AJE modding

This thread is very instructive about the main flaws of AJE:

http://www.ageod-forum.com/showthread.php?25930-Pompey-s-forces-can-t-win&p=253410#post253410

 

These weaknesses are so important I begin to consider necessary to create an open mod on the main AJE scenarios to offer a better gaming experience.

 

1) I will begin by testing a new battle system. Under the hood, the test results done several days ago on the official forum are demonstrating the battle results to be heavily flawed. If the subject isn’t more disputed for now, I guess it’s the result of both the complexity and hidden aspect of the battle engine and the fact many of not all scenarios are competition between very similar forces organized mainly around heavy infantry.

 

The first step will be to improve the tests, by creating testing scenarios based on this excellent model, to determine the right values.

 

2) The AI is exceptionally poor. If the 1.0 version has allowed to believe AI was good, it was an optical illusion due to the flawed battle system, that created so huge bonus for Caesarian and Optimates faction AI was boosted. The 1.01 version has lowered these bonus, and the AI is displaying the same fundamental flaws  then ain former AGE games. The AI ay show here and there operational brilliance, but most often, it’s unable to understand the map. The most obvious example is Pontus, that in 1.0 landed in Sicilia. The only AI change with the 1.01 version has been done to lower Pontus AI interest for Sicilia; that’s why it’s now invading regularly Tunisia. More and more posts and reports are made from PBEM games.

Each time  new exe version is released, each time AI events need to be modified, partially or totally. If not, AI events will weaken AI rather than improving it. AI events in AJE are amateurish to say the last, based on the wrong presumption AI is thinking like an human mind, but at least, they  had some correlation with the 1.0 version. the changes introduced in the 1.01 must be tied with changes in AI events.

 

That is easier to do. I may create the FY or SVF AI for AJE.

3) Last , the main scenarios are heavily unbalanced. To just take the example of Pompey, they are reasons why the Civil War didn’t ended with the death of Pompey, and why Caesar was assassinated in 44. If Pompey had a weaker military tool, the opposition to caesar ascension was much stronger  than determined by the scenario.

 

That’s a lot of work 🙂

 

Ideas welcome ; -)

Progress of the day

I’ve worked mainly on FY today. Changes aren’t spectacular but the AI has been adapted to the new exe version and should be on part with the FY AI in 1.07 version. A few minor glitches have been fixed too, and I’ve cleaned a bit the BAKU events.
The new version should come tomorrow or Friday, as I plan to revise a few more things.

 

I should release a new SVF version, taking the last official version as basis, Sunday.

New version of Fatal Years 1.08

This version fixes the bug discovered today affecting the absorbfaction command, a big one precipitating the release of this new version. AS announced, Nizhny Tagil has now railroad level transport and a very few more small glitches have been fixed.
Beware: the AI is certainly less worthy than in the 1.07 as the new official version has modified the AI behaviour. I need  a few days to assess and adapt AI events to the new configuration.

Railroad at Nizhny Tagil

The next version will support railroad movement between Nizhny Tagil and Ekaterinburg.

The first Russian steam locomotive was constructed at Nizhny Tagil in 1833, and the father-and-son engineers who developed it, Ye.A. and M.Ye. Cherepanov (Черепанов), were in 1956 commemorated by an 8m bronze statue (executed by sculptor A.S. Kondratyev and architect A.V. Sotnikov) which stands in the center of the Theatrical Square in the heart of downtown.

 

I’ve done a little gfx change of course (I’m not gifted)

There are yet many heavy geographical errors in the RUS map. I’m undertaking a few changes when possible, but I’m missing the minimal expertise in gfx to work them best.

For those interested in WW2, this book:

http://www.amazon.com/Tankograd-Formation-Company-Cheliabinsk-1900s-1950s/dp/0230208878

is a must to understand how Soviet Union built so may tanks. Cheliabinsk ( Ekaterinburg) and Nizhny Tagil were both major production sites during the war.

AbsorbFaction command is bugged

The current FY  1.08 version is suffering a AGE engine bug that has surfaced today, concerning the AbsorbFaction command:

 

Description

Will absorb the faction corresponding to SubjectedFacTag into the currently selected faction. A ChangeRegionOwner is done on each region of the Subjected, meaning legitimate ownerships are transferred (including structures). The unconditional transfer of all structures may be added to the command syntax in the future.

PercMatAssets

Percentage of the stockpile of material assets (excluding conscripts) that is given to the beneficiary faction.

