Who needs engineering science in the early game?

[Peter Ebbesen]

This is a semi-serious question, since the question has an obvious answer: Anybody planning on an early game war using their own fleet rather than federation allies and focusing on science over alloys.

More generally, it is obviously useful to everybody in the early game. Whether you are planning war or peace, Engineering has valuable building and starbase upgrades early on.

But does everybody need a substantial output of engineering in the early game? Is it truly the most important of the three science disciplines in general? - I would argue not, and have the last few weeks been playing an AAR game teaching the readers one extreme counterexample, where society is the most important discipline and engineering the least, in Galactic Pacification for Dummies.

Even so I was amazed when through no deliberate action of my own, it just happened to come out that way as a natural fluctuation of vassal Prospectorium subsidies and own engineering production, I learned a curious Stellaris fact: If you have between 0 and -1 net science output it is displayed as +-0.

My Priesthood Tech Build is a font of weird game balance, and even this particular game, which is the third iteration of it playing out in 3.11 after having played it once in each of the two tech betas, comes up with surprises every now and then.

----

And that made me think. Specifically, it made me reflect on some of my assumptions: I know that in the past it was often claimed on the forum and on Reddit that Engineering research was the bees knees, the forum consensus being that yes, it was in general the most important discipline, and this is part of what I am reacting against with my AAR, since I found that to be erroneous as a general rule..

But am I simply, earnestly, wrong in that assumption? Behind the times? Has the general opinion moved on towards my own position, which is more, "it depends, for some builds engineering is most important, for others society, for yet others physics, - it all depends on circumstances, really"? - or perhaps to another position entirely?


So what say you?

• How important is engineering science to you compared to the other disciplines in the early game?
• How important is engineering science to you in general?
• Do you have other thoughts on the balance of unity to science in the early game? Are there certain minimums you need?

 

[CaesarVincens]

I weight all three areas more or less equally, though I tend to ignore most weapon techs until the mid game anyway.

I think your first point, that engineering is key to early war is probably most apt. If you aren't going to war, most engineering techs can be left aside until later.

 

[DamnedLackOfTropicalFruit]

I think while it may have been true that engineering was the most important discipline for early war, it's definitely not true anymore.

Ever since 3.6, physics is way more important for early rushing. Power is much more scarce, so the reactor techs aren't skippable anymore. Blue lasers is the biggest improvement in alloy efficency of any tier 1 tech, and it unlocks disruptors at T2, which in 3.6 are no longer a meme and are instead a top-tier option that don't need an engineering weapon to complement them.

On the society side of things, I'm not sure what update moved the first naval capacity tech to T1, but obviously that's a pretty big change for early war. The first fleet capacity tech (which grants the Fleet Supremacy edict for +100 starting experience for +10% damage) was also moved, though that's a lot more niche, since you need to build your navy after finishing the tech and the bonus obsoletes after your fleet finishes its first real fight.
Other than that, the only meaningful change in recent years has been the addition and buffs to catalytic converters, which makes the food techs more valuable for some builds.

Then on the flip side, two of the starting engineering techs ate huge nerfs in 3.6. Corvette hull tech was previously +100 hull, it is now +20. Similarly, flak cannons (the starter PD) are better kinetic weapons than coilguns (the upgraded mass driver), so coilgun tech is only worth picking up to unlock autocannons in tier 2.
Speaking of picking up techs to go for higher tiers, the tech changes in 3.11 make cruiser rush waaaaay slower, which consequently makes strikecraft tech much less important, so that's two more engineering lines that are largely skippable for early aggression.

Engineering is still the largest tech tree, and it still has a lot of important techs regardless of what you're doing, though you can skip way more of it than you used to be able to.

 

[Anaximatter]

The only reason to prioritise weapons research in the early game is to guarantee that no other empires have spawned anywhere near you.

To the questions:

• Roughly equally. Still important for economy techs down the road.
• Going to vary based on the game and build but it is essential to get to, for example, orbital rings, alloy boosting buildings, ecus. Otherwise you aren't going to be able to survive the crisis.
• I just try and get a feel for how quickly I want to progress toward traditions and techs and increment based upon this through the game.

I don't generally plan or micro too hard because I find that either I get stomped by an AI early, or I am going to become the most powerful empire by the time the crisis shows up. I have very limited time to actually play in, so if I spend loads of time microing to then find my first contact and nearest neighbour is a devouring swarm, I've wasted a large proportion of the time I have to play in.

I know you have outlined ways to deal with this in the past, I have found these ways too difficult to implement. I have found it less time consuming to just restart.

 

[Critical Ethics]

I think this is an artefact of Engineering having way too many techs. If I'm pumping Physics or Society it's because I want something specific and I can be pretty sure I'll get it in a few rolls. For engineering I could triple my research and it could still be decades before I get the mineral plants or whatever I'm looking for.

