Mining and Legions

Trey

Morthif said:

Trey said:

Sarapis said:

Watchman said:

Don't suppose we could open up another place to get wood - between mining and housing the price has seemingly permanently doubled. Unless that was the intent, of course..

6100 wood on the comm market at or below 60 gold!

100,000 wood and change in my private estate

this is why we can't have nice things.

Dude, I'm trying to get enough to recreate the Chateau de Versailles I don't give a fuck about your mines brah.

Medi

Tharvis said:

Islands are also still a steady and reliable supply of wood at 26 per.

Until you said something -_-

Skye

Yeah, shut up Tharvis.  

Draekar

Wood seems to be pretty reasonable, especially now mine building has levelled off a bit. I have seen it stacking up in a few shops and managed to gather up a few thousand over a couple real days without any effort at all

I am also pretty happy with all the changes @Sarapis and co have made since implementing this system, they seem to be constantly improving the way it works and I really like the way it has shaped up now.

I still don't understand the spawning at all. It was odd when there was nothing over the weekend and it dwindled down a lot then suddenly there was coal across the majority of the rooms, then that died down over the following days and a similar roll out happened with Iron. I think this is where the misconception that @Rangor mentioned about the chatter of set spawn times after the serenade came from.

 I am happy that this isn't the case, thanks for clarifying @Makarios

Taraus

Skye said:

Yeah, shut up Tharvis.  

Inuad

@Draekar:

Sarapis said:

Inuad said:

Makarios said:

I don't imagine we're going to reveal the exact criteria under which new lodes spawn, but suffice it to say they do not spawn at a set time.

As it seems lodes are spawned based on relative production of commodities (putting out as much of x commodity as the Garden wants at any particular point), is it possible that there would be times when there are no new lodes at all, of any kind? Or more drastically, times when there are no mines and no lodes at all?

Yep. If people just mine out everything, that could happen for awhile. It's necessary in order to control the overall rate of production of comms.

Got a bit of a plan we're working on here to potentially ameliorate the no lodes thing.

Lode generation seems almost entirely based on how much is produced in a given time frame. It may also have something to do with available commodities, though @Sarapis has consistently suggested that the mining system is meant to roughly replicate the village system in the amount of commodities which were in the economy at any given time.

So if the Garden decides that 10k coal is meant to be pumped into the system in a RL day, but available mines extract 30k coal, it seems as though there would be a gap of about two RL days before new coal lodes would appear.

Keep in mind this is my best guess as to how the system roughly works. I'm sure there's a bit more variation and nuance. I'm sure @Makarios or @Sarapis could give a vague affirmative or negative on it.

Sarapis

Inuad said:

@Draekar:

Sarapis said:

Inuad said:

Makarios said:

I don't imagine we're going to reveal the exact criteria under which new lodes spawn, but suffice it to say they do not spawn at a set time.

As it seems lodes are spawned based on relative production of commodities (putting out as much of x commodity as the Garden wants at any particular point), is it possible that there would be times when there are no new lodes at all, of any kind? Or more drastically, times when there are no mines and no lodes at all?

Yep. If people just mine out everything, that could happen for awhile. It's necessary in order to control the overall rate of production of comms.

Got a bit of a plan we're working on here to potentially ameliorate the no lodes thing.

Lode generation seems almost entirely based on how much is produced in a given time frame. It may also have something to do with available commodities, though @Sarapis has consistently suggested that the mining system is meant to roughly replicate the village system in the amount of commodities which were in the economy at any given time.

So if the Garden decides that 10k coal is meant to be pumped into the system in a RL day, but available mines extract 30k coal, it seems as though there would be a gap of about two RL days before new coal lodes would appear.

Keep in mind this is my best guess as to how the system roughly works. I'm sure there's a bit more variation and nuance. I'm sure @Makarios or @Sarapis could give a vague affirmative or negative on it.

Close enough to be more or less correct, yep.

