Prod3 in mid/late-game science-pack production?

[someone1337]

Am I the only one, who considers using Prod3 Modules in science pack production before launching the first few rockets?
At least it seems that noone else seems to use prod3 mods in science, at least noone, who posts their setups online at the stage of still having belts and only at around 1 science/sec.

So my question is: why not?
Is it just not viable - which at least for purple and yellow science seems illogical, as prod3 should be able to save quite a few input resources...
Or does just noone posts belted mid/late-game 1 science/sec setups?
Or am I just not able to google such setups?

[zOldBulldog]

I don't know about others, but I have yet to feel like I need any faster science. It is usually the other stuff, like scaling things up, that end up being my bottleneck.

But then, I have been using Nilaus 'Base in a Book's designs for science lines and they are all amazingly well balanced.

[Hedning1390]

Productivity 3 is very expensive. Up to 20 p2 modules can be used in the labs without you having to pay extra (you need 20 for the p3 in the rocket silo), so turning them into p3 is a lot of investment for a small benefit. The time between when you can build modules and you have finished the rocket silo research isn't that great. Especially if you consider the P3 research it is right before the rocket silo, so you may just have a few rocket shooting speed research left. Not much time to get back the investment.

That said a lot of people do delay their first rocket and bluild lots of modules. So it's probably a case of you not being able to google. I know katherineofsky made a "megabase" series where she moduled up pretty much everything before even starting to build the rockets.

As for why they aren't in blueprints posted online that might just be a case of most posted designs being bad. Take a tip from me and don't bother with other people's blueprints. Build your own.

[Aeternus]

It largely depends on what you want to do. Personally I do prefer to have the higher tier production assigned PM3's before going with rockets. Stuff like low density structure, rocket control mods, blue + yellow + purple science and of course the rocket platform itself all have expensive production, so you're saving a lot of materials by prod-modding those. Before white science you'll be going through iron and copper mines fairly fast.

When transitioning to a true megabase I prefer to prod-mod everything but the kitchen sink (since i don't need 140% water) to keep material flow to a minimum. And produce looooots of stinky stinky pollution. Only thing not prod-modded are mines, once you get the higher mining productivity bonuses the difference those modules make is negligable.

[Hedning1390]

once you get the higher mining productivity bonuses the difference those modules make is negligable.

Once you get high mining productivity the effective speed is still high, so you might as well put in some productivity modules.

[Frightning]

once you get the higher mining productivity bonuses the difference those modules make is negligable.

Once you get high mining productivity the effective speed is still high, so you might as well put in some productivity modules.

Except the prod bonus from the modules is additive with the one from mining productivity research, not multiplicative. Meaning at a certain point, the modules will decrease resource production rate for almost no gain in resource yield (+30% prod when you already have like +150% prod isn't much of a gain in yield, and losing 45% speed for it is a big deal, not to mention prod modules are resource intensive and result in much higher energy cost and pollution output).

[MeduSalem]

once you get the higher mining productivity bonuses the difference those modules make is negligable.

Once you get high mining productivity the effective speed is still high, so you might as well put in some productivity modules.

Except the prod bonus from the modules is additive with the one from mining productivity research, not multiplicative. Meaning at a certain point, the modules will decrease resource production rate for almost no gain in resource yield (+30% prod when you already have like +150% prod isn't much of a gain in yield, and losing 45% speed for it is a big deal, not to mention prod modules are resource intensive and result in much higher energy cost and pollution output).

Exactly. Electric Mining Drills are the only reasonable machines to put in some cheap Efficiency Modules 1 ... to reduce energy consumption and the likeliness of Biters wanting to chew through outposts.


For everything else I say PM3/SM3-becconize the crap out of it... to cut down on the resource consumption and overall amount of infrastructure necessary.

I start with that as soon as possible... beginning at the Labs, then the Science Packs, then their ingredients, and so on until reaching the smelters. Using it stage by stage I free up more and more resources to be dumped into Circuit making, which results in faster module crafting in parallel to constantly growing Science output. Usually I have enough modules lying around by the time I am ready for the space science pack.

[Hedning1390]

once you get the higher mining productivity bonuses the difference those modules make is negligable.

