How to fix module balance

[Frightning]

jcranmer is right on the money: The reason prod>others is because of geometric scaling when used at multiple places in a production chain. The only reason speed sees use is because of beacons and how they speed modules interact with productivity. Honestly, as much I as don't like the idea of removing things from the game, the simplest solution to prod+spd being king is to get rid of beacons. Without beacons, you now actually have reasons to consider not just using prod in buildings, and then the idea of making energy efficient setups that are also fast (e.g. 3xEff3+1xSpd3) becomes something worth considering relative to prod modules. A secondary reason removing beacons would be good for the game is that the way beacons currently work is enforcing a particular layout for endgame production: alternating rows of beacons and assemblers. This is because it's the most efficient layout for stacking beacons on assemblers. Each beacon (sans those on the edges) affects 8 assemblers and each assembler is affected by 8 beacons too. There is no other layout that comes close to that level of efficiency. Without beacons, now layouts for endgame production will be much more varied.

[thereaverofdarkness]

Productivity is always inherently better than the other modules, and that's because the productivity across the entire chain becomes geometric. If you have modules that reduce the power consumption of each building by 40%, your entire factory uses 40% less power. If you have modules that give a 40% speed boost to every building, your factory gets a 40% speed boost. If you have modules that give a 40% productivity boost to every building, your factory gets a 2000% productivity boost. Even if that productivity boost comes with a 6× energy cost, that's still a 3× boost to items produced per unit energy.

Anyone who comes to understand the geometric nature of productivity then knows there's no point using anything other than productivity modules in machines and speed modules in beacons. A geometric function always beats out a linear function at some point, so making productivity cost more in speed, power, and pollution doesn't change the tradeoff, it just pushes the breakeven point a little further.

Yes. I wouldn't say it's better to use productivity at all levels, rather that it's better to use productivity at higher levels. The higher the level, the more gain there is to using productivity.

So here's a random idea: what if productivity modules could only be used in buildings that required some form of active cooling using heat pipes? I don't know if mods can do this right now, but it does give more reason to use the heat system which I think is underused in general. It also adds extra constraints in building high-productivity factories which I suspect would be enjoyable to try to work out rather than annoying to deal with, but I obviously haven't play-tested.

Interesting idea. I'll have to think more on this.




I also agree with removing beacons. I think they can work, but I don't think they do work at current. Modules, vanilla or our suggested attributes, are generally better balanced for use without beacons. This is especially true with productivity, as currently one of the biggest strengths of productivity is the way you can use beacons to offset its drawbacks.

[Hedning1390]

Without beacons, you now actually have reasons to consider not just using prod in buildings, and then the idea of making energy efficient setups that are also fast (e.g. 3xEff3+1xSpd3) becomes something worth considering relative to prod modules.

How would removing 8/12 modules affecting each assembler make the remaining 4 anything but even more precious? Removing beacons may make you consider using speed in some assemblers, but wouldn't make you choose efficiency. In fact since by removing beacons you are nerfing the entire factory you would need less power making the eff modules less important. By removing beacons you would also make people hit their ups cap much sooner both in terms of items produced and game play time.

I think removing beacons could work (although not for making eff modules better), but you would then have to multiply items produced and consumed by each recipe to compensate, eg instead of gears being 1 gear for 2 plate make it 10 gear for 20 plate, or find some other way to keep the logistics system taxed so that a single belt or bot can't service all of your assemblers.

[thereaverofdarkness]

I came up with an idea to partially solve the productivity problem!

Divide the productivity bonus by the level of the output. For example, if you put them in mining drills, you get the full bonus because the ore is a level 1 product. If you put it in furnaces for making iron plates, the bonus is cut in half because those are a level 2 product. In furnaces making steel it's cut to one third because those are a level 3 product.

If we consider petroleum gas to be a level 2 product, then high tech science is a level 7 product because you can trace its inputs back through a maximum of 6 other processes (processing unit < red circuit < green circuit < copper wire < copper plate < copper ore). If you trace through batteries to get to oil, or if you trace through processing units for sulfuric acid to oil, you get 5 intermediates. You can trace through speed module and into red circuit, following the first path to also get 6 intermediates, or you can trace either way through red circuits and through plastic to oil to get 5 intermediates. The shortest path is copper wire < copper plate < copper ore. You determine the level of the final product by the largest number of intermediate levels that reach it. So when you put productivity into high-tech science production, you get 1/7th of the bonus.

