DCL evolution and proposition

[Grendeldef]

Yup, that belongs more to the wishlist for developers than DCL options but as a conclusion the possibilities of easily fixing the issues are limited or non-existent so I'm still in favour of 'roll something less than average on a map that's not always std 8/16 Pangaea' and we're good to go.
Participation is still voluntary and there're no prizes that I'm aware of so whether a map was interesting and/or approachable can be seen on feedback & results and the host can only wish for active gaming. It's not the hosts' fault if the game turns out as crap from T100+.

In short, I'm not too worried that any single map will turn people away from the series.

 

[IronfighterXXX]

Finally I have a proposition. For me right now the DCL are a bit too quick, but if people are still interested into having 3 a month I propose to structure it that way:
First map of the month will be on the easy side. Either because is naturally very good or because the map is modified to help the player. For example a map with salts and mountain or you could start with Optics on a water map.
Second map of the month will always be unmodified and I'll try to give an average map for medium difficulty.
Third map of the month will be on the hard side either due to a terrible map or the AI receiving more bonuses (Deity+).

I very like that idea, especially that I posted almost exectly the same some time ago

But Deity+ just makes the already formulaic difficulty of Deity even more oppressive without allowing for imaginative new strategies, I'd argue. It will force the players to turtle even more than they do now. Bear in mind that more than 50% of wins on the DCL have been SVs.

I think that for experimenting with new strategies are two others maps and who know maybe very hard difficulty will make new strategies.
I really don't see any problem with ,,more than 50% of wins on the DCL have been SVs" - DomV isn't in any way better than peacefully one.
You like Domv and don't like very hard map, some people like SV and very hard map - really no problem (and I speak with position of player who like almost the same DomV or peacefully one)

 

[consentient]

it's unreasonable to expect the hosts(s) to play the potential maps to any sort of serious degree in advance.

I don't have any kind of expectations of people. I am sure that people, in all situations, in the end, end up doing what they think is best. But the point I was trying to make is that if the host does play a little way in, they will have a better idea. If they have a better idea, they can be MORE sure about the difficulty level. If not the host, then we could have testers, if anyone would be interested in that. However, that brings in extra subjectivities that might not be conducive to a good map.

You like Domv and don't like very hard map, some people like SV and very hard map - really no problem (and I speak with position of player who like almost the same DomV or peacefully one)

I quite like peaceful play as you can see from the spreadsheet. I just don't want to be forced to take a peaceful VC in order to survive because instead of Deity, I'm playing on a level 2-3x as difficult.

I enjoyed #14 and am replaying #10, both of which are hard, but possible for me. If I have no possibility to win, it's not enjoyable.

 

[Sessy]

It's good to be able to modify maps and tweak difficulty that way, such as in the venice map, but that makes it less of a challenge. I'm wondering if it would be possible to do maps with tweaks just on the advanced settings (strategic, legendary, sea level, VC disabled etc.)

Also would like to see more jungle starts for hard maps.

 

[Grendeldef]

You should also test the current Danemark map if you want to be challenged and I'll be interested to know what you think of it.

I played few turns the DCL #27 and some thoughts.

Spoiler :

I don't think this works as intended in respect of making the challenge harder. What it does though, is make the start binary highly encouraging peaceful hide-in-the-corner play. Either one gets steamrolled without any tools available to resist or enjoys tons of caravan beakers.

I just jumped in the game and before meeting anyone I saw CB & horsemen barbs so I went safety first as much as possible. No DoWing, no worker steals from CSs etc, 4C Liberty start to keep it roughly the same as previous games for testing purposes. I even hard build 1 & bought 2 workers which usually is very much against my religion.

First Pikeman came on T32, Trebuchet T38 and Cusco was in 15 pop by T55 so any sort of early wars are out of the picture. On the other hand if one manages to keep the peace the tech pace is highly accelerated and when spying starts on T74 which is where I ended my stint very quick science victories are possible while domination is practically excluded. Sure, late dom is still very viable but not something one would actively pursue from the start. This is catered for peaceful (early) play.

 

[BlackWizard]

Difficulty can be added many different ways. These are my propositions to make game little harder

1) Make worker stealing harder by giving all AI's one or two horseman or knights.

2) Give AI's some other boosts like 1-3 extra settlers

 

[Acken]

I played few turns the DCL #27 and some thoughts.

