why prod modules are important

[Khagan]

With modules, you can also boot strap them, which one frequently does, even in speed runs. Ie, you place PM1, and as production / tech ramps up, you replace them with PM2 (reusing the PM1s of course), and then finally PM3s. Upgrade BPs FTW.

Yes, and even if you don't actually do it this way, the cost-benefit calculations should usually assume that you were. That is, instead of calculating the extra production and cost of PM3 compared to nothing, compare them to PM2; similarly, for PM2, assume you already have PM1.

Maybe there is something to be said for efficiency modules in miners. 2 efficiency module 2 in each miner and you cut the energy and pollution down to 20%.

3 efficiency 1 does the same job for a fraction of the cost. EM1 are not silly if you really want to keep pollution down, and since mining patches are often on the edges of defended territory they are a particularly desirable place to do so. But even if you also want efficiency for the power savings (presumably because you are not using nuclear power) higher level EMs are just not worth the cost.

[aka13]

As for building new mines you should start up the map editor, put down a huge patch of ore and design a big mining outpost with rail station and defenses and all the bells an whistles included. Then take a blueprint of that. Go back to the game and you can establish a new mine with one click + connecting the railroad, build the train station and place the first roboport. Easily done with a personal roboport, a roboport in a spidertron or a roboport in a train if you have a mod for that active. Automation is the name of the game.

Still annoying work but placing the actual miners is the least of the problem. If you build every mining outpost to match the ore patch then you are wasting a lot of time doing boring stuff.

Certainly, that speeds things up. I have went further, and only build mining stations "inline" into my rail segments, so that I do not need to clear any land or trees, or lay landfill for that.
Nevertheless, it's still either "send a spidertron, wait 5 years, send a train, wait for spidertron to arrive, work with cumbersome and small radar window, move spidertron around, also cumbersome, all conveyor between mine-smelter-station also more cumbersome, because you only can work with either blueprints, or copy paste, none of the nice things you get when laying them yourself. Also, don't forget - you have to click the whole route to base for the spidertron with waypoints by hand.


99% of the time setting up the outpost yourself + the supply train will be faster and more "fun" if you will. Less cumbersome limitations, faster interaction, etc.

So yeah, usually I have more prod3 than I can use, and not enough outposts I should be using. I know, that there are a lot of great mods to mitigate that, AAI vehicles, as stated before, for example, but I am nurturing this factory to be the perfect start for the expansion.

[mrvn]

Starting again with one miner and a single prod 3 module, that miner has an effective speed of (1 + 10%) * (1 - 15%) = 0,935. To compensate we need to reach an effective speed of 1,1 (i.e. base mining speed plus prod bonus), thus add 0,165 effective speed, which is 0,165 / 0,935 ~ 0,1764 extra miners.

This seems a bit backwards and unrealistic. You have the resources for build a PM3 module but you didn't build enough miners to cover all of the ore patch yet? But your math seems reasonable.

Of course this is more theoretical than practical.
Sometimes the mathematicial approach seems backwards at first glance, but if we want to compare various setups, they must be consistent in some metric, for which I chose rate of output. Otherwise we're comparing apples and oranges. And even if your ore fields are covered completely, you may need an estimate on how many new ore fields you need.

I was about to argue about potential energy savings using this setup, but 90kW * (1 + 3 * 80%) * (1 + 81,82%) ~ 556kW is more than the energy consumption of the beaconed setup from my earlier post which uses 90kW * (1 + 3 * 80% + 3 * 60% / 2) + 2 * 480kW / 8 = 507kW per miner and there's still room to add one efficiency module of you are so inclined.

In conclusion, the more miners variant uses roughly 10% more energy than the beaconed variant. Maybe not so theoretical after all, if there are things you'd trade this energy inefficiency for...

Wow, I didn't realize that beacon setups aren't just cheaper for the same output/s but also save energy when compared against the same use of productivity modules.

As for the efficiency module: If you use nuclear power is trivial to make. If you use solar then power costs time to place the panels. So it's a question whether you rather want to place boring solar panels or boring mining outposts. Personally I think I have never used an efficiency module. But seeing the pollution numbers of mining drills in detail I might try it out for comparison in some mining outposts and compare the amount of bitter attacks with and without. Next time I play with aliens ...