I believe this would still be highly advantageous at the top end, and possibly some mid stages would actually do worse than at lower levels, but overall there would still be advantage to using productivity at the high end, though its extreme potency would be drastically reduced. If someone could run the numbers to find out how much infrastructure savings you get from putting productivity in the rocket silo using these new values, that would be great. Maybe I will later but my brain is foggy and tired right now.

Here's the productivity module attributes we're running with:
+10% productivity
+20% energy consumption
-15% speed
Since rocket part is a level 8 product, it's productivity bonus per module is: +1.25%

Don't bother considering the effects of adding speed modules, since we're going with them being multiplicative with the speed reduction from productivity.

[Frightning]

Without beacons, you now actually have reasons to consider not just using prod in buildings, and then the idea of making energy efficient setups that are also fast (e.g. 3xEff3+1xSpd3) becomes something worth considering relative to prod modules.

How would removing 8/12 modules affecting each assembler make the remaining 4 anything but even more precious? Removing beacons may make you consider using speed in some assemblers, but wouldn't make you choose efficiency. In fact since by removing beacons you are nerfing the entire factory you would need less power making the eff modules less important. By removing beacons you would also make people hit their ups cap much sooner both in terms of items produced and game play time.

I think removing beacons could work (although not for making eff modules better), but you would then have to multiply items produced and consumed by each recipe to compensate, eg instead of gears being 1 gear for 2 plate make it 10 gear for 20 plate, or find some other way to keep the logistics system taxed so that a single belt or bot can't service all of your assemblers.

The main reason people have zero incentive to touch efficiency for late game is because you can abuse prod modules with speed beacons to get the productivity bonus and fast/compact production at the same time...and the energy cost per item is actually better than prod modules alone, despite the beacons. In a world without beacons, prod modules would be hard to justify because of their speed, energy, and pollution modifiers. The main reason to use them would be to stretch the supply of a scarce resource on maps without a general abundance of all resources. So what about speed and efficiency? 4xSpd3 in Assem 3s gives +200% speed=300% speed in total for +280% energy consumption=380% of base energy consumption. You can get that same throughput with 2 Assem3s at 20% energy consumption cost with 3xEff3+1xSpd3, 150% speed on each, effectively 40% the energy cost of 1 assem3. Unless you're really hurting for UPS (and using solar specifically), I expect that 3xEff3+1xSpd3 would become the norm in that world.

I also expect that people that mixing prod and speed in the assembler would also start being a more common practice (eg. 3xProd3+1xSpd3 might be another common approach; though energy intensive and highly pollutive).

To solve production problem, perhaps they could add some space science that increases crafting speed (multiplicative with speed effects from modules). At least this way of endgame boosting of factory UPS efficiency doesn't eliminate 99% of practical layouts like beacons do, and could actually exceed beacons with enough research being done.

[bobucles]

The main reason people have zero incentive to touch efficiency for late game

is because energy is not the most limited resource. Energy scales as far as the player wants it to scale. SPM is limited by production, and Prod/speed will always give more production.

[BlakeMW]

Beacons make productivity modules much better but not having beacons would just increase mixing of prod and speed and increase the amount of lower tier module use.

The main reason to use beacons is productivity3 modules are almost prohibitively expensive, making them do 11x while only roughly doubling the cost of the setup is a great way to reduce the effective cost. But lower tier mixed modules or pure productivity have a similar ROI to beaconized speed 3, you don't get as much productivity multiplier but it's profitable to do. A simple and common example is shoving 2xprod1 modules in labs, that 8% bonus amounts to a 7.4% resource cost reduction in techs.
Say it requires 60 prod1 modules to prodify the labs setup, this would cost 7440 resources. Let's say it costs 10000 resources total to prodify the labs setup, because you also had to build extra labs and power.

