The Spice Thread

[Hived]

Then they probably don't have enough workers. low difficulty level....

hmmm... similar problem, when starting a "deity-game", so it's not (only) difficulty:
Just wanted to look, how AI develops and "waited" 100 turns. Then I opened the WB, to have a closer look.


Summary:

- Two AI spice-firms were founded. One of them in turn 88. (so i suppose, the other players haven't harvested yet)

- The two "founders" were on the last two ranks (excluding my "humble" faction)... they simply had not enough space to expand, what means enough "other tiles" to develop first.

- Everyone had the necessary techs to harvest (they were quite advanced at all: Lansraad was founded in 13191AD)

- Alia, who was spied during the game by my scouts developed nearly all other tiles (built several cottages, mines and stuff), just ignoring the spice lying in front of her "house". That means, she developed also "non-ressource-tiles" before going for spice!

- in turn 100 there were overall five(!) tiles harvested on the whole map. All by the Harkonnen (the only guy with only four cities). He also has developed every other tile in his borders... making a sum of 21 developed tiles for a population of 6

Don't know if it makes a difference, when starting a game on "monarch", but I won't give it a try until the next patch... sorry

That would break the game I think, making all other resources unimportant. And why should spice be giving happiness or health??!? Ordinary citizen/workers can't afford to buy significant quantities of spice. Only the incredibly wealthy can become addicted. Most of the population never gets more than trace amounts, they aren't longer living or easier to keep in line because of it. The spice all gets exported, not used for local consumption.

I don't think it will be a „game-breaker“. Strategic ressource doesn't mean, that you loose the game, when not having it. Other ressources give health or happiness bonuses, too. Have you ever lost a vanilla game, because you had no gold and no rice?! I don't think so.
Maybe we understand fusetechs suggestion in different ways and that's the problem. IMO it should be read: As soon as you have one spice, all cities gain +1health +1 happy (just like gold AND rice in vanilla). More spice doesn't expand that bonus, so you do NOT get +2,+2 having 2 spice and so on... in that case, it would defenitly be much too strong and a „game-breaker“!

Strategic ressource also doesn't mean to me, that it is needed to construct units (although I think, that's the regular term). I think, you can also declare ressources as „strategic“ without creating units, that require it. If not, it would be possible to add a non-essential bonus unit (such as war elephants via ebony in vanilla... maybe with even more prerequisites, such as religion: Jihad+spice=Fanatics or something).

The only reason for declaring it as „strategic“ should be imo, that the AI develops it earlier and maybe settles more „coastal“ (hopefully...).

...and story-relation:
do you remember the Fremen-spice-orgies?! I don't think, that inhabitants ON Dune had no access to spice. Remember Paul saying "It's in everything here"?! (don't know the exact words, as I saw the film in German) ...and even the inhabitants of a Harkonnen city might be "happy", that they won't be slapped to death, when bringing their spice.

All-in-all, spice is still too useless to AI-players IMO... somehow it must be "upgraded" maybe by giving it even more commerce-output when being worked by citizens, but giving "only" +2 for the spice-firm. I can even think of upgrading the c-output without cutting the spice-firm... or by increasing the hammer output. Most other ressources (esp. the non-water-ress) give also a bonus of 6 or 7, when upgraded, but they also provide you happieness or health. Yes, I know, +3c of the firm, but for other ressources there is no "danger" of disappearance.

How do other players think about that?

So far... hope you don't feel bothered... this mod is way tooo good, that it cannot founder on stupid AI-behavior... that's just my opinion.

Greetz, Hived!

 

[davidlallen]

All-in-all, spice is still too useless to AI-players IMO... somehow it must be "upgraded" maybe by giving it even more commerce-output when being worked by citizens, but giving "only" +2 for the spice-firm.

Thanks for the input. We are trying to balance two different things, with one hand tied behind our back. (Sounds painful.) On the one hand, we want a good human player to get a lot of benefit from spice, but not too much. If we increase the yield of spice much more, then the amount of commerce and beakers will be too high and the human player will progress too rapidly. On the other hand, we recognize that the AI may make bad choices about which plots to develop first. So there is a gap between what the AI can do, and what the human player can do.

The problem is that the AI for workers is very hard to understand and modify. We would like to somehow make the AI prioritize spice harvesters more, without making it too easy for the human player. I hope that we can learn more about the AI and modify it, but that is very slow going.

 

[Lord Tirian]

Is it possible to add other player-wide effects depending on the amount of spice you harvest at a given time (possibly scaling with map size)?

Such effects could model that *you*, the player, and the high-ranking staff ingest the spice and gain benefits from that.

