Tracking Issue: Recipe balance megathread.

[notnotmelon]

What is the problem?

This tracking issue exists in order to resolve the following three common complaints when navigating through recipe chains in pY:

"Why does this recipe exist? I would never use this."

or

"This recipe is too strong, I would never use the alternatives"

or

"This TURD upgrade is too weak. Who would choose this?"

Please list the following:

• What is the internal name of the recipe/TURD path? You can find this by hovering the recipe and pressing ctrl+shift+f
• Why do you feel that it is underpowered/overpowered?
• What suggestion would you make to fix the issue?

Reports are intended to be concise. If you find yourself writing an essay, it may be more useful to open a dedicated bug ticket.

[BeaulieuT]

Fawogae recipe using manure seems too expensive to ever justify...fawogae is literally free other than space, and since it stacks to only 50, I always preferred to make it on site instead of shipping. Sometimes I wanted to make it in bulk, but the manure cost seemed expensive for the small extra gains.

I don't remember how much fawogae you get off the top of my head, so maybe it's actually fine and i went too ZI crazy. But it would be nice to see either better output for the manure, or maybe a different fawogae recipe for more options since fawoge is used so much.

[sunrosa]

I don't think you can make one item spoil into multiple items, so the manure would have to spoil into an intermediary, wouldn't it?

[protocol-1903]

Moondrop Co2 TURD is powerful, perhaps too powerful. Making it a moondrop consuming recipe instead would make it more involved and less free, while also allowing for a "sink" for moondrops

[notnotmelon]

Moondrop Co2 TURD is powerful, perhaps too powerful. Making it a moondrop consuming recipe instead would make it more involved and less free, while also allowing for a "sink" for moondrops

I imagine there will be pitchforks if this change is implemented.

[chunkybanana]

Every tholins-to-gas recipe is absurdly underpowered, even deuterium. I could see three interesting ways to fix this:

• Add some unique products only primarily extractible from tholins (maybe from others as a byproduct) - does require significantly changing chains though
• instead of having tholins->single gas recipes, have one/a few tholins->multiple gases recipes, which while quite powerful if using all the gas outputs, are generally worse than regular production chains if you're voiding most of the outputs - i.e. good but only usable as a supplement
• Add more advanced processing recipes for some gases/products that double output at the cost of a small amount of tholins

all of these are going to be a bit tricky to balance, and all three could be involved in a solution.

[EigensheepLambda]

You can throw all/nearly all of the phagnot gas bladder recipes into the pile as well. Not as underpowered as the tholins stuff, but still pretty underpowered. Kill two birds with one stone and have competitive gas-producing recipes that require both somehow? Feed the phagnots tholins to get the bladders in the first place, or require bladders to "process" the tholins (possibly with byproduct gasses, as chunkybanana mentioned, since it's now a gross alien biology process)?

[EigensheepLambda]

The higher tier coal processing chains are terrible in the sense that they can't compete with the basic crushing process. Like quartz, they could produce mandatory intermediate products (purified coal?) that also can produce huge amounts of coal or other useful carbon-y things. Or, they could just be pruned from the tech tree.

[chunkybanana]

Moondrop Co2 TURD is powerful, perhaps too powerful. Making it a moondrop consuming recipe instead would make it more involved and less free, while also allowing for a "sink" for moondrops

I imagine there will be pitchforks if this change is implemented.

I agree that making it not easily ZI would basically remove all of its benefits. Doesn't mean it can't be nerfed, or at least made more interesting though. One thing I thought of was making it output trace amounts of some other (justifiably "extracted") solid that can't as easily be voided, idk what would be both somewhat justifiable and work well balance-wise though, especially considering it's unlocked potentially before the player has green circuits. Alternatively it could be made to require water barrels(and output barrels), although that might be too annoying.

Could also be moved to py science 1 to facilitate this, although that might also annoy people.

[protocol-1903]

Moondrop Co2 TURD is powerful, perhaps too powerful. Making it a moondrop consuming recipe instead would make it more involved and less free, while also allowing for a "sink" for moondrops

I imagine there will be pitchforks if this change is implemented.

I agree that making it not easily ZI would basically remove all of its benefits. Doesn't mean it can't be nerfed, or at least made more interesting though. One thing I thought of was making it output trace amounts of some other (justifiably "extracted") solid that can't as easily be voided, idk what would be both somewhat justifiable and work well balance-wise though, especially considering it's unlocked potentially before the player has green circuits. Alternatively it could be made to require water barrels(and output barrels), although that might be too annoying.

Could also be moved to py science 1 to facilitate this, although that might also annoy people.