The rocket silo research costs 310,000 resources, so the investment in the prodified labs setup is around 3% of the cost of Rocket Silo research. This means, unless your labs setup is much bigger than the ~30 or so labs I assume you're going to save resources by prodifying it. Rocket Silo is a big chunk of the tech tree up to Rocket Silo, but still a minority. It's possible that prod2'ing your labs could make economic sense - you spend 4x as much for only +2% extra productivity so it'd be close, but it definitely pay off if getting optional techs (nuclear, infinity research and such).

Now a clever person might say: but what if you were to take those 10000 resources and built mining drills, furnaces, assemblers. Might you not get more resources? You might, or you might not, it is entirely possible for prod1 modules to be "producing" more free resources per second than the same investment in miners+furnaces+assemblers+etc can (this is definitely true of Military Science packs and any higher throughput recipe), meaning even putting aside resource stretching it makes strict economic sense to get prod1 modules in certain recipes. But even if that weren't the case, module production is highly automatable and deployment is very fast, so it can potentially make sense in terms of saving time.

The point is, productivity modules actually pass this "cold hard maths" test, deploying them costs less than the alternatives in some cases. This doesn't come close to being true for even eff1 modules, unless the competition is solar/accu rather than coal or nuclear (eff1 modules are often more cost-effective than Solar/Accu).
Speed modules also don't really pass the cold hard maths test, they could theoretically: boosting the production of a building by 20% could make sense if that building cost more than 5x as much as a speed1 module. This is almost true of Assembler 3 and is true of Centrifuge - it's cheaper to speed boost centrifuges than build more. But both those buildings are optional and for the most part it's cheaper to build more than to use speed modules. But when a building has expensive productivity modules added, it becomes expensive enough that it's cheaper to speed boost than to build out.

But anyway, that's the point. That productivity modules have an economic argument for their use being sometimes the most effective use of resources. Efficiency modules do not and speed only really has an argument in the case of boosting productivity. None of this is intrinsic, like if eff1 modules happened to cost 1/10th as much they'd be economical to use.

[Hedning1390]

The main reason to use them would be to stretch the supply of a scarce resource on maps without a general abundance of all resources. So what about speed and efficiency? 4xSpd3 in Assem 3s gives +200% speed=300% speed in total for +280% energy consumption=380% of base energy consumption. You can get that same throughput with 2 Assem3s at 20% energy consumption cost with 3xEff3+1xSpd3, 150% speed on each, effectively 40% the energy cost of 1 assem3. Unless you're really hurting for UPS (and using solar specifically), I expect that 3xEff3+1xSpd3 would become the norm in that world.

The main reason to use productivity is not to save ore. It is to reduce the number of lower tier assemblers and furnaces. If I only have 4 slots to work with I wouldn't waste 3 of them on efficiency. I would use them for productivity. On lower tier products I may use maybe 1 productivity and 3 speed. On higher tier products I would still use max productivity and just build a ton more production buildings of that type.

Power is not a limiting factor, especially not when you slow down production and reduce power consumption compared to today. (Today power consumption is +880% + the cost of beacons which in the classic 8 for 8 pattern means an effective power cost of 1200%.)

Divide the productivity bonus by the level of the output.

You're just making every factory worse. And by doing that you'd have to rebalance all recipes otherwise people's factories would be like 90% furnaces, which isn't fun. Maybe it could work, but is it worth all that effort?

[thereaverofdarkness]

You're just making every factory worse. And by doing that you'd have to rebalance all recipes otherwise people's factories would be like 90% furnaces, which isn't fun. Maybe it could work, but is it worth all that effort?

Yes it will make the factories worse. I thought we were in agreement here that productivity modules are obscenely overpowered. It therefore follows that if you nerf them reasonably, the output of a factory previously depending on them will be greatly diminished. I don't see the problem here. I do understand people who got used to all the free resources will feel cheated, but you have to decide whether you want to balance the game or leave the problem in order to save people's feelings. Ultimately, they could just not upgrade their version if they didn't like the new modules, or mod the old ones back.