The individual techs are at least as useful as the other groups but yeah I can see the argument that less reliable = less valuable. Early game getting the right techs is way more valuable than getting more techs.

 

[Nakkivene]

And here I was, thinking "Me, certainly" and hoping this thread would tell where and how.

 

[Dr. B]

There are some key economic techs there, like mineral boosing techs and the armor techs that enable the +1 building slot techs. You certainly want those.

 

[Peter Ebbesen]

Engineering is still the largest tech tree, and it still has a lot of important techs regardless of what you're doing, though you can skip way more of it than you used to be able to.

I think this is an artefact of Engineering having way too many techs. If I'm pumping Physics or Society it's because I want something specific and I can be pretty sure I'll get it in a few rolls. For engineering I could triple my research and it could still be decades before I get the mineral plants or whatever I'm looking for.

Where does this idea that Engineering is the largest tech tree come from?

Society is the largest of the disciplines, having far more techs than any of the two others, and to make matters worse it its heavily frontloaded being especially numerous in the early tech tiers.

Of core techs (the base 00_<discipline>_tech.txt files), society has 138 distinct techs that aren't disabled, and engineering has a mere 84. This is why 00_soc_tech.txt is 100.647 bytes long compared to 00_eng_tech.txt being 54.400 bytes long. (And 00_phys_tech.txt is a mere 39.018 bytes long.) There are a few - a very few - in the core that are mutually exclusive (e.g. biological vs machine version of same tech), but not enough to change the picture.

Likewise all the DLC and other tech files taken together doesn't do much to change this general picture, because most DLC has either a few techs that aren't mutually exclusive or no techs at all, and the only two that have many that aren't mutually exclusive are First Contact (12 physics, 6 society, 4 engineering) and Megacorp (Megastructures adds 6 physics, 4 society, and 2 engineering)

The Society tech tree is massive compared to both engineering and physics, which is why players who pass up on important early society gateway techs like Planetary Unification risk greatly hindering their long-term development because the sheer number of techs once they break into tier 2 means that even common techs aren't that common until your science investment is large enough that you are churning society techs fast.

 

[Peter Ebbesen]

And here I was, thinking "Me, certainly" and hoping this thread would tell where and how.

1. Scout rapidly and extensively with corvettes
2. Survey useful avenues of expansion first, not just systems nearby. Do not waste time investigating anomalies early; Do so later with higher level scientists
3. Expand rapidly seizing choke-points and planets
4. Colonize everything
5. Concentrate population on high habitability planets; Get more species in your empire to cover more climates, if possible
6. Specialize your planets
7. Build lots of research labs
8. Run an economy that can afford the above
9. Keep empire size under control relative to the size of your economy
10. Stack scientists on the council if you have Paragons DLC

DONE. Next question?

At the risk of tooting my own horn, my Galactic Pacification for Dummies teaching AAR won't teach you how to produce lots of engineering research early, but if you are a player at a stage of Stellaris knowledge development, where you need to learn how to expand and build an economy capable of doing that, you might benefit from some of the lessons taught with regards to building a strong economy via rapid peaceful expansion, as it shows one way of doing so and goes through some of the strategic thinking involved.

 

[trent.layell]

Where does this idea that Engineering is the largest tech tree come from?

I was always told and it's always been my experience that the Engineering tree is normally the last one to complete. Perhaps the total cost of all Engineering technologies is higher than all the Society ones. For most of my games, it's Physics then Society then Engineering. I thought that was one of the reasons why Natural Engineers used to be a recommended trait pick.

 

[theBigTurnip385]

Engineering costs more research to finish.
Society has more techs to research.

The engineering techs hold an important role in getting minerals online, however minerals are sub-optimal in the early game and investing in them is a mistake.

Catalytic gains huge advantage from society research.

and Catalytic + Trade only needs society as trade output is driven via unity not tech.

Anglers has a huge advantage here, however you need to understand that Anglers doesn't scale well and gets worse over time, as food comes from vassals and a consumer goods worker is better on an Arcology as you get 6 per district as opposed to 1 pearl diver per district.

Slaves negate early tech economic boosts due to the slaves 40% output bonus from domination so they can give you an early boost.
You are limited to either 2 origins or the slaver guild civic for early game slaves.

a Trade based economy has little reliance on tech to boost early game worker economy, so it still has a lot of flexibility, Trade builds are driven by unity and two traditions.
statecraft + mercantile.

This is why prosperity preachers are just so good, it's a shame you can't run them in Peter's UOR build.

 

[Peter Ebbesen]

This is why prosperity preachers are just so good, it's a shame you can't run them in Peter's UOR build.

Yes, that would be bonkers, and for one extra reason in addition to those you list.

