Module Reballance

[Holy-Fire]

I'll try to give an example.

Let's assume the following:

1. In your chain A->B->C->D->E->F->G->H->I->J, A takes 1 iron / second, and each machine takes 1 item of the previous machine per second. Until machiine J outputs a product at a rate of 1 per second.

2. Each prod module gives +10% prod, but with a speed penalty so it still has the same output rate. (Not a fan of the speed penalty thing, but the same applies without it and it's simpler so let's go with that). It also gives +100% energy penalty.

3. You use coal for energy. (There is a phase in the game where that's cheaper than solar. Even if solar is cheaper, it still has a cost due to the need to waste time placing panels. The same will apply with different numbers).

4. Each machine consumes 200W, which requires 0.05 units of coal per second.

Now: If you craft 10 prod modules and put one in each machine, you still get one end product per second, but instead of spending 1 iron per second, you only spend 0.9^10 = 0.35 iron. You saved 0.65 iron / s.

You will also need an extra 2000W of energy for these modules. This will cost 0.5 coal / second. Since coal is about as expensive as iron, you're better off this way.

If you try to do the same with a chain of length 20, you'll find that you now save 0.88 iron / s, but the extra cost is 1 coal / s. With the longer chain, you actually lose out by putting prod modules.

(The optimum will be to put modules only on some of the machines, perferrably the later ones).

If the energy penalty is +50% instead, you still gain by putting prod modules in this longer chain.

So there is a ratio between energy penalty and productivity which gives breakeven between prod module and nothing at all. This ratio changes with the length of the chain, but not as much as you think, and not in the way you think. Keeping it break even for a typical chain will give a good balance.

Of course, the correct comparison is not prod module vs. nothing, but prod. module vs efficiency module.

Which leaves us with the point that with large energy penalty that stacks multiplicatively, limiting prod modules is good, even for the purists among us.

For some formulas:

If you have a 1-1 chain of length n, where normally you spend A resources per second, then using prod modules that save x% of the input of each machine, saves you a total of

A * (1- (1-0.01x)^n)

per second. Note that as n increases, this approaches the fixed number A (you can't save more than you spend).

If the normal energy consumption is B and the modules make +y% energy consumption, the extra energy required is:

B*0.01y*n

As n increases, this increases more and more.

Furthermore, if the cost of a module is C, the one-time cost for placing the modules is

C*n

If each unit of energy is equivalent to e units of resources, the ROI you get from modules is:

(A * (1-0.01x)^n / n - 0.01By) / C

Note that this decreases with a larger n.

The same principles apply for chains more complicated than 1-1. (In general, the ratio between resource cost and time to craft for different items is very important).

[libik]

What about turn down customability a bit for more balance control?

Level 2 factory can have 1/1/1 slots and level 3 factory 2/2/2, in each group you can add modules of different groups (so the best factory can have 2 productivity, 2 speed, 2effectivity modules).

Also I think if we keep it on the rule "different groups", it can scale even with future adding of new modules. If you have 10 types of modules, you can just choose 3 of them and insert it into factory like here : 2/2/2.

It should prevent players for "massing" one type of modules and get exponencially advantage from that.

[Balthazar]

The problem is that the production mechanics are too simplistic. Think of an ultra simplistic smart phone rpg; no health bars, no mana, no notion of damage; you simply cast a spell and it kills an enemy. If a game like that had sockets for equipment, what kind of stats could they offer? Damage buffs, defense, hp doesn't exist in the game, so theres no potential for variety. That's what the problem here is.

Like a lot of others in this thread i love the idea of modules, but i think it's speed and efficiency that needs to go, because as established, they cannot compete with productivity modules, and i think productivity modules are the interesting ones, they just need alternatives added.

Instead of seperating stuff into intermediate recipies, you could add a refinement value to the recipie; making gears requires a very low refinement value while advanced electronics requires a very high one. If the machine does not >= the refinement for the recipie, the machine is unable to do the recipie, and having productivity reduce the refinement value for the machine; this way you limit what recipies can use productivity but in a more dynamic fashion. You can then add modules that increases the refinement value, giving machines access to more advanced recipies.

That's the core of my idea at least, so now i'm just gonna these two other changes that could be made to the system to add more variety to modules...

I mentioned the idea of a high quality output before; add a module to a machine and it has a small chance of producing a different output, like steel bar has a 3% chance to instead output a high quality steel bar, and have these high quality materials necessary for higher end machines and equipment, like submachine gun of penetration +1 or assembly machine 4 This gives you a choice between making lots of low grade materials with productivity, or getting less but high grade materials that are needed in other stuff.

Modules that alter the stats of output items would be hella cool, like +50 HP in your wall making machine = better walls, increased size of magazines in your ammo assembler, reduced power/pollution, etc etc. although i'm assuming this would be a lot more work to do. I kind of dismissed this at first because i assumed it would clog up your inventory, but as long as all of your items are from the same machine they should be stackable anyway, so that shouldnt be an issue.

Anyway, thoughts/additional sugestions?

[ssilk]

Hm. Well, right direction.

What I like with that idea is this
- advanced items can't use much productivity modules
- "special mix of modules" instead of new receipts.

But I have a problem with the amount of new items you generate like so. They are just more different items, not new stuff. From game aspect this brings no advancements.

I add a very new idea. Perhaps it is usefull.
What if from a certain level, we need to built more advanced research? The current research is simple. What, if I need to test something, to research it. And testing means: testing all combinations.

