Succession Game IX - Don't throw those bloody spears at me!

[Maquiladora]

Well thats the only thing im worried about, any SAP2 SLIC that doesnt work with the playtest. I thought Tamerlin was playing SAP2 with the playtest?

[HuangShang]

it was so late night
sory fo creating the OTing

so if EPW dont show up, i can post my turn soon, but i never made the updates according to the turns... it shouldnt matter really

heres the teaser, i figured navy are more powerful so im gonna get an ironclad to add to the Ulundi garrison, but we have been plagued and thus Ulundi has rioted for 2 turns (ya, 2 plagues in a row)... talk bout slow tech...

[Tamerlin]

Originally posted by Maquiladora
Well thats the only thing im worried about, any SAP2 SLIC that doesnt work with the playtest. I thought Tamerlin was playing SAP2 with the playtest?

No, I am not playing the SAP2 with the Playtest version because there are things I don't want to find in my SAP2 games (spending resources in trade is worthless in a playtest game for example) and because I don't really agree with some of the current changes in CtP2. I have also tweaked my original SAP2 configuration, I have added the unit Updater2 and some of Martin's files (BetterAI, etc.), I have suppressed a few of the most obvious cheats (spoilers as far as I am concerned). This is really the game I want to play.

Besides that, I don't have the time to play these days, I have not played a PC game since three months... at least.

[Maquiladora]

Originally posted by Tamerlin
No, I am not playing the SAP2 with the Playtest version because there are things I don't want to find in my SAP2 games (spending resources in trade is worthless in a playtest game for example) and because I don't really agree with some of the current changes in CtP2. I have also tweaked my original SAP2 configuration, I have added the unit Updater2 and some of Martin's files (BetterAI, etc.), I have suppressed a few of the most obvious cheats (spoilers as far as I am concerned). This is really the game I want to play.

Besides that, I don't have the time to play these days, I have not played a PC game since three months... at least.

Well I just tried SAP+GoodMod+Updater2 but the playtest doesnt like Goodmod or Updater2 and I really wanted to use those, theyd be really interesting, especially in a succession game.

So I think the 1.11 patch+SAP+Goodmod+Updater2 is what we should use. I vote for that anyway.

[Tamerlin]

Originally posted by Maquiladora
So I think the 1.11 patch+SAP+Goodmod+Updater2 is what we should use. I vote for that anyway.

That is OK with me. I can send the latest Updater2 files (with the extended "upgrade paths") to everyone interested.

[Tamerlin]

Apolyton Pack 2.0 Updater2

Needless to wait, here are the files needed to install the Updater2. Please, read carefully the Readme provided with the Updater2. There might be some differences because of GoodMod but the informations given should allow you to install the Updater without any difficulty.

You should not replace the file named APOL_main.slc with the one provided in the Zip file. I suppose that the line #include "APOL_Updater2.slc" should rather be added at the end of the GM1_APOL_script.slc file.

I guess that you just have to copy and paste the Updater2 related strings in the APOL_Updater2.str at the end of the APOL_main_str.txt in your language folder as the latter file is already included in Goodmod.

Is that correct Maquiladora?

[Tamerlin]

Here are the Updater2 files...

[Marko Polo]

Originally posted by Tamerlin


No, I am not playing the SAP2 with the Playtest version because there are things I don't want to find in my SAP2 games (spending resources in trade is worthless in a playtest game for example)

I was wondering why playtest requires so many caravans for a trade route and asked about it. I was told that I had to divide the number by 10 with this line in APOL_const.thx:

CARAVAN_COEF 0.03
=>
CARAVAN_COEF 0.003

This line is at the end of the file. That fixes the problem.

Now, what other problems there are in the playtest version? I find it worrysome if people start to reject it for various reasons. Shouldn't we discuss about the issues and try to find some kind of consensus what we actually want the new CTP2 look like? Otherwise all the hard work a lot of programmers are doing and have already done is going to be all wasted.

[Maquiladora]

Re: Apolyton Pack 2.0 Updater2

Originally posted by Tamerlin
Is that correct Maquiladora?

Dont ask me, you made it Yeah it sounds right to me though.

Im gonna make a clean playtest install and try SAP+Goodmod+Updater2 again, since im not sure, ive had it installed a long time.

Now, what other problems there are in the playtest version? I find it worrysome if people start to reject it for various reasons. Shouldn't we discuss about the issues and try to find some kind of consensus what we actually want the new CTP2 look like? Otherwise all the hard work a lot of programmers are doing and have already done is going to be all wasted.

Well of course i agree we should report every bug, but this thread isnt the place. I want us to pick the most stable option for our succession game so we dont have to stop or as has happened before, completely start again.

Also I dont really see the compatibility issues with mods as real problems, I mean theyre kind of expected, especially with really complex slics. If there are compatibility issues with SAP we're only skipping the playtest so we can get the game started straight away too, without waiting for a new playtest.