Moving it to Py1 or later would be a good solution. Its just way too powerful to get early game so theres no consideration for the alterantive options. I dont know of anyone that picks the other options other than a challenge run, which isnt how it should be.

[notnotmelon]

Have you seen the new buffs to moondrops paths 1 and 2? They are significantly stronger compared to 1.1

[BeaulieuT]

I personally really love the co2 moondrop turd, its one of my absolute favorites. That being said I'm not opposed to it being changed, but I don't know if it needs a nerf right now. I wouldn't say it's too strong, as once you get co2 from biomass, co2 is easy to get ZI in a lot of ways. I think it's prety strong early game, and then after co2 from biomass it's just a very nice QoL / build simplifying TURD. Personally I think it's in a good spot.

I would suggest buffing the other moondrop turds to be more competitive, which it sounds like you guys have already done (haven't looked at the updated ones yet).

[BeaulieuT]

I think that there is room for a higher tier coke recipe that unlocks maybe around chem. Red hot coke is great, but after that there's not really a better coke recipe. Coke was one of my top 3 (maybe even top 1) shipped items, and it would be nice to have a complex route that gives great yield. I saw someone mention that higher tier coal is useless, and I 100% agree. Maybe somehow higher tier coal can also tie into better coke production to give that route a use.

As it currently stands, when I needed more coke, I went to a large coal patch and just spammed the cheapest raw coal -> coke recipe because it was essentially free. Feels a little weird that in py 3 I was using the early game coke recipe because it was kind of the best option.

[protocol-1903]

I think that there is room for a higher tier coke recipe that unlocks maybe around chem. Red hot coke is great, but after that there's not really a better coke recipe. Coke was one of my top 3 (maybe even top 1) shipped items, and it would be nice to have a complex route that gives great yield. I saw someone mention that higher tier coal is useless, and I 100% agree. Maybe somehow higher tier coal can also tie into better coke production to give that route a use.

As it currently stands, when I needed more coke, I went to a large coal patch and just spammed the cheapest raw coal -> coke recipe because it was essentially free. Feels a little weird that in py 3 I was using the early game coke recipe because it was kind of the best option.

Perhaps late game coal recipes should add new alt recipes that supplement early game coal recipes to create better resource chains

[EigensheepLambda]

One of the "improved" doped silicon recipes adds arsenic and it's a trap and should probably have the required amount of arsenic cut down by 90% or something silly like that. (Sorry for being annoying vague in these- haven't migrated my save yet to check internal, or uh, even external recipe names directly.)

[darko-dll]

The most concise answer would be to just install YAFC. Look at random items in Not Enough Items Explorer, many of which have recipes that are <10% efficient compared to the top pick, at which point it's sensible to ask why are they in the game?

One counterexample is plastic, where the "worst" recipe, the initial one with syngas and aromatics, is 20% as efficient as the top one. But I still use it frequently for small amounts if both syngas and aromatic pipes are nearby because of its simplicity. So, perhaps inefficient recipes are okay if they take 1-2 inputs of very common ingredients.

Another aspect that matters to me is realism. Does the recipe make sense, or can it teach us something about real-life industrial processes? Great, I don't mind it being long or complicated. OTOH I'm never going to turn chitin into geothermal water because of how nonsensical it seems.

Sometimes a bad recipe becomes okay with enough productivity modules. Sometimes it's a way to get rid of excesses in a no-void run (especially animal parts).

So anyway,

• Any recipes that involve drilling fluids, other than to unlock a bitumen seep, are not worth it. Also I was never tempted to use any of the coalbed gas recipes
• Big mines: uranium and phosphorus are good, lead is okay, others are meh, copper and iron are bad (why pay drill heads when you can mine them for free)
• Recipes unlocked by Mass creature production in utility science are not worth it
• Power generation deserves its own post; for example the recipe "dt-fusion" made my eyes fall out at how OP it seems (haven't actually built it yet)
• This in turn makes nucleosynthesis recipes too easy
• Recipes "fatty-acids" and "fatty-acids-2" are pointless when there's "oleochemicals-to-fatty-acids"
• "coaldust-ash" is the worst recipe in the game
• ... stopping here for the sake of brevity.

[protocol-1903]

The most concise answer would be to just install YAFC. Look at random items in Not Enough Items Explorer, many of which have recipes that are <10% efficient compared to the top pick, at which point it's sensible to ask why are they in the game?

While YAFC can be a good tool to organize and view recipes since Factoriopedia cant do shit with py, it's recipe rating system is unreliable at best and downright misleading at worst. It shouldn't be taken into account when considering recipes.

