why prod modules are important

[blazespinnaker]

Adding the prod mods was a pretty inspired idea by Wube, and I hope they don't get deprecated going forward. One of the things I find a bit silly sometimes about Factorio is the copy/paste aspect of it. Complex factories are fun puzzles to figure out, but slamming down endless repeats of the same thing doesn't take a lot of thought or really any creativity. It's just repetition.

Prod modules encourage using speed modules and powerful, dense factories. Without prod modules, speed modules don't make a lot of sense, as you might as well just copy/paste the underlying machines as speed modules start to get expensive. speed modules are multipliers for prod modules and make the resource cost for the speed modules more realistic.

Prod modules are also a bit of a reward for building complex production lines. If it takes 5 different steps to build some item, hey no problem, that's an excuse to add prod modules to each step of the way and increase the utilization efficiency of the raw resrouces. When you take prod modules out of intermediates, you lose that reward, and the complex factory building makes for more pain and less fun.

I realize prod modules make balancing a bit more complex, but removing them also makes the game more static. There are fewer ways to build the factories and you lose the dynamic behavior as you progress, rejigging your factories to make them ever more concentrated and efficient is kinda fun.

I know there are some folks who seem to think removing prod modules from the game is a good idea, but hopefully they'll reflect on that a bit and maybe rethink that perspective. Removing prod modules to incentivize more complex factories is cool, removing them because your having trouble figuring out balancing, is not.

One suggestion for balancing, is go half under half over. Produce less for lower level / no PMs, and more for higher level PMs. Low level PMs throughout intermediates to achieve potential is a simple puzzle relative to some of the more complex ones you find in factorio.

Please, give Prod modules a chance.

Cheers

[FuryoftheStars]

Have there been discussions somewhere from the devs about removing them? Just trying to figure out where this is coming from.

That said, while I wouldn’t advocate their removal, I do actually feel the opposite from you. I feel like the modules (or perhaps more so beacons?) are little bit too magicky for me and don’t typically like or use them, whereas copy/paste is perfectly fine to me because I don’t want to go through the boring routine of having to manually build the same thing I already did over, and over, and over, and over, and over, and over….

[blazespinnaker]

Well, the good news is that prod modules actually don't make much sense if you're just trying to launch a rocket. The resource cost versus what they deliver doesn't really pay off unless you're in it for post launch play. Maybe a few in labs or the rocket silo and some other expensive intermediates, but that's about it.

If you're interested in continuing to play past a rocket launch (say maximizing science per minute), than higher level PM modules make more interesting puzzles to continue with, rather than just duping what you did pre-rocket launch. Their high resource cost also pays off quite nicely after a few more rocket launches.

I'm not sure what you mean by 'boring routine'. With PM modules and beacons there are a lot of new and different ways to build an optimal factory, sometimes it can be fairly complex. Factorio is, if anything, about trying to optimize. Modules add the opportunity to do that even further.

[mrvn]

Damn those prod modules. I've have about 1000 steal furnaces and now I have to throw them all away and build electrical furnaces instead. What a waste of resource and time.

[blazespinnaker]

Damn those prod modules. I've have about 1000 steal furnaces and now I have to throw them all away and build electrical furnaces instead. What a waste of resource and time.

Someone did an analysis once, PM3s take awhile to pay off. 3000 raw + 4025 petro for 1 PM module. Unless you're building for the long haul, you're going to use these pretty sparingly.

[FuryoftheStars]

Hmm, I guess I just don’t agree with the “more interesting puzzles to continue with” part. Regarding productivity modules specifically (as I don’t believe they work with beacons), to me, it’s just a way to get the same output with less machines.

[blazespinnaker]

Well, they're puzzles as much as anything in Factorio is a puzzle.

Consider the resource cost of both prod and speed 3 are quite a bit. Now, imagine how you might build the highest producing factory which requires the fewest amount of modules. It's actually very tricky to find the optimium. It requires careful placement of beacons such that they are shared / effecting the greatest amount of assemblers as possible.

One advantage of this particular puzzle is that it strongly encourages footprint optimal factories. That is, dense factories which do not sprawl, which is an interesting problem in and of itself.

[mmmPI]

I think depending on the settings of the game prod modules range from pretty bad to absolute must have with all sorts of consideration in between.

With default settings and the objective of launching a rocket as fast as possible, there only use seem to be PM3 in the silo from what i've seen faster player do.

For a long game like 200 hours+ , i end up placing PM3 everywhere i can except mining drill and pumpjack , because with default settings at some point i feel all i do is building outpost to feed an always hungry machine and they mutiply the yield of each outpost when used in the factory further down the chain. The longer the game, the more ressources are put into science rather than into new buildings or modules in % of total ressources, this ratio is made even higher if you build (very) slowly production chain with large output or many steps of production like those in overhaul mods.

Before the time where they are widely available i usually have to make builds to adapt to my production rate/stock of modules. So if need new furnaces array or new refineries i'll either surround them with beacon, or surround a beacon with machines, or have a row of machine between 2 row of beacon, or having one row of beacon and 1 row of machine on each side to benefit from it depending if at the moment i have more speed modules or productivity modules in chests.

