科技

科技

科技

工程學研究

物理學研究

社會學研究

Technology in Stellaris is divided into 3 research areas with each area corresponding with one of the research resources:  Engineering,  物理學研究（Physics Research） and  社會學研究（Society Research）. Additionally, each of the ~300 techs belongs to one of 12 subcategories divided between the areas (most appearing pre-dominantly in one area though not exclusively).

The user interaction aspect utilizes a card shuffle approach rather than a traditional tech tree presentation, thereby introducing an element of semi-randomness into the system.

Research areas & fields

 請幫助核對或更新這個段落，最近核對為3.3版本。

Technology interface. Every slot here can have one research project active at a time and every research area has its own slot.

There are three different main research areas in-game, with each area corresponding with one of the research resources. Additionally, each research area has multiple subcategories, for a total of 12 such subcategories, and every tech belongs to one of these subcategories.

The areas and their subcategories are as follows:

Engineering research

•  工業技術（Industry）: Mineral production and storage, robots, building construction.
•  材料科學（Materials）: Armor, strategic resources.
•  推進力學（Propulsion）: Kinetic and explosive weapons, thrusters.
•  宇航技術（Voidcraft）: Ship types and hulls, starbases, strike craft.

Physics research

•  計算技術（Computing）: Science Labs, research, science ships, ship computers, point-defense, sensors.
•  力場操控（Field Manipulation）: Power Plants, shields, some strategic resources.
•  粒子物理（Particles）: Ship reactors, energy weapons, FTL, some strategic resources.

Society research

•  生物學（Biology）: Food production, leader lifespan and policies, species modification, army types, some strategic resources.
•  軍事理論（Military Theory）: Fleet command limit, naval capacity, army buildings and statistics, claim cost.
•  新世界理論（New Worlds）: Tile blocker clearance, terraforming, starbase capacity, administrative capacity.
•  治國術（Statecraft）: Unity, edicts, leader pool and recruitment, planetary and empire capitals.
•  靈能理論（Psionics）: Exotic mid- to late- game technologies. Unlocked through the 靈能理論（Psionic Theory） technology.

The card system

 請幫助核對或更新這個段落，最近核對為3.3版本。

The "card" system. Every time a research project is to be initiated, the game will draw a hand of cards like this one from which the player then can pick one technology to research.

Although the underlying technology system is based on the familiar tech tree structure, it does away with its standard interaction/presentation in favor of the card shuffle approach. This injects a semi-random element into research, making it somewhat less predictable as well as more non-linear.

Whenever the player is prompted to initiate a research project, the game gives them a "hand of cards" consisting of a semi-random selection of technologies available for research which are picked randomly from a hidden "deck of cards" consisting of all valid research options that can appear.

Technologies are divided into 6 levels (T0 to T5) of advancement. Higher tier technologies require obtaining a certain number of technologies from the preceding tier in the same area of research (set at 6 for all tiers) before they become available as research alternatives.

Every time the player meets the requirements for a given technology to become obtainable, the technology in question is added into the main "card deck" and will then have a chance of appearing in the "card hand" whenever the game draws a new hand for the player to pick from. There are one basic parameter affecting a tech's chances of appearing in a given card hand:

• Weight – A base value denoting the relative chance (out of all the valid options) of a particular technology to be drawn as an alternative. This may be further modified by an empire's composition (ethics, civics, etc.) and its current state (scientist's traits, past tech, etc.).

To further increase variation, a tech which appeared as an alternative in the previous card hand will have its weight set to 0 in the next prompt.

As a rule of thumb, the more advanced a technology is, the lower its weight (and the higher its cost). Systematically important techs (e.g. terraforming) have a higher weight (typically between 150% and double normal) to help ensure empires get a chance to research them, whereas rare technologies have significantly lower weight (typically one-eighth normal without the matching expertise). Some factors increase the weight of rare techs:

• Being in a level 3 or higher Research Cooperative federation with another member who knows the tech doubles the weight.
• The Technological Ascendancy ascension perk increases the weight of rare technologies by 50%.

Example

Let's assume the game draws a card hand of 3 techs: Armored Torpedoes, Autocannons, and  異星天然氣精煉（Exotic Gas Refining）. Additionally, the lead engineering scientist has a  推進力學（Propulsion） field expertise (matching Armored Torpedoes) and the previous card hand had Autocannons as an available option.

[math]\displaystyle{ \text{Total weight} = \text{Armored Torpedoes weight}\, \cdot\, \text{expertise bonus} +\, \text{Autocannons weight}\, +\, \text{Exotic Gas Refining weight} }[/math]

[math]\displaystyle{ \text{Total weight} = 75 \cdot 1.25\, +\, \frac{75}{2}\, +\, 170 = 93.75 +\, 37.5 +\, 170 = 301.25 }[/math]

If these were the only three valid techs and only one tech were to be drawn, then each tech's chances of being drawn would be 29.1%, 12.5%, and 56.4% respectively (rounded to 1 decimal place). Realistically there are usually many more valid picks and so the probability of picking any given tech is significantly lower than the figures given here.