When Dimensional Worship was introduced in 3.10, the Prosperity Preachers got a double-strength Astral Minister (+0.2 physics, +0.2 TV per effective level) compared to the regular empire Astral Minister (+0.2 physics per effective level) with the argument that this was due to megacorp temples only having half as many priests, with the other half being managers, so it evened out. We could imagine the physics bonus going to the priests, so each priest provided the same physics for regular and megacorps, with the TV going to the managers as their bonus from the minister that had - for convenience of mechanics - been assigned to the priest. This made sense.

Then in 3.11 the changes to Prosperity Preachers making them even more powerful for trade builds, though arguably less valuable for many non-trade empires gaining them with a megachurch branch, and additionally temples and arcologies now gave megacorps give the same amount of preachers as regular empires get. So, logically, the corporate astral minister should now provide the same bonus as the regular version, but he still provides the double-strength version, with hilarious consequences for the trade value of extreme dimensional worship megachurch builds.

 

[Ramidel]

• How important is engineering science to you compared to the other disciplines in the early game?
• How important is engineering science to you in general?
• Do you have other thoughts on the balance of unity to science in the early game? Are there certain minimums you need?

Anyone who wants robots (including all machine intelligences) might want to put the pedal to the engineering metal too. Being able to create custom labor bots is a valuable way to expand your workforce in the early game, especially if you're not running Slaver Guilds.

So the answer is, if I'm a spiritualist who's not planning an early game war, I'm not going to care too much about engineering in the early game (though obviously there's only a limited extent to which I can "focus on the other areas"). If I've got threatening neighbors, I'll look for defensive pacts if possible to keep from becoming prey; if I've got genocidal neighbors, then I am now planning for an early-game war. Otherwise, it's not as important as Society.

However! I still want to advance engineering to get ready for the midgame, because at some point I'm gonna want a fleet. Even if I don't want to fight a war because I'm a megacorp who's busy doing business with everyone, I'll still probably want a fleet to charge my vote in the GalCom. I'm gonna want bigger startbases with longer trade range. I'm gonna want minerals. And I'm sure as hell going to want ecumenopolis and megastructure tech, which is even more important if I'm going the peaceful tall builder route! Engineering is a very important enabler for peace, as well as providing the most important tools of war.

Early game, I ram the dial hard toward Unity for most games - I want to get started on my traditions and ascension, and doubly so if I'm doing a trade build. That said, after my first ascension perk, and after I feel that I'm comfortable with the pace of my Unity growth, I start workng on research. (I get comfortable earlier if I'm running a trade build, because most of my unity will come from trade rather than administration.)

 

[Oculument]

Where does this idea that Engineering is the largest tech tree come from?

It comes from the number of repeatable techs at the end of the game, not the early game.

 

[theBigTurnip385]

Anyone who wants robots (including all machine intelligences) might want to put the pedal to the engineering metal too. Being able to create custom labor bots is a valuable way to expand your workforce in the early game, especially if you're not running Slaver Guilds.

So the answer is, if I'm a spiritualist who's not planning an early game war, I'm not going to care too much about engineering in the early game (though obviously there's only a limited extent to which I can "focus on the other areas"). If I've got threatening neighbors, I'll look for defensive pacts if possible to keep from becoming prey; if I've got genocidal neighbors, then I am now planning for an early-game war. Otherwise, it's not as important as Society.

However! I still want to advance engineering to get ready for the midgame, because at some point I'm gonna want a fleet. Even if I don't want to fight a war because I'm a megacorp who's busy doing business with everyone, I'll still probably want a fleet to charge my vote in the GalCom. I'm gonna want bigger startbases with longer trade range. I'm gonna want minerals. And I'm sure as hell going to want ecumenopolis and megastructure tech, which is even more important if I'm going the peaceful tall builder route! Engineering is a very important enabler for peace, as well as providing the most important tools of war.

Early game, I ram the dial hard toward Unity for most games - I want to get started on my traditions and ascension, and doubly so if I'm doing a trade build. That said, after my first ascension perk, and after I feel that I'm comfortable with the pace of my Unity growth, I start workng on research. (I get comfortable earlier if I'm running a trade build, because most of my unity will come from trade rather than administration.)

worker Robots have such a long pay off and a terrible output compared to slaves that they are hard to justify, and their habitability is getting nerfed in the next patch making them even worse, if they don't make much sense now they are going to be complete garbage in the next patch.

And robot tech is gained via council agendas and synth ascension.

If you want synth ascension and you like playing mega-corp just run permanent employment and synth ascend the zombies which removes the zombie trait. Syncretic Origin is quite good at synth ascension and an enslaved chattel zombie outproduces a worker robot even with a -25% output penalty. Robots get so few output buffs.
You can come out of mega-corp after you've ascended and have all those synth pops without ever wasting alloys on robots.


