Productivity modules: where do you draw the line?

[greep]

So, I've been using productivity modules more and more as I've played the game, but there's certain breakpoints where they start to get less and less useful. At the furnace level they don't give much resources back, and at the mine level, they absolutely explode pollution while also not giving much back. Personally, I use them in everything but mines. I'm curious how much other people use them.

[TN_Creator]

I'm not yet very experienced in the game (90hrs) but I never used them. If I need to boost some production, I use speed modules, as their behavior is simpler.

[Neotix]

Setup low resources in map generation and you will bless productivity modules.

[BlakeMW]

I pretty much always use them in green circuits, gears, blue circuits and everything oil related. If I'm doing a logistics build I use them in every furnace/assembler/chemical plant which will take them. If I'm doing a belts build I might not bother with furnances and low yield recipes like red circuits.

However on Death Worlds I often use them in everything including electric miner drills. Once you have achieved full evolution (and have massive laser defenses) there is no more harm in releasing vast quantities of pollution and expanding to acquire more ore is kind of tedious because you have to beat back the biters and establish new laser walls, so I just make a strip mining setup with prod3 miner drills and beacons:


This does tend to require large ore buffers because miner drills running at +155% speed and +30% productivity put out quite a lot of ore. Using productivity 3 modules without speed beacons to multiply the output is really not an option, the speed beacons make the whole thing a lot more affordable. But the beacons do result in a strip of unmined ore every 11 tiles, even if you don't go back and finish off that strip you still get a +18% productivity boost and if you do go back and mine the strips you can get the full potential output.

[Tnarg]

I don't use them till I get to rocket silos which gets 40% productivity. Then I work back as needed.but never far back up the production chain.

[albatrosv13]

deleted

[saturn7]

I use them (almost) everywhere, together with beacons with speed modules.

And a huge solar farm to power the beast.

[KatherineOfSky]

I tend to only use them in end-game products, like blue chips (Processing units), Rocket control units, etc. They use up so much power & speed, that I often balance them with either a speed module or efficiency module.

I do, however, use them in Labs and purple science production with a beacon around to keep the speed up.

[Xterminator]

I think it depends on your current needs. I find they are good for smelting if your low on ore or for making expensive products like processing units or something. Also great to use in all stages of oil processing if you are low on oil.

However if I do use them, I almost use beacons with Speed Modules to counter the slowness of the productivity ones.

[Archalieus]

On everything that can take productivity modules. Drills, furnaces, assembly machines, everything. At this point, the factory has actually produced more product than there exists raw resource. The other main reason is due to belt maximization and the actual speed limitations on inserters. Without productivity, several assembly machines in the factory can't be emptied fast enough, as the arms can't offload the product fast enough. The solution? More productivity.

I also don't use any speed beacons whatsoever.

[albatrosv13]

deleted

[Reviire]

I use them in everything, but I make sure the craft speed still has a positive bonus, so for furnaces it's beacon(s) with speed modules, with assemblers, 3 productivity + 1 speed

[siggboy]

The only line that should be drawn -- where Productivity Modules are objectively worse than Speed Modules -- is on low yield Oil Wells. Somebody on Reddit has simulated it and it turns out that below ~93% yield it's better to use Speed 3 instead of Productivity 3 in your oil pumps.

When I can't be bothered to craft them in bulk I only use them in very high value products: Rocket Parts (that's the most obvious one), components of Rocket Parts, Flying Robot Frames (and everything on that production chain), Science 3 maybe, Science 4 to save Alien Artifacts.

However, productivity modules early in your production chain (e.g. on furnaces, refineries) have absolutely no effect if your resource needs are covered. If you put out 4k plates per second, but only need 2k per second, another 20% productivity on your furnaces does absolutely nothing for your factory.

I think it's always advised to put them into the super high value products that require a long and expensive production chain in front of them (like Rocket Parts, as mentioned). Mostly not to save the raw resources (ores, oil), but mainly because it amplifies the entire part of the factory required to make the stuff by 40% (if you use 4x Prod 3 in the Rocket Silo). Another ore outpost is found quickly, making 40% more of everything needed to craft Rocket Parts is far more difficult.

[BlakeMW]

A couple of days ago I determined that if you're going to do the crazy thing of putting 12 speed3 beacons around a single depleted oil well, you get significantly more production by putting 2x prod3 modules in the pumpjack than 2x speed3 modules - although naive maths indicates 1x prod3 and 1x speed3 is the best, it's actually the worst. Both in game simulation and more sophisticated maths demonstrate that 12x speed3 beacon + 2x prod3 modules maximizes the output of a depleted oil well.
Not that I actually ever do 12x speed beacons around a single pumpjack, but the claims based on naive maths (and I've seen the same claim several times before) bothered me enough to test it .