It's not about whether or not a recipe is comparable in difficulty and ouput, or if it's 'efficient' compared to others, but if it's difficulty and output are representative of the stage of the game. Early game recipes should be simple and crappy. Late game recipes should be complex yet efficient. That's what this thread is about.

One counterexample is plastic, where the "worst" recipe, the initial one with syngas and aromatics, is 20% as efficient as the top one. But I still use it frequently for small amounts if both syngas and aromatic pipes are nearby because of its simplicity. So, perhaps inefficient recipes are okay if they take 1-2 inputs of very common ingredients.

Plastic having an easy but low output recipe in the early game makes sense, as long as later recipes are better but harder.

Another aspect that matters to me is realism. Does the recipe make sense, or can it teach us something about real-life industrial processes? Great, I don't mind it being long or complicated. OTOH I'm never going to turn chitin into geothermal water because of how nonsensical it seems.

Whether or not you use an obscure recipe is up to you. All that matters is that the option is there. Most of the recent additions and more obscure recipes were added from Hardmode, since people like playing with pseudo-voiding instead of throwing everything into vents and burners. It makes a low-void or no-void run possible without installing Hardmode.

Sometimes a bad recipe becomes okay with enough productivity modules. Sometimes it's a way to get rid of excesses in a no-void run (especially animal parts).

Productivity modules should not be considered when balancing recipes. Some players use them, but interestingly with py the percentage appears lower than in normal Factorio. They should, however, be taken into account for recipe chains that may become net positive at high productivity, which is a later issue.

[BeaulieuT]

The most concise answer would be to just install YAFC. Look at random items in Not Enough Items Explorer, many of which have recipes that are <10% efficient compared to the top pick, at which point it's sensible to ask why are they in the game?

One counterexample is plastic, where the "worst" recipe, the initial one with syngas and aromatics, is 20% as efficient as the top one. But I still use it frequently for small amounts if both syngas and aromatic pipes are nearby because of its simplicity. So, perhaps inefficient recipes are okay if they take 1-2 inputs of very common ingredients.

Another aspect that matters to me is realism. Does the recipe make sense, or can it teach us something about real-life industrial processes? Great, I don't mind it being long or complicated. OTOH I'm never going to turn chitin into geothermal water because of how nonsensical it seems.

Sometimes a bad recipe becomes okay with enough productivity modules. Sometimes it's a way to get rid of excesses in a no-void run (especially animal parts).

So anyway,

* Any recipes that involve drilling fluids, other than to unlock a bitumen seep, are not worth it. Also I was never tempted to use any of the coalbed gas recipes

* Big mines: uranium and phosphorus are good, lead is okay, others are meh, copper and iron are bad (why pay drill heads when you can mine them for free)

* Recipes unlocked by Mass creature production in utility science are not worth it

* Power generation deserves its own post; for example the recipe "dt-fusion" made my eyes fall out at how OP it seems (haven't actually built it yet)

* This in turn makes nucleosynthesis recipes too easy

* Recipes "fatty-acids" and "fatty-acids-2" are pointless when there's "oleochemicals-to-fatty-acids"

* "coaldust-ash" is the worst recipe in the game

* ... stopping here for the sake of brevity.

I want to offer a counter to a couple of your points just from my personal experience.

1. I think the type of recipe of: "pay a small resource cost for a large amount of free resource" actually is a good option, given the numbers are tuned correctly. (Think pumpjack fluids that need small parts, coalbed gas, sand extractors needing iron sticks, etc). For I while I didn't use these recipes because they weren't "free", but recently I started using them (around py3) and I think there's actually a lot of value to them. They usually let you produce things in a very small footprint, the resource cost is usually pretty manageable, and they can allow you to save on train shipments and produce things on site. Could some of these recipes use a buff, like the coalbed gas? Sure I think so. But I don't think they're useless recipes by any means.

For reference, I uses coalbed gas to make coal gas, I used pumpjacks to make bulk saline, and I started using sand extractors quite a bit.

2. I like the big mines, other than iron and copper I agree aren't really useful. But I love using drill heads instead of drilling fluids they're great.

3. I think coaldust to ash is just a way to void coaldust, if you look at it like that it's a fantastic recipe.

4. also I think it's OK if early recipes are outright outclassed by later recipes; these recipes still have a use case, that being you have to use them until you get the better recipe. I think these are different than bad recipes that have no use case, ones that you should absolutely never use (think high tier coal processing).

[the-s4uce]

Fawogae recipe using manure seems too expensive to ever justify...fawogae is literally free other than space, and since it stacks to only 50, I always preferred to make it on site instead of shipping.