In fact I'd like to talk about how player feelings go into a lot of these things you guys have been talking about, such as FPS limits and power production. Ultimately, players are going to keep building a bigger factory until the game physically restrains them and prevents it from getting any larger. Not every player will, but there will always be the ones who make it to that point. If we raise the size ceiling, those players will just keep building a bigger factory and launch rockets faster, and make their SO/spouse even more lonely. Where they reach the limit is an arbitrary and emergent point which we can't really predict before we see where it is. Had the game originally had balanced modules, that point may have been reached with a smaller factory, but it would inevitably have been reached at a different point. And if someone had proposed making a change to those factory dynamics, someone would have complained that the change alters the factory size ceiling, as if it's some immutable fact of the game that shouldn't be contested. What's really happening is that change pushes people outside of their comfort zone.

You don't want solar power nerfed because you're used to relying on it, and nerfing it would feel like taking something away from you. But if it had been nerfed in the beginning, you wouldn't have placed 5x as many panels and hated every minute of it. At some point you would have become bored with it and found another way to enjoy the game, or quit the game, or perhaps wrote up a suggestion on the forums for how to improve FPS with power production. Or maybe you'd have found solutions to save power. Who knows, maybe you'd have discovered a way to build your factory even more powerful than you currently think is possible.

Nobody wants a thing nerfed, no matter how obscenely overpowered it is. This is basic psychology. At its core, the nerf comes off as an attack, and people don't rationalize it before determining the threat should be eliminated. If they eventually do rationalize it, they approach it with the intention of proving that their initial thinking was reasonable. They finally may work out in their minds the result post-nerf and compare its net power to pre-nerf, and feel it is sub-optimal, ignoring that they aren't the only player who will receive the nerf. They also will fail to spot tricks that enable a higher power output, which previously weren't necessary or weren't potent enough to compete with the thing which was previously overpowered.

When people are forced to have their thing nerfed, they rage about it and some people quit the game, or they mod it to have the nerf reverted. But those who continue playing post-nerf almost invariably wind up in a game world where everything is fine after all, maybe better than before, with brilliant solutions to existing problems--whether they choose to believe it or not.


I don't see a problem with nerfing/fixing productivity modules. If you want an alternative, try suggesting a buff that makes efficiency's effect "geometric", and see if it doesn't sound obscenely overpowered.

[dragontamer5788]

Yes it will make the factories worse. I thought we were in agreement here that productivity modules are obscenely overpowered.

Are they really? It costs thousands of iron and copper to build PM3 modules. At mega-base sizes, that resource expenditure is just a few minutes, but at "before the first rocket" sizes, there won't be too many uses of PM3 modules outside of science labs and the rocket silo.

Part of the question is 'what part of the game are we fixing balance' ?? At endgame levels with nearly infinite resources, the primary issue is to build the fastest setups possible that uses the least amount of mines. Because setting up mines is the most tedious part of the endgame. So productivity 3 is naturally going to be the ideal endgame strategy.

[BlakeMW]

The other aspect is sometimes it's okay for some things in a game to be relatively good or relatively useless. This allows players to discover the good strategies.

Productivity modules are good but not obscenely overpowered in the context of a speedrun. An in-depth understanding of the game is required to use them optimally.
Speed modules are convenient and are useful for empowering productivity modules which otherwise would be too expensive to use.
Eff1 modules and to a much smaller degree eff2 modules support a legitimate playstyle even if they aren't as strictly cost effective as the pollute and murder approach to the game. The best thing to say about them is of all the tier1 modules you notice their effect by far the most.
Eff3 modules are used for Power Armor II. It's something.

The game lets you develop things, and maybe it just turns out that the efficiency modules path was a bit of a dead end, but at least useful for that really nice power armor.

Games where everything is unambiguously useful can feel more lifeless than those which have a mix of useful and useless stuff. Maybe even the slowdown capsule is okay just to taunt players into trying to find a use for it (I know I tried...).

[thereaverofdarkness]

Games where everything is unambiguously useful can feel more lifeless than those which have a mix of useful and useless stuff. Maybe even the slowdown capsule is okay just to taunt players into trying to find a use for it (I know I tried...).

Not really, because you still have to find a strategy that works with it. Just because it's useful when used properly doesn't mean it'll work for you when you apply it for the first time.

