M&T 3.0 Dev Diary #9 - Path to Modernity, v2
M&T 3.0 开发日志 #9 - 通向现代之路，第二版
FireKahuna, Mar 1, 2020
Welcome to the newest version of M&T, still 3.0, but with enough work you could think otherwise. 3.0 is the next version of M&T, and 3.0’s probably worth reintroducing considering it's been so long since the last update on how its systems are progressing along. But do not despair, the second great dev diary drought is over. In response to this agonising delay, we’ve prepared a hybrid dev diary: One part a Trade/Production diary on the new systems (yes, third iteration) we’ve designed, and one part a summary of all the other new features, which will receive their own dev diaries in the near future. I do hope that you enjoy this toblerone-sized dev diary.
欢迎来到M&T的最新版本。它还是3.0，但是随着工作量的增加，你或许会对它刮目相看。M&T的下一个版本确实仍然是3.0，但考虑到上一次的开发进度更新距离现在已经过去了很长一段时间，这个3.0版本或许值得我们从头开始再介绍一遍。不要绝望，第二次开发日志大饥荒已经结束了。为了补偿这段令人心焦的拖延期，我们准备了一篇融合日志：前一部分关于我们设计的新版贸易/生产系统（没错，第三版了）；后一部分关于所有其他新系统，这些系统会在不久的将来以更多开发日志的形式进行详细的说明。我希望这篇巧克力式分块日志可以让你满意。
Note: The first two sections, Trade and Production, are written personally by KJH, lead programmer on economy.
注：前两个小节，也就是贸易和生产，由我们的经济系统主程序员KJH亲自撰写。
Trade
贸易
How goods travel between different regions of the world has changed rather drastically as development progressed. That's because for a long while I was focused not on creating a fun system for players to interact with but was focused on instead creating an elaborate and intricate simulation. You could barely interact with it, and it was so delicate and unstable that one wrong move and it fell apart like a house of cards. It was a mistake on my part which cost me both time and emotions.
随着开发进度不断推进，我们在很大程度上改变了货物在不同地区间运输的方式。这是因为，在之前的很长一段时间里，我的工作目标都是创造一个逻辑缜密，错综复杂的模拟系统，而非为玩家提供一个有趣的，可以与之互动的系统。你很难与原来的系统产生互动，而它也太脆弱，太不稳定；只要一步出错，它就玉碎珠沉了。我的锅，我为这个错误多花了不少的时间和感情。
The main focus of new trade, therefore, was in addressing those problems that I just pointed out. I made it so that you can interact with it, and I made it resilient so that it can withstand a lot of abuse. Let's start with sectors.
所以，新贸易系统的关注点就是解决我上面提到的这些问题。我的改动让你可以与它互动，并且把它做得更加稳健，不管你怎么玩弄它，它都不会轻易崩溃了。让我们从贸易域开始说起吧。
A sector is a new addition to the new trade, which is basically just a set of provinces. It is defined as a set of provinces that are within the same tradenode and share the same nationality. Sectors avoid the need for each province to conduct foreign trade on their own by grouping provinces together. After that, each sector can do trade with other sectors, which greatly reduces the number of possible trade links.
贸易域 sector 是加入新贸易系统的一个新元素，由一组省份组成。它的定义是：同一个国家拥有的，在同一个贸易区内的一组省份。通过把省份分组，贸易域避免了让每一个省份单独进行国际贸易的麻烦。而且，贸易域之间也可以相互贸易，极大地减少了可能出现的贸易链接。
A sector can trade with another sector when it has a trade link between the two. A trade link is created and maintained by states. It is created when there exists an unbroken chain of merchants placed in tradenodes between two sectors.
一个贸易域和另一个贸易域之间如果存在贸易链接 trade link，就可以互相贸易。贸易链接由国家创建并维持。要创建贸易链接，需要国家在两个贸易域之间的所有贸易节点上存在商人团，中间节点不可空缺。
For example, suppose there exists 3 nodes A, B and C. There also exists 3 sectors Z, Y, and X, where Z is in A, Y in B, and X in C. If 3 nodes A, B and C have vanilla trade connections that's set up like A -> B -> C. If owner of sector Z has placed a merchant in both A and B, then trade link is formed between sectors Z and Y. If the owner of sector Z has placed merchants in all 3 nodes of A, B and C, and if it is also the owner of sector Y, then trade link is formed between sectors Z and X through sector Y. If the owner of sector Z has placed merchants in all 3 nodes but does not own any node B, then a trade link cannot be formed between sector Z and X.
举个栗子：我们假设有三个贸易节点，A、B和C。还有三个贸易域，Z、Y和X。Z在A里，Y在B里，X在C里。再假设三个贸易节点的原版流向是A -> B -> C。如果Z贸易域的拥有者在A和B上都有商人团，那么Z和Y之间就建立了贸易链接。如果Z贸易域的拥有者在A、B、C三个节点上都有商人团，而且它还拥有Y贸易域，那么Z和X之间就通过贸易域Y建立了贸易链接。如果Z贸易域的拥有者在三个节点上都有商人团，但是在B节点没有任何省份/贸易域，那么在Z和X之间就没法建立贸易链接。
This change, which puts the burden of opening and maintaining trade routes on the state, is a radical change from the previous model of provinces forming trade routes on their own. It meant that trade routes are more dynamic and less stable. Therefore, I also had to change how trade itself is conducted once trade links are established.
在之前的模型中，省份会自己建立贸易路线；而与之相比，现在的模型由国家接管维持贸易链接的负担，这可以说是一项根本性的改动。这意味着，贸易路线现在更加动态，也更加稳健。我还改变了贸易链接创建后，贸易本身的运作方式。

