Cradle 1.2

[hexagonian]

Originally posted by Martin Gühmann
The problem with your and Wes' settings as well is that the player is kept from negative clicks and the AI as well, but the human will try everything to stay on zero cklicks for all sliders. But the AI will use them to cure happiness by using positive clicks.

I just made a comparison on We's files and mine. I do not know if he updated this from when I downloaded it, but I have noticed that he does not have any allowance for entertainers to be used, even as an option. So at least from that standpoint, the AI in MedMod HAS to use the sliders to adjust for happiness. I know in Cradle that I reduced the overall percentages for entertainers, but the option is still there.

Originally posted by Martin Gühmann
The problem is that this AI strategy will cause the AI falling behind you in techs and economy. It is only a question of time. Also with my minimum happiness settings the AI use the sliders to fight againts riots.

Are you saying that the first option that the AI uses in a happiness crunch is a global slider adjustment? And that it will max oux sliders to get a major happiness bonus, at the expense of an efficent empire. Remember that extra happiness will cut down on crime, so the loss may nor be so great as thought.

Another thing in MedMod was that Wes did not do, IMO, is give the AI a big enough production/food/commerce/science boost on the higher levels. Maybe bumping up the numbers in DiffDb.txt will compensate fo any slider movement that the AI will do. Because I had reduced the benefits of what each slider will give to a player, in terms of a percentage, even if the AI does max out the sliders, it may not actually lose too much because of the bonuses it gets from DiffDb.txt.

I haven't noted many riots in Cradle, though I will keep an eye on it. Most have been, I suspect, revolts due to slave/garrison issues.

Originally posted by Martin Gühmann
The slic file that adds in riot cities a building that will increase happiness is now working only a small modification that will make it only work for the AIs is missing.

Maybe a happiness cheat for the AI could push the delvelopement of AI civs.
-Martin

That would be a good solution to compensate for the problem.

[Martin Gühmann]

Originally posted by hexagonian
Are you saying that the first option that the AI uses in a happiness crunch is a global slider adjustment? And that it will max oux sliders to get a major happiness bonus, at the expense of an efficent empire. Remember that extra happiness will cut down on crime, so the loss may nor be so great as thought.

That's exactly my point. My minimum happiness level for AI cities is now set to 65 (from 75 in ApolytonPack). The AI uses the sliders to reach this goal on the price of an efficient empire. I max out the sliders in Apolyton Pack every time and the gain is so high that I can catch the AI sometime in the later game. I expect also the lost of scince and production of maxing out the sliders concerning happiness is significant. It could be much better if it max out the sliders.

Riots and slave uprisings are two different things. A city that riots just stop production, growth and so on. A city in that a slave uprising happens becomes indipendent and falls to the Barbarians or founds a new nation.

I hope an happy timer for unhappy cities will solve the problem.

-Martin

[hexagonian]

Originally posted by Martin Gühmann
That's exactly my point. My minimum happiness level for AI cities is now set to 65 (from 75 in ApolytonPack). The AI uses the sliders to reach this goal on the price of an efficient empire. I max out the sliders in Apolyton Pack every time and the gain is so high that I can catch the AI sometime in the later game. I expect also the lost of scince and production of maxing out the sliders concerning happiness is significant. It could be much better if it max out the sliders.
-Martin

I'm still a little confused - this may be stating the obvious, but if you set the minimun to 65, are you now saying that the AI uses the sliders in an efficient manner. When I set up the minimum happiness number in my setup, I made that number 75, because that is the threshold in Cradle at which a city can remain and not fall into a riot state. (the threshold in MedMod and Apolyton Pack is 72) I thought that dropping the number to something lower would cause the AI to ignore happiness, if a city was in a riot, and that the city would remain in a riot state. And I do think that there is less margin for error in MedMod/Cradle than in the Apolyton Pack for maintaining happiness, because issues like pollution cause unhappiness earlier (though Dale may have used the pollution settings from MedMod/Cradle).

Something to take into account in the Apolyton Pack is the fact that (and I do not know this, but it will be worth checking into) Dale also may not have made the slider adjustments in govern.txt., because it is those numbers that adjust the happiness penalties for using the sliders. And I think he used the const.txt numbers from MedMod, which control the amout of gain/loss of each slider - those I know allow for greater gains to be made than in Cradle. By altering those numbers, you can limit how effective the sliders are.

This whole issue got me to thinking about a related issue and that is how accurate are the numbers as reflected in the cheat mode for AI civs. I do monitor them on a regular basis, just to get an idea how many cities the AI is building, and how its science rates are holding up. I believe that those numbers are raw, in that they will not reflect the built-in AI bonuses set up in DiffDb.txt, but I am not sure (and I want to believe that they are raw, based on what I have seen). It would be nice to have a clearer picture of those numbers.