For example, the prophetic effects of spice could increase the vision radius of your cities - for each 4 spice resources you control, the vision radius increases by one tile. This can be a huge strategic benefit, allowing you to anticipate attacks way earlier.

Similarly, it could add an effect to the culture rate, perhaps include buildings that produce extra culture per spice (like +1 culture/3 spice), this would go along the same direction and represent the political influence of the spice.

Finally, it might be interest to have an UN-like council, but one that's counting your spice instead of your population - representing your interstellar influence instead your influence on Arrakis. Such a council could have effects on the homeworld screen costs and unit availability and give the whole thing a very "epic" feel that drives home that Dune isn't just about the planet but also about interstellar political intrigue - where Spice is power.

Cheers, LT.

 

[Deliverator]

Hi there, LT.

I've posted some ideas for buildings that offer scalable benefits from the spice here. I'm patiently waiting for the idea to be implemented. Perhaps I should post in SDK requests since david has lots to do.

 

[Hived]

Thanks for the input. We are trying to balance two different things, with one hand tied behind our back. (Sounds painful.) On the one hand, we want a good human player to get a lot of benefit from spice, but not too much. If we increase the yield of spice much more, then the amount of commerce and beakers will be too high and the human player will progress too rapidly.

I agree to that... that's why maybe an increase in hammers might be a better choice... or no increase at all

On the other hand, we recognize that the AI may make bad choices about which plots to develop first. So there is a gap between what the AI can do, and what the human player can do.

Yes, so my first objective as programmer would be, to get the AI to harvest earlier. I think it's okay, if they improve other ress. first... also wind-traps are okay (or better: wished), human player develops them first, too... but seeing, what Alia or the Baron were doing, really hurts... and destroys much of the so well built up dune-feeling (just my opinion).

The problem is that the AI for workers is very hard to understand and modify. We would like to somehow make the AI prioritize spice harvesters more, without making it too easy for the human player. I hope that we can learn more about the AI and modify it, but that is very slow going.

Yes, I understand the problem... I can also imagine, that manipulating the workers "brain" is more difficult then changing the things he develops.
What about the "strategic-ressource" idea? Is there an option to declare ressources as "strategic" (if possible without attaching units to it)? And if, will it change AI behavior?
If that works, it would be a fine method of changing "nothing" to human players but changing AI worker behavior... so dune wars will finally be the battle for spice, everybody expects

Hope I don't bother you, I just try to improve your mod

Greetz, Hived!

P.S.: I really love this Mod and as I know, that there are still things to improve, I just wait, how it develops

 

[Ahriman]

Hived, again thanks very much for the feedback.

in turn 100 there were overall five(!) tiles harvested on the whole map

The AI always builds harvesters in my games, though I don't know how much by turn 100.
Maybe you're just expecting too much too soon.

I think there is no reason for spice to be important in the early game. In a logical sense, in the early game there isn't even any offword trade, so spice is not very valuable.

I don't think it will be a „game-breaker

Here is why it would break the game (if each spice gives a bonus). Health and happiness are very binary in nature; they only have any impact when you have less than some threshold. So your aim is not a maximizing function, but a satisficing function. That is, you need just enough so that the constraints are not binding.

If you can get all the health and happy you need from spice alone, then every other resource in the game basically just became unimportant. Which makes the game incredibly boring, and breaks the balance between resources.

More spice doesn't expand that bonus

If you're just addnig a happy/health bonus to a single spice resource, that is even worse; it is trivially easy for every faction to get a single spice resource. So all you're really doing is making a global increase in health and happiness, and maknig the game easier, and making happiness and healthiness less important as mechanics. And you aren't even adding any marginal incentive to seek out additional spice tiles - which I thought was your whole point.

do you remember the Fremen-spice-orgies

Fremen are different than ordinary populations, and do have significant spice in their food and diet, but still not enough to significantly increase their lives, or to addict them.
And the orgies come from the transformed water of life, not from spice consumption.

All-in-all, spice is still too useless to AI-players IMO

The best way to improve spice is to provide more buildings that have cumulative bonuses depending on the total empire-wide number of spice resources you have, not to slap a bunch of bonuses onto just having a single resource.

it must be "upgraded" maybe by giving it even more commerce-output when being worked by citizens

I have no problem with this. Its now down to ~+4 commerce. I'd like to see it move back up to 6, with spice industry tech (ie not in the early game).
The one thing we do want to avoid is big early game commerce differences across factions based on the very very random distribution of spice in the early game.

Yes, so my first objective as programmer would be, to get the AI to harvest earlier

I don't see this as a priority.