I think it's fine, more of a tradeoff between either doing super huge ZI fawogae or compact manure fawogae setup
Though I don't mind if manure cost is somewhat decreased, ZI is a pretty good thing on itself

[notnotmelon]

Using YAFC cost in order to balance recipes is dangerous.

YAFC cost does not account for:

• Void recipes that intentionally delete things. (example: coal dust -> ash)
• "Progression" recipes that are strong when unlocked but have a bad YAFC cost due to being outclassed in the future. (example: aromatics -> plastic)
• Recipes that are expensive but simplify logistics. (example: iron stick -> sand)

[darko-dll]

1. For I while I didn't use these recipes because they weren't "free", but recently I started using them (around py3) and I think there's actually a lot of value to them.

Drilling fluids type 3 and 4 get kinda expensive for what they provide, and then the pumpjacks and ground borer excavations take small parts and drill heads on top of this. But fair point - other players might have different preferences.

I think coaldust to ash is just a way to void coaldust, if you look at it like that it's a fantastic recipe.

I mentioned it because just burning the coal dust as fuel would at least give you twelve ash instead of one.

[darko-dll]

Using YAFC cost in order to balance recipes is dangerous.

I wouldn't follow its advices literally, but if it says that a recipe is hugely inefficient it seems worth taking a look? Case in point - tholins, as mentioned previously, are all rated very low for a good reason.

Btw, YAFC hates nuclear power so it doesn't like "phosphoric-acid" recipe because it produces hydrofluoric acid which can only be used for nuclear power (either as HF or Fluorine). As a result, it loves phagnot gas bladders > phosphine gas > phosphoric acid. I'm undecided if this assessment makes sense.

YAFC cost does not account for:

• "Progression" recipes that are strong when unlocked but have a bad YAFC cost due to being outclassed in the future. (example: aromatics -> plastic)

It has both Cost and Current cost (at current milestones) which is used in evaluations. Of course it's up to the user to interpret those results.

• Recipes that are expensive but simplify logistics. (example: iron stick -> sand)

Funny you should mention it, YAFC loves that recipe and prefers it to soil -> sand/biomass/limestone/coarse fraction for some reason (I provided it ways to get rid of each byproduct)

Bottom line, I look at it as a tool similar to GPS while driving. Its recommendations make sense majority of the time but I wouldn't drive into a lake if it told me to.

[notnotmelon]

Thanks to everyone for the great engagement on this thread. It's very helpful to see more perspectives.
Would like to make a reminder that some of the changes resulting from this thread will only be visible in pySEx in order to not break existing AE bases.

[BeaulieuT]

I have one more suggestion I completely forgot about....what is the point of hydrogen peroxide to sulfuric acid? Worst recipe in the game, never once have I actually seriously considered using it. I don't even know what you would change to make it better, I'm convinced that's a meme recipe there to taunt me every time I want more sulfuric acid.

[EigensheepLambda]

Small balance thought that the above just reminded me of: it would be nice if there was some other use for hydrofluoric acid, even if it's a glorified void (hugely lossy way to turn it into something else earlier in the phosphine gas chain, maybe) just to avoid actual voiding, for those morally opposed.) Maybe it could be used in the navens->gold process or some other such AL chemistry.

[EigensheepLambda]

Sort of related, an extra way to dump phosphorous acid (? I think, running from memory again) to unblock the production of phosphine gas would be useful in the post-pv late game- quantum dots demand so much that it makes it back up on the acid without a decent way to unblock it without direct voiding. (The antelope folding turd was my solution to this, removing use of the base dot recipe entirely, so there's a way out already, but it's kind of drastic. Reducing the phosphine input to dots by maybe ~half would probably solve it in practice too.)

[DrAfdch]

Suggestion: add another latergamish recipe for the Iron Nexelit Antimony alloy, maybe using molten nexelit; like there is a recipe for the Lead Antimony Alloy using molten lead. Maybe not even reduce any costs, just increase the crafting speed for the INA alloy so there aren't a stupid amount of smelters.

[Sirn3]

Request: Gearbox MK04 requires way too many Special small parts. 500 per. Can we reduce it? Maybe by half? If it was brought down to 180 that would be 3x what is required in the Gearbox MK03 recipe. Either that or have an additive to 2x the Special small part recipe.