[mmmPI]

That's not what the math mean, There was a comparaison made between 2 different specific setup where one repay itself faster , it's completly different, on a limited amount of ore on the map you will have to change your mining drill more often and potentially run out of ore before you can repay them. Which is already a requirement when you have a beacon setup. And is made even worse when you have a design with beacon AND overlapping mining drill.

How often does it occur that all the mining drill end up depleted at the same time ? never you always have some working and some not working due to depletion or output ful, you also have to acount for beacon not always able to provide their full bonus. You can easily account for 5 to 10% of losses there which is already the difference in energy between 2 setups using the same SPEED BONUS from beacon.

If you account for the design made by Nidan, which is either 2prod3 and 1 speed2 in a mining drill , or 2 beacon with 3 speed 2 and 3 pm3 in the mining drill, then you have a terribly inefficient setup (sorry Nidan) but if you use 2 beacon you place in them 4 modules not just 3, as it's way cheaper to place a lower level of module rathen than a higher level of module. Nidan's design are illustrative of 2 different setup made to have the same speed, not illustrative of how beacons setups do vs non beacon setup.

As suggested by Nidan one can add an effiency module, or a speed module which seem to be the prefered option from previous user. In this case one has to add more effect than the only 3 speed 2 module in Nidan's test setup.


Concluding that "beacon setups are cheaper" is wrong not only because the comparaison is only between 2 different setup and not at all beacons setup in general . But also because you may need more ore total to repay for the cost depending on which module you use and which setup you use, mining productivity bonus are only generated on raw ore from the map. They may or may not repay for themselves before you run out not or can't replace them fast enough.

The correct wording would be : between setup 1 and setup 2 one repay itself faster providing you change your mining drill more often, the one.the one that repays for itself in 6h52 has beacon the other one 7h40 min or so. The difference in energy cost is debatable as one would require to remove every beacon that power depleted mining drill immediatly to achieve a net gain.

Any different combinaison of beacon / module would give a different answer.

[jodokus31]

Maybe there is something to be said for efficiency modules in miners. 2 efficiency module 2 in each miner and you cut the energy and pollution down to 20%.

3 efficiency 1 does the same job for a fraction of the cost. EM1 are not silly if you really want to keep pollution down, and since mining patches are often on the edges of defended territory they are a particularly desirable place to do so. But even if you also want efficiency for the power savings (presumably because you are not using nuclear power) higher level EMs are just not worth the cost.

Efficiency modules 1 are even considered OP for Deathworlds by some people. At least in the beginning, where aliens are a real threat.
I even created a mod to rebalance that for my own challenge to make it less a no-brainer:
https://mods.factorio.com/mod/spice-rack-pollution

But, it's a certain defensive playstyle, which not all people prefer.

[blazespinnaker]

Doing DW without eff modules is real craziness.

Some guy did a speed run, but he pre-selected the perfect map for it. Which is way cheating when it comes to DW, imho.

Krastorio has this cool machine which sucks up pollution. I also like what they do with trees, but the nuclear thing is irritating and the mod seems like just more recipies and not cohesive like SE.

[mrvn]

I really miss that I can't plant trees in vanilla. After the start wood becomes just annoying. Burning it causes pollution which you might want to avoid. And you don't need that many wooden poles or chests. Would be nice if trees could be transplanted or even more created where I want them.

[Danjen]

What I'd love to see is combat modules for turrets. Eg something that reduces attack speed but increases damage, or shoots faster and reduces damage, etc. It might make the game feel more like tower defence, but being able to specialize turrets a bit more is a fun idea (and I'm pretty sure some mods already do this)

[mrvn]

What I'd love to see is combat modules for turrets. Eg something that reduces attack speed but increases damage, or shoots faster and reduces damage, etc. It might make the game feel more like tower defence, but being able to specialize turrets a bit more is a fun idea (and I'm pretty sure some mods already do this)

I can't remember what mod it was but it added sniper turrets. Slow repetition but greater range. And research to increase range. Because you have to stay ahead of the evolution of worms increasing their range and killing your turret creep.