As i said ill try the playtest again since i changed a lot of files myself im not convinced it isnt messed up somewhere in my install

[Tamerlin]

Originally posted by Marko Polo


I was wondering why playtest requires so many caravans for a trade route and asked about it. I was told that I had to divide the number by 10 with this line in APOL_const.thx:

CARAVAN_COEF 0.03
=>
CARAVAN_COEF 0.003

This line is at the end of the file. That fixes the problem.

Now, what other problems there are in the playtest version? I find it worrysome if people start to reject it for various reasons. Shouldn't we discuss about the issues and try to find some kind of consensus what we actually want the new CTP2 look like? Otherwise all the hard work a lot of programmers are doing and have already done is going to be all wasted.

As far as I am concerned I have already listed the things I am complaining about. But as I have not played a CtP2 game since a long time now I must admit that I don't remember exactly what are the things that were bothering me. I should play another playtest game with the latest build. I don't know when I will find the time but I should do it.

What I don't like in the way the coders are working is that there are some major changes that have been encoded (like the way trade is working) where I would have liked the changes to be optional so that a player can choose at the beginning of the game which part he wants and which part he does not want.

I really like the original gameplay and I would have prefered the coders to correct all the bugs and work on a good AI before implementing changes to the gameplay. It is far more important, IMO, to teach the AI how to use sea transports and aerial units, how to manage its cities and how to place the good terrain improvements at the right place.

I know that this is the most difficult part of the project and that the work of the coders is also very important for the future of CtP2 but I would have liked the playtest to be as close as possible to the original game or the SAP2.

[HuangShang]

wait so we think we've already lost?

anyway, i won't be able to finish/post my turn until my monday

[Maquiladora]

I really like the original gameplay and I would have prefered the coders to correct all the bugs and work on a good AI before implementing changes to the gameplay. It is far more important, IMO, to teach the AI how to use sea transports and aerial units, how to manage its cities and how to place the good terrain improvements at the right place.

Come on Tamerlin theyre doing it in their own free time, so they can choose what they want to fix or add. The sea transport is gonna be a big job so its not gonna be a weekend job, but things like the specialist and unit tabs in the national manager are really useful additions that were (relatively) quick to do. In the end everything needs doing at one time or another, so it will be complete in the same amount of time.

If i had my way id have everyone working to fix MP and PBEM, in fact MP is in worse condition than it was in 1.0 but i dont kick up a stink or abandon hope in the project, i just find a way i can help.

wait so we think we've already lost?

anyway, i won't be able to finish/post my turn until my monday

Well its certainly looking like we have lost but dont finish playing, anything could happen.

[Martin Gühmann]

Originally posted by Tamerlin
What I don't like in the way the coders are working is that there are some major changes that have been encoded (like the way trade is working) where I would have liked the changes to be optional so that a player can choose at the beginning of the game which part he wants and which part he does not want.

Oh yeah, let's add a very big screen to the SPnewGameWindow, from were you can choose whether you use, trade route pathing as it should be, or as it was bugged in CTP2 out of the box. If a feature is missing in CTP2 that was present in CTP1, and noone annouced it explicitely that it was stripped, then it is a bug. In CTP1 you may get an advance when you capture a city, but that was missing in CTP2. And of course for each thing we add we should add another entry on that big huge list, use original AI instead of new better AI, just another entry as suggestion.

In fact trade should be now fixed at least the Caravan requirements should be ok now thanks to Flinx. The only thing I could do for mod compartibilty is to move the one tenth factor from const.txt to the source code. But don't ask for removing the improved trade route pathing engine.

And another thing I don't do is updating any mods, that are up to the creators of the mods. And especially in the case of ApolytonPack I think this is superflous. What is the suptitle of ApolytonPack? Yes, CTP2 the way it should be. Well this might be right for the means we had in those days when ApolytonPack was created. But with the source code, we can say ApolytonPack is neither super nor the way CTP2 should be. But it was a trial to come close to the way it should be part. So this means in the end, everything that was intended as fixes and workaround goes into the source code for sure. And what are additions might find there way into the source code project, or are kept as mod. Well and another thing is the capture city advance all I did was to fill a gap in the code, I am not responsible that someone screwed up the const.txt of ApolytonPack, that turns the advance gain on city capture into a 100% chance. The default game, Cradle and CTC don't have this problem, so which mod shall I break Cradle and Call to Conquest or MedPack and World of War? Well I go for the way it was supposed to and that means not breaking Cradle and Call to Conquest.

What changes to the gameplay did we? Well of course we cannot exclude to add bugs of our own. That was true for trade, by the way it was necessary to fix this as the AI suffers from the problem more than the human player. And the other problem is the unplayability on maps with more then one continent.

Originally posted by Tamerlin
I really like the original gameplay and I would have prefered the coders to correct all the bugs and work on a good AI before implementing changes to the gameplay. It is far more important, IMO, to teach the AI how to use sea transports and aerial units, how to manage its cities and how to place the good terrain improvements at the right place.

