WIP: Data Dive on comparing buildings/Jobs

[strangebloke]

https://docs.google.com/spreadsheets/d/1CKgTb0N7Cr3LoCyU2rd7bHGCxlgSFVOvQ11uuOS_FpA/edit?usp=sharing

So I know that I am by no means the first to do this, however, this is an attempt to factor how effective various jobs/buildings are in the new patch. I'm not an expert, and I've definitely misunderstood certain mechanics. But still, please review and create your own, improved versions.

How to Use: You can't edit it, (I fear vandals) so duplicate your own copy to change values. The key pages that control the various comparisons are the 'energy conversion' page, which establishes how much 'energy' every producible commodity is worth, and the 'living standard' page, which calculates the output of every job, when pop upkeeps are accounted for. Just copy the relevent column from that sheet.

Method: This sheet converts everything to energy, and then calculates the total energy produced/consumed by all buildings/jobs. I didn't look at hive minds or synthetic empire since they'd need their own comparison sheet. I also didn't develop a scheme for comparing more esoteric things like planetary modifiers, since TBH those are too hard to evaluate. If you're on a planet with +59 food production, +15% food production is great.

For the energy conversion, I used the market base sale value for all the resources it applies to.

For other resources, (unity, amenties, research, and housing) I had to use guesswork. Ammenities and housing you need a certain number of, and don't usually want a ton extra. Unity and research you can never have too much of, but that doesn't mean that they're always worth choosing over other things.

The 'market of ideas' policy trades 0.5 energy for 0.15 unity, so 3.33 is the figure I used. This probably leads to an overvaluing of unity, but, well, see my conclusions below. Research I just valued at 1 energy per research, since they're approximately equally common in unoccupied systems. Lots of things generate amenities and something else, so I have it as being pretty cheap at 0.25 energy. Conversely, things that generate housing pretty much only generate housing, so I rate housing as pretty valuable.

Cautions: This kind of analysis says nothing about what buildings are actually most useful. A generator district might be lower value than an agriculture district, but you can do a lot more with energy overflow than you can with food overflow. If you need consumer goods and have spare crystals, using the crystals to boost consumer goods production is much better than selling those crystals for energy and then buying consumer goods, even if the higher-level civilian industries buildings are not super good value.

So figure out what you need (alloys, consumer goods, housing, etc.) and then consult this. Figure out why a given building is rated lowly before you turn away.

Conclusions: Clerks suck! There's so many other ways of generating amenities that are so much better, producing loads more money, unity, or housing. Clerks are worse than all other energy/amenity generators, except in two ways. They're workers (meaning that in a stratified economy they're rather cheap) and they produce trade, not energy, which makes them a little more flexible.

Districts in general are a great expenditure of your resources, and agriculture districts are the best of the basic ones. Eucemonopolis districts are, very simply, insane.

Cheers! Please let me know about any mathematical or data-based errors in my spreadsheet.

 

[AlanC9]

Interesting; confirms a lot of what people have been saying about clerks, for sure.

But it makes me think that it might be even more useful to rate everything in terms of what fraction of a pop it takes to produce.

 

[strangebloke]

Interesting; confirms a lot of what people have been saying about clerks, for sure.

But it makes me think that it might be even more useful to rate everything in terms of what fraction of a pop it takes to produce.

Aha! feedback!

Well, I think you have to do this step first, to determine what the most efficient ways of producing various things are, and even then certain things don't really translate well to that kind of paradigm.

for example, city districts provide 5 housing per pop, but paradise domes produce 6 housing without any pops. Which is better?

...I'd argue that this chart shows that under most circumstances, the city district is preferable.

There is an analyis on reddit that looks at this, right here:

https://www.reddit.com/r/Stellaris/comments/a9hd3e/relative_value_of_resources/

I've put those numbers into my sheet for now, because its interesting.

Anyway, a few more conclusions:

• Merchants and prosperity teachers are the most efficient energy producers, but following them is the lowly technician. However, since you can sell literally anything to get energy, you might be better off just farming or selling motes or whatever.
• 1 unity is a lot easier to get than 3 energy, so the marketplace of ideas trade policy is not a good idea unless you're really trying to skyrocket unity production.
• In real games, minerals end up getting bought up like crazy on the galactic market, and therefore I'd argue that mining districts are better than they appear.

 

[Eled the Worm Tamer]

I think you are overlooking one thing with clerks: They allow you to produce consimer goods or unity without needing minerals.

If you are playing tall, with a small area and a natural mineral cap on claimed teritory not having to use minerals that can go to alloy production is very usefull.

 

[strangebloke]

I think you are overlooking one thing with clerks: They allow you to produce consimer goods or unity without needing minerals.

If you are playing tall, with a small area and a natural mineral cap on claimed teritory not having to use minerals that can go to alloy production is very usefull.