[FuryoftheStars]

What I'd love to see is combat modules for turrets. Eg something that reduces attack speed but increases damage, or shoots faster and reduces damage, etc. It might make the game feel more like tower defence, but being able to specialize turrets a bit more is a fun idea (and I'm pretty sure some mods already do this)

I can't remember what mod it was but it added sniper turrets. Slow repetition but greater range. And research to increase range. Because you have to stay ahead of the evolution of worms increasing their range and killing your turret creep.

Long Range Turret by Yehn (there’s another one out there that appears to be someone’s personal edit of Yehn’s, but doesn’t give credit or explain the differences).

[Jane23s]

I'd like to add some mod for turrets to reduce damage or to make shots faster.

[mrvn]

I'd like to add some mod for turrets to reduce damage or to make shots faster.

Love modules. Like productivity but every time the bar reaches 100% a bullet is replaced by a poppy flower.

[mmmPI]

I'd like to add some mod for turrets to reduce damage or to make shots faster.

Do you mean module or mod like from the mod portal ?

module for turret could definitly be more useful than productivity module in mining drill as they would spare ammunition which are composed of multiple ore compared to placing them in mining drill which is the worst place for productivity module.

[joonazan]

I'm surprised that no one in this thread compared putting speed modules in miners vs productivity modules.

The following calculations are assuming that miners are moved to a new patch when the old one depletes. This does ignore the benefit that the patches last longer with prod modules but that is pretty insignificant. Even when resources are scarce, do you really care about getting 15% more out of a patch?

First, prod modules alone never pay themselves back, as they reduce the output of the mine. This has been "fixed" with speed beacons, so let's use twelve beacons because while unrealistic, that is the most favorable situation for productivity modules.

The beacons give a speed bonus of + 6. We can choose between + 1.5 speed or - 0.45 speed and + 0.3 productivity. The benefit of the latter is that there is already so much speed that the added speed barely makes a difference. And indeed we find that the speed moduled miner outputs 8.5 times its normal output vs the 8.515 of the productivity moduled miner.

However, the picture changes with mining productivity research. How much research is needed for speed to be better than prod?

Let r be 1 + the mining productivity from research.
By solving when (7+ 1.5) * r = (7 - 0.45) * (r + 0.3), we can figure out the tech level at which speed is equally good as prod.
The result is 1.00769, or less than 1% of mining productivity research.

And indeed, with a typical 30% mining productivity, the speed side is 11.05 and the prod side is 10.48. Also note that prod3 consumes more energy and pollutes more.

How do you find the machine where you should insert productivity modules? Resource flow. Put the modules in the machine that has the biggest recipe cost divided by crafting time.

[joonazan]

An interesting corollary of my previous post is that prod modules are better for speed than speed modules when beacons give a speed bonus of 6 or more. For non-mining recipes, that is. This explains why serious builds put them everywhere. One could say that they should be nerfed.

[mmmPI]

I'm surprised that no one in this thread compared putting speed modules in miners vs productivity modules.

The following calculations are assuming that miners are moved to a new patch when the old one depletes. This does ignore the benefit that the patches last longer with prod modules but that is pretty insignificant. Even when resources are scarce, do you really care about getting 15% more out of a patch?

First, prod modules alone never pay themselves back, as they reduce the output of the mine. This has been "fixed" with speed beacons, so let's use twelve beacons because while unrealistic, that is the most favorable situation for productivity modules.

Part of the previous discussion concluded with the obvious fact that speed beaconed mining drill setup are faster than non-beacon setup given the same PM3 in a mining drill which makes the first one faster than the 2nd one to repay for itself counting only the "free ore from productivity bonus" , not the most useful conclusion i agree

However regarding the previous discussion it is not correct to say "prod module alone never pay themselves back", they do despite lowering the output of the mine, this was (imo) a "twisted" metric introduced previously, where you only count the extra ressources generated from the productivity module as the ressources that can repay the cost of the module. It makes total sense then to use 12 beacons i agree on your conclusion, if you are to follow the "twisted" metric, why not doing it properly

