the 6th and create the last three: 10th, 11th, and 12th. The 6th (main form and drop down menus) needs redoing because I have seriously modified some forms since I recorded that video in December of 2009. The last 3 training videos are for naval movement, naval combat, and production/politics (e.g., declarations of war, neutrality pacts, and aligning minors).
Historical Video, Music, and Sound Effects
I have all the files I need as WAV files. What’s needed is for me to insert calls into the sequence of play to activate these 3 glitz elements. I really wanted to get to inserting the sound effects last month, but was unable to find the time. I’ll see if I can get that done in the first half of April.
Web Site
Nothing new.
Marketing
Nothing new.
----------------------
Below is a short history of me trying to fix bugs in the supply routine.
The 4 screenshots show:
Upper Left - Graziani, a secondary supply source, supplying Rommel and other units in the desert. Included in the list of supplied units are those that Rommel himself supplies. Note Graziani’s supply path tracing overseas to Athens and from there via rail to a primary supply source (Trieste) in Italy.
Upper Right - Rommel acting as a tertiary supply source.
Lower Left - Yeremenko out of supply. Yeremenko is under the Russian La-5FN fighter with a range of 3. Because Yeremenko is a hex experiencing Storm weather (shown under the heading Cost in the Path information), his basic supply path is only 2 hexes. It takes him 3 basic path hexes to reach the rail line NW of Kalinin. From there it would cost zero basic path hexes to reach Moscow because he could use rail lines all the way. But he first has to avoid the enemy ZOC exerted by the 5-4 German INF to its NE, so Yeremenko is out of supply. As are all the other units he could otherwise supply, including Timoshenko, who would be a tertiary supply source if Yeremenko were a viable secondary supply source.
Lower Right - MacArthur acting as a supply source for the Free French (note the title above the list of supply sources). MacArthur has two supply paths. One is overseas back to a primary supply source in the US. The other is the one shown, which traces overland to a city in Australia. That is a Mixed supply path since for the Free French it uses both US and Commonwealth supply sources. The two supply sources might both be needed. While the one to the US could supply the units of all countries that cooperate with the US, it uses an overseas path so units themselves would not be able to use it if they trace overseas to MacArthur (only 1 overseas link is permitted in a supply path). The path to an Australia city is overland, so it doesn’t impose that restriction. But because it traces to a Commonwealth primary supply source, it cannot be used by units belonging to minor countries aligned to the US (e.g., Philippines units).
1081 Supply 9.03.05 Post #4 Rob W. #80 January 18, 2012
Several supply problems, notably tertiary supply. There is a saved game from Aaron for testing tertiary supply as well as one from Rob.
May 2, 2012 - Revise the code for finding overland and overseas supply for tertiary sources [currently tertiary supply sources are not calculated]. Three saved games available: CVPSupply and SupplySCS do not find Rommel as a tertiary supply source; TertiarySupply does not find Guderian and Balbao as Tertiary supply sources.
August 24, 2012 - Added code for finding Tertiary supply by land and sea for major powers.
August 27, 2012 - Tertiary supply for majors partially works. It needs to differentiate between paths that include aligned minor and/or cooperating major powers and those that do not.
August 27, 2012 - There is a bug with units in coastal hexes not being able to find overseas supply.
September 2, 2012 - Fixed a couple of bugs with determining supply for units in coastal hexes.
February 10, 2013 - Thomas has another saved game.
March 28, 2013 - In CVPSupply Rommel is not in supply. That appears to be because Graziani is not listed as a secondary supply source for Germany. The 3 units that can trace to Rommel are not listed as in supply or out of supply. The problems with the last is almost certainly because the units are tracing to Rommel but ConnIndex = LastConnIndex, which was producing an infinite loop.
March 28, 2013 - Fixed a bug so secondary supply sources that trace overseas to a primary are added to the list of supply sources for cooperating major powers. In this case, Graziani in North Africa is now a valid supply source for German units. That puts one of the German units in supply, but Rommel (which is adjacent to Graziani) is shown as out of supply. He should be a tertiary supply source. The code reports that he is found to be a tertiary supply source, but he doesn’t show up on the supply report as such.