One other thing I noted (unrelated to above). I finally made a concerted effort to get an early Wonder unit in my setup (Hammurabi - in Code of Hammurabi). When I got it, I got a huge spike in my military - to the point that I had the most powerful military in the game. In looking at the numbers in the cheat mode, my military is not the largest in terms of numbers. So does the ranking system take the strength of the units into account?

[kaigun]

Few days away for a bref holiday in Kyushu but back to the game to try and answer some queries...Firstly I altered your map definition for gigantic map after looking at the ApolUL Const file your original sorry didnt work this did
MAP_SIZE_SMALL 26 52 2
MAP_SIZE_MEDIUM 48 96 2
MAP_SIZE_LARGE 64 128 2
MAP_SIZE_GIGANTIC 140 280 2
ust one entry
To create new AIs I just altered the userprofile text it caused no messages for me as far as I can remember.....Im getting alot of new empires even though some are conquered Its not an answer to the stability ? ..Possibly remove morale effects from the AI(not morale) until acertain critical level...historically cities seethed with discontent but civil wars and the creation of new nations were infrequent ..and cities would only rise up @with the hope of support ...say a trigger when 40% of the cities morale fails that they secede and form anew nation ...single city empires are unviable.......especially when defended as they oft are by the hordes of coracles......
Well Im going to try to change the morale andf revoltb settings
any suggestions gratefully accepted....
Just a reminder to any casual reader despite the above Cradle is excellent and in my opinion the best variant on CTP2....it makes it fun and with what Im reading onCiv3 and MOO3 it should have a long life ahead of it..........especially if people keep tinkering with it

[hexagonian]

Originally posted by kaigun
First I altered your map definition for gigantic map after looking at the APOlUL Const file your original sorry didnt work this did
MAP_SIZE_SMALL 26 52 2
MAP_SIZE_MEDIUM 48 96 2
MAP_SIZE_LARGE 64 128 2
MAP_SIZE_GIGANTIC 140 280 2

D'oh, that's why that file was there!!!! Thanks for the fix kaigun. It will be in the next update, but if others cannot wait, make the change in CRAB_const.txt - this is the file for the large map setup.

In my game, the fix I attempted to make regarding coracles acting as the rebellious slaves did not work as hoped. I do not have any other ideas as to how to fix this either. On the good side, the Barbarian civ had built a settler though, so it is acting like civ. And your fix for the addition of new civs is a viable option too.

Some minor upcoming fixes and adjustments...

I am going through the DiffDb.txt file to hopefully present better numbers for the bonuses that the AI will get. What I want to do is have no dropoff on the bonuses, so once you catch an AI civ, it should also stay more competitive throughout the entire course of the game.

Also, after looking at the stats for the WatchTower Tile Improvement, I am increasing its vision from 2 tiles to 3 on flatland and 4 on hilly/mountainous terrain - giving you superior vision against any unit for that time period. Some of my units had a vision of 2, so purchasing the Watch Tower seemed to be a waste of PW. Watch Towers will also have a defensive bonus of 25%

I am also considering removing the POW function for all Ancient Wonder units, since they also have the enslave ability, making them too powerful, IMO. Since the Medieval (and beyond) units do not have the enslave ability, this will be their special ability.

Some minor adjustments to the build lists will be made too.

[Martin Gühmann]

In my APOL and MM2 games I saw every time one AI that has a higher grows rate than the other AIs. The reason is that this particular AI is the owner of the Ramayana that increases the happiness for its empire by 3. In my AI monitor game I have a second AI that is still behind the first AI but it doesn't fall back, the gap between them is constant, because the second AI has also happiness bonus of three, because of martial law instead of the Ramayana. I saw both civs maxing out the sliders. The solution would be to give every AI city a happiness bonus, every turn for one turn.

Another issue that I noticed with the sliders is that only civs at the beginning are using the sliders concerning maxing them out respective production, food and gold. Also less advanced civ are using my max settings. So if the AIs change their strategies they don't max out the sliders fortunatly not respective happiness, either.


The problem in Apolyton Pack is that Dale used the const.txt form MedPack including the riot level at 72 and the strategies.txt from Cradle including with its minimum happiness settings at 75. That was the problem why the AI didn't used the sliders well.

For the entertainers from the strategies.txt settings the AI will use them irrespective if they are necessary or not.

And at least I wrote a script file that gives the AI forts in order to close the gaps in its road network. Here is the Cradle version of that file.