Sarapis

Ok, so here's what we're thinking. I wanted to share it with you guys first, however.

1) A new kind of sort of mini-lode (call it mini-lode for the purposes of this discussion) will exist. There will be a profusion of these.
2) Mini-lodes will be of a certain raw comm type (iron ore, platinum ore, raw stone, etc).
3) Mini-lodes will have, just like lodes, a lode size and so on.
4) However, you'll have to mine multiple shards or splinters or whatever of a lode in order to get a single raw comm. Basically, mining is slowed down dramatically.
5) Because of that, squad xp gain is also slowed down proportionately.
6) Squads could not level up to higher than level 2 in these mini-lodes.
7) There'd be no rare minerals gained from mining these.

So basically, their purpose would be to let you effectively lose money at mining, over time, in order to level up your squads from 1 to 2 and thus have some value in an attack (level 1 squads having no value at all in an attack). 

Our thoughts are that this will let people effectively pay to, slowly, have their squads leveled up to level 2, but it's not unlimited like a shop. It's highly rate-limited, which prevents people from just slamming through hundreds of squads or whatever and thus buying the right to just assault people over and over.

Thoughts?

Kez

My thoughts would be to make the mini-lodes just coal, stone, and bone, and maybe carbon and obsidian? and perhaps remove those commodities from the normal lodes.

Melodie

Would these mini-lodes impact the normal lodes spawning at all? I assume not since there's so many restrictions with them, but just wanted to be clear.

Medi

I like things how they are. With more people being discouraged from/bored with mining there have been things like small coal lodes left open for a day or two. Maybe just adjust things so that more small lodes spawn? Perhaps, specifically coal/stone/carbon? Since these comms come in large quantities anyway the mines aren't high-value targets and, in my opinion, are an easy in to the mining world.

Skye

I'm with Medi on this one. I've lucked out by mining mostly the low-demand things. If you're willing to drop a large mine on it (essentially mining at a loss), you'll level up pretty fast without much problem. It's only people doing silly things like putting small mines on huge loads of like iron and such that tend to get curbstomped. 

Inuad

Sarapis said:

Ok, so here's what we're thinking. I wanted to share it with you guys first, however.

1) A new kind of sort of mini-lode (call it mini-lode for the purposes of this discussion) will exist. There will be a profusion of these.
2) Mini-lodes will be of a certain raw comm type (iron ore, platinum ore, raw stone, etc).
3) Mini-lodes will have, just like lodes, a lode size and so on.
4) However, you'll have to mine multiple shards or splinters or whatever of a lode in order to get a single raw comm. Basically, mining is slowed down dramatically.
5) Because of that, squad xp gain is also slowed down proportionately.
6) Squads could not level up to higher than level 2 in these mini-lodes.
7) There'd be no rare minerals gained from mining these.

So basically, their purpose would be to let you effectively lose money at mining, over time, in order to level up your squads from 1 to 2 and thus have some value in an attack (level 1 squads having no value at all in an attack). 

Our thoughts are that this will let people effectively pay to, slowly, have their squads leveled up to level 2, but it's not unlimited like a shop. It's highly rate-limited, which prevents people from just slamming through hundreds of squads or whatever and thus buying the right to just assault people over and over.

Thoughts?

My main concern, as @Melodie asked, would be how these would influence normal lode spawns. In addition, I would worry that they would crowd out normal lodes in the very limited space around a camp. Perhaps add a zone from 7-10 spaces from a mine which is where these would spawn? Otherwise, if they count against normal lode spawn rates, or take up the spaces that a normal lode would spawn, it would deter established miners (especially with a level cap on the lodes).

I think that if you're limited the level gain, the production rate could be increased slightly. Unless squads only gained xp when enough shards/splinters were mined to equal an actual comm. On the other hand, I suppose it's equally fair to think that, given that mining is a long term 'game', having someone set up a mini-mine on a mini-lode would take maybe a few days or even a week to level up their squads.