March 29, 2013 - Fixed the supply problems for Rommel, but now he appears twice in the list of supply sources (sigh). The changes I just made allow for HQs (and cities) to serve as multiple conduits for supply. For instance, it might lead back to a cooperating major power’s primary supply source and be useful for units belonging to both major powers - but not provide supply to units belonging to aligned minors. Or the HQ might trace to a capital of an aligned minor and then by rail to one of its own primary supply sources. In that case it could supply units belonging to itself and the aligned minor, but not those belonging to cooperating major powers. There is yet a third case, where the HQ uses both cooperating major power supply sources and aligned minor supply sources in its path back to a primary supply source. Then the HQ could only supply its own units. The code now searches for all these possibilities and correctly reports multiple supply paths in the Supply Sources and Paths form.
March 29, 2013 - Rommel’s supply is now perfect. There were duplicate entries since Rommel can supply both German and Italian units. I just had to make sure only one was displayed, depending on whether the player is looking at German or Italian supply. There are still German units OOS in a coastal hex of the Black Sea which should be able to trace supply overseas to von Bock who is in supply and also in a coastal hex of the Black Sea. Mao is listed incorrectly OOS as a secondary supply source; as a unit, Mao is in supply. The Japanese HQs in China are OOS when they should be able to trace a rail path to a port and hence overseas to Japan. Curiously, another Japanese HQ in Burma does just that successfully to be in supply, even though he has to go through 3 sea areas.
March 29, 2013 - Fixed the problem with the Japanese HQs. The program now displays 3 routes for Terauchi, who is in China. One goes overseas to Tokyo, one goes to a city in Korea, and one goes to a city in Manchuria. This is all correct. The overland routes to the minor country cities enable units belonging to those minor countries to use Terauchi for supply even if they have to trace overseas to reach him.
March 30, 2013 - The German units are correctly OOS. CWIF showed them as in supply but that’s wrong. HQs in coastal hexes can receive supply from overseas but they cannot send supply overseas. So von Bock cannot send supply over the Black Sea to the German units in a coastal hex. It took me 2 days to figure this out last September and another couple of hours this month to remember that I had previously deduced that the CWIF code for the rule was incorrect. Arrrgh!
March 30, 2013 - In addition to Mao not being listed as a secondary supply source, there are units from Korea, Formosa, and the Netherlands which are incorrectly shown as OOS. All 3 of the minor country units should be able to trace supply overseas to a primary supply source. In the case of the Japanese controlled minor country units, they can also trace overland to a city in Japan. Something is wrong with some minor country units finding supply from their controlling major powers. There is also a 7-4 German INF in Estonia which should be able to trace to the port Parnu and from there overseas to Kiel.
March 30, 2013 - Mao is listed as a secondary supply source for the USSR. The Nationalist Chinese HQs are listed as secondary supply sources for the Communist China.
March 30, 2013 - Fixed the problem with the minor country units aligned to Japan. There was a missing code fragment needed to assign a major power’s primary supply sources as useable by its aligned minor country’s units.
March 31, 2013 - The corrections I made for Rob’s saved game caused new problems in Aaron’s. I took out those changes and now supply is correctly found for the German 7-4 INF. Corrected the bugs in both saved games concerning HQs that can reach the capitals of aligned minors overland as well as a primary supply source overseas. The pointers for the former were interfering with the pointers for the latter.
March 31, 2013 - The Communist Chinese are also showing the Nationalist Chinese cities as primary supply sources. Solved the problem with the Nationalist Chinese cities showing up as Communist Chinese primary supply source. Also fixed the bug with Mao appearing in the USSR secondary supply sources list.
March 31, 2013 - Stilwell does not show up in the list of United States secondary supply sources. Fixed the problem with the Nationalist Chinese HQs showing up as Communist Chinese secondary supply sources.
March 31, 2013 - Added Stilwell to the list of US HQs but he still doesn’t show up as a valid US secondary supply source when he traces supply to a primary Nationalist Chinese supply source (e.g., a Chinese city). Added Stilwell to the list of valid US supply sources when he can trace a rail path to a primary Nationalist Chinese supply source.
March 31, 2013 - Duplicate MacArthur entries are in the list of valid secondary supply sources for the Free France. Fixed; the problem was not in the calculations, but instead in the display of multiple paths.
April 1, 2013 - Fixed the problem with Mao. He was a special case since all other minor country HQs belong to countries that are aligned to a major power. The normal processing of major power secondary sources checks for them having supply. Mao needed his own personal set of code.
[image]local://upfiles/16701/03326BC7D8874558B9BA7AA149AF954A.jpg[/image]
|
|
|
|