Research alternatives

 請幫助核對或更新這個段落，最近核對為3.3版本。

Each card hand drawn will contain 3 research alternatives, which may be further increased by 1 for each of the following:

• Scientific Revolution unity ambition
• Self-Evolving Logic technology
• Science Division / Mindlink / Research Path Optimization discovery tradition
•  天然神經網絡（Natural Neural Network） civic
•  靜態研究分析（Static Research Analysis） civic
•  研究合作組織（Research Cooperative） level 2

Additionally, some research cards may be gained as permanent alternatives. They appear below the regular alternatives and are distinguished from them by a golden border surrounding the tech card's bezel. These alternatives are usually acquired (along with some partial progress) through post-battle debris analysis, from special projects or event chains. These options will always appear as available options when drawing a new card hand and will remain as such until completed.

If need be, it is possible to change an on-going research project at any point – without incurring penalties – in order to continue researching it at a later time. It is important to note, however, that the saved progress does not scale with changes occurring to technology costs (for better or for worse). This tends to lead to the tech's completion percentage getting lower as the empire continues to further exceed its administrative capacity.

As opposed to a permanent alternative, there is no guarantee that the technology in question will appear in the next card hand drawn.

Research output

Each research area employs the use of a different research resource.

Each of these individual resources has a base production of 5 research points a month. This can be further increased by pops working the appropriate Jobs or research stations on the galactic map.

The research production may then be modified by the following:

 請幫助核對或更新這個表格，最近核對為1.9版本。

	Population	Governance	Planet	Technology
+25%	"Pleased" modifier	"Precursor Data Cache" modifier	—	—
+20%	博學多識（Erudite）  合成人（Synthetic）	—	Research Assistance (improved) Magnetic Miracle Spirit of Science	—
+15%	—	"Curator Insight" modifier	—	—
+10%	聰慧（Intelligent） Logic engines  靈能（Psionic） Living standards – Academic privilege	Analytical  知識分子（Intellectual）	Research Assistance Uncertain History Observatory	—
+5%	Latent psionic	—	—	—
–10%	、​ Labor Camps Ghost Signal	"AIs Patched" modifier	—	—
–20%	AI – Servitude "Stellar Culture Shock (early space age)" modifier "War Protester" modifier "Withdrawal Symptoms" modifier	—	—	—
–25%	奴性（Serviles） "Enraged" modifier Ghost Signal (2) Ghost Signal (machine 4)	—	—	—
–40%	Droid AI – Outlawed Ghost Signal (3) Ghost Signal (machine 5) "Stellar Culture Shock (atomic age)" modifier	—	—	—
–50%	Battle Thralls	—	—	—
–55%	Ghost Signal (4)	—	—	—
–60%	Living standards – Chemical Bliss "Stellar Culture Shock (machine age)" modifier	—	—	—
–70%	Ghost Signal (5)	—	—	—
–75%	精神閹割（Nerve Stapled） Chattel Slavery Domestic Servitude	—	—	—
–80%	Robot "Stellar Culture Shock (industrial age)" modifier	—	—	—
–100%	"Stellar Culture Shock" modifier	—	—	—

Additionally, each of the resources may also have the following modifiers affecting them (these apply only to a single resource rather than to all of them):

 請幫助核對或更新這個表格，最近核對為1.9版本。

	Population	Governance	Planet	Technology
+20%	—	—	不穩定的構造（Unstable Tectonics）	—
+15%	天生工程師（Natural Engineers）	—	—	—
–30%	—	"Subterranean Tech Provision" modifier	—	—

 請幫助核對或更新這個表格，最近核對為1.9版本。

	Population	Governance	Planet	Technology
+20%	—	暗物質（Dark Matter）	狂野之風（Wild Storms）	—
+15%	天生物理學家（Natural Physicists）	—	—	—
+10%	—	"Prismatic Lenses" modifier	—	—
+5%	—	—	強力磁場（Strong Magnetic Field）	—

 請幫助核對或更新這個表格，最近核對為1.9版本。

	Population	Governance	Planet	Technology
+25%	—	—	Odd Animal Life  泰坦巨獸（Titanic Life）	—
+20%	—	"Pharmaceutical Innovation" modifier	鬧事的動物（Hostile Fauna）	—
+15%	天生社會學家（Natural Sociologists）	—	—	—
+10%	—	"Well Rested Scientists" modifier	致幻大氣（Atmospheric Hallucinogen） Filtered Atmospheric Hallucinogen Subterranean Civilization	—
–10%	—	"Loss of Personnel" modifier	AI Failure	—
–20%	—	—	Unproductive Without Experiments	—

Stored research

Research output which is not being used for research at the time is being accumulated as stored research. Alternatively, it may also include research resources gained from other avenues such as debris analysis, anomalies, events, and etc.

This method allows for a more "relaxed" research by eliminating the loss of research points due to forgetting to pick a research project while, at the same time, also simulating accelerated periods of research (as opposed to sudden "jumps") caused by newfound knowledge sources. The current storage levels can be seen by hovering over the appropriate resources.