It is so, that the right combination for new receipts is randomly generated at the start. Now we have to research it. We have new labs for that.

So we want research for example special walls.
- the game tells, that the research takes for each test a piece of wall, and it knows that the receipt includes a combination of red potion plus a green and red module.
- a main lab (lab network) tells the labs, which combination it should test.
- the lab requests the needed items via logistics
- there are now two possibilities:
-- the labs need to make all tests
-- the labs can stop, when finding the right combination, which is in overall after half of the tests.
- so or so: at the end we have the right combination of new ingredients/modules, to create new items. In short: we just need to set the new receipt and put in more ingredients/modules


I don't know if this is better. But I know that more items need also more receipts, not a hidden system.

[Balthazar]

I have a problem with the amount of new items you generate like so. They are just more different items, not new stuff. From game aspect this brings no advancements.

No, the new recipes doesn't add anything, it's in the way these are acquired that changes the game, i was just giving an example. My point is there needs to be more mechanics to the actual production in order to allow more types of modules that can compete with productivity, allowing higher grade materials is one way to do this.

High grade factories would require you to sort the materials made and send them off to different places, to me that's something that adds to the game, and something i'd like to work on.

So we want research for example special walls.
- the game tells, that the research takes for each test a piece of wall, and it knows that the receipt includes a combination of red potion plus a green and red module.
- a main lab (lab network) tells the labs, which combination it should test.
- the lab requests the needed items via logistics
- there are now two possibilities:
-- the labs need to make all tests
-- the labs can stop, when finding the right combination, which is in overall after half of the tests.
- so or so: at the end we have the right combination of new ingredients/modules, to create new items. In short: we just need to set the new receipt and put in more ingredients/modules

I wasn't talking about research at all ssilk, i'm not sure what this would accomplish or if i'm even understanding it right

[ssilk]

I understand what you want to achieve, but it think the way how the game works doesn't allow to create new kinds of materials without a definition. And it will be very confusing, because technically they are different items, even if they might look the same. Oh, we'll. then my thoughts jumped a bit: The research lays then on hand: higher grade materials need to be researched first. The research shrinks the number of new possible receipts (combinations of modules) to some predefined. Hm. Perhaps this idea with random receipts isn't the best. But what comes out must be ordinary receipts.

And the way how the item is then produced depends on the recipt. You may need assemblies with 6 ingredients and more or some special new item as catalysator, or or or... There is a bunch of possibilities.

I want to add one more thought: assumed, we need to create rockets. Assumed, that rockets can't be transported as item, only as entity. We need many of them and we need many different types.
Now what would really fix me on screen Is a factory, where I need to produce the parts of the rockets also as entity.
And now this thought: think to a production street of a car: there are many stations and what kind of car comes out, depends on what is built into it. What kind of motor, air condition, servo, nice leather seats... Etc. Think further: in Star Wars we have completly automated factories, also in games like Portal. They "shift" a construction plate in steps forward. Either inserters add some parts, or they are processed (connected, heated, hammered, etc). Small items are connected to big items, big to bigger until we have a working rocket, which then must be transported to a start ramp, filled with stuff and fired?

How many steps do we need to achive a "perfect" rocket, which is able to return also abnd bring the colonists down?

And to come back to your point: a rocket, which consists of many different parts, which are consists of many different items, that thing can be assumed to have different characteristics, depending on what it is built and how. A simple wheel not. Or not that easy.

It is the same as if you have chillies, they are hot and an expert can tell you the difference between hundreds sorts of chillies. But what counts in the end is, how much of them you put in your food.

[Dark_star]

I hope this can include changing products back into their sub-products. example break down Brick Wall back into stone bricks or iron gears back into iron plates.
Nice to have a "reverse Assembling machine", put products into it to be changed back into sub-products with cost of 1. energy & 2. some lost to sub-items .

Thanks for a great game

[Holy-Fire]

Another thought I'd like to reiterate - The suggestion of multiplicative energy consumption penalty really balances out the dynamics. Let's say each prod module 3 triples (or quadruples, or whatever) energy consumption. So Normally you'd want to put maybe 1-2 modules, otherwise the extra consumption just wouldn't be worth it (especially when a green module can cut consumption by 70% or so, again multiplicatively). If in some other situation prod modules are more effective (for whatever reason) then you can bump that to 2-3 modules... But you won't get into a situation where filling the machine completely with prod modules beats all other options hands down. So you don't need to worry about "Fine tuning the balance for every scenario", it happens naturally.

Still hoping the developers won't break everything.

[Zourin]

Productivity modules are fine..

If you watch Malkasphia's Lets Play, he puts productivity modules in everything. While he's getting good results, he's producing so much pollution he's almost visibly mutating the Biters to be larger and more aggressive. Productivity modules are fine and 'nerfing' them isn't warranted for something that carries significant risks in overuse. Having 4-5 miners each with 71 pollution each is enough to significantly mutate biters to spawn more larges and mediums within very small windows of time.

I only use productivity modules in chip factories, steel production, and science labs. Speed modules go into Science Pack facilities and Weapons Facilities. Effectivity modules go in everything else to keep my footprint and power draw under control. Any Biter spawner location under my pollution halo is slated for removal as a 'threat'.

If anything, i'd say that the power draw effects on pollution are a bit too pronounced. When an Electric Miner with two productivity modules goes from 9 pollution to 70 (enough for seven more miners), there's something a bit out of whack.