Another point to make; it is not necessarily good for the AI to try to change their AI to make them prioritize spice in the early game. They arguably get much better economy bonuses from building windtraps, wells, dew collectors, cottages and mines than they do from building harvesters in the early game.

 

[Hived]

Hived, again thanks very much for the feedback.

You're welcome. As long as I can help you, I am happy
Also thank YOU for your patience and fast reaction to a „noobs“ idea.

The AI always builds harvesters in my games, though I don't know how much by turn 100.
Maybe you're just expecting too much too soon.

I'am definitely an „early-player“. In vanilla I often cancel a game soon after I know who will probably win it... so „modern-tech-games“ are not typical for me. That might be a part of the problem I have. But I also think we have different demands on this game. When I think of Dune, I think automatically of spiceharvest, harvest-sabotaging Harkonnen, guerilla-Fremen blowing up Harkonnen harvesters... or smuggling spice to the guild. IMO a game called „dune-wars“ should be themed on battle for spice, such as the dune-game-series from Westwood or the boardgame of 1979 or the film with Sting.

I think there is no reason for spice to be important in the early game. In a logical sense, in the early game there isn't even any offword trade, so spice is not very valuable.

For most of the factions (imo all but fremen) the contact to „outer space“ is most vital: They have to sell spice, to get solaris and buy weapons, to survive on dune. IMO „offworld trade“ starts as soon as spice is harvested/a spice firm is founded. The tech „offworld trade“ in the game means IMO the continuous trade between the faction and their homeworld (or another planet, depending on selected ressource). Also I cannot imagine, that they build up a weapon-factory first, before training the soldiers. No, their weapons were brought to planet from outside! (or even the soldiers themselves were brought in).
By the way, that's one of the major challenges you have, with this mod: Civ is originally a game of development: tech, ressources, cities, religions, governments and improvements, all are developed from “nothing”. Dune on the other hand starts with the “end” of ressource-development, cities and technology. Only a few facets are developed here: on the tech-side the golden path e.g., the fremen religion/utopia. You already have implemented them greatly into your mod. Also the “rest” of the techtree (and it is quite a huge one) was built by you very well. So many dune-related things have been added in the right situation, never leaving gameplay-results behind. Maybe I'm simply fixed on the film/board-game too much
A simple scenario could “fix” my problem in that case

Here is why it would break the game (if each spice gives a bonus). Health and happiness are very binary in nature; they only have any impact when you have less than some threshold. So your aim is not a maximizing function, but a satisficing function. That is, you need just enough so that the constraints are not binding.

I know this. That's why I just „vote“ for a regular bonus: if you have spice +1,+1, else not.

If you can get all the health and happy you need from spice alone, then every other resource in the game basically just became unimportant. Which makes the game incredibly boring, and breaks the balance between resources.

I totally agree

If you're just addnig a happy/health bonus to a single spice resource, that is even worse; it is trivially easy for every faction to get a single spice resource. So all you're really doing is making a global increase in health and happiness, and maknig the game easier, and making happiness and healthiness less important as mechanics. And you aren't even adding any marginal incentive to seek out additional spice tiles - which I thought was your whole point.

That was not my aim. It could also come along with reducing the starting health and happy by 1, so that it is „relatived“. Of course this doesn't solve the „one spice is enough“ problem, but maybe it makes AI harvest earlier... at least before they improve ALL other tiles. Also I wouldn't be sure, if AI will be satisfied having one spice, just because of „knowing“ that further doesn't give any more bonus, in vanilla they also build several gold mines, if possible.

Fremen are different than ordinary populations, and do have significant spice in their food and diet, but still not enough to significantly increase their lives, or to addict them.
And the orgies come from the transformed water of life, not from spice consumption.

I think, that's because they don't get the refined material. They even craft baskets out of spice-parts. Maybe it can be compared with drinking thinned beer and hard schnaps
Hmmm... was it really the water of life?! Okay, I have to admit, that I am not that firm about that anymore.

The best way to improve spice is to provide more buildings that have cumulative bonuses depending on the total empire-wide number of spice resources you have, not to slap a bunch of bonuses onto just having a single resource.

That's only interesting in later game ...and won't make them harvest earlier

I have no problem with this. Its now down to ~+4 commerce. I'd like to see it move back up to 6, with spice industry tech (ie not in the early game).
The one thing we do want to avoid is big early game commerce differences across factions based on the very very random distribution of spice in the early game.

Yes, that's right... I don't want to „overpower“ spice either... I just hope, that the opponent not just starts to harvest, when „having nothing else to do“, because spice should be at least more important than an ordinary (maybe even not manned) mine on a tile without ressources.