Spoiler :

I don't think this works as intended in respect of making the challenge harder. What it does though, is make the start binary highly encouraging peaceful hide-in-the-corner play. Either one gets steamrolled without any tools available to resist or enjoys tons of caravan beakers.

I just jumped in the game and before meeting anyone I saw CB & horsemen barbs so I went safety first as much as possible. No DoWing, no worker steals from CSs etc, 4C Liberty start to keep it roughly the same as previous games for testing purposes. I even hard build 1 & bought 2 workers which usually is very much against my religion.

First Pikeman came on T32, Trebuchet T38 and Cusco was in 15 pop by T55 so any sort of early wars are out of the picture. On the other hand if one manages to keep the peace the tech pace is highly accelerated and when spying starts on T74 which is where I ended my stint very quick science victories are possible while domination is practically excluded. Sure, late dom is still very viable but not something one would actively pursue from the start. This is catered for peaceful (early) play.

Well I noted on the thread that Science shouldn't be too hard, and this is by design. Since Science is a turtle strategy that isn't influenced by the AI performance there is no way to make it hard beside increasing the threat of being attacked and dying. Victory conditions are unbalanced. That's just a fact. Science and Diplo are easier than the other two. I can put someone next to Shaka next time in the middle of a flat desert. But aren't people tired of that trick ?
Those wanting a challenge should attempt domi/culture in #27.

 

[nmp0098]

I enjoy the map tinkering. I wouldn't complain if the schedule slowed down for the summer, but 10 days between games isn't too much.

For the easy maps, I'd probably rather have the changes focused on balancing food and production without going crazy on the bonus resources. Adding a lot of rivers really makes the game easier too.

For the harder maps, I'd rather fight god-awful terrain than AI bonuses. Feel free to flatten hills, delete luxes, and turn the starting location into a bayou, regardless of opponents. Those are by far the hardest games for me. Picking all warmongers for a game would be fun too.

If you were ever up for it, I would love to play an Earth map with extra civs and true start locations for whichever civs make the cut.

 

[consentient]

Difficulty can be added many different ways. These are my propositions to make game little harder

1) Make worker stealing harder by giving all AI's one or two horseman or knights.

2) Give AI's some other boosts like 1-3 extra settlers

I can promise you that the AI will not keep those horses at home. They will use them to scout and do barb quests etc.

Giving the AI extra settlers is something I tried myself and they all ICS like crazy and...once again, human is forced to go peaceful tradition.

I think that the Celts map was about perfect for balanced challenges.

 

[stormtrooper412]

Like I said, someone should try tweaking with the game code to make workers "unstealable" to everyone except barbs, i.e. you can only get a worker if it's been captured by barbs

 

[consentient]

Like I said, someone should try tweaking with the game code to make workers "unstealable" to everyone except barbs, i.e. you can only get a worker if it's been captured by barbs

That would be interesting. No worker steals, no baiting. Gives even more reason to go to war early while making it more difficult. Interesting...

 

[Grendeldef]

Like I said, someone should try tweaking with the game code to make workers "unstealable" to everyone except barbs, i.e. you can only get a worker if it's been captured by barbs

This should be easy enough to do as there probably is a binary option between military & non-military units. If a worker is a military unit it just gets killed much like the AI taking player's workers on player's lands. Problem at least partially solved. In regards to barbs something optional could be implemented.

 

[Grendeldef]

Well I noted on the thread that Science shouldn't be too hard, and this is by design. Since Science is a turtle strategy that isn't influenced by the AI performance there is no way to make it hard beside increasing the threat of being attacked and dying. Victory conditions are unbalanced. That's just a fact. Science and Diplo are easier than the other two. I can put someone next to Shaka next time in the middle of a flat desert. But aren't people tired of that trick ?

Yeah, my bad - I just read the OP with attributes of 'Deity+' & 'Very hard' and was expecting something else. With all VCs enabled this just didn't seem that after few turns if one manages to stay out of wars; the barbs were refreshingly annoying instead of the usual free xp.
Shaka has a tendency of making the games more interesting but he also has the unfortunate side effect of making the early game situational do or die and I assume that he's not the most welcomed of neighbours and using him as the std diffuculty booster would soon become boring.

As for additional difficulty I'm still all for cooking the maps if someone's willing to do that - make starting positions worse by removing surplus food etc tiles, no mountain for the capital, preferrably not a coastal either etc while possibly boosting nearby locations especially in vicinity of other civs. Basically an incentive for expanding towards each other for all parties. It doesn't have to a jungle start to be harder than average nor does the difficulty have to rely on aggressive neighbours.