I use prod module 2 when i don't have prod module 3 yet, or when my production rate of PM3 is still low, upgrading them as soon as i can , same for prod module 1, but even less as i'm pretty sure detailed math would show that PM1 and PM2 are very cheap compared to PM3.

I also like to use the higher tier of productivity module available to me in the first and last machine of a production lane that is not active all the time and/or is not properly supplied with ressources for continuous production, thinking in extreme case only the last machine function because everything is backed up, or only the first machine function because there is not enough input, for example the iron gear for the mall, never enough of them when all belts are being made, but only 1 machine slowly making them when the mall is almost filled. Others would take speed module, until i can replace some of the assembly in the lane with beacons to speed those that are left and then they all receive PM3.

I like those puzzle, and design decision that they create, and it always feel good to me using the "max level", it feel like the early hours of manually mining the coal and wood are long gone

[ickputzdirwech]

Just wanted to add that while PM3s are really expensive, it is always worth it crafting four of them for the rocket silo. They pay of even for the first rocket.

Also PM1s are relatively cheap. Their pay of time is quite short for most things. For example PM1s are cheaper than processing units. So as soon if they produce one extra, they payed of.

When talking about investment cost there are two more things to consider:

• when building a base that is supposed to run “forever” the initial material cost becomes infinitely small compared to the resources required for science.
• when you invest in productivity modules you need fewer machines producing the ingredients for the recipes you use them on. This can lower the initial investment cost significantly as it ripples through all production steps.

[mrvn]

when you invest in productivity modules you need fewer machines

And when you are talking mega-bases it's all about the rockets vs. ups. Sure you can just spam twice the blueprints around but then your UPS takes a dive.

[jodokus31]

Well, the good news is that prod modules actually don't make much sense if you're just trying to launch a rocket. The resource cost versus what they deliver doesn't really pay off unless you're in it for post launch play. Maybe a few in labs or the rocket silo and some other expensive intermediates, but that's about it.

Prod modules, esp. mk1 make totally sense before rocket launch. Like you said, rocket silo (mk3 even), labs, green circuits, gear wheels, blue circuits, yellow/purple science, but I wouldn't call it "not much sense".
Even, if it only lowers the amount of mining and smelting lanes (f.e. full yellow belts), because it's usually more effort to get more lanes from a mining field + smelting column, unless you play around in creative mode.

[mrvn]

People keep talking about the cost of prod modules. But consider this:

Put in miners they produces 20% more ore (and oil although speed is probably better there). Imagine taking those 20% and using it exclusively to produce the infrastructure to make more modules and the modules themself. That will grow exponentially. Put prod modules in electric furnaces for 20% more plates for a total of 44% gain. The exponential growth has gotten more than twice as steep. Add it to copper wire and steal and many others and the gain just keeps growing.

Other than the initial investment to get the process started it wouldn't cost you anything, the modules produce themself. Well, except for the time for you to set it up and expand.

[mmmPI]

I'd rather put only 4 PM3 in the silo and get 40% productivity on everything before rather than placing them in all miners and furnaces to get a 44% bonus raw material input that will require more machine to process. Maybe in labs too, starting from the end of the production chain makes more sense than the opposite ...

Also when using many modules you expect the game to last long, and in this case research the mining productivity quite a lot, in this case adding productivity module only has diminishing return contrary to adding speed modules in both mining drill and pumpjack.

But really when you're in for a very long game, imo one is better off placing efficiency modules 1 in the miners, to reduce the ups cost of nuclear power, or the tedious of many solar as with a lot of productivty bonus, the mining drill don't require speed module contrary to pumpjacks.

[mrvn]

Obviously when you build the silo you add the modules. That is a no brainer and already pays for itself on that alone. As mentioned. And while working up to the silo adding them to the research labs is a good idea too. But that wasn't the point I was making.

Most people say producing the modules costs too many resources and argue about the resources they have to divert to make the modules. But in all that they totally ignore the fact that the fraction they divert is, other than at the start, all resources that wouldn't even exist if not for modules. Whether that is because the rocket silo uses fewer resources for the rocket or because miners produce more ore. It's just easier to argue from the miners side of the equation because there it's clear it's 20% more ore for every miner you equip with modules.

As for the time investment for building the infrastructure for making modules... That probably balances more quickly than most people think with the ore fields lasting longer. Therefore you have to kill fewer aliens and build new mining outposts less frequently. That is if you don't divert all the extra resources gained by modules to make more modules. At some point you produce modules as fast as you can place them.

As for the modules being less effective with mining bonuses that is only when seen as a percentage of the total. In ore counts it remains 20% of the base mining speed. You always get the same amount of extra ore from module.

[Amarula]

"There is no one true way." - Mercedes Lackey, Heralds of Valdemar
There are so many different ways to use modules and beacons (including not using them) and I love the challenge of trying different ways of playing, figuring out how the different options work, what the benefits and the hassles are, and getting, at the end, to choose how I like to play. So I agree with blazespinnaker, keep the modules and their puzzles for those who want those options.