I also feel that maybe the output of a mini-lode should be masked to prospect, both to prevent larger mining players from both constructing a mine and from assaulting/taking a highly desired mini-lode during an actual lode drought (say, iron isn't spawning normal lodes currently). If the intention is more to allow people to level up their squads, I think not knowing what you're building on would be fine (PROSPECT would show something like "There is large mini-lode here that you'd guess is 0% mined out." Basically replace the commodity line with 'mini-lode'.)

Inuad

Medi said:

I like things how they are. With more people being discouraged from/bored with mining there have been things like small coal lodes left open for a day or two. Maybe just adjust things so that more small lodes spawn? Perhaps, specifically coal/stone/carbon? Since these comms come in large quantities anyway the mines aren't high-value targets and, in my opinion, are an easy in to the mining world.

The issue is that big miners might still jump on a small lode of stone or gems or other 'low value' targets if there aren't any high value lodes (like iron or coal, yes, I consider coal high value). There's also every potential that a more powerful miner would swoop in, steal your 'low value' target, and give it to a citymate.

Adding something that has even less value (with level caps, production rate caps, etc) would be more likely to deter the stronger miners and have a higher chance of letting a new miner at least get their squads up a bit.

Silas

Mini-lodes for otherwise high-yield, low value commodities such as coal and carbon would be great, but I think Inuad's suggestion of having these spawn separately, in a 7-10 room radius of the mining camps, would be fantastic. I don't think they're necessary, as such, in that I still think it's easy enough for new miners to get their foot in the door by constructing large mines on smaller coal lodes, which mine very quickly and are not particularly valuable, but if you want to add an entry level it's a great idea for it.

If the mini-lodes would block or interfere with actual lode-spawning in any way, though, I would prefer not to see them. We've all seen how quickly Achaea can be strip-mined since mining went in, and I don't think adding an entry-level that would exacerbate the problem of availability of lodes would be a good idea at all.

Siduri

Sarapis said:

Ok, so here's what we're thinking. I wanted to share it with you guys first, however.

1) A new kind of sort of mini-lode (call it mini-lode for the purposes of this discussion) will exist. There will be a profusion of these.
2) Mini-lodes will be of a certain raw comm type (iron ore, platinum ore, raw stone, etc).
3) Mini-lodes will have, just like lodes, a lode size and so on.
4) However, you'll have to mine multiple shards or splinters or whatever of a lode in order to get a single raw comm. Basically, mining is slowed down dramatically.
5) Because of that, squad xp gain is also slowed down proportionately.
6) Squads could not level up to higher than level 2 in these mini-lodes.
7) There'd be no rare minerals gained from mining these.

So basically, their purpose would be to let you effectively lose money at mining, over time, in order to level up your squads from 1 to 2 and thus have some value in an attack (level 1 squads having no value at all in an attack). 

Our thoughts are that this will let people effectively pay to, slowly, have their squads leveled up to level 2, but it's not unlimited like a shop. It's highly rate-limited, which prevents people from just slamming through hundreds of squads or whatever and thus buying the right to just assault people over and over.

Thoughts?

If there is a level cap, I'm not sure the experience gain should be slowed that much. I mean, could get the same exp for a shard than for a raw unit, just... without being profitable.

Alternatively, if it allows to levelup further than 2, keep the slowed down exp rate and you've got yourself the perfect gold sink, all while forcing people to be smarter about their troop use, because it too so long to train them.

Lisbethae

To be honest, the only thing that makes mining profitable at present is the rare minerals. Now, I rarely have had the chance to be on anything other than a coal mine, but still. 

Now, I can't even get a lode to mine so that my miners can level up, putting me further behind. 

I feel that I -need- to mine to keep my Trade Ministry afloat, otherwise, I don't think I'd bother. The frustrations are getting pretty numerous.