I can see that!

I actually do call this out in my first post: Clerks are a bit more flexible than technicians, and they're cheaper to maintain than specialists.

This sheet doesn't really consider things like caps and economy though. It's purely a look at the total value generated per job or per slot.

For energy, Clerks are always a worse option than technicians. They additionally have a problem of not getting scaling multipliers like technicians do. There's a building that grants +15% technician energy, then improves to +25%, there's the edict, there's several researches... and the clerk was behind to begin with.

Overall, they're bad compared to other options, but there are still instances where you'd build them anyway.

 

[Wolfgang I]

If houseing is really valuable but clerks are not very good do you think that agrarian idyll+slavery is the way to go to get the most out of your planets?
You lose city houseing but I felt no need for cities in the games I tried it(stacking slavery houseing reduction with traits). I got one ecumenopolis early game in the 1st pacifist game I tried(Frist League) and a second one once I started insulting FEs though. In my current 2nd pacifist game I switched to agrarian idyll after getting the arcology perk which is not normal agrarian idyll gameplay I guess(I have not actually used the edict though it might be disabled) .

 

[strangebloke]

If houseing is really valuable but clerks are not very good do you think that agrarian idyll+slavery is the way to go to get the most out of your planets?
You lose city houseing but I felt no need for cities in the games I tried it(stacking slavery houseing reduction with traits). I got one ecumenopolis early game in the 1st pacifist game I tried(Frist League) and a second one once I started insulting FEs though. In my current 2nd pacifist game I switched to agrarian idyll after getting the arcology perk which is not normal agrarian idyll gameplay I guess(I have not actually used the edict though it might be disabled) .

Well, I don't think this limited-purpose spreadsheet really does a great job covering the viability of an entire playthrough strategy, but I can't resist a good numbers challenge:

Decent conditions, no civics, no slaves, 'balanced' resource value(value based on opportunity cost, not market value):
city value: 2.5 per slot, 2.5 per job
generator value: 5.25 per slot, 2.63 per job
mining value: 5.25 per slot, 2.63 per job
Farming value: 5.67 per slot, 2.84 per job

Stratified conditions, Agrarian Idyll, all slaves, 'balanced' resource value(value based on opportunity cost, not market value):
city value: 2.75 per slot, 2.75 per job
generator value: 3.5 per slot, 1.75 per job
mining value: 3.5 per slot, 1.75 per job
Farming value: 3.42 per slot, 3.42 per job

So yeah, this leads to a huge increase in overall production although it needs to be said that this isn't a fair comparison, because obviously using slaves with stratified conditions is going to negatively impact happiness, and therefore production.

Also, we all already knew that slavery and agrarian idyll were good for pruduction, so this isn't a shocker. It is a solid strategy though, and the sheet comfirms it.

I'm personally just impressed by 'efficient bureaucracy' and technocracy. Science Directors are nearly twice as good as regular admins, and giving a flat +1 unity (~roughly 2 energy in my estimation) to every scientist is crazy. The efficient bureucracy part is crazy because if you're keeping yourself at or above cap, you've essentially given yourself

• -6.0% Tech cost
• -10% Tradition adoption cost
• -20% Campaign (Certain sorts of Edicts) cost
• -20% Leader Upkeep cost
• -20% Leader Cost

which is pretty insane for a single civic. You can really get to max tech quickly with intelligent fanatic materialists, and if you also have technocracy you'll be picking up traditions left and right.

 

[RoverStorm]

From my own games, I must point out that minerals are highly likely to suffer shortages, and are very important to focus. This is likely because ANY production building can indirectly become energy from selling, and there is a dedicated building that provides farmer jobs, but NOT one for minerals. Meaning the only minerals your galaxy is ever going to get is from mining districts, the mining +15/25% building, and space stations! Also ringworlds are penultimate sources of energy and food, but they can't give minerals.

Meanwhile, minerals are a key ingredient for consumer goods and alloys. Trade value, lategame, becomes really valuable simply because you can turn a quarter of it into consumer goods.

 

[strangebloke]

From my own games, I must point out that minerals are highly likely to suffer shortages, and are very important to focus. This is likely because ANY production building can indirectly become energy from selling, and there is a dedicated building that provides farmer jobs, but NOT one for minerals. Meaning the only minerals your galaxy is ever going to get is from mining districts, the mining +15/25% building, and space stations! Also ringworlds are penultimate sources of energy and food, but they can't give minerals.

Meanwhile, minerals are a key ingredient for consumer goods and alloys. Trade value, lategame, becomes really valuable simply because you can turn a quarter of it into consumer goods.

Currently the two ways of looking at things are "Everything as energy" or "everything as fractional pops" but I suppose that you could look at things from the perspective of "how hard is it to get 'x.'"

But I don't really favor that because that's going to vary a lot from playthrough to playthrough.