The previous discussion also tried to show that Producitivity module 3 in mining drill are a waste compared to using ressources to research the mining productivity bonus. This 'maybe' unless you are having a very hard time finding new ore patch which could make one player eager to get a bit less than 30% more ressources out of the ore patch. ( but more than 15 %?) or if you have a very low number of mining drill like 1 mining drill on the extreme case, then the cost of the 3 productivy module is cheap compared to the cost of 1 level of mining productivity research ( but also in this case you'd still be researching mining productivity bonus anyway).


the actual amount of "more ressources" is not so straightforward to compare as one would need to subtract the cost of the modules to the total amount of ore in the patch, which is not the same if you are re-using the same modules in several patches overtime . If the ore yield is very low, one could be required to move the mining drill several time before being able to repay for the cost of the modules but if you re-use the same modules many times on rich yield ore patch, then the cost trend towards 0. But in this case there is still another calculation to be made with the amount of "more resssources" one would have gotten from the mining productivity research. This is also difficult to math out because in a real game you can do 0 or 1 or 2 or 3 or whatever number of research without changing mining drill, which then means that the amount of "more ressource" you get from doing the research depend on the number of drill you have. The more mining drill at the same time = the less amount of ore per patch consumed per research = the more "free ressources" for each patch overtime.

( to illustrate the relations if you trend everything toward very high made up number to the point where 1 level of mining productivity research only require 1% of the ore in the ore patch, and if you keep doubling the amount of mining drill between each research to keep this at 1% despite increase in cost for research, at the end of the life cycle of 1 mining drill when it's depleted, the last 1% of raw ore from the map would have been mined with a productivity bonus from research that would be an additionnal 99 % or so compared to the the 1rst 1% of raw ore mined on the patch)

I think it then becomes natural to compare productivty module versus speed module i agree they are much better if you are discarding the component of "how much time is required to drain an ore patch" /"how often human intervention is required" which is the only reason why one would use productivy module in mining drill in the first place; which then makes no sense to use speed beacon like it was done in the previous discussion, i hope i'm not the only one noticing

The beacons give a speed bonus of + 6. We can choose between + 1.5 speed or - 0.45 speed and + 0.3 productivity. The benefit of the latter is that there is already so much speed that the added speed barely makes a difference. And indeed we find that the speed moduled miner outputs 8.5 times its normal output vs the 8.515 of the productivity moduled miner.

However, the picture changes with mining productivity research. How much research is needed for speed to be better than prod?

Let r be 1 + the mining productivity from research.
By solving when (7+ 1.5) * r = (7 - 0.45) * (r + 0.3), we can figure out the tech level at which speed is equally good as prod.
The result is 1.00769, or less than 1% of mining productivity research.

Those math would show why SPEED 3 modules are used in build optimized for UPS, because those do mining drill=>train wagon direct insertion,trying to minimize the number of entities thus seeking for higher ouput per mining drill to require less of them running at the same time if understand correctly. "how much ore per second is produced per mining drill ?" comparing the ouput of 2 maxed in speed3-beacon, one with productivity module in the mining drill, one with speed module in the mining drill and then asking when is the productivity research enough to make one better than the other ?

I'm saying this because i'm not sure it's possible to fit 12 beacon per mining drill using train wagon direct insertion, following your math reasonning, if we were to use less beacon, say only up to +4 speed then the solving would be when ( 5+1.5)* r = ( 5-0.45 ) * ( r + 0.3) with a result of around 0.7 ( https://www.wolframalpha.com/input?i=%2 ... r%2B0.3%29 ) i'm not quite sure what to make from that it shows that if you use less beacon then you need even less research for speed module to be better than productivity module 3 ?

Although i agree with your final conclusion that speed module 3 are better than productivty module 3 when placed in mining drill and looking at the max output per mining drill as metric i think there is an error in the way the problem is made into equation. because if we solve it with no beacon at all we find around 0.0085 (https://www.wolframalpha.com/input?i=%2 ... %2B+0.3%29) which contradict the 8.5 vs 8.515 calculation.

[mrvn]

I'm surprised that no one in this thread compared putting speed modules in miners vs productivity modules.