Once you get high mining productivity the effective speed is still high, so you might as well put in some productivity modules.

Except the prod bonus from the modules is additive with the one from mining productivity research, not multiplicative. Meaning at a certain point, the modules will decrease resource production rate for almost no gain in resource yield (+30% prod when you already have like +150% prod isn't much of a gain in yield, and losing 45% speed for it is a big deal, not to mention prod modules are resource intensive and result in much higher energy cost and pollution output).

Except I was talking about high mining productivity. Benefit isn't great, but neither is cost:
-Lower speed is 0 cost if the alternative is 45% of your miners standing idle because the belts are saturated with just a few of them working anyway (yes I know people usually bot mine at that level).
-Power for miners is also a lower and lower fraction of your total factory power cost the higher mining productivity you get.
-Cost in resources is also negligible. 10k of all science packs is more expensive than 1k modules and to have reached high mining productivity you would have made many 10k of science packs as well as many more modules for the rest of your factory. As a fraction of your overall spending it is not noticeable.
-Pollution? Please...

No, the only actual cost is the extra effort of lugging the modules to the build site and inserting them in the miners since they are effectively quadrupling the inventory space demand of the miners. And considering effort of bot v belt mining I can plonk down a general belt blueprint that is guaranteed to claim the entire field, while bot mining is not, although bot mining has free construction bot infrastructure which is a great bonus. Inventory wise bots and belts take similar space, so in total it is a tossup imo.

[Aeternus]

once you get the higher mining productivity bonuses the difference those modules make is negligable.

Once you get high mining productivity the effective speed is still high, so you might as well put in some productivity modules.

Except the prod bonus from the modules is additive with the one from mining productivity research, not multiplicative. Meaning at a certain point, the modules will decrease resource production rate for almost no gain in resource yield (+30% prod when you already have like +150% prod isn't much of a gain in yield, and losing 45% speed for it is a big deal, not to mention prod modules are resource intensive and result in much higher energy cost and pollution output).

Except I was talking about high mining productivity. Benefit isn't great, but neither is cost:
-Lower speed is 0 cost if the alternative is 45% of your miners standing idle because the belts are saturated with just a few of them working anyway (yes I know people usually bot mine at that level).
-Power for miners is also a lower and lower fraction of your total factory power cost the higher mining productivity you get.
-Cost in resources is also negligible. 10k of all science packs is more expensive than 1k modules and to have reached high mining productivity you would have made many 10k of science packs as well as many more modules for the rest of your factory. As a fraction of your overall spending it is not noticeable.
-Pollution? Please...

It depends of the size of the mine. A small field that can't scrape out a full train fast enough might benefit from speed modules instead. A huge field can be prodmodded without a production interrupt, but for 30% more productivity when I'm already at 800%? Seems rather pointless. And yea, pollution is a minor concern. Mines are hands down the heaviest base polluters - and pollution too consumes some all-important UPS.

[Hedning1390]

Yes, like I said it's small gain for small cost. I certainly have no consistency when it comes to mining. Some mines have prod modules, some don't and it's not really for any rational reason.

800% is a fun one since that is when you need exactly 1 miner per furnace if you are using full prod modules.

As for pollution I don't know. Does it really matter if it is saturated red anyway? At some point it doesn't really seem to spread much further, just kind of builds up in place like if there was a pollution throughput limit. Maybe it's that with enough spawners in a chunk no amount of pollution can get past it even if the flow is infinite. Like I said I don't really know.

[MeduSalem]

As for pollution I don't know. Does it really matter if it is saturated red anyway? At some point it doesn't really seem to spread much further, just kind of builds up in place like if there was a pollution throughput limit. Maybe it's that with enough spawners in a chunk no amount of pollution can get past it even if the flow is infinite. Like I said I don't really know.

In my experience it does matter. The biters are usually looking for heavy pollution sources... So I'd rather have them attack my mainbase then the outposts.

It is not like you don't need to defend the outposts at all when using efficiency modules, but due to the modules the biters seem to be much less interested in the outposts than they are in the mainbase.

For me that is a considerable convenience factor knowing I have to check up less if the defense of the outposts are still completely intact.


