I agree with Myles.  I like the concept of needing swords in the city where I am raising the company of swordsmen.  To have swords instantly transferred across a continent, when it takes a unit months to walk that distance, really detracts from the game for me.  Re the OP's comments, if I leave my core empire weakly defended and send my army off to chase wild geese, then I deserve to be smacked around by an invader!

Actually, I suggest that we have warehouses of different sizes.  A small one can be cheap to build and can hold maybe 50 bushels of grain or 150 swords or 50 breastplates or some combination there of.  The larger ones have to be built and can hold more.  If you have more than can be held in your warehouse, then you lose some of the stored item thru attrition ot weather, vermine, theft and spoilage.  After all, if you pile the grain outside then the rats eat some, and if you leave your swords there they will rust and people like me will swipe some.

Ok....do you you three (Myles, cleflar and Raven X) like Civ, Cities XL and that type of games which are a bit like economy simulators?

I agree completely. Non-rare resources could be scavaged in small-med amounts from around cities even without a deposit near by. Things like Mithril though you would have to actually find a source. And I really dont want to be able to produce units using Mithril armor and swords if I dont have a resource deposit and haven't negotiated a traded with someone.

I really like civ, but I don't consider it even close to an economy simulator. I haven't played Cities XL, but I do really like the Anno series.

No resource storing! When I think about resource storing, it makes me dizzy. I want to see Elemental a bit more hardcorish, but come on - do we want a fantasy capitalism simulation? Imagine this: you have a mine. This mine can mine (pun intended) a fixed number of units of ore (UO) every day. Mine won't store it's ores anywhere - when a city needs a specific resource, this mine will mine it (I hate the word mine) and send it to the specified city. The longer the destination city is, the longer it will take for caravan to reach it's destination (is should be relatively fast). Of course whole system should be automatized.

To make it a little bit more comprehensible, I will try to walk you through a typical production chain:

1. We want to train an army of Drizzts (let's say 30x of them). Each one of them needs following 'items'/resources:

• Training (City needs to have a special (upgraded) barracks built).
• Magical sword (ie. two scimitars) & armor (provided by upgraded blacksmith who has access to mana crystals and iron).
• Recruits (let's assume we can use a typical Joe here; every city has people, so we have a nice ).

2. We press the button to create the Drizzts. Now the game looks for all needed resources and whether or not we own them.

3. If we don't have - let's say - mana crystals, the game should inform us that we don't have a specific resource, thus we won't be able to train Drizzts.

4. Of course - as you can see in the picture - we have all the needed ingredients. Now comes the hardest part.

5. Before training of Drizzts can begin in Iyawa, all the items (apart from 'training', obviously) have to be on place. One may argue that Drizzts could start training before they got their hands on magical armament, but not to confuse the player too much, we require all pieces to be in one place (the training zone).

6. First, we need people. We require 30 recruits. 30 recruits will be taken from the global number of citizens. Every city will contribute in this recruitment relatively to it's size. That is why Macira will provide more people then Safat or Tikolego. We define that every day max. 10 new recruits can be found (ie. transported to Iyawa) - despite of the placement of Iyawa (is it the center of our empire, or the outskirts). As you see, recruiting soldiers isn't hard at all and it forces the player to have enough of big cities to provide recruits for huge armies.
Indirect cost: 30 citizens, 3 days (final)

7. Second, we need iron ore. Miners can mine 50 units of ore (iron) every day. 10 UO allows you to forge 10 swords, 10 scimitars, 8 shields or 5 chain vests. For all the equipment, we need 90 UO (30 swords = 30 UO, 30 chain vests = 60 UO). So the mine needs 2 days to send them. We assume (for simplicity's sake) that the mine won't send the shipment, if it won't have enough ore packed. Of course the ore has to be transported. The distance is small, thus the shipment will take only one day. I assume that distance from Safat to Macira will take ~1 day (remember that the map is quite zoomed out); if you would like to transport it from Tikolego to Macira, it would take 2 days.
Indirect cost: 90 units of ore, 2 days (mining) + 1 day (shipment) = 3 days

8. Third, mana crystals. This case is similar to iron mine, so I will just get straight to the point.
Indirect cost: 15 units of crystals, 2 days (mining) + 1 day (shipment) = 3 days