The following calculations are assuming that miners are moved to a new patch when the old one depletes. This does ignore the benefit that the patches last longer with prod modules but that is pretty insignificant. Even when resources are scarce, do you really care about getting 15% more out of a patch?

First, prod modules alone never pay themselves back, as they reduce the output of the mine. This has been "fixed" with speed beacons, so let's use twelve beacons because while unrealistic, that is the most favorable situation for productivity modules.

The beacons give a speed bonus of + 6. We can choose between + 1.5 speed or - 0.45 speed and + 0.3 productivity. The benefit of the latter is that there is already so much speed that the added speed barely makes a difference. And indeed we find that the speed moduled miner outputs 8.5 times its normal output vs the 8.515 of the productivity moduled miner.

However, the picture changes with mining productivity research. How much research is needed for speed to be better than prod?

Let r be 1 + the mining productivity from research.
By solving when (7+ 1.5) * r = (7 - 0.45) * (r + 0.3), we can figure out the tech level at which speed is equally good as prod.
The result is 1.00769, or less than 1% of mining productivity research.

And indeed, with a typical 30% mining productivity, the speed side is 11.05 and the prod side is 10.48. Also note that prod3 consumes more energy and pollutes more.

How do you find the machine where you should insert productivity modules? Resource flow. Put the modules in the machine that has the biggest recipe cost divided by crafting time.

Especially when resources are scarce people care to get 15% more out of a patch. If you have a million bucks you might not worry about an extra 150k. But if you only have 100 bucks an extra 15 will be huge. The difference between paying the rent and having some money left over for food.


Two things you are forgetting in you math:

1) Speed beacons don't give you more ore, they only give you ore faster (and deplete the patch that much faster). If you have the ore patches available then just building more mining drills is cheaper. And overall the amount of ore you get is the same. So they never pay for themself no matter what. They only become a cost saving when they help you save prod modules. Because speed beacons are cheaper than more miners with prod modules.

Note: Same goes for furnaces or assemblers. Cost wise building more of them is cheaper than speed beacons. That only changes with prod modules.

2) The difference between prod modules and mining research is that a) prod modules reduce speed (which you counter with speed modules), b) prod modules are per miner and c) mining research costs more for each level.

By using prod modules you can save 3 levels of mining research. For a given amount of miners there will always be a level at which the mining research is more expensive than the prod+speed modules. Or for a given mining research level there is a number where prod modules are cheaper if you have less miners than that and research is cheaper if you have more miners. That number just keeps on rising with each research level.

[mmmPI]

Another significant drawback of the prod modules in mining drill is that it increase the pollution per ore by a lot contrary to the mining productivity research.

There is no alternative for productivity in assembly machines, chemplants or refinieries than productivity module while there is a superior alternative for mining drill productivity, doing the research and using no module, or using speed module, or using 2 efficiency 1 module

[SoShootMe]

Another significant drawback of the prod modules in mining drill is that it increase the pollution per ore by a lot contrary to the mining productivity research.

True, but producing the science packs in order to research mining productivity bonus also creates (quite a lot of) pollution, and after Mining productivity 1 they are fairly/very expensive technologies.

[mmmPI]

Another significant drawback of the prod modules in mining drill is that it increase the pollution per ore by a lot contrary to the mining productivity research.

True, but producing the science packs in order to research mining productivity bonus also creates (quite a lot of) pollution, and after Mining productivity 1 they are fairly/very expensive technologies.

It would create even more pollution if one had placed PM3 in its mining drill while doing the research !

The base pollution of a mining drill is much higher than the base production of most other machines, ( except boilers 30/m ) as it is 10/m base for an electric mining drill, vs 2/M for assembly lvl 3, Oil refinery is 6/M, chemical plant 4/m. No pollution on labs, 1/m for electric furnace, 4/m for steel furnace.

So placing productivity module there ( in mining drill) is where it will generate the most extra pollution. while it could be avoided with only 3 level of research, which cost less than 300 prod modules. If you plan to never have more than 100 miners at the same time maybe doing 3 level of research is more costly, otherwise it's better to do 3 level of research as it would require less ore, each responsible for less pollution