——————–

I use this command to simulate the Baku putsch in Azerbaijan,Red possessions in this area being transferred to Whites. I’ve set the PercMatAssets to 0, ie no transfer of material assets. Conscripts are indeed excluded, as all the RED conscript pool is transferred to Southern Whites

Here the the script log. I’ve placed 2 test events in bold before and after the baku putsch event (in italic):

Line 153937: SelectFaction, selected: Communists
Line 153938: Started processing event: evt_nam_Renforts_RED_start_eval
Line 153938: Event already referenced, current occurences 1 Max allowed: 6 <>
Line 153941: Min date evaluated: 1918/06/01 converted to turn 4043 current turn 4046 True
Line 153942: Max date evaluated: 1919/12/30 converted to turn 4081 current turn 4046 True
Line 153944: EvalConscript: Communists Conscript: 555 against 550 – Operator: >= Result: True
Line 153946: Entering triggered actions for event evt_nam_Renforts_RED_start_eval
Line 153950: Finished processing event: evt_nam_Renforts_RED_start_eval
Line 153950: ——————————
Line 153954: SelectFaction, selected: Southern Whites
Line 153955: Started processing event: evt_nam_Baku_putsch3
Line 153955: Event already referenced, current occurences 0 Max allowed: 1 <>
Line 153959: Min date evaluated: 1918/07/01 converted to turn 4045 current turn 4046 True
Line 153960: Max date evaluated: 1918/07/12 converted to turn 4046 current turn 4046 True
Line 153961: EvalEvent evaluated: Checking evt_nam_Baku_captured 0 equal to 0 True
Line 153962: EvalEvent evaluated: Checking evt_nam_RED_Tzaritsyn_or_Baku_Tracker 2 equal to 0 False
Line 153971: SelectFaction, selected: Southern Whites
Line 153976: SelectFaction, selected: Southern Whites
Line 153981: SelectFaction, selected: Turkey
Line 153985: Finished processing event: evt_nam_Baku_putsch3
Line 153985: ——————————
Line 153988: SelectFaction, selected: Southern Whites
Line 153989: Started processing event: evt_nam_Baku_putsch23
Line 153989: Event already referenced, current occurences 0 Max allowed: 1 <>
Line 153993: Min date evaluated: 1918/07/15 converted to turn 4046 current turn 4046 True
Line 153994: Max date evaluated: 1918/10/12 converted to turn 4052 current turn 4046 True
Line 153995: EvalEvent evaluated: Checking evt_nam_Baku_captured 0 equal to 0 True
Line 153996: EvalEvent evaluated: Checking evt_nam_RED_Tzaritsyn_or_Baku_Tracker 2 equal to 0 False
Line 154007: SelectFaction, selected: Southern Whites
Line 154013: SelectFaction, selected: Southern Whites
Line 154018: SelectFaction, selected: Turkey
Line 154022: Finished processing event: evt_nam_Baku_putsch23
Line 154022: ——————————
Line 154024: SelectFaction, selected: Southern Whites
Line 154025: Started processing event: evt_nam_Baku_putsch
Line 154025: Event already referenced, current occurences 0 Max allowed: 1 <>
Line 154029: Min date evaluated: 1918/07/01 converted to turn 4045 current turn 4046 True
Line 154030: Max date evaluated: 1918/07/12 converted to turn 4046 current turn 4046 True
Line 154031: EvalEvent evaluated: Checking evt_nam_Baku_captured 0 equal to 0 True
Line 154032: EvalEvent evaluated: Checking evt_nam_RED_Tzaritsyn_or_Baku_Tracker evt_nam_RED_Tzaritsyn_or_Baku_Tracker 2 at least equal to 1 True
Line 154033: Probability evaluated: Probability 30 rolled 68 False
Line 154042: SelectFaction, selected: Southern Whites
Line 154048: SelectFaction, selected: Southern Whites
Line 154053: SelectFaction, selected: Turkey
Line 154056: Finished processing event: evt_nam_Baku_putsch
Line 154056: ——————————
Line 154059: SelectFaction, selected: Southern Whites
Line 154060: Started processing event: evt_nam_Baku_putsch2
Line 154060: Event already referenced, current occurences 0 Max allowed: 1 <>
Line 154063: Min date evaluated: 1918/07/15 converted to turn 4046 current turn 4046 True
Line 154064: Max date evaluated: 1918/10/12 converted to turn 4052 current turn 4046 True
Line 154065: EvalEvent evaluated: Checking evt_nam_Baku_captured 0 equal to 0 True
Line 154066: EvalEvent evaluated: Checking evt_nam_RED_Tzaritsyn_or_Baku_Tracker evt_nam_RED_Tzaritsyn_or_Baku_Tracker 2 at least equal to 1 True
Line 154067: EvalEvent evaluated: Checking evt_nam_Baku_putsch 0 equal to 0 True
Line 154068: Probability evaluated: Probability 75 rolled 42 True
Line 154070: Entering triggered actions for event evt_nam_Baku_putsch2
Line 154072: Select Unique Unit, found: Communists Baku Bolshevik Corps Petrov Sailors in Baku
Line 154073: RemoveUnqUnit, operation done on Petrov Sailors in region Baku Removed entirely the unit.
Line 154076: SelectFaction, selected: Southern Whites
Line 154078: AbsorbFaction Communists has been absorbed by Southern Whites using these values: PercMatAssets, PercImmatAssets, PercLoy, PercMC, PercColPen, PercLand, PercNav, AreaUID 0 0 100 100 0 100 0 51
Line 154081: SelectFaction, selected: Southern Whites
Line 154083: Set Faction Relationships: Southern Whites TUR -100
Line 154086: SelectFaction, selected: Turkey
Line 154088: Set Faction Relationships: Turkey WHI -100
Line 154090: Finished processing event: evt_nam_Baku_putsch2
Line 154090: ——————————
Line 154092: SelectFaction, selected: Communists
Line 154093: Started processing event: evt_nam_Renforts_RED_start_eval2
Line 154093: This event is not yet referenced, base number of allowed occurences 6
Line 154096: Min date evaluated: 1918/06/01 converted to turn 4043 current turn 4046 True
Line 154097: Max date evaluated: 1919/12/30 converted to turn 4081 current turn 4046 True
Line 154099: EvalConscript: Communists Conscript: 0 against 550 – Operator: >= Result: False
Line 154105: Finished processing event: evt_nam_Renforts_RED_start_eval2
Line 154105: ——————————