9. Blacksmith comes as the forth. In order to start making the magical armament, he need all the ingredients in his shack, thus he has to wait 3 days (max. of all shipments). One blacksmith is able to forge 3 magical armors/3 magical weapons a day. Remember: we are always talking about 1 mana extractor, 1 mine and 1 blacksmith; more will provide us with more stuff, but there will be limit - let's say you can build max. 3 mana extractors, 3 mines, but as many blacksmiths as you wish. That is quite obvious rule - you can't extract some resource too much, but you can make as many factories as you wish (providing you have the resources available). Let's say we have 3 blacksmiths. 3 blacksmiths = 9 magical armors/weapons a day. We need 30 weapons (let's assume our Drizzt's will be a poor man's type, wielding only one blade) & 30 vests, that equals 60 magical items. 60/9 = 6.67 = 7 (of course we have to round up). So it will take 7 days (a friggin' week!) to forge these items. Don't you forget about sending it over to Iyawa. The time, of course, 1 day.
Indirect cost: 3 days (shipment) + 7 days (forging) + 1 day (shipment #2) = 11 days

10. Ufff... we have been waiting 11 days and finally we have all the parts in one place. Now the last part. Each barracks (you can build as many as you need) can take 10 recruits. The training to become a Drizzt clone takes 7 days (a week). Let's say we have 2 barracks built in our beautiful city of Iyawa - it will take 14 days to train 30 drizzts (2 barracks + 1). What's important is that we will get 20 drizzts after 18 days, so if you order to train 100 units (of any type) you won't have to wait 1 month . In other words: it's incremental build.
Time cost: Max(all resources) + training = 11 + 14 = 25 days
Resource cost: 90 units of iron + 15 units of crystals

HOW THE HECK YOU WANT TO MAKE IT SO A TYPICAL JOE WILL UNDERSTAND THIS ROCKET SCIENCE?!111!one

Simple .
Everything I have presented will be fully transparent to the player! Yes, that's what I am saying. He only chooses the type, amount of units he wants to train, and he gets the estimated time. If he wants to know more info, he can always hover over the time and check the description:

• Resource cost: 90 units of iron, 10 units of mana shard
  
• Human resources cost: 30 citizens
• Transport time: 4 days (total)
• Manufacturing time: 7 days (total)
• Training time: 14 days (total)

Ok, so we have 1 more problem. Let's say I desperately need to build some defenses in my cities. That's where priority system comes into play. There will be 4 levels of priority: 0 stars, 1 star, 2 stars, 3 stars. The more stars, the higher the priority. If we suddenly add new units we want to build, then all the resource chain switches to provide enough resources for our needs. What does that mean? Mines will select new caravan, load it with ores and send it to blacksmith as a priority package. All other (lower priority) tasks are stalled (if they are conflicting with the important task). If two task are of the same priority, FIFO will start to work.

The last thing I would like to talk about is the resources sharing system. In the aforementioned example I assumed a simple case when we have one city with barracks, one with crystals, one with blacksmith and one with mine. But what when we have several cities having such facilities built? There's an easy answer, but I know that it would be quite a feat to write a fast algorithm for that. On the other side, nowadays we have so fast CPUs and efficient algorithms, that I don't think it's anywhere impossible. Keep in mind, that a player won't know anything about this system - he will just enjoy the game.

So, that's how I would define the algorithm:

1. Select the city in which the units are to be trained as the destination place (we don't allow training units in a few cities).
2. Find all the starting places from which we would have to obtain resources. Obviously recruits are already done (read 6th point above), so we are left with mines, crystal extractors etc.
3. Divide resources into such that can be immediately delivered to the destination (aka simple resource), and such that has to be processed (complex resource).
4. Simple resources: find production places closest to the destination. Select the closest one and find if it can deliver all the goods in the lowest possible time (it means: extraction for 1 day). If it can't, calculate the time it needs to extract all needed resources. Sort the production facilities (the rest of them) so that the closest ones (the shortest travel time) are on top. Check the delay in the travel time (comparing to the best place) and calculate whether or not it's worth choosing this location as sub-resource-supplier. If it is, than select it and check if now the shipment will be fulfilled. If not - either wait longer for extraction, or take the next supplier - and so on...
5. If case of complex resources the case is much more complicated. I don't want to define yet another algorithm, so just a short explanation:

1. Find all the start locations (mines).
2. Find all the processing locations (blacksmiths).
3. Find all the extra processing locations (super hyper something).
4. Create a matrix of nodes and interconnections.
5. Use some fancy graph algorithms or something like this and calculate
6. Get the best path and block/reserve the spaces according to priority.