[70000hp]

Suggestion: Make Iron 1 take less oxygen
Reason: It needs an entire electrolyzer per 2/3 of a BOF, making it entirely unviable until logistics science. If the oxygen requirement was lowered to say, one electrolyzer per 2 BOFs, or even just 1:1, you could have a much more reasonable trade-off between power consumption and yield, making the decision a lot more interesting, instead of the clear choice of not using it

[lleylar]

Every tholins-to-gas recipe is absurdly underpowered, even deuterium. I could see three interesting ways to fix this:

• Add some unique products only primarily extractible from tholins (maybe from others as a byproduct) - does require significantly changing chains though
• instead of having tholins->single gas recipes, have one/a few tholins->multiple gases recipes, which while quite powerful if using all the gas outputs, are generally worse than regular production chains if you're voiding most of the outputs - i.e. good but only usable as a supplement
• Add more advanced processing recipes for some gases/products that double output at the cost of a small amount of tholins

all of these are going to be a bit tricky to balance, and all three could be involved in a solution.

Solved today in pyanodon/pypetroleumhandling@13a6ebd

[limelou]

The anthracene -> gasoline recipe makes coke early game a lot easier, enough that it looks like it outclasses red hot coke

Red hot coke (Logistics Science):

Tar processing (Automation Science):

I think tar processing should be strictly worse for coke - since otherwise there's no reason to use it at all. It's also a huge increase in fuel value (around 2.5x if I did my math right) from automation science tech)

Removing the anthracene recipe seems to make red hot coke slightly better again - though tar processing does have all the other useful products

Tar processing coke should probably be a little weaker so there's a reason to set up RHC

[O5MO]

Grenades should take gunpowder instead of coal.
It is not realistic nor balanced for grenades to use the vanilla recipe, and the bullets already take gunpowder, and gunpowder isn't much more expensive.
It could also be beneficial to move Military 2 earlier in the tech tree, right after py1, so it would be before or at the same time as bots for use in deforestation, because right now it has a logistic science requirement (while not taking any logistic science to research and the recipes don't use any related ingredients)

[notnotmelon]

Request: Gearbox MK04 requires way too many Special small parts. 500 per. Can we reduce it? Maybe by half? If it was brought down to 180 that would be 3x what is required in the Gearbox MK03 recipe. Either that or have an additive to 2x the Special small part recipe.

Added!

[AFranticTypist]

The recipe to use advanced sand castings to cast iron gear wheels has no use. Gear wheels are only used in the creation of small parts, and the gear-casting recipe comes at the same time as the small-part-casting recipe, its roughly the same cost as the small-part-casting recipe, so there's no point in ever casting the gears first and then assembling the parts.

Probably we need more uses for gear wheels.

[protocol-1903]

Probably we need more uses for gear wheels.

Perhaps they should be added to more of the advanced mechanical/parts recipes in bulk, but eventually phased out for bulk casting recipes later on

[notnotmelon]

The recipe to use advanced sand castings to cast iron gear wheels has no use. Gear wheels are only used in the creation of small parts, and the gear-casting recipe comes at the same time as the small-part-casting recipe, its roughly the same cost as the small-part-casting recipe, so there's no point in ever casting the gears first and then assembling the parts.

Probably we need more uses for gear wheels.

I disagree. Casting of small parts is prod science, gear wheels is chemical science. Thus there is a reason for this recipe's existence.
Additionally, I think its really funny that gears are useless in py.

[notnotmelon]

Grenades should take gunpowder instead of coal. It is not realistic nor balanced for grenades to use the vanilla recipe, and the bullets already take gunpowder, and gunpowder isn't much more expensive. It could also be beneficial to move Military 2 earlier in the tech tree, right after py1, so it would be before or at the same time as bots for use in deforestation, because right now it has a logistic science requirement (while not taking any logistic science to research and the recipes don't use any related ingredients)

Added!

[jigsaw3xD]

""This recipe is too strong, I would never use the alternatives""

@notnotmelon 's EAFF mod's stack inserters (along with belt stacking) unlocking at logi science pretty much break the inserter progression in the modpack.

to make a comparison, fast inserters unlock at logi with a speed of 2.31 ips and bulk inserters unlock at py2 with a speed of 4.62 ips. stack inserters unlock at logi with a baffling 18.57 ips costing slightly more expensive than fast inserters to craft. and no, the increased power draw doesn't cut it as at that point of the game it's fairly easy to spam 500MW blocks of free self-sustained salt power.

potential solution: move stack inserter farther than bulks (in py3 or further) and make their cost reflect the sheer buff in speed they provide (bulk's 4.62 ips vs stack's 20 ips (it's 20 at that point cos they get the first inserter capacity buff from bulks researched).

[AFranticTypist]

Right now stack inserters are better than bulk inserters in every way, and I agree something should probably be looked at there. However, I think this is a more nuanced problem than just the progression of unlocks. Belt stacking is a new concept in 2.0 and the player is first exposed to it with sap extractors, very early. From this point on everyone realizes "I want to stack stuff", but if the availability of stacking inserters is waaay down the tech tree at py2 that's.... not a fun span of time to keep a core game concept locked down.