On the flip side of the argument, I find it frustrating when I gravitate to a thing on the basis that it suits my style, or I find its purpose aesthetically pleasing, but I discover after many hours of trying to force it to work that it just plain doesn't work. I like everything to work, and to be viable. They don't all have to be balanced, but they should occupy a niche where they operate better than anything else at a specific task. If the best something can do is done better in every way by something else, then it has no purpose in the game except to stumble new players.

That being said, I think my original suggested attributes cause all modules to be usable, viable, even useful at times. Just because one player believes that some of those modules aren't worth using doesn't mean others would feel that way. In my original suggested attributes, I can see a use for every module:

Efficiency 1: fill every slot with these to get -60% or -80% energy expenditure to easily run on a small amount of power and generate very little pollution. Especially useful with harder power mods.

Efficiency 2: you can use 2 of these to get something all the way down to -80%, useful either if it has only two module slots or if you wanted to put productivity (which has only a small energy penalty) into the third slot. It's more expensive but useful if you're trying to squeeze your pollution output down as low as possible.

Efficiency 3: very expensive but useful for trying to squeeze in some speed modules and still keep the pollution/energy cost low. Works well when using productivity and beacons. The reduced number of efficiency modules needed substantially expands the number of other modules you can have without the pollution or energy cost overriding the rest of the factory's savings.

Speed 1: fill every slot with these so you just don't have to build as many things. Everything just runs faster now, with the same energy efficiency. Also good mixed with productivity.

Speed 2: less energy efficient, but packs in more speed per module. These are useful for combining with productivity when module slots are limited, while cheap enough to be used in the mid-range parts of your factory.

Speed 3: lowest energy efficiency and expensive. These are best used at the top parts of your production chain with beacons where they really go well with productivity, and at high enough production levels their cost is easily offset by their savings. You could also use them to purposely generate extra pollution if for any reason you wanted to do that.

Productivity 1: fill every slot with these to simply make resources last longer. Everything will go slower, but you can easily offset that by just building more things, or just accept the reduced speed and don't consume as fast. You can also combine them well with speed to offset the speed penalty, or with efficiency to offset the energy penalty. They also work fine by themselves since the penalties are pretty weak. These are cheap enough to be used on the lowest levels of production.

Productivity 2: these go great in the mid-ranges of your factory where they save enough resources to justify their cost. They combine well with speed modules but the speed penalty still isn't too strong to worry about.

Productivity 3: these are best used at the high end of your production chain along with beacons, so you can mix in speed modules. Once you get some speed in there, their speed penalty is very minor. The combo will cost a lot of power but as it's only on a small part of your factory, it's not very important for most setups. If you're using a low power/low pollution setup, you can mix in 2-3 efficiency modules to offset the power cost. You can also reduce the number of speed modules. You might find several speed modules beneficial, but you really can offset the speed penalty with very few speed modules.

[Frightning]

The main reason people have zero incentive to touch efficiency for late game

is because energy is not the most limited resource. Energy scales as far as the player wants it to scale. SPM is limited by production, and Prod/speed will always give more production.

So does building more assemblers, but people use beaconized setups instead. The reason they do so is because prod's geometric stacking when used on as many tiers of a high level production chain as possible saves on materials used per high tier item made significantly. This material savings comes at a pretty limited cost when one abuses the 8x8 layout we all know thanks to favorable additive interaction of prod+spd. Without beacons, you would need a LOT more assemblers to get a reasonable level of throughput with prod modules, and remember, those modules (tier 3) are VERY expensive. So I think you'd either see people doing 1xSpd3 with the rest of the slots Prod3 so that you get a decent speed per assembler, or people wouldn't bother w/ prod3 because their amortization time would now be FAR longer (another reason beaconized prod+spd is heavily used is that it actually has reasonably fast amortization time, on the order of less than 10 hours, without beacons that time will probably be more like 30-40 hours).

[Hedning1390]

I thought we were in agreement here that productivity modules are obscenely overpowered.

I never agreed to that. I think efficiency modules are underpowered but producticity are well balanced with the amount of speed that is available andthe current recipes. Here's what I said earlier:

It rebalances the factory so that late game your factory isn't 90% boring furnaces and 10% interesting chains. I think it is cool that early game you focus more on digging and smelting, and later you focus more on the high tech stuff.