You need engineering for better colonies(arcologies) and ships.

 

[Ramidel]

compared to slaves

Implicitly assuming that the player empire can use slaves when that's hardly a valid assumption - it requires one of two specific ethics that the majority of player empires do not use. If I'm not running an authoritarian or xenophobe then guess what - slaves aren't an option.

if they don't make much sense now they are going to be complete garbage in the next patch.

A fairly minor habitability malus on most planets, compared to non-machines off their homeworld, or still a plus compared to climate-match. Ultimately fairly minor.

And robot tech is gained via council agendas and synth ascension.

Oh, clearly I'm only building robots if I'm going for synth ascension. Thanks for telling me that I'm not doing what I'm doing XD

If I'm Materialist, then I want robots to keep the faction happy even if I'm going psionic materialist (and yes, there are reasons to do this even if it's an oddball build). If I'm going cybernetic, then I want to develop the robot tech sometime anyway because it gives me extra genetic points, and getting started early means I can assemble more pops. Yes, I know, they're not the most efficient pops early on. Fine. They're still pops, which are still the ultimate resource in this game. Early on, you want as many butts in seats as you can get (and again, pop assembly doesn't compete with pop growth). They can fill worker jobs and produce minerals, freeing up your organics'

Ultimately, you're making a lot of implicit assumptions based on how you play that may not hold true for everyone's strategy. If the individual assumptions don't hold, then your argument is unsound. There is more than one valid strategy for Stellaris.

 

[andreeright]

One of the most difficult but indescribably interesting games I've ever known

 

[fourteenfour]

This is a semi-serious question, since the question has an obvious answer: Anybody planning on an early game war using their own fleet rather than federation allies and focusing on science over alloys.

So what say you?

• How important is engineering science to you compared to the other disciplines in the early game?
• How important is engineering science to you in general?
• Do you have other thoughts on the balance of unity to science in the early game? Are there certain minimums you need?

They all have important technologies.

Early game

• Physics, energy capacitors, research complexes, and disruptors
• Society, hydroponics, capital buildings, and navy size
• Engineering, mineral purification, crystals and gases, and starholds

I think what comes next is both Physics and Engineering end up in that cycle of not picking an important tech because a more important one is available at the same time. Society rarely has this issue and I tend to back fill a lot of the blocker technologies to quickly have a new roll hunting for the few technologies I just want - namely arcane deciphering and studies as everything else is about my fleets and navy.

So I never seem to see one area as more important than another but I do accept that physics and engineering are far more likely to present more that one choice each roll that I need both. I would love a way to lock single technologies as always unlocked so I don't have to wait for them to randomly appear again

 

[Altruist]

How important is engineering science to you compared to the other disciplines in the early game?

Early on engineering is THE most important tech field in my games because successfully beelining to cruisers and whirlwinds equals survival, safety and soon enough the ability to go into offfense and conquest.

But this is rather specifiic to my game settings which ensure that most AI-neighbours are much stronger than my starting entity of wannabe warmongers can honestly hope to become in the first 40-50 years. This is also more connected with getting the right tech cards than with pure tech researching power. Although the latter helps if one is unlucky with the first.

How important is engineering science to you in general?

If, during the game, I have the choice between more tech pts or a bonus in engineering or society... I usually choose society. Not only is society extremely important for economic development, there is also really lots to research plus usually the most extra-researches. The latter especially when going the genetic way.

 

[Thiend]

worker Robots have such a long pay off and a terrible output compared to slaves that they are hard to justify, and their habitability is getting nerfed in the next patch making them even worse, if they don't make much sense now they are going to be complete garbage in the next patch.

You've made the same mistake a lot of people do when evaluating slaves. Slaves are not at all good, robots are good. This is because robots (literal robots, not even getting into droids or synths) are crappy additional pops and slaves are crappy nonadditional pops. When you enslave a pop, you are trading a high-functioning pop for a low-functioning pop with worse output and stability loss. When you build a robot, you are trading some pop job time and alloy for an entirely new pop.

Depending on game speed settings robots can be good or bad (increased relative build time, reduced absolute time post-build for each robot). Slaves are always bad, as the only two scenarios in which slaves do anything at all that free pops wouldn't do better are:
1. Servant pops covering amenities without a building at worse pop efficiency (almost never, you could shuffle pops around, use amenity buildings, and employ more of the actual job they're providing amenities for with the freed pops unless you're somehow totally out of planets)
2. Livestock when you're out of the district for that basic resource (also almost never, if that was a scenario that could come about without you turning every planet into an ecumenepolis nonslaving empires wouldn't be playable).
So, in short, slaves are useless. Robots are good. Ideally this will be changed so that authoritarian has more going for it compared to egalitarian.