I don't see this as a priority.

sadly ...so what priorities do you have?!

Another point to make; it is not necessarily good for the AI to try to change their AI to make them prioritize spice in the early game. They arguably get much better economy bonuses from building windtraps, wells, dew collectors, cottages and mines than they do from building harvesters in the early game.

And that's the point! On the one hand, you seem to be afraid of spice becoming a too strong ressource, on the other even YOU write, that the results even from „normal“ tiles are higher, than from spice. As cities begin with low population and don't grow that fast, it is just normal, that spice is not harvested, when most other tiles bring in better results.

So far... maybe I resume my „deity game“ to become either convinced and finally shut up or become even more rebellious, when seeing, that all the others improve most of their other territory first, too.

Greetz, Hived

 

[Slvynn]

Have fun with Your Deity game , Hived.
i think spice still should have some improvemsts (as bunus per spice tile for certain city buildings, and as benefactor in offworld trade (thats a good direction of Yours)
Simply there is no hurry, because it should be really accurately weighted, to preserve balance and not to create a can of worms.
Old horse is better, and you'll see that game is very hard, really, at highter levels more late in the game.
So enjoy game , and another advice - just to win 1 epic game of any level. Till the end. Meeting "Win" conditions.
As i told before, DW have different development curve and difficulty curve from BTS, but its close enough , and harder alot than FFH.
You can often go nuts at start of the game, but that also weakening you, because you lose awareness. Caution. And then you meet really tough challenge, so this game is good when played further.

 

[davidlallen]

By the way, that's one of the major challenges you have, with this mod: Civ is originally a game of development: tech, ressources, cities, religions, governments and improvements, all are developed from “nothing”. Dune on the other hand starts with the “end” of ressource-development, cities and technology. Only a few facets are developed here: on the tech-side the golden path e.g., the fremen religion/utopia.

I agree with you that Civ is a game about development. I agree with you that the Dune books, Dune 2000 game, and movie are at the end of development where almost all the technology exists. However, the Dune Wars mod for Civ is not about the end of development. If it were, all civs would start with shield generators and heavy carryalls. The Dune Wars civ game is an alternate reality where Arrakis has been shut off from the rest of the universe by some apocalypse. We have not really detailed this out but you can see some concepts in this thread. So part of the point is to progress through building (rebuilding) this technology.

 

[Ahriman]

IMO a game called „dune-wars“ should be themed on battle for spice, such as the dune-game-series from Westwood or the boardgame of 1979 or the film with Sting.

I think what you want is not terribly feasible within the Civ engine. This is still Civilization, we still have the same mechanics (cultural borders, improvements, resources, weird espionage) and the same AI to deal with (particularly for using espionage and sabotage).
Civ is much more about infrastructure, land-terrain and open warfare around cities. It doesn't deal well with trynig to focus on resources way out of cultural borders, or sabotage/espionage, or battling for resources.
Civilization is about cities (tautological). The Civ AI is designed to contest cities. Its not really feasible to get away from that.

And that's the point!

If we make spice better than normal tiles (in particular if we made it better than water tiles) then the AI would risk starving its cities, and the whole game would risk devolving into a crapshoot depending on how much spice you managed to get near you.

Spice is developed by spice blows, which are highly random, particularly in the early game.

 

[Hived]

Have fun with Your Deity game , Hived.

Just "kicked" it. Waited further 50 turns and realized, that they really improve very much (but at least not all) "normal" terrain before getting the spice.

You can often go nuts at start of the game, but that also weakening you, because you lose awareness. Caution. And then you meet really tough challenge, so this game is good when played further.

Okay... I'll give it a new try. Knowing now, that it's based on an alternate story, the fact doesn't bother me longer that strong.

I agree with you that Civ is a game about development. I agree with you that the Dune books, Dune 2000 game, and movie are at the end of development where almost all the technology exists. However, the Dune Wars mod for Civ is not about the end of development. If it were, all civs would start with shield generators and heavy carryalls. The Dune Wars civ game is an alternate reality where Arrakis has been shut off from the rest of the universe by some apocalypse. We have not really detailed this out but you can see some concepts in this thread. So part of the point is to progress through building (rebuilding) this technology.

Holy one, you should have shown me this thread before! As the last post there was in July, I haven't "seen" it yet
In that case it is totally agreeable, that they first have to build up new cities and develop from "zero"... also the fact, that spice is less useful (and as result not harvested) early makes sense, when assuming, that a (completly) different story is behind it, where maybe contact to the guild has been lost (at first)!