Ok, that's all. Red signing out...

P.S. I got suggestion addicted .

Sometimes yes, sometimes no, depends on the game. I haven't played Cities XL yet either but I loved Sim City back in the day. I stopped playing the Civ games after 2 or 3. The Civ games are good at a lot of things but the unrealistic battle models just killed the game for me. For grand strategy these days I bounce back and forth between Empire Total War, Hearts of Iron 3, and Gal Civ 2.

I haven't played a good fantasy empire game in a long long time and I'm really hoping Elemental will have what it takes to fill that void.

After reading red1939's suggestion on how it could work, it really drove home how complex the system would be. Managing that manually seems rather painful. If it will all be automated, then all of the complexity seems a little much. I'm in agreement with the "no resource storing" camp after seeing some more of the detail.

I like Capitalism 2, which is a lot like an economic simulator.

I loathe the Civ model where one iron mine somehow supplies a transcontinential empire with inifinite supplies of iron such that all its manufacturing capacity can go full out and never run out. It makes additional strategic resource nodes useless except to trade. Its so unrealistic that its ludicrous.

That said, maybe storage isn't needed. If you have a linked Iron mine, it can provide X iron a turn. When you make things that get a speed bonus (using the present resource model) from Iron, some of the iron is used to get that bonus. When you run out of iron during that turn, the bonus goes away for items that can't draw on the iron.

Hooking up more iron or improving your mines through research lets you make more things at an accelerated speed, so that problem goes away. But it also doesn't require resource storage or things like that.

What do you think?

Let me present another model.

Because while I don't like warehouse management in my games, I do like supply lines and logistics management as a battle strategy. Is it just going to sit there when you don't need it for swords? No, the people are going to demand it for making buildings, utensils, farming tools, etc. So what is in "storage" is simply a reserve.

Each initial town area is built with limited "storage". It should cap out early and easily and the player should rarely think or worry about it.

Instead what is used for calculations is how much of a needed resource is being DELIVERED via caravan. If you don't have enough materials coming in via caravan then you have a shortage which means production is slower or more expensive.

So why would you have storage at all? To stop caravan harassment from totally shutting down the economy.

To use a real world example I give the most important modern day resource: Oil

We have strategic petroleum reserves, but they aren't used at all day to day. Instead what is being delivered via "Caravans" determines the price of what you produce using oil. If there is an interrruption in supply only then does the U.S. consumer care about our "reserves".

So unless your caravans are besieged you would be managing supply lines and not warehousing.

That is my counter proposal.

Unfortunately without resource storing the underdog is at even a larger disadvantage in a game.  As the map becomes more dominated by the strong players the weaker players will lose control of their mines and be more bottled up inside their towns.  At least with the option for storing resources they can hold out longer against sieges and even more likely to use stored resources for breaking that siege.   Without resources being stored the underdogs will fall more quickly.

When you guys say resource storing, are you talking about stockpiling in general, or stockpiling in a particular city?

This is bad... what do you do in peacetime when you aren't building soldiers? And then in wartime, when you need to build soldiers faster than you can mine the ore for their swords? There has to be some kind of stockpiling system. If travel times are going to be taken into account in production, then the stockpiling should be done at some predefined point in the supply chain, if not, then store them in the grain silo in the sky for every city's use. (I would prefer storage in a specific location though).

I think that red1939's model looks pretty good though, if you just include some overflow storage .

Of course, I'm a rocket scientist with a minor in economics, so I might be biased towards a little complexity here and there. (seriously, I'm a 3rd year student majoring in aeronautical and astronautical engineering with a minor in economics at The Ohio State University)

No amount of gold in the world is going to let me build houses if there is no one to buy wood from. No amount of gold in the world is going to let me build more swords than I have iron for if there is no one to sell me iron. I hate the idea of turning gold into some ubiquitous speed-enhancing elixir that lets you build/train/produce things that you don't even have the requisite resources for to begin with. Doing so severely decreases the worth and individuality of resources, even if gold costs are high.

IMO, if you want finite resources, then you need local resource storage. If SD decides to get rid of resource storage, then IMO they'd be better off getting rid of finite resources, too, and taking the Civ IV approach. That wouldn't make me happy at all, but as far as I can tell any 'middle ground' between the two approaches are necessarily contrived.