On hindsight the DCL #27 could've been trickier on Continents but that might just be me & my dislike of Pangaea.

 

[Light Cleric]

I don't want to comment too much since I've only played a couple of the DCL maps and didn't play particularly well on two of them; I went a long stretch of not playing the game so, unsurprisingly, I kinda forgot how to not play like crap

EDIT: I guess I also have a different style than most Deity players; I only worker-steal from CSes because imo worker stealing a major AI makes them "potato", i.e. play like garbage the rest of the game. I don't like to farm XP on city-states or do Pillage/Repair stuff either. I'm not condemning anyone who does that, you can do w/e you want; I'm just saying that's why my perspective is different and I feel like avoiding this makes the game more challenging. This is mostly in response to the earlier suggestion of making all workers unstealable(presumably even from CS) which could be interesting.

That said, the Deity+ bonuses felt utterly ridiculous. I think consentient picked the perfect word when he said "oppressive". It basically left science unchanged and made the most tedious VCs on Deity even more tedious. I only played to see if I could make headway with Berserkers for fun but

Spoiler :

there's like no iron

I think a lot of people remember the "psychotic" Fall Patch G&K Era AI. I am not advocating for that at all, but imo that AI performed better than this one does. It spent its gold, it understood that early aggression was when it was at its best, it kept the player honest so they couldn't just beeline NC every game but it still gave you some options. I think an AI somewhere in-between that and the current AI would make things much more challenging. I mention this because, if I remember correctly, those changes were done by changing XML values and not actually reprogramming the AI so they could be replicated. I just know next to nothing about XML values beyond the basic ones that determine personalities.

Also I realize this is a bit of a tangent but it is related to strategies; am I the only one who would like to see the default map size increased by like 10-20%? I feel like maps are simply too cramped and force you into 3-4 cities even when you don't factor in the AI's free Deity settler.

 

[Grendeldef]

Also I realize this is a bit of a tangent but it is related to strategies; am I the only one who would like to see the default map size increased by like 10-20%? I feel like maps are simply too cramped and force you into 3-4 cities even when you don't factor in the AI's free Deity settler.

Nope, I'm there with you and in the past when I had time/interest to roll maps to play I often used low sea level on all 8/16 maps and used quite a few 10/20 maps removing 1 or 2 AIs and occasionally using high sea level. Nothing too drastic but there's a noticeable difference and moreover it slows down the early pace which I prefer.
Whether it made the game that much harder I'm not sure but achieving winning times close to T200 mark was rarer and in general and more importantly, I enjoyed the games more - the RP gamer in me sticks like tar.

 

[nmp0098]

If I understand what I've read in the modding section correctly, the AI actually can be set to higher numerical difficulty values than Deity. This can be adjusted dynamically too. The idea is to increase the AI's bonuses beyond Deity as the eras progress, or use an Immortal start and ramp up the difficulty to beyond Deity right away.

Some AI are notably better at winning than others, and it seems like there's a lot of room to improve the AI in general by tinkering with their flavoring. If I'm not mistaken, this is all read when the game is initialized and then the final flavor values are embedded in the save. So a map could be cooked up where every AI went Alex on city-states, expanded like Hiawatha, chased science like Sejong, or went military like Shaka. Then the map could be played by people with unmodified xml files.

I like the idea of workers and settlers taking damage instead of being captured. It would take away one of the great equalizers in the game. Getting a settler killed by a barbarian would be devastating, far more so than being able to recapture it.

My own idea for a war rebalance is to make archers, CBs, and XBs ineffective against cities, or maybe just cities with walls. They would damage the unit in the city instead, and that unit would contribute more to the city's defense. Only siege and melee units would be capable of cracking open a city until gats. I think that change would not only rebalance warfare, but help the AI justify the units it actually builds. It's a drastic change though, and I'd be playing a very different game then everyone else if I even felt like investing that much effort in making a mod.

Finally, I'm willing to be a guinea pig for some cooked maps. I have a good shot at losing if the map is super hard and I have no problem sharing details of my beat.