I think what you want is not terribly feasible within the Civ engine. This is still Civilization, we still have the same mechanics (cultural borders, improvements, resources, weird espionage) and the same AI to deal with (particularly for using espionage and sabotage).
Civ is much more about infrastructure, land-terrain and open warfare around cities. It doesn't deal well with trynig to focus on resources way out of cultural borders, or sabotage/espionage, or battling for resources.
Civilization is about cities (tautological). The Civ AI is designed to contest cities. Its not really feasible to get away from that.

Yes, that's the problem, I've mentioned above: Civ develops from zero, Dune not.

If we make spice better than normal tiles (in particular if we made it better than water tiles) then the AI would risk starving its cities, and the whole game would risk devolving into a crapshoot depending on how much spice you managed to get near you.

Hmmm... I hoped, that Firaxis taught the AI, to look at all the three components (food, hammer, commerce) when founding cities... but I got no idea, if they really did.

Spice is developed by spice blows, which are highly random, particularly in the early game.

Right... but in vanilla you can also start in "tundra woods" on an island or in a vast flood-plain-paradise... thats a matter of luck

Anyway... The link to the "dunewars-story-thread" above made me "trust" again in the mod

Nevertheless I hope, my data of "AI-spice-tile-improvement-behavior" was useful, as it might be good, if they harvest at least before improving "not-yet-needed" tiles...
Probably this change has lower priority than others. Seeing, what has happened from 1.6.1 to 1.6.3, I guess, that there is still much work (e.g. fill the Dune-o-pedia... or find a suitable story and add it to Dune-o-pedia )

Greetz, Hived.

Edith says: Gosh it's late.... edith shall be quite and leave here master in peace!

 

[Stochastic]

First off this is an awesome mod, you've put me into a civ addiction relapse just in time for final exams...

I wanted to point out an exploit with spice that I discovered. I successfully traded 2 gpt for 1 spice, something that does not seem right. Is there some way to weight the value of spice so the AI won't accept such a net negative deal?

 

[davidlallen]

Welcome to civfanatics, and welcome to Dune Wars!

This is a good point, the AI is giving away 4 gold per turn in exchange for 2. We will look into that.

 

[Ahriman]

Worse, its giving away *commerce* to get gold. Commerce gets scaled by bonus buildings; diplomatic gold does not.

 

[Deliverator]

Civ4BonusInfos.xml has a tag called iAITradeModifier which apparently defines how much the AI values the resource, the higher the number, the more the value. I'm guessing that "value" means that they demand more gold for it, not just that they want to get it more.

Our current values seem to be a bit random:

Groundwater = 25
Polar Ice = 25
Stravidium = 15
Crystal = 10
Uranium = 10
Spice = 5
Everything else = 0

In vanilla BTS, Copper, Iron and Aluminium have this value set to 10, Oil is 20, Uranium is 30.

This seems to be all you've got to make civs place a higher value on specific resources, so I suggest we review iAITradeModifier for our bonuses, perhaps ramp up Spice to 25/30. Also, following the lead from vanilla, Nitrates which is a strategic resource should probably be set to 10.

 

[Hived]

Changing this value could also reason in a better harvest-behavior (just as i define better ). Making AI probably improve spice-tiles, before "regular-non-ressource-tiles".

I'd give it a try

 

[Deliverator]

Changing this value could also reason in a better harvest-behavior (just as i define better ). Making AI probably improve spice-tiles, before "regular-non-ressource-tiles".

I'm not sure, but I think this value only controls how much value the AI places on a resource in diplomatic negotiations.

 

[Ahriman]

How about:

Groundwater = 25 (I guess the idea is to prevent trading for this? But the resource does nothing - maybe we should remove this all together?)
Polar Ice = 25
Stravidium = 15
Crystal = 10
Uranium = 10
Spice = 15
Nitrates = 10
Everything else = 5

 

[Slvynn]

I think that Greenhouses should be unlocked by Fresh Water resources. Then it makes sensce abit.

 

[keldath]

hey,

i think the spice shuold be placed at a high level of trade by the ai as you suggested,

the ai shouldn't be keen to trade spice, maybe 30 is a good num.

I agree with you that Civ is a game about development. I agree with you that the Dune books, Dune 2000 game, and movie are at the end of development where almost all the technology exists. However, the Dune Wars mod for Civ is not about the end of development. If it were, all civs would start with shield generators and heavy carryalls. The Dune Wars civ game is an alternate reality where Arrakis has been shut off from the rest of the universe by some apocalypse. We have not really detailed this out but you can see some concepts in this thread. So part of the point is to progress through building (rebuilding) this technology.

uap, we must remember that in order to make it a noon boring mod, we needed to adapt the story line to dune as much as we can, alternative is a good definition.