In general I feel infrastructure upgrades don't come fast enough, so I'd rather see bulk inserters show up at Logi and Stack inserters be available at Py2/Chem.

Edit: Alternatively, there might be design space for a "mechanical bulk inserter" at Automation tech, which would share the 1x2 form factor of other stack inserters, be more expensive than mechanical inserters, but grant access to stacking mechanics. Given it would be a palate swap of the existing 1x2 inserters it might not even be that hard to whip up. With something like this on board it would be easier to push the outperforming stack inserter further up the tree.

So progression like:

Base: mechanical inserters
Auto: bulk mech inserter
Py1: inserter / long inserter
Logi: fast inserter
Py2: bulk inserter
Chem or Py3: stack inserter
Py4: ??? (THE ULTIMATE INSERTER)

This bumps up the normal "inserter" 1 tech tier, but I don't think that's a huge loss as mech inserters outperform them well past the Logi era.

[BeaulieuT]

I also wanted to reply about stack / bulk inserters. I just barely reached the stack inserter unlock, so I haven't used them extensively yet. But I am firmly of the opinion that belt stacking and stack inserters are generally a good thing. I think by the time you unlock stack inserters, inserter throughput is WAY less of a limiting factor than it is in early game. I think in just about every scenario I've come across, I can reach max throughput with fast inserters. So I think the "inserter progression" part of the pack has already been largely solved by this point.

I think the power cost and ingridient cost of stack inserters is expensive enough that I want to use them sparingly, not spam them everywhere. To me, a 500 power plant to power 500 inserters sounds pretty expensive (power plants are not cheap to build). At the point of unlocking them, I think my entire base runs on 600MW for reference, so doubling my entire power usage for 500 inserters is rough.

Also, I think the number of scenarios that you would want to really max out your throughput of items is pretty limited, mostly to ore processing. Also having a crazy number of insertions per second isn't really a big deal if you don't have any machines that can utilize that high speed, and I can't think of any scenarios (other than train loading) where having a huge number of insertions / s would be game breaking. You're still limited by the building's crafting speed.

Lastly I think UPS is a real consideration here, and if you can achieve the same exact throughput with 5 blue inserters vs 1 stack inserter, I think it's actually very beneficial to have stack inserters in order to minimize UPS impact. Early game inserter puzzles are interesting because buildings are so small you can only fit 9 inserters max around it, but later on they're big enough that if you really want to max out throughput it's very easy to fit enough inserters to do so. Therefore, I don't think stack inserters actually change all that much personally.

I'm not opposed to balance changes on them, but personally I don't think they're game breaking at all and I personally REALLY appreciate how much belt stacking buffs sushi belts and makes them much more viable at large throughput. I think they're an interesting addition to the game.

[jigsaw3xD]

In general I feel infrastructure upgrades don't come fast enough

It's not just about the time of unlocks tho, it's also about spreading unlocks out throughout the whole run. Unlocking every inserter in the first 200h and then not having anything new for the next 800h is a major downer.

Stacking is basically just a fancy way to double the belt throughput and offering it at the very start of the game (which is very well balanced without it) not only breaks the established progression but also isn't even needed (I have a 10SPM logi base and I doesn't even need stacking or red belts, 95% of materials need less than half of a non-stacked belt including sap, ironically).

Alas, my point isn't about stacking per-se. Machine stacking only doubles the belt throughput, inserters just straight up octuples (that's a x8) it. I wouldn't've have such strong feelings if stacking was unlocked at logi without the inserters.

[jigsaw3xD]

I think in just about every scenario I've come across, I can reach max throughput with fast inserters.

Also, I think the number of scenarios that you would want to really max out your throughput of items is pretty limited

While it is indeed pretty limited, imo stack inserters break the one that matters the most. One of the major bottlenecks in every gigantic factorio overhaul is the speed with which you load/unload the train wagons. So while i agree that in a situation of inserting from belt to a machine stack inserters don't really matter, they do break the game when instead of 2.4x6=14.4 IPS you can load your wagons at 18.6x6=116.6 IPS.

[BeaulieuT]

I think in just about every scenario I've come across, I can reach max throughput with fast inserters.

Also, I think the number of scenarios that you would want to really max out your throughput of items is pretty limited

While it is indeed pretty limited, imo stack inserters break the one that matters the most. One of the major bottlenecks in every gigantic factorio overhaul is the speed with which you load/unload the train wagons. So while i agree that in a situation of inserting from belt to a machine stack inserters don't really matter, they do break the game when instead of 2.4x6=14.4 IPS you can load your wagons at 18.6x6=116.6 IPS.