 

[Nigel_Tufnel2]

I agree with many above comments. I enjoy the variety that different sea levels and civ/CS numbers and ratios can provide. The Rome DCL was a great example. The cramped nature of many games is good for TR, and using the settle your cap next to the AI's cap or expo trick - but I enjoy being able to really scout the land and have a variety of chokepoint expansions, etc.

I also think adding a few medals or special quests add a layer of complexity and variety. Such as religious goals/quests, no worker stealing, no coups, settle a NW, eliminate a certain civ (you can have hints in the description that are vague but make sense once you play a while), no Rationalism, etc.

I don't recall Random Personalities being tried in the DCL or ICL... Always knowing what you are likely to get with someone is an advantage, IMO. Of course, they overdid the Piety flavor last Fall, so that is sort of a variant...

These are all doable in the base game. I used to spend a lot of time in the Mods section years ago, and that is an entirely different set of ideas/issues. Very fascinating, and it really opens up the game. Certainly ramping AI attributes has been attempted by now? It seems like such an obvious approach. If you ever feel bored by the base game, take a spin with some mods.

Here were the medals I put into ICL 21, England:

Gold: Clear your continent the fastest.
stormtrooper412 - about t160

Silver: Get 8 Ship of the Line into "The Southern Sea" (you'll know it when you find it!).
no medalists so far

Bronze: Settle a NW on your continent.
Nigel_Tufnel, stormtrooper412, Gustavus Rex

 

[nmp0098]

Smart AI Mod: http://forums.civfanatics.com/showthread.php?t=521955

Buffed AI Mod (uses per era modifier to dynamically adjust difficulty, and includes Smart AI Mod and some other changes): http://forums.civfanatics.com/showthread.php?t=504220

 

[Acken]

That said, the Deity+ bonuses felt utterly ridiculous. I think consentient picked the perfect word when he said "oppressive". It basically left science unchanged and made the most tedious VCs on Deity even more tedious.

What do people mean when they say that ? Because to me it is the exact same thing people playing Immortal say when asked about Deity. So the purpose of making it as tough as a level "9" difficulty sounds achieved when I hear this. Sure there are some stuff I wasn't happy with either (like them getting a CS ally on T0). And science victory IS harder as there is a possibility for an AI to win science before T300. This doesn't happen regularly on Deity. But that is all a SV is, beating a time limit.

I'd also like an AI that doesn't get frontload bonus and instead can tech correctly all game long. That doesn't exist with the base version of the game. Also regarding science victory it would still play the same as that victory ignores the AI performance. The only thing affecting science victory are warmongers. And the only truly aggressive guy is Shaka.

I mean I don't necessarily want to keep doing games like #27. But I also hear people saying the game is too easy and the AI too much of a pushover. There are only two options right now to make things harder. Give you a poor start and/or give the AI more power. If people truly prefer Jungle maps and have Shaka has a mandatory neighbor, like #10, that's what I'll do but I would think that stuff would get old very quickly.... and science would still be the same if you can fight the initial attack.

I think a lot of people remember the "psychotic" Fall Patch G&K Era AI. I am not advocating for that at all, but imo that AI performed better than this one does. It spent its gold, it understood that early aggression was when it was at its best, it kept the player honest so they couldn't just beeline NC every game but it still gave you some options. I think an AI somewhere in-between that and the current AI would make things much more challenging. I mention this because, if I remember correctly, those changes were done by changing XML values and not actually reprogramming the AI so they could be replicated. I just know next to nothing about XML values beyond the basic ones that determine personalities.

I don't really know what exactly changed for the AI to be way more aggressive in BNW. This would require modding though. If there is a way to tweak flavors with commands in the firetuner I'm all ears.

 

[nmp0098]

I don't have Firetuner, but adjusting the AI flavoring just requires making a backup folder of the files and an ASCII editor. The hard part is knowing what to change, and what those changes actually do. I'm certainly not suggesting that anyone knows what to do, but I do have some file comparison tools if someone had a copy of the Fall Patch AI files or could tell me where to find them. I'm fairly certain that all of the xml files are read when the map is rolled, tweaked slightly by a randomizer, then embedded in the save file like the raging barbs option.

Edit: I realized that I might be more familiar than most with this architecture. All of the game mechanics are hard-coded in the .dll's, but all the parameters that those mechanics use are pulled out into separate files and stored in xml format. There are files for more than just leader flavoring. These parameters influence build order in cities, military planning, etc. and are very powerful without digging into the source code. They don't effect game mechanics at all, just AI decision making.