-Martin

[janilxx]

I am still playing Apol but it is about 1920AD now and I am totally the best nation in the game. I have not got tanks etc nowadays units yet but there is no doubt that I couldn't win the game. This means that the rest of the game will be totally boring.

So it is time to get Cradle

But you are still updating Cradle so do you have any kind of estimate when 1.2 will be "ready"? I wouldn't like to be "betatester" anymore but I want to play working version. (Maybe I got too much testing with medpack in spring).

[hexagonian]

Originally posted by janilxx
But you are still updating Cradle so do you have any kind of estimate when 1.2 will be "ready"? I wouldn't like to be "betatester" anymore but I want to play working version. (Maybe I got too much testing with medpack in spring).

Is Cradle playable - and is it able to offer a good stable game...????

Yes I believe so, if you play the game on the hardest levels with the Barbs on the highest settings. I also try to play the game without exploiting the known limitations in the AI - other players may not play that way and thus have an easier time of it. In my mind, the game is 98% complete. Things like the tech tree, units, wonders, city improvements have been set in stone for some time now. There are times that I will make adjustments to those files, but they are very minor and should not affect the overall game. Most of the additional tweaks are more subtle, as they relate to the AI.

Is Cradle finished...???

That's a harder question to answer, because of the ongoing addition of SLIC files, as they are created - a good example of this is Martian's Good Mod, which actually added a whole new level in the Cradle game. I liked the concepts behind it, so I have included it, but it is still open (at least in my mind) for evaluation, whether the benefits need to be toned down. It's the same with the POW coding for the Wonder units. A great concept, and I had set it up for all Wonder units, but it made them somewhat unbalancing, because the Ancient Wonder units also had the enslave ability. I usually did not make an effort to go for the Wonder units, but I managed to pull one out and was able to use it too effectively, IMO. So I fixed it.
I did not benefit from the pool of playtesters that Wes had. I designed the Mod to counter my own tendencies, but if your style is different, you may see issues that I may have missed. And there are certain aspects of CTP2 which cannot be overwritten in any mod, so there are limitations that are common to all. (like the Barbarian slave revolts that are led by Coracles in landlocked cities). And by nature, I am a habitual tweaker, so I know I will continually be trying to improve performance, which means more updates.

I am also one that wants players to take what I have done and modify the areas where they see shortcomings themselves. I've always intended Cradle to be open-ended. If you like playing Ancient/Medieval based Mods, Cradle was set up to be that platform for you (though I had taken steps to extend the game into the Modern Age) One thing that is nice about the Forums is that most of what I have done to the files have come from suggestions from other players. I couldn't have done anything at all, if not for their helpfulness. I've also tried to pass on that helpfulness to others, based on what I have discovered, so that they can make their own adjustments.

The problem is that players will also bring in different expectations into what they expect a Mod to do, when in fact the creator may not think that those issues are important to that Mod. For example, since the AI is locked into a percentage of how many naval units to build, I usually play on more land based maps. Playing with fewer civs makes for a more peaceful game, so someone may find conquest easier in that setup.

And irregardless, there are inherent limitations in all civ-style games. I do not know how many times I have said this, but the caliber of player is a lot better than a few years ago, because of places like Apolyton, where strategy can be discussed, limitations can be posted, and thus exploited. CTP2 did suffer from a below average AI, but it is partially due to the fact that players are very good too, having cut their teeth on other civ-style game. The bar is a lot higher now, and frankly, I'm not sure any gaming companies have caught up yet..

[Kull]

Hex: I went to download the latest Cradle updates, but get a "Server Error" when clicking on the "Cradle 9/17" and "Martin's Good Mod (graphcs/sounds only)" links. Two others worked fine ("Units" and "Update 9/21"). Looking forward to testing all the new changes!

[Kull]

"Never Mind!"

[Martin Gühmann]

Originally posted by hexagonian
That's a harder question to answer, because of the ongoing addition of SLIC files, as they are created

Yes and here is another slic file. It is the commerce improvements for AIs slic in a Cradle version.

-Martin

[hexagonian]

All of Martin's recent SLICs that have been adapted for Cradle have been bundled into a single ZIP file at my site so if you want them, head over there for your one-stop shopping needs.

This last one (AI Commerce Boost) looks especially promising, as any file that helps the AI is a welcome addition!

Haven't had too much time lately to play though...

[player1]

Hey, hexagonian!
Try this slic
PW-cheat for AI v1.01
It works good in combination with Martins Commerce for AI SLIC.

[player1]

Althgouh, Craddle is for Acient-Medivial civs, this code might be handy
in Modern Times:
Infrastructure Fix code v1.01

[player1]

For AI happines problem, look my post in Martin's Commerce for AI thread.