Xinna

I don't like the idea of mini-lodes. There are enough comms out there already, and I wouldn't want less real lodes.

As others have said, you can already level soldiers pretty easily by throwing a large mine on a bad lode, like a small coal or bone lode.

Sarapis

I have no idea where any of you got any idea that mini-lodes would crowd out normal lodes. It's certainly not in the high-level design I posted! 

Lisbethae

Sarapis said:

I have no idea where any of you got any idea that mini-lodes would crowd out normal lodes. It's certainly not in the high-level design I posted! 

I think probably because it wasn't -explicitly- ruled out. I am constantly amazed at the quick grip people in Achaea have for keeping rules, exceptions and possibilities in their heads. 

Sarapis

Clowns dropping from skyships wasn't explicitly ruled out either!

Tecton

Sarapis said:

Clowns dropping from skyships wasn't explicitly ruled out either!

Man, I told you, we're keeping that under our hats until phase 2!

Sarapis

Shit. Beans = spilled.

Sarathai

I'm totally down for skyship warfare.

Just saying.

You know, in case.

Ahmet

Tecton said:

Sarapis said:

Clowns dropping from skyships wasn't explicitly ruled out either!

Man, I told you, we're keeping that under our hats until phase 2!

Clowns... things falling from the sky...

Wait... is that how Jesters get their things? Do they have like... some supersecret fleet of skyships just on standby to provide things for them under the guise of wishing?

Skye

on topic though, can we get MINE SHOW # to also include the date at which the mine was constructed/taken over? A little late down the line but it'd be helpful for record purposes. Sometimes we forget.

Delios

**This is my first forum post EVER, and it's pretty long, so please don't get mad at me

I was thinking of an idea along the line of mini-lodes, but I'm not sure why you would want to decrease the rate of soldier experience gain. It sounds like these are already expected to be a guaranteed net gold sink, so adding a rate limiter on experience gain on top of that handicap seems unnecessary.

The general problem, as I understand it, is that:
    - People are frustrated with the lack of nodes
    - You want to limit the flow of commodities into the system

I'm probably missing something here, but overall the entire system is controlled by two main variables:
    - Lode spawn rate
    - Lode spawn size

To increase the number of available nodes, so that everyone who wants to participate, can, the general idea is that you want to add some pretty low value nodes. But what if you just add another continuous variable as an input to the system, that being the rate at which minerals are extracted from the system (basically the 'shard' size, above) - it seems like you already have a knob for this, since extraction rates for different commodity types vary. Let's call that the 'purity' of a node. Basically if a node has a purity of .5, it takes just as long to mine in "miner hours" (# of miners in the mine * # of hours they were there), so experience for soldiers would be the same, but you only would get half the commodities someone mining a pure node of the same size gets. So, why give the same amount of experience? If you don't the people who happen to find the good nodes are basically safe from anyone who isn't also on a good node, because they would be leveling their soldiers faster.

So now we have: spawn rate, node size, and node purity

I saw on some post that you guys like suggestions to be fleshed out, so I tried to fully flesh out how you might calculate each of this. So...

First, lets just start out by saying you check if you should spawn a new node every minute. Also on every minute, you say that you want to on average add dQ of some commodity to the system (like 1 iron per minute or something). So, on each time stamp, you add dQ to a total Q you keep track of.