I think I disagree here. I played my first py run in 1.1 and without cranes, and instead loaded my trains with 12 inserters (both sides of the wagon). Train loading speed was never once a bottleneck in my entire playthrough, but I did start to have severe UPS issues due to the inserter spam.

I started a new run with cranes, and I don't feel like they're any different than just running 12 inserters per train. Basically train loading is a binary question: are you using items faster than you are unloading them? Of yes, you're limited. If no, you're not. If you're able to unload items faster than you can use them, the unload speed doesn't reaaallyy matter past that point.

[error0664]

i found a minor balance issue on auogs
https://github.com/pyanodon/pyalienlife/blob/master/prototypes/recipes/auog/recipes-auog-raising.lua

raising from pups in auog-maturing
between base and red there is a odd productivity drop
base = 0,75 mature/pup (4pup to 2-4mature)
red = 0,625 mature/pup (8pup to 2-8mature)
green = 0,666 (12pup to 4-12mature)
chem = 0,8125 (16pup to 10-16mature)
py = 0,9 (20pup to 16-20mature)

ususally adding ingredients / go up in tech not only increase speed but also productivity a bit - in this case, adding auog food1 in base receipts actually drops productivity by a little bit making that recipe less appealing (actually base is even quite good compared to green)

suggestion for smoother progression:
base = 0,625 (4 to 1-4)
red = 0,6875 (8 to 3-8)
green = 0,75 (12 to 6-12)
chem = 0,8125 (unchanged)
py = 0,9 (unchanged)

[BeaulieuT]

I think in just about every scenario I've come across, I can reach max throughput with fast inserters.

Also, I think the number of scenarios that you would want to really max out your throughput of items is pretty limited

While it is indeed pretty limited, imo stack inserters break the one that matters the most. One of the major bottlenecks in every gigantic factorio overhaul is the speed with which you load/unload the train wagons. So while i agree that in a situation of inserting from belt to a machine stack inserters don't really matter, they do break the game when instead of 2.4x6=14.4 IPS you can load your wagons at 18.6x6=116.6 IPS.

Ok I have to apologize, I just started using stack inserters and I see your point. They are pretty strong lol. Personally I really enjoy them so I wouldn't mind if they stay as they are now, but I can fully understand the arguement for tweaks on them.

I don't think they're bad, I think they're a great addition to the game. But they do make bulk inserters look a bit sad. I think the 1MW cost is a good idea though to prevent the stack inserters from being too spammable.

[jigsaw3xD]

Alternatively, there might be design space for a "mechanical bulk inserter" at Automation tech, which would share the 1x2 form factor of other stack inserters, be more expensive than mechanical inserters, but grant access to stacking mechanics.

I honestly don't know why you'd be so adamant to bring stack inserters so very early. Vanilla pY's mechanical inserters have such low speed that throughput into smaller entities (like AMs that are 3x3) becomes a puzzle in itself and adding stack inserters to that era would break that puzzle. Rn, pre-logi, inserter progression feels very well balanced, you get a new better inserter just around the time when you start to actually need it in production, so just throwing a stack inserter in there would very much mess with that.

I like the idea of making stack inserters their own specialized type with their own upgrade chain (mechanical stack into yellow into blue into final orange), it'd potentially would solve lots of problems: it'd make stack inserters introducible in the early game without breaking anything (yeah logi is still early game), it'd massively nerf stack inserters as they're now and it'd give players more stuff to upgrade way later.

Here's how things are now:

base - mechanical 0.7 (items per second)
auto (middle) - yellow 1.2
py1 (middle) - long 1.2
logi (start) - fast 2.4
py 2 (middle?) - bluk 4.6
and
logi (start) - stack 18.5

I'd propose to add only the stacking tech without any inserters at the start of Auto (so Extractor machines could start stacking before a player can); add the first mechanical stack inserter at the start of Logi while giving them bulk's 4.6 speed (at the cost of much higher power draw and a 2x1 footprint); and then you can place however many tiers later. The problem with placing the mechanical stack inserter before Logi would be that the max speed you can give it is fast's 2.4 (to not break the existing balancing) which I don't think worth it (tho you definitely can place a mechanical 2.4 stack inserter in the middle of py1 right after the long ones). Too further add to this, the first (mechanical) tier of stack inserter should retain its current power draw and later tiers should demand even more.

[AFranticTypist]

I honestly don't know why you'd be so adamant to bring stack inserters so very early.

Because I like stacking. I see things stacked, I want stacky.