[TN_Creator]

I never thought about using speed beacons for depleted oil wells, it will be very usefull as I play with RSO and very few wells! thanks.

So the ideal setup is Productivity in the pumpjack and Speed in the beacons?

[greep]

Nah it's still speed3 all around for pumpjacks, since said above situation doesn't happen in a real game (if you have one oil spot, you're not going to waste 14 modules on one oil spot, and if you have more you can't fit the beacons).

[untrioctium]

I use them in pretty much everything now except miners (though I will throw them in stone miners). The addition of productivity down the chain gets ridiculous; for example, a new T3 productivity module normally costs 2000 copper ore, 1083 iron ore, 190 coal, and 592.5 petroleum gas, but assuming if you have T3s in all of the spots in the chain where it's supported (and assuming my calculation script isn't off), it goes to about 641/463/102/319 on average. The copper savings are drastic, nearly 67%. There is a speed hit, but I normally space my factories and design around places to insert beacons so that a factory is hit by four of them. Rockets are the obvious big savers in this process, going from 77.5k copper ore and 89.1k iron ore to only 14.1k and 26.3k, a savings of 81% and 70% respectively.

I know diminishing returns and all that, but nothing pleases me more than pulling up my production stats and seeing 20% more plates being produced than ore being consumed.

[siggboy]

I use them in pretty much everything now except miners (though I will throw them in stone miners). The addition of productivity down the chain gets ridiculous; for example, a new T3 productivity module normally costs 2000 copper ore, 1083 iron ore, 190 coal, and 592.5 petroleum gas, but assuming if you have T3s in all of the spots in the chain where it's supported (and assuming my calculation script isn't off), it goes to about 641/463/102/319 on average.

This is true, and it's due to the fact that the production bonus is multiplicative, a.k.a. exponential. The longer your production chain, the crazier it gets. Like compounding returns when you invest money (which is why the smart money is on ETFs that charge 0.1% fee instead of managed funds that charge 2%, because 2% doesn't sound like much but when you leave the money in the respective funds for 20 years, the difference becomes really big).

However, keep in mind that even if you save 1.4k copper ore when making a single T3 module, it's not really that much in the context of a large scale base; and that's not even taking into account that the modules themselves have to amortize as well.

Let's say your smelting setup produces 10k plates per minute; that's about 290 furnaces. Putting 1x Productivity 3 into all of them will increase production by 1k per minute, saving the same amount of ore. However it costs you about 1.5k * 290 = 435k materials (ores + petroleum) to make all of these modules IF you already have the full discount (from even more modules) in the entire production chain! After 435 minutes (over 7 hours) you still haven't earned back your initial investment, not even talking about all the other modules that are unmentioned in my example...

Getting 1k ore/minute (the amount you saved in the example above) requires not even a medium size quarry. That might be a big deal in the early game, but not for a base that already puts out 10k plates per minute for as long as 7 hours, which is the time it takes to amortize the initial investment.

I'll take yet another stab at this:

A furnace puts out ~35 ore/minute. A prod 3 modules costs ~3500 ore (very roughly, lumping all required materials together), completely disregarding the infrastructure and energy required. So your furnace is working for 100 minutes to produce the material required for the prod 3 module that you just inserted. Practically, you're disabling the furnace for 100 minutes (it's still consuming ore and energy during that time) to get a ~3.5 ore/minute bonus after that. Does that sound like a good deal to you? It's a very long term investment that only makes sense in large scale operations that run for a good while.

[BlakeMW]

Putting 1x Productivity 3 into all of them will increase production by 1k per minute..

Thing is that's not what you do, you use alternating rows of furnaces and speed beacons, this multiplies the effect of each productivity module as it is in a furnace running at +200% speed, this dramatically reduces the time to amortize (it's kind of like the difference between a machine running 25% of the time, vs 100% of the time - the investment doing more per second pays off faster). The speed beacons themselves are pretty cheap, you can actually use speed2 to get most the benefits, but even speed3 are pretty cheap when their benefit is being applied at 50% to 8 machines. I think the time to recoup the investment ends up something like 1 hour or maybe 2, in any case nothing to a factory which is going to run for 20+ hours.

[siggboy]

Blake, not doing the detail math on this now, as I'm too lazy (and probably too dumb). In principle you're right of course, but my example didn't feature beacons.

If you add beacons to the setup, you have yet another array of modules that need to amortize as well, plus the substantial amount of MW you need for the beacons (Al Gore would NOT approve of the use of beacons, I'm sure ).

Also I don't agree when you say that the effect of the productivity module is "multiplied". 1x Prod 3 saves you 10% ore, no matter how fast the machine runs. That part of the effect is not multiplied at all, it's always 10%. The amortization occurs faster, of course.

Anyways, it's a good point that you made and I admit that I didn't think of beacons when I wrote up my example.