Spawn rate: People are complaining primarily about not being able to find available nodes, so obviously node availability should be incorporated here - if no nodes are open, let's jack of the chance of a new node appearing! But lets say there are already tons of nodes around - like when people were complaining about there being a lot of 'useless' nodes - well, the problem there was that the quantity flowing into the system was lower than ideal, so the total quantity you want to add to the system should also be an input - if you want to add a lot of quantity, there should be a higher chance of adding a new node. Both of these seem to be sigmoid type functions, ideally, so what I have for this is something along the lines of:

N = number of open nodes
Q = quantity of commodity you ultimately want in the system
C_1, C_2, C_3, C_4 = some random constants to tweak stuff, say (2, 5, 2, 10) to start
max_rate = the fastest average rate you want to add nodes to the system (one per hour = 1/60, etc.)
min_rate = the slowest average rate you want to add nodes (one per day = 1/(60*24))

probability of spawning a new node = ( (1-1/(1+exp(-C_1*N + C_2)) + ( 1/(1+exp(-C_3*Q+C_4)) * (max_rate - min_rate) + min_rate

Let's call the above term W. You throw that probability through some function that spits out true with probability W. If it's true, you throw a new node somewhere in the game. If not, just set Q = Q + dQ, and carry on.

So, if you do end up building a node, you now need to throw down a node with some size S and purity P (stick with me on the purity). 

For the size of the node, I imagine you guys are just sampling from some kind of distribution, but lets just say (where rand(0,10) chooses a random number from 0 to 10):

S = rand(0, 10) > 3 + rand(0,10) > 8 //Say 0 is small, 1 is medium, 2 is large

and now, for purity: the less quantity you want to add to the system (Q), the less pure your nodes should generally be. So, that's also sounding sigmoid ish:

P = (1/(1+exp(-C_5 * Q + C_6))) * (max_purity - min_purity) + min_purity

So, when those get plopped down, you would say that some new quantity of possible commodity entered the system, so you have to update your Q:

Q = Q - S * P

And you now have one more node in the system, so:

N = N + 1

Also, you can keep track of an estimate of gold that will flow out of the system. If R is the current price of the commodity on the comm market, or just average trade price in general:

dQ * R = gold flowing out of the system per minute

All things being equal, you should just be able to fiddle around with everything else and leave dQ the same, and you then have pretty good control of lodes popping up that take into account the things people have complained about. I know you guys probably have something more complex going on behind the scenes, but I was just thinking about it over lunch and thought I'd post this. I've also yet to find a free node that I want, soooo this post is entirely to try and get something through that increases the chance that I can find a free node

Also, how did this Silas guy manage to sit on a large iron node with a small mine for two months? That was weird.

Xer

Silas's small mine was bugged. Whenever anyone wanted to attack it, it would say another mine was already attacking it. I already outlined in the report how I think it'd be reproduced. Not sure if it was ever fixed.

Inuad

Skye said:

on topic though, can we get MINE SHOW # to also include the date at which the mine was constructed/taken over? A little late down the line but it'd be helpful for record purposes. Sometimes we forget.

As long as we are talking about MINE SHOW, it would be nice to also include the lode size the mine is on.

It would also be nice if deployed squads had more organization. Either a long list, one item per line, or some kind of color coding (e.g. miners are tan, heavies are red, etc).

On the other side, I know someone had already requested LEGION LIST show which mine a legion was deployed in. If that kind of thing is being considered, it might also be nice to have a LEGION LIST MINES, which would arrange your legions by mine. LEGION LIST <mine id> might also be nice.

Sarapis

Delios said:

**This is my first forum post EVER, and it's pretty long, so please don't get mad at me

I was thinking of an idea along the line of mini-lodes, but I'm not sure why you would want to decrease the rate of soldier experience gain. It sounds like these are already expected to be a guaranteed net gold sink, so adding a rate limiter on experience gain on top of that handicap seems unnecessary.

The general problem, as I understand it, is that:
    - People are frustrated with the lack of nodes
    - You want to limit the flow of commodities into the system

I'm probably missing something here, but overall the entire system is controlled by two main variables:
    - Lode spawn rate
    - Lode spawn size

To increase the number of available nodes, so that everyone who wants to participate, can, the general idea is that you want to add some pretty low value nodes. But what if you just add another continuous variable as an input to the system, that being the rate at which minerals are extracted from the system (basically the 'shard' size, above) - it seems like you already have a knob for this, since extraction rates for different commodity types vary. Let's call that the 'purity' of a node. Basically if a node has a purity of .5, it takes just as long to mine in "miner hours" (# of miners in the mine * # of hours they were there), so experience for soldiers would be the same, but you only would get half the commodities someone mining a pure node of the same size gets. So, why give the same amount of experience? If you don't the people who happen to find the good nodes are basically safe from anyone who isn't also on a good node, because they would be leveling their soldiers faster.