As you may see, the Red has 555 conscript before and 0 after. The Southern Whites conscript pool jumps from 35 to 625 the same turns as show another events a thousand lines later.

That’s sad.  I’m going to write a new version of the events using the command to avoid the bug…

News of the day

I’ve worked today on FY 1.08, fixing a few messages. I’ve mainly done some tweaking on the gamelogic setting files for retreat and battles. Not yet sure of the results. The Siberian AI will receive too some minor fixing. The new “hard AI” is doing some actions better and other worse, especially in the first turns in Far East.

Finland: there’s a very interesting post on the Another official website:

http://www.ageod-forum.com/showthread.php?25459-Finland-and-Yuildech&p=253074&viewfull=1#post253074

even if I’m disagreeing with the very end, the in game conclusions. There’s a large difference between history lessons and game lessons. History lessons are always marred by the impossibility to test alternative outcomes, the “what-ifs”. Between determinism and other schools of thinking emphasizing more ability to change results, debate is raging since…the beginning and will never cease.

A game, by nature, is a what-if. As soon the first turn is resolved the situation will not be exactly the historical one. Even if the game is a perfect simulation, a game isn’t bound to history. Of course, all the design is tied to the delimitation of the plausible outcome. And this delimitation is a subjective choice, of course. However, there are several other factors that can’t be forgotten.

Let’s take the Finland example:  what if Bolsheviks had been weaker in 1920? Would Finland have left untouched the possibility to grab Karelia? What if Mannerheim side would have won the elections in 1919? Are we bound to consider poll results in 1919 unavoidable? What if France and Great Britain had committed themselves more in Baltic area?

Finland evolution during these years was partially depending on the overall situation. Deciding to disallow Finns to enter war after the 1919 election is to my taste a bit too deterministic, that’s why I prefer to reduce the chances of a full-scale war after.

Then a game isn’t destined to place us in the same place than a historian, ie on a neutral point of view. Players are acting like the side they have chosen to play and design must concentrate on this. In a RCW, I would say it’s much less important for historicity to determine what Finland could have done than portray WHite and Red hopes and fears about a possible Finnish intervention. The key point in RCW is showing both White dilemma about and frightening for Reds. You can’t do that if Finland can’t enter the game after a certain date.

Now, what’s lacking in FY, that has an event for Karelia in 1919 and full-scale Finnish intervention, is the negotiations between Russian factions and minorities on a more systemic manner, something I plan for 1.09 version. 🙂

 

SVF 2.0: I should begin tomorrow the SVF 2.0 adaptation to the last AACW version , published a few days ago with the new official exe. Berto, you may follow your current game, as I think a few days will be mandatory to check myself how the AI performs with the new official exe.

Some FY results by players

A Southern White players has reached victory in early  1924 against Reds, thanks to his skills and Kronstadt event, that was in FY about the same than in offcial RUS. 1.08 version has restricted occurence of the revolt to allow it only if Red NM is very high ( Historical situation of victory) or very low ( all seeming lost for Reds, Anarchists are seizing their chance). These conditions will eveolve.

In the Russian AAr played as Southern Whites too Winter 1919 is marked by the victories of Makhno who is menacing Rostov

 

http://rghost.ru/41220030/image.png

http://rghost.ru/41219971/image.png

 

Small fix for FY 1.08

I’ve uploaded a new version of the GC scenario ( GrandCampaign zip). It removes a feature from the official version, that was never a part of FY and creates large gameplay problems. The change will only be in effect in new game. At least, I’ve now got back my own script file ( I had lost the precedent one, so I’ve been forced to use the offcial one as a blueprint for my own, hence the recent and hopefully past troubles).

Fatal Years 1.08 new version

I’ve uploaded a new version, fixing some problems created by the regeneration of the scenario file, improving AI ( AI agents as described yesterday, a few tweaks to Siberian, etc).
BTW, the retreat routine seems to have suffered in the last RUS version, as some regions chosen aren’t the more probable. Maybe I’m wrong on this one, but I keep this point on scrutiny.

Fatal Years, Windows 8 and Asus EP 121

For those interested, I’ve installed FY on a ASUS EP 121, ie a tabletPC supporting both touch and stylus, equipped with Windows 8.

It works, even with fingers. the pinch gesture for zooming/dezooming  is even recognized, even if it remains a bit cumbersome.