Where they reach the limit is an arbitrary and emergent point which we can't really predict before we see where it is. Had the game originally had balanced modules, that point may have been reached with a smaller factory, but it would inevitably have been reached at a different point.

Hitting a hardware limit is not good. By being smart we can fill the game with many different and interesting challenges so that reaching that limit takes as much time as possible without artificially lengthening the time with grindy tasks such as building millions of solar panels.

Perhaps I wouldn't have suffered though the solar panels if they were weak from the start. Perhaps you are right that I would instead have stopped playing or played less. Those are not good alternatives imo. Games should be fun, not drive players away by having boring gameplay, or like I said earlier:

The question shouldn't be "is improving the efficiency of my factory too boring to be worth it?" It should be "is this fun option more efficient than this other fun option?".

As for not wanting things nerfed you should search for my posts on the bots v belts discussion. I am very much in favor of nerfing requester chests. I like things balanced. I have nothing against nerfs in general. General rule of thumb is nerf things that stand out as being very powerful and buff things that stand out as being weak. I think in this case it is the efficiency module that needs buffing, because nerfing productivity is a massive undertaking to keep balanced as you would completely change the late game. There are so many things that can go wrong, some of which I have already mentioned.

If you want to nerf solar panels that's fine. But you should find a way to nerf them that doesn't just make building them more repetitive. Things should have pros and cons, but being boring should never be a con.

[jcranmer]

Nobody wants a thing nerfed, no matter how obscenely overpowered it is.

That's not true. Obscenely overpowered things generally have very strong consensus that they need to be nerfed. The old god modules of Bob's Mods is a good example. I'm not aware of anyone who campaigned against nerfing them or ripping them out, and most of the people strongly arguing for ripping them out did so after playing with them at least once.

What people complain about is seeing their play styles get regarded as wrong and demanding that they change their play style. If something is objectively far and away the most powerful mechanism, calling it overpowered and ripping it out is going to generate backlash. And there's a good reason: the choice of mechanisms is ultimately about relative balance, and you can nerf something in relative balance by buffing alternates. Alternatively, you can often please many people by replacing the nerfed item with something that is more interesting.

One of the challenges with balancing modules is that there is a dichotomy between midgame (from initial science automation to launching your first rocket) and endgame (after rocket launch). Collecting resources or expanding power is a minor challenge in the midgame, but in the endgame these tasks are merely grind. Modules are available in the midgame, and the game very weakly tries to entice you to use them then, but as a matter of effect, their use is largely limited to the endgame. What balances well in the midgame often doesn't balance well in the endgame, and vice versa; alien artifacts were a good example of just how messed up the balance gets.

Productivity modules and beacons aren't overpowered. Instead, they're largely rewards that reduce the endgame grind that aren't effective in the midgame. The problem is that efficiency modules are underpowered, and that's arguably a combination of the power saving bonus having an easy-to-reach cap and efficiency-1 modules being balanced more for midgame, rendering higher-tier efficiencies useless. Nerfing productivity by making the bonuses weaker and the costs higher is not a fun-inducing nerf, nor would removing beacons be a fun-inducing change.

You make a big deal about how the situation would be different if these things were never in the game in the first place, but that's not really an argument. The fun of a game isn't determined solely based on the contents of the most up-to-date version of the game; the evolution of the game also plays a major role.

[bobucles]

So does building more assemblers, but people use beaconized setups instead. The reason they do so is because prod's geometric stacking when used on as many tiers of a high level production chain as possible saves on materials used per high tier item made significantly.

But speed modules aren't just for prod3 boosting. Plenty of beacons and speed modules get used on items that can't benefit from productivity modules. I'm sure you've seen at least one module factory that's loaded with speed beacons through and through. Even if the resource efficiency is silly, a speed beacon bases is still the best way to cram a huge amount of production power into a small footprint. Speed beacons become especially important with smelters because speed beaconing up to 400% or more will absolutely shrink one of the largest portions of your base.