1. The system wouldn't be so complex. What I described above is the internal workings of the algorithm. You really don't have to think about every step of the chain - this is what will be done automagically by the game. There are just a few simple things you have to keep in mind, while managing your empire (all of these steps are quite similar to the real-life simulation, so you shouldn't have problem remembering/understanding them):

• If you can, try to put all facilities in one city (mine, forge & training facility). This way you minimize the travel cost.
• The more forges (for example) you build, the faster you can expect the goods to be available. Remember not too build too many of them as the maintenance costs.
• Don't put all your facilities on the opposite sides of your empire.
• You are basically limited only to the placement of the resources (ex. ore). So, if you have two training facilities, than try to build processing processing facilities near them.
• If you want to build something faster, than just set the priority - the best combination will be calculated by the computer.
• It's a really easy to understand system, hence it's even hard for me to make any more point explaining it .

2. It was an ad-hoc proposal - I didn't consider all the possibilities, solutions etc. Guys in Stardock can really make something much more useful and easier to understand.

3. While I understand that storing has it's pros, I really would like to see something different this time. I mean, resource storing is the easy way out. You extracted some materials and bam - you have them stored for eternity; but what about building storage houses? Another problem is, how should forges operate? If you store resources, then forges use some of it, but how are the goods transported? I won't lie to you, I have an idea of a functional storage system, but it looks like an original no-storage one, thus I see no point in revamping the terribly dumb idea (useful in some games, of course), into making it work in more complex environment.

As I see it, the storage system leads to a game when every resource (forges, crystals, etc.) is global. You build a few mines, which store as many minerals as you want, so you can use it anytime, anywhere. You also build a few forges that can pump out a specific number of ... 'forge points' which in return will be automatically changed into armament in every city, any time... Forge points?? Does it make any sense? No. So, maybe we require that in every city with training facilities, have to have a forge (and any other processing facility). Said that, is there any reason for a forge to be in a city without a training facility? Probably not. Let's remove the forges then, and put them into some upgraded training facilities - one less building to choose from, and one less thing that can confuse a player. It's too drastic? Then, just one forge for the whole empire? Come on - it's a global system!

Ok, ok - we can't store forge points, but we state that a forge can make a fixed number of items, every day. Such forge should be global of course (not to require too much thinking from our players), so how do we state to which town will the resources come first? Do we appoint each forge to a town (can be a different location)? We require a forge to be built in a town with training facilities (deja vu?)? Or we make some priority queue... What!? But, that's basically the same thing that you wrote before, in the NO-STORAGE system! I hope you get what I am getting at. A storage sys. would have be a no-storage sys. without distance (real life) issue, and with a infinite storage place (virtually impossible).

The main issue with the storage system, it's hard to change anything in it, without literally revamping it, thus making it more like the counter-proposal. There are a few other issues with such 'game feature'. Before I get to that, let me assume that we don't have any storage houses in the game - ie. more complexity. In SS (storage system), opposing to NSS (no-storage system), you really don't have to guard (at least very fiercely) your mining facilities, as even if they are taken, you have enough of stored minerals, that you can pump out units for quite a few turn, before recapturing the city. I mean, the resources should be something more than a simple number - this is your extra edge over an enemy, not a number that will increase to some cosmic value.

There is another aspect of it: there will be many resources ranging from iron ore, through blood crystals to available forges. How do you plan to put all the info on the screen? Iron ore: 40 pieces, blood crystal: 24 pieces, etc. It will be really horrible for a player to calculate all this stuff. In NSS it's simple - you have some daily inflow of resource, which you can easily determine, whether it's big inflow, or small. Basing on that you can assume how much you can produce. Of course, you could do such calculations in the SS, but how about, if you have a chain of 4 elements? Iron ore -> weapon, iron ore -> armor, crystal -> amulet, etc. People will still would like to know these values in NSS, so the game should allow them to view how much ore a mine can produce per day and such. Still, most of the info a player would need, would be in the tooltips. There would be a number of days needed to complete the task.

I know that the no-storage system isn't perfect, and shares similar problems like the typical Starcraft one. Never the less, I see it as a system that adds some interesting new aspects to the game (priorities, travel times) and feels much more natural than infinite-space thingie. I understand that - as it is now - the NSS is too complex (at least for many people), but I would like to ask devs not to choose the easy way out (ie. implement a simple global resource system). Please, try to make it rewarding when a player captures a resource. Try to imitate the reality (to some reasonable degree), so that player can know that the location and number of facilities IS important. Try to make the system with enough tooltips/tools, so that a player can manage a more complex/interesting system without spending 10 min. before building a unit.