But that said my outposts were never that far enough away from my mainbase that the huge red blob from my mainbase didn't also reach much further than the outposts.

[dragontamer5788]

You should always build 4x Prod3 for the first rocket, under optimal strategy. Its more efficient to build 72 rocket parts + 4 Prod3 modules rather than 100 rocket parts. Math is simple: 28 rocket parts (aka: 280 Rocket Fuel + 280 LDS + 280 Compute Units) is way more expensive than 4x Prod3.

The question therefore: is how to most efficiently use the 4x Prod3 modules before the rocket? 20x Prod2 modules in your labs implies ~10 boosted labs for a good chunk of the game. So I think that's a good idea. Maybe 80x Prod1 modules is better, hard to say really.

20x Prod2 modules will turn into the final 4x Prod3 modules (for the rocket) most quickly. So that's probably the optimal run, it just seems easier to manage. Managing 80x Prod1 would require a lot of clicking and probably would slow down a speedrun.

[BlakeMW]

It is totally worthwhile using prod3 modules in science labs.

Okay here's the thing: Rocket Silo research costs 266k Iron+Copper in terms of the science packs (with no prod modifier)

A tier3 module costs 3.1k Iron+Copper

So if you reduce the cost of Rocket Silo by 1/1.2 then you have freed up 44k, that would buy 14 tier3 modules or enough to prod3 mod 7 labs which is actually quite a lot of research power once speed-beacons are applied.

Now it does get more complex if you want to complete the research quickly and when you're using prod modules in lower tiers the overall cost reduces and lower tier prod modules have a quicker ROI - at least until you can do some serious speed beacon boosting - but overall it's not a question of if you should use prod3 modules, but how you should use prod3 modules in conjunction with prod2 and prod1 modules.

2x prod1 modules in assembler 2 is a very good strategy with quick ROI on many recipes, easily quick enough to make sense in a speedrun even if you never pull the modules out. So this is actually the baseline which should be compared against. Then you should upgrade high throughput recipes with higher tier modules boosted by speed beacons (you must have speed beacons, though they can use lower tier speed modules). Green circuits, blue circuits, production science packs, high tech packs and labs doing high tech science are the most useful places to put prod3 modules to minimize research cost.

[Kryptos]

Yes, like I said it's small gain for small cost. I certainly have no consistency when it comes to mining. Some mines have prod modules, some don't and it's not really for any rational reason.

800% is a fun one since that is when you need exactly 1 miner per furnace if you are using full prod modules.

As for pollution I don't know. Does it really matter if it is saturated red anyway? At some point it doesn't really seem to spread much further, just kind of builds up in place like if there was a pollution throughput limit. Maybe it's that with enough spawners in a chunk no amount of pollution can get past it even if the flow is infinite. Like I said I don't really know.

I have seen pollution go all the way to the edge of the generated zone. At a certain point, where you're overwhelming the biter's ability to suck it up, you'll see your pollution go offscreen. Then stay there, and build up. Eventually, it will all be red, and it won't matter if you pollute any more or less, for similar reasons to why you don't see a 30% productivity bump against a base +800% mining productivity research.

[Jap2.0]

Yes, like I said it's small gain for small cost. I certainly have no consistency when it comes to mining. Some mines have prod modules, some don't and it's not really for any rational reason.

800% is a fun one since that is when you need exactly 1 miner per furnace if you are using full prod modules.

As for pollution I don't know. Does it really matter if it is saturated red anyway? At some point it doesn't really seem to spread much further, just kind of builds up in place like if there was a pollution throughput limit. Maybe it's that with enough spawners in a chunk no amount of pollution can get past it even if the flow is infinite. Like I said I don't really know.

I have seen pollution go all the way to the edge of the generated zone. At a certain point, where you're overwhelming the biter's ability to suck it up, you'll see your pollution go offscreen. Then stay there, and build up. Eventually, it will all be red, and it won't matter if you pollute any more or less, for similar reasons to why you don't see a 30% productivity bump against a base +800% mining productivity research.

No, it'll generate more land and spread offscreen.

[bobucles]

Yep. Pollution will spawn more chunks as it spreads into unscouted tiles. Land absorption is a very tiny factor, so pollution will spread until it hits a forest or more biter nests.