[jodokus31]

..starting from the end of the production chain makes more sense than the opposite ...

That and also fast recipes like green circuits and gear wheels.
Blue science and red circuits is f.e. not that profitable, because the recipe is quite long and the resources are not that valuable compared.
But of course at some point in the megabase, you want to prod almost everything possible

Overall I would say, its not a question whether to use prod modules, but in which priority/order
f.e. miners are probably the last thing, and here I would go with mining prod instead.
prods in pump jacks is also quite questionable overall. Maybe for the first cycles, when yield is still high and you are low on oil on the map.
smelting is also quite late.

[mrvn]

prods in pump jacks is also quite questionable overall. Maybe for the first cycles, when yield is still high and you are low on oil on the map.
smelting is also quite late.

With prod modules the yield decreases less quickly but you also get less oil/s and adding beacons isn't always possible to counter the speed reduction. But I think you might be right and you get more oil out of a well with prod modules, at the start.

But at some point the oil/s with prod modules will be less than the minimum flow rate + speed modules and then preserving the yield becomes irrelevant. The exact point in time when switching from prod top speed modules would depend on the formula for how the yield decreases, if it ever is better to use prod in the first place. Might even differ from oil well to oil well.

[mmmPI]

Most people say producing the modules costs too many resources and argue about the resources they have to divert to make the modules. But in all that they totally ignore the fact that the fraction they divert is, other than at the start, all resources that wouldn't even exist if not for modules. Whether that is because the rocket silo uses fewer resources for the rocket or because miners produce more ore. It's just easier to argue from the miners side of the equation because there it's clear it's 20% more ore for every miner you equip with modules.

I disagree, it is very possible to divert 80% of your ressources to produce module which is much more than what the productivity bonuses would give you as "bonus ore" it depend on your setup .


If a miner has 100K ore and already 50% productivity due to the research, you expect 150K ore from it. If you add 2 PM3 modules, you add 20% on the original value, which means 100+20+50 = 170K expected ore total. While 150K+20% would be 180K ore. This is what i call diminishing return.

It also doesn't work this way if you have 100K ore on the map and you place 10 miners on it, and you give them all 2 PM3 modules. You use 20 modules total, but only raise the quantity by 20%

[blazespinnaker]

"There is no one true way." - Mercedes Lackey, Heralds of Valdemar
There are so many different ways to use modules and beacons (including not using them) and I love the challenge of trying different ways of playing, figuring out how the different options work, what the benefits and the hassles are, and getting, at the end, to choose how I like to play. So I agree with blazespinnaker, keep the modules and their puzzles for those who want those options.

Yeah, exactly. Prod modules make the game more open ended. Removing them from intermediates just reduces the amount of paths you can potentially take. I get that they boost production a lot, but this can be balanced by just increasing resource usage on later game materials, or maybe tweaking PM bonus a bit. PM2 maybe could be +5 instead of +6.

[mrvn]

Most people say producing the modules costs too many resources and argue about the resources they have to divert to make the modules. But in all that they totally ignore the fact that the fraction they divert is, other than at the start, all resources that wouldn't even exist if not for modules. Whether that is because the rocket silo uses fewer resources for the rocket or because miners produce more ore. It's just easier to argue from the miners side of the equation because there it's clear it's 20% more ore for every miner you equip with modules.

I disagree, it is very possible to divert 80% of your ressources to produce module which is much more than what the productivity bonuses would give you as "bonus ore" it depend on your setup .


If a miner has 100K ore and already 50% productivity due to the research, you expect 150K ore from it. If you add 2 PM3 modules, you add 20% on the original value, which means 100+20+50 = 170K expected ore total. While 150K+20% would be 180K ore. This is what i call diminishing return.

It also doesn't work this way if you have 100K ore on the map and you place 10 miners on it, and you give them all 2 PM3 modules. You use 20 modules total, but only raise the quantity by 20%

Again you fail to read what I write.

As for the modules being less effective with mining bonuses that is only when seen as a percentage of the total. In ore counts it remains 20% of the base mining speed. You always get the same amount of extra ore from module.

For 100k ore patch you get 20% more ore == 20K. That is totally expected. Productivity bonus is additive and not multiplactive. Othwerwise it would be 100k + 1 module = 110k, 110k + second module = 132k. Or 100k + first 10% bonus = 110k, 110k + second 10% bonus = 132k and so on.

No matter what productivity bonus (or how many miners) a 100k ore patch will always give you 20k extra ore due to the prod modules. And those 20k you can diver towards building more modules without loosing anything.

But productivity modules are multiplative across recipe chains. So it's 120% ore from the miners. Multiply by 120% for plates = 144%. Multiply by 140% for copper cables = 2.02. Multiply by 140% for electronic circuits = 2.82. Multiply by 140% for advanced circuits = 3.95. Multiply by 140% for processing units = 5.53. If you divert 80% of the produced processing units (blue cards) then you still get 10% more blue cards out of it going by copper. Only 80% going by iron. So more or less breaking even with the resource needing a bit more iron and less copper.

Suddenly diverting 80% to module production doesn't seem so insane anymore, does it?