I think your proposal is broadly okay.

[BeaulieuT]

I like stacking too. Bots are so strong, it's nice to give belts a buff in my opinion.

[jigsaw3xD]

I am not against stacking. As a thing machines could do, sure sap extractors don't need it (my 10SPM logi base works on less than half a belt of sap) but in some cases the early game would benefit from it imo. After py1 wood starts to be in pretty high demand all over the base. Mud is in very high demand when you make sand, the fact that a belt could hold 30 of it instead of 15 is a godsend for landfill production (which you need a lot when playing with Alien Biomes lol). Limestone going 30 would mean you can produce more limestone tiles faster (which shouldn't be an hard thing to make). Stacking of these particular items doesn't harm the early game's limitations, they're more so an annoyance than a challenge. I am very much in favor of unlocking just stacking early (rn it's at logi btw).

Also another thought on this topic, if you want to give stacking a real practical use make casting units (that make plates from molten metals) spew stacked plates. It'd be sorta OP but in terms of things that you would actually want to be stacked by machines on your existing belts, the t2/3 recipes of plates are like the number one item to receive the x2 speed buff on the belt.

I am also not even against stack inserters. Like i said before it's a fancy way to doubling belt's speed and pY doesn't have new belt tiers so it's definitely not being over the top in that regard. It's also the only way to potentially get meaningful train loading/unloading speed. I only dislike how current stack inserters break the existing delicate balance of inserter progression, that is all.

[notnotmelon]

Thanks again for all the comments.

[protocol-1903]

Perhaps, is there a way to have different tiers of stack inserters 'max out' their belt stacking? That way, certain stack inserters could be introduced earlier but top out faster, while belt stacking can be further researched to augment entities with built-in loaders, it won't automatically buff that respective stack inserter.

[darko-dll]

Some thoughts on late-game balance, still from version 1.1

• Why do most numal recipes, including MK2-MK4 uplifting, require bedding? They are water animals. Same with vonixes, would giant burrowing worms need it?

• Recipe u236-u237 (Nuclear power 2, prod science): after bombing U236 with a LOT of neutrons, eventually it turns into ten times more U237, but the recipe needs to be repeated 1000 times. Is this intentional? screenshot
  Perhaps the 10% chance to make U237 is a typo that's supposed to be 0.1%, like in u237-pu238? Or is it because the same isotope is both input and output, the only such recipe where this is true?

• Molten fluoride with Pa-233 creates up to 27x more UF6 than what was provided as input, essentially turning thorium and lithium into uranium. Is this intentional? screenshot
  Also how does U233 make UF6 by itself, shouldn't it be mixed with fluorine or something?

• Purex raffinate 1->2->3 chain unlocked by Nuclear Power 3 in utility science doesn't result in anything better than what was possible since chemical science (purex-raffinate-vitrification). Here is a screenshot of what you can get from 1M of Purex raffinate 1. Americium and curium can be obtained much more easily with earlier recipes; anecdotally by that point I already had several full deposits of each. Not sure what would be a better end result there.

• Non-conductive phazogen recipe is not very useful because by the time it's needed (py4) you can make denatured seismite which takes prod modules, and is also required for other py4 stuff. Until then it's best to make negasium which also takes prod modules.

• Please take a look at Simik MK3 and MK4 uplifting. Overall they require ~1M logistic science and ~500k red chips. This is because each attempt requires Unknown and exotic DNA samples which also needs Simik codex. They are both very expensive in needed amounts.

[Hematin]

The recipe that uses a compressor to turn refined natural gas into condensates currently accepts productivity modules, which allows for a 2-building loop that can make an abundance of natural gas products without needing to extract any natural gas after seeding.

[AFranticTypist]

The Mukmoux manure recipes are terrible. 2 Mukmoux food, 30 ralesia seeds, 30 fawogae, 2 bedding, for 5 manure in 90 seconds. Meanwhile, the tier-1 Auog manure recipe is 20 flora and 10 moss for 5.5 manure in 100 seconds. It gets no better at higher levels.

You could buff the 'moo poop recipe by 10 and it would still be questionable. More interesting would be is if it produced the current amount of manure but also manure bacteria as an alternate route for that intermediate.

[Hematin]

The yaedol TURD (Humidity Control) is extremely potent for generating wastewater and biomass from nothing. In terms of space, it rivals both kicalk/wood routes for biomass without any input cost while also trivializing wastewater/urea/fertilizer production chains. This recipe produces spores at 27x the speed of the basic recipe, so one avenue for rebalancing might be to drastically reduce the speed, requiring a large array of spore collectors if players wish to use this route for ZI power/wastewater production.