The expenditure rate of stored research can be expressed as follows:

[math]\displaystyle{ \text{Stored research expenditure length} = \frac{\text{Research resource stored} }{ \text{Research resource produced}} }[/math]

Example

An empire produces +24 Engineering research a month and has just finished debris analysis granting it +1000 Engineering research (stored research).

[math]\displaystyle{ \text{Stored research expenditure length} = \frac{\text{1000}}{ \text{24}} =\, \text{41.67 months} }[/math]

The empire will supplement its Engineering research by an additional +24 research points for the first 41 months and by another +16 research points in the following month, after which the Engineering stored research bank will completely deplete.

Research speed

 請幫助核對或更新這個段落，最近核對為3.3版本。

Research speed represents a given expertise in a particular research area or field leading to shortened research time of related technologies.

Research speed boosts the perceived research production in the following manner:

[math]\displaystyle{ \text{Perceived research production} = \text{Research resource produced}\, \cdot\, \sum\text{Research speed modifiers} }[/math]

The most immediate sources of research speed modifiers are the 3 lead scientists. As the applied bonus depends on a scientist's skill level and overall traits, it is recommended to assign newly recruited scientists to command a science ship. Scientists on exploration duty tend to gain experience faster and, if one is in luck, they may also gain a new trait beneficial to research while leveling up.

Example

An empire is a  唯物主義（Materialist） and produces +24 Engineering research a month. It has a lead engineering scientist with a  推進力學（Propulsion） field expertise while researching Armored Torpedoes.

[math]\displaystyle{ \text{Perceived research production} = 24\, \cdot\, (1\, +\, 0.05\, +\, 0.15)\, = 28.8 }[/math]

Note: A research area without an acting lead scientist receives a penalty to research speed.

Research Speed can be affected by the following:

Source	Research Speed
Scientific Revolution edict	+20%
Nanite Actuators edict	+10%
Scientific Leap agenda	+10%
Discovery tradition tree finish	+10%
Technological Ascendancy ascension perk	+10%
極端唯物主義（Fanatic Materialist） ethic	+10%
唯物主義（Materialist） ethic	+5%
管理型人工智慧（Administrative AI） technology	+5%
正電子人工智慧 Positronic AI 正電子人工智慧（Positronic AI） technology	+5%
博學多識（Erudite） ruler trait	+5%
Miniature Galaxy relic	+5%
研究合作組織（Research Cooperative） level 2	+5%
Science Nexus megastructure (stage I)	+5%
Science Nexus megastructure (stage II)	+10%
Science Nexus megastructure (completed)	+15%
The Singularity Processor empire modifier	+5%
The Mirror of Knowledge empire modifier	+10%
Curator Insight empire modifier	+10%
Covenant: Whispers of the Void empire modifier	+10%
Spurred by the Past empire modifier	+5%
Full Circle empire modifier	+5%
Cover Around, Goes Around empire modifier	+5%
Minor Research Sanctions resolution	-5%
Moderate Research Sanctions resolution	-10%
Major Research Sanctions resolution	-15%

Technology cost

 請幫助核對或更新這個段落，最近核對為3.3版本。

As a game progresses, research option choices will begin to include progressively more advanced and correspondingly more expensive technologies.

Each technology has a base cost that applies as long as an empire does not exceed its Administrative Capacity. Once that happens, the research cost will be increased by 0.4% for each point of exceeded Administrative Capacity. This can be fixed by employing more bureaucrats.

Research progress

 請幫助核對或更新這個段落，最近核對為3.3版本。

In general, research time can be summarized as the following formula:

[math]\displaystyle{ \text{Research time} = \sum\frac{\text{Research cost} -\, \text{Partial research done} }{ \text{Monthly research progress} } }[/math]

Research progress is one of the key elements affecting the time it takes to discover new technologies. It is formulated as follows:

[math]\displaystyle{ \text{Monthly research progress} = \text{Research resource produced}\, \cdot\, \sum\text{Research speed modifiers}\, +\, \text{min}[\text{Research resource stored},\text{Research resource produced}] }[/math]

While it is possible for the monthly research progress to remain constant during the course of researching a certain technology, it is more than likely to fluctuate – for better or for worse – due to consequences of various actions (such as exploration, empire development, expansion, etc.). This, in turn, may increase or decrease the total research time of a given tech.

Example

The empire from the previous examples (same setup) has just finished debris analysis granting it +10% research progress in Armored Torpedoes along with +1000 Engineering research (stored research).

This setup will generate the following monthly research progress:

[math]\displaystyle{ \text{Monthly research progress} = 24\, \cdot\, (1\, +\, 0.05\, +\, 0.15)\, +\, \text{min}[1000, 24] = 52.8 }[/math]

Assuming no further changes will impact the above setup, the total research time of the tech would be:

[math]\displaystyle{ \text{Research time} = \frac{1862\, -\, 1862 \cdot\, 0.1 }{ 52.8 } = \text{31.74 months} }[/math]

Note: Partial research can refer to debris analysis, previous research, event rewards and etc.

參考資料

遊戲概念