So now we have: spawn rate, node size, and node purity

I saw on some post that you guys like suggestions to be fleshed out, so I tried to fully flesh out how you might calculate each of this. So...

First, lets just start out by saying you check if you should spawn a new node every minute. Also on every minute, you say that you want to on average add dQ of some commodity to the system (like 1 iron per minute or something). So, on each time stamp, you add dQ to a total Q you keep track of.

Spawn rate: People are complaining primarily about not being able to find available nodes, so obviously node availability should be incorporated here - if no nodes are open, let's jack of the chance of a new node appearing! But lets say there are already tons of nodes around - like when people were complaining about there being a lot of 'useless' nodes - well, the problem there was that the quantity flowing into the system was lower than ideal, so the total quantity you want to add to the system should also be an input - if you want to add a lot of quantity, there should be a higher chance of adding a new node. Both of these seem to be sigmoid type functions, ideally, so what I have for this is something along the lines of:

N = number of open nodes
Q = quantity of commodity you ultimately want in the system
C_1, C_2, C_3, C_4 = some random constants to tweak stuff, say (2, 5, 2, 10) to start
max_rate = the fastest average rate you want to add nodes to the system (one per hour = 1/60, etc.)
min_rate = the slowest average rate you want to add nodes (one per day = 1/(60*24))

probability of spawning a new node = ( (1-1/(1+exp(-C_1*N + C_2)) + ( 1/(1+exp(-C_3*Q+C_4)) * (max_rate - min_rate) + min_rate

Let's call the above term W. You throw that probability through some function that spits out true with probability W. If it's true, you throw a new node somewhere in the game. If not, just set Q = Q + dQ, and carry on.

So, if you do end up building a node, you now need to throw down a node with some size S and purity P (stick with me on the purity). 

For the size of the node, I imagine you guys are just sampling from some kind of distribution, but lets just say (where rand(0,10) chooses a random number from 0 to 10):

S = rand(0, 10) > 3 + rand(0,10) > 8 //Say 0 is small, 1 is medium, 2 is large

and now, for purity: the less quantity you want to add to the system (Q), the less pure your nodes should generally be. So, that's also sounding sigmoid ish:

P = (1/(1+exp(-C_5 * Q + C_6))) * (max_purity - min_purity) + min_purity

So, when those get plopped down, you would say that some new quantity of possible commodity entered the system, so you have to update your Q:

Q = Q - S * P

And you now have one more node in the system, so:

N = N + 1

Also, you can keep track of an estimate of gold that will flow out of the system. If R is the current price of the commodity on the comm market, or just average trade price in general:

dQ * R = gold flowing out of the system per minute

All things being equal, you should just be able to fiddle around with everything else and leave dQ the same, and you then have pretty good control of lodes popping up that take into account the things people have complained about. I know you guys probably have something more complex going on behind the scenes, but I was just thinking about it over lunch and thought I'd post this. I've also yet to find a free node that I want, soooo this post is entirely to try and get something through that increases the chance that I can find a free node

That's a lot of words to say "You could vary the time to extract a comm!"  

(Seriously though, I appreciate it when people take the time to drill down into their ideas in general.)

Not something we're going to do though. I think it takes things a bit too far in terms of complexity and figuring out how to turn a profit consistently. We'll be adding in mini-lodes, which are effectively very small lodes in which it takes longer to extract a particular comm, though they will be a completely money-losing prospect. The difference is that it's very easy to tell which is which this way.