It's not all about multiplying prod3, even if that is the best use of a speed beacon build.

[Frightning]

So does building more assemblers, but people use beaconized setups instead. The reason they do so is because prod's geometric stacking when used on as many tiers of a high level production chain as possible saves on materials used per high tier item made significantly.

But speed modules aren't just for prod3 boosting. Plenty of beacons and speed modules get used on items that can't benefit from productivity modules. I'm sure you've seen at least one module factory that's loaded with speed beacons through and through. Even if the resource efficiency is silly, a speed beacon bases is still the best way to cram a huge amount of production power into a small footprint. Speed beacons become especially important with smelters because speed beaconing up to 400% or more will absolutely shrink one of the largest portions of your base.

It's not all about multiplying prod3, even if that is the best use of a speed beacon build.

And that is an example of speed modules serving their purpose, helping you make stuff faster/in a more compact layout. That's what you're paying for when you make those modules and use them in buildings. The beacons still enforced more or less the same 8x8 layout (at least if one is trying to be efficient for something that needs to be mass produced), which is why I think beacons need to go (and be replaced by the space science research I proposed earlier; a crafting speed research option that is multiplicative with speed modifiers from modules). Thing is, in world without beacons, using 3xEff3+1xSpd3 becomes a hell of a lot more attractive because you no longer have to deal with Eff modules being (basically) pointless in beacons and in general for beaconized setups (already did the math that shows this, 2x as many assem3s for the same throughput and far lower energy usage compared to 4xSpd3).

[thereaverofdarkness]

I agree with all that.

I'm on the fence about whether or not beacons should go. Maybe the important thing for the early game is to provide balance and a variety of opportunities so everything has its place. Maybe the best thing for the late game is to let megabases have as much room to expand as possible. Perhaps at megabase level the module balance isn't important and it doesn't matter if some module types never get used by master players--as long as they get used by some people at some stages of the game.

Removing beacons would seem to improve balance, but it also doesn't seem entirely necessary to preserve a fun gameplay experience. The existence of beacons sure isn't preventing me from using efficiency modules, because I play differently from those of you who build megabases. Efficiency 3 needs a buff because I play the perfect style for it and I can't make it work for me, because what it does best is done better in every way by efficiency 2. But if its value were boosted significantly to -75% or something like that, I would use them. I can't speak for others, but that's high enough to be valuable to me.

I think every module should have its place, and new players and veteran players alike should always feel like modules offer choices, even if some players work out the exact math and determine the "correct choice". As long as the choice is subjective in a normal playthrough, I'd call that good.

[Frightning]

I agree with all that.

I'm on the fence about whether or not beacons should go. Maybe the important thing for the early game is to provide balance and a variety of opportunities so everything has its place. Maybe the best thing for the late game is to let megabases have as much room to expand as possible. Perhaps at megabase level the module balance isn't important and it doesn't matter if some module types never get used by master players--as long as they get used by some people at some stages of the game.

Removing beacons would seem to improve balance, but it also doesn't seem entirely necessary to preserve a fun gameplay experience. The existence of beacons sure isn't preventing me from using efficiency modules, because I play differently from those of you who build megabases. Efficiency 3 needs a buff because I play the perfect style for it and I can't make it work for me, because what it does best is done better in every way by efficiency 2. But if its value were boosted significantly to -75% or something like that, I would use them. I can't speak for others, but that's high enough to be valuable to me.

I think every module should have its place, and new players and veteran players alike should always feel like modules offer choices, even if some players work out the exact math and determine the "correct choice". As long as the choice is subjective in a normal playthrough, I'd call that good.

Well, if you like to use 2xEff2 and 3xEff1 a lot for min. energy use than for your Assem3s, use the 3xEff3+1xSpd3 setup, it's actually more energy efficient per item made (13.(3)% base energy cost per item). Beacons can go and we could still have something for the people that are going to push the limits of what the game can handle performance-wise. That's why I suggested new infinite research for crafting speed (this also has some very interesting choices because as crafting speed goes up, the optimal number of assemblers per belt changes for instance, ditto which inserters are needed/how many etc.). Still allows for small footprint production, and does so in a way that doesn't enforce a particular layout.