AGEOD Forums

Hi Calvinus,

Playing a 2 player pbem game and have got out of synch by October 1914. We were very carefull so not sure how it has happend but we can see different results and unit positions in both our games

Have looked for differences in the game files and there are a few . For example, see below the references for the same Austrian reinforcement. Does the id make a difference with regard to orders ?

1st game
BeginReinforcement
Id=769
Type=RT_Unit
Applied=No
EntryTurnNumber=4
Placement=0
Nation=2011
Parameter=AUS_n_343
EndReinforcement

2nd game
BeginReinforcement
Id=772
Type=RT_Unit
Applied=No
EntryTurnNumber=4
Placement=0
Nation=2011
Parameter=AUS_n_343
EndReinforcement

Rob

For Reinforcements the Id should make no difference.
Are the differencies in units placement due to reinforcements or to retreats after battle or to movements execution?

Edit: is the difference in reinf. IDs the only one between the two .SAV files?

Hi calvinus, I have the files if you want to have a look...

The differences seem to be where units are in close contact with the enemy so maybe they tried to move to the same location or fought a battle with two different outcomes in both games. Units that are away from the front and are moving about seem ok. I would have to do some more checking to confirm.

And are you sure you haven't replied the save / ord issue we discussed in the other thread?

Yep, we were very carefull. What we can do is synch after the October 1914 diplomacy phase and then conduct our moves/battles and then check again for any discrepancies.

Yes please do. Many thanks, Calvinus.

Found what could be a problem.

Events phase Early August 1914: CP draws 4 cards. Plays one card for technology bonus, sec_030.

Diplomacy phase Early August 1914: The CP save file is correct in that it no longer shows CP holding sec_030. However the Entente save file shows that CP still has sec_030 in his hand. Also the Entente save file does not show the technology bonus turn rule for the CP.

Do event card plays get written to the .ord file ?

Mmmhh... it should. Ok I will check this in debug mode. Did entente PC displayed the event play notification window of sec_030?

Any news about the forces deployment issue?

ok thanks. Can't recall if the message window came up or not. I will see if I can reproduce again.

Just to clarify the situation in case it makes a difference. Both the CP and Entente drew and played sec_030 in the event phase. Only in their own game files was the turn rule recorded. The Entente and CP both thought the other player still had the card in their hand. Hope thats clear

Not looked at battle results yet. Thats next.

RGA wrote:Just to clarify the situation in case it makes a difference. Both the CP and Entente drew and played sec_030 in the event phase. Only in their own game files was the turn rule recorded. The Entente and CP both thought the other player still had the card in their hand. Hope thats clear

I experienced the same bug a couple of times. Indeed it's extremely difficult to fix...

Anyhow, when one player uses the event card, this is notified to the opponent(s) and the game engine processes the event effects. I tested this stuff with no problems, but now you're reporting this issue. So in the end two bugs to fix.

RGA wrote:Just to clarify the situation in case it makes a difference. Both the CP and Entente drew and played sec_030 in the event phase. Only in their own game files was the turn rule recorded. The Entente and CP both thought the other player still had the card in their hand. Hope thats clear

Well, in fact you pointed out a big issue to fix...

When playing in single player mode, AI first draws all events and chooses which ones to keep in hand, then discards other events and reshuffles. Thus human player can draw and proceed. (Invert AI/human) draw order according to play order, if needed, of course.

In PBEM mode this procedure is not possible!!!
Because both players draw events in the same phase!!!
So it may happen that both players draw and keep the same card.
Also, I suspect that there's some constraint coded in the game: that event is not in enemy's hand, so it can't be played by the enemy...
That means: on your PC (allies), SEC_030 is in your hand, you parse the ORD that tells CP play SEC_030, your PC refuses to activate the event because it knows the event is in your hand instead!

Very very difficult to fix!

All event card are unique or some are uniques others not?

I think the better should be avoid the initial pool from what the players can choice being the same so that is not possible to choice the same events. THe game use a random routine for generate the event pool of the player?

Some event cards are duplicated, but every card has a unique ID. The only solution is to split the deck into N sub-decks (N = number of players) before the draw procedure in PBEM mode takes place, as suggested by Mitra...

Side effects: the sub-decks could be not enough for card draws...

Random rountine, yes, but all random factors are always synchronized in PBEM.

I know, true random doesn't exist in informatic, i'm programmer too

The points not expended in a turn for event draws could be cumulated for another event turn, or the AI could take automatically the missing events when it had the info about the ohter players events from the ord file.

Mitra wrote:The points not expended in a turn for event draws could be cumulated for another event turn, or the AI could take automatically the missing events when it had the info about the ohter players events from the ord file.

They're lost.

Bug of Events Draw fixed and tested, as described above, but I confirm that the deck is not enough for all players drawing in the same time...

For example, if both CP and Allies chose warplan option 'Events', CPs draw much more events as compared to Allies because CP move first...

Is not better disable this option from the warplans for PBEM?

I had that thought this morning, Calvinus is there anyway you can disable event cards when playing in pbem mode?

ZardozLord wrote:I had that thought this morning, Calvinus is there anyway you can disable event cards when playing in pbem mode?

Not all the events, but the option "events" from the warplans, so is not possible have a big number of events from where choice.

You could have a house rule to not pick events.

Calvinus,

Instead of releasing official patch 1.08N could you release an updated beta patch so that these fixes could be tested and see if there are other issues ?

Rob

[quote="Mitra"]I know, true random doesn't exist in informatic, i'm programmer too
[quote]

Mee too

Mitra wrote:I know, true random doesn't exist in informatic, i'm programmer too

The points not expended in a turn for event draws could be cumulated for another event turn, or the AI could take automatically the missing events when it had the info about the ohter players events from the ord file.

I'm a programmer as well..... is that what attracts us to this game

what sector? Mine is ERP (SAP R/3).

ERP as well. JD Edwards (now owned by Oracle)

RGA wrote:Calvinus,

Instead of releasing official patch 1.08N could you release an updated beta patch so that these fixes could be tested and see if there are other issues ?

Rob

I fixed all reported bugs, except the loss of synch in forces deployment you reported in this thread. Some news?

Next week I will publish the 1.08N as public beta.

Important news: I finally reproduced several different outcomes in PBEM battles, than could led to the different deployments in PBEM as reported by RGA. I will fix these PBEM bugs as soon as possible, so to make 1.08N finalized.

Found and fixed!!!

1.08N is now really upcoming!

good so we wlll restart our 4 players battle for final tests.