I know that this is the most difficult part of the project and that the work of the coders is also very important for the future of CtP2 but I would have liked the playtest to be as close as possible to the original game or the SAP2.

As Maq stated, coding is done in our free time, and that the little bugs and little add ons can be done on a weekend. But as you stated the heavy AI stuff is more difficuilt and need more time and it is more boring to test.

Here some piece of code concerning AI terrain improvements:

//----------------------------------------------------------------------------
//
// Name       : CityData::NeedMoreFood
//
// Description: Checks whether a new food terrain improvement should be 
//              built in the area of the according city.
//
// Parameters : -
//
// Globals    : g_player:            List of players
//              g_thePopDB:          The pop database
//              g_theCitySizeDB      The city size database
//              
//
// Returns    : Whether a new food terrain improvement should be built 
//              in the area of the city.
//
// Remark(s)  : -
//
//----------------------------------------------------------------------------
bool CityData::NeedMoreFood(sint32 bonusFood, bool considerFoodOnlyFromTerrain){

	sint32 turnsForOnePop;
	if(GetGrowthRate() != 0){
		turnsForOnePop = k_PEOPLE_PER_POPULATION/GetGrowthRate();
	}
	else{
		turnsForOnePop = k_PEOPLE_PER_POPULATION;
	}

	sint32 maxFoodFromTerrain = GetMaxFoodFromTerrain() + bonusFood;
	sint32 currentFood = (sint32)(GetProducedFood() + bonusFood);

	double foodBonus;
	buildingutil_GetFoodPercent(GetEffectiveBuildings(), foodBonus);
	maxFoodFromTerrain += (sint32)(maxFoodFromTerrain * foodBonus);
	currentFood += (sint32)(bonusFood * foodBonus);

	maxFoodFromTerrain += sint32(ceil(((double)maxFoodFromTerrain * 
		double(double(wonderutil_GetIncreaseFoodAllCities(
			g_player[m_owner]->m_builtWonders)) / 100.0))));

	currentFood += sint32(ceil(((double)bonusFood * 
		double(double(wonderutil_GetIncreaseFoodAllCities(
			g_player[m_owner]->m_builtWonders)) / 100.0))));

	if(!considerFoodOnlyFromTerrain && m_specialistDBIndex[POP_FARMER] >= 0) {
		maxFoodFromTerrain += FarmerCount() *
			g_thePopDB->Get(m_specialistDBIndex[POP_FARMER])->GetFood();
	}

	const CitySizeRecord *nextRec = NULL;
	if(m_workerPartialUtilizationIndex >= 0 
	&& m_workerPartialUtilizationIndex < g_theCitySizeDB->NumRecords()){
		nextRec = g_theCitySizeDB->Get(m_workerPartialUtilizationIndex);
	} 
	else{
		nextRec = g_theCitySizeDB->Get(g_theCitySizeDB->NumRecords()-1);
	}

	sint32 foodForNextRing;
	if(nextRec){
		foodForNextRing = (sint32)((nextRec->GetPopulation() - SlaveCount()) * GetFoodRequiredPerCitizen());
	}
	else{
		foodForNextRing = (sint32)((PopCount() - SlaveCount()) * GetFoodRequiredPerCitizen());
	}

	sint32 foodForNextPop = (sint32)((PopCount() + 1 - SlaveCount()) * GetFoodRequiredPerCitizen());

	const CitySizeRecord *rec = g_theCitySizeDB->Get(m_sizeIndex);
	sint32 maxPop = rec->GetBaseMaxPop() + GetBuildingMaxPopIncrease();

	const StrategyRecord & strategy = Diplomat::GetDiplomat(m_owner).GetCurrentStrategy();

	sint32 popDistance;
	strategy.GetStopBuildingFoodBeforePopMax(popDistance);

	sint32 timePerPop;
	strategy.GetTurnsAcceptedForOnePop(timePerPop);

	return((turnsForOnePop/GetOvercrowdingCoefficient() > timePerPop
	&&      PopCount() < maxPop - popDistance
	&&      !GetIsRioting())
	||     (maxFoodFromTerrain < foodForNextRing)
	||     (currentFood < foodForNextPop));

}

The whole code just determines wheather the AI should consider to build a food tile improvement or another type of improvement. And from the complexity of the code you cannot be sure whether it does what it should do. And another thing is that such a function for the decision whether, a production tile improvement or a commerce tile improvement should be built is still missing.

-Martin

[Tamerlin]

Hey guys, I know that coding is done during your free time and, though there are a few choices I don't agree with, you all have done a great job. I don't blame any coder of anything. I also know that an optional entry for each change is not possible and it would also be ridiculous.

[Tamerlin]

Originally posted by Martin Gühmann
Here some piece of code concerning AI terrain improvements: ...

That's a good beginning, this is the kind of things the AI really needs.