In this example, Fars has merchants in four surrounding nodes. This empowers them to trade with all other sectors in those nodes if relations permit. Also worth noting that their home node does not need a merchant, providing a guarantee of intersector trade in a tag's home node.
在这个例子里，法尔斯在周围四个节点有商人团。只要关系允许，这让他们可以与周围的所有贸易域开展贸易。顺带一提，他们的贸易本埠所在节点不需要商人团，每个国家都可以在本埠所在节点进行贸易域间贸易。
Previous model worked based on a concept of incremental optimization. Basically it stored a variable whose value represented the total trade volume, and it incrementally optimized that value by either increasing or decreasing it based on price differences of trading actors. The key problem of this model was that it takes time for sufficient trade volume to accumulate and therefore was unsuited for an environment where links were opened and closed in unpredictable fashion.
之前的模型基于增量优化。简单说来，它会把总贸易量存储为一个变量，然后根据贸易价格差渐进地增加或减少这个变量。这个模型的关键问题在于，要积累足够的贸易量需要一定的时间，而这样的模型不适合现在贸易链接可以随意开启和关闭的环境。
New model works based on making bids and then resolving those bids by matching them with offers. Basically, I made it so that sectors no longer care about their previous business transactions and made them buy and sell all they want with whomever they can find that gives a good deal. This model proved to be far more stable and resilient, as even if one trade route suddenly got closed it could still try to find another sector to trade with.
新模型则基于出价，然后处理与出价相关的报价。简单说来，我让贸易域不再关心以前的交易，而是让它们只要觉得足够划算，就向任何人购买和出售任何商品。经证实，这个模型更加稳定，强健，就算突然有一条贸易路线关闭了，它也能尝试寻找另一个贸易域来开展贸易。

The stockpile of a sector is the same for all, yet is distributed across the provinces for storage. This allows the capture of stockpiles, and provides essential data integrity.
贸易域的仓储没有改动，但现在它分布在各个省份中。现在你可以夺取（敌人的）仓储货物，也提供了至关重要的数据完整性。

Production
生产
If trade is about the movement of goods, production is how those goods are produced in the first place. Like trade, production too has seen a good amount of changes during its course of development. Key changes to note are changes in how industry size is determined, how it shifts, and how it interacts with property.
如果说贸易是关于货物流动的系统，那么生产就是关于这些货物起初是如何生产的系统。与贸易一样，在开发过程中，生产系统的改动也不少。如何决定产业规模，产业规模如何增减变化，它又如何与资产互动，这些方面的改动都值得一提。
Industry size is a value that represents the sheer size of scale of a given industry. Before, it used to be directly tied to how much wealth is stored up in a given industry's treasury. If industry profits, then it will accumulate wealth and therefore increase in size. If the opposite, then size would have decreased.
某个特定产业的规模就是它的产业规模 Industry size。过去，它曾经与某个产业拥有的财富直接关联。如果某个产业盈利了，那么它就会积累财富，于是产业规模就会增加；反之则产业规模就会缩减。
The problem arose when the concept of separate treasuries for each industry became a goner. As part of property rework, it was changed such that all industry profit and losses, rather than accumulating in their own separate treasuries, would just be fed back to property owners based on their property share. It was a nice change, but it meant a lot of production code and concepts had to be reworked. I tried many different things, but in the end concluded that just having industry size to be its own thing not tied to anything else was the best approach. Now, industry size is not a composite or result of some other set of variables. It just stands on its own as its own thing.
而问题在于，产业会进入持续萎缩的恶性循环。作为资产重做的一部分，现在所有的产业盈利或亏损都不在产业本身的财富中积累了，取而代之的是，它会让财富根据占有份额，流入各自拥有者（译注：社会阶层）的口袋。这个改动确实是个好改动，但这也意味着生产系统的很多代码和设计都需要重做。我尝试了很多种办法，但最后的结论是，产业规模最好还是作为一个独立的值，而不应该与任何东西挂钩。现在，产业规模不再是一个复合值，也不是其他变量的结果；它就是它自己。

An example of the freefloating nature of size. Note that 'Skill' is a WIP stat, and thus set to 100 in the interregnum.
产业规模非依赖性的一个例子。注意：这里的生产技术 Skill数值还有待开发，所以只是设成100的占位符。
That did lead to a new set of problems. If industry size is its own thing, then I had to rework how it changes based on market situations, as I could no longer just rely on change in treasury size as a result of profit and losses to automatically handle industry size changes for me. I again tried many things, and settled on a simple script. Basically, all industries now record any change in size that they made to themselves, and record how much change in profit has occurred after it made that change. Then, they calculate how much they'll change in size and in which direction next year by simply multiplying previous size change with previous profit change. If one of the two was negative, then industry will try to shrink. If both were positive or if both were negative, then industry will try to expand.
但这也带来了一系列新问题。如果产业规模是一个独立的值，那么它根据市场情况扩张或收缩的逻辑也需要重做，因为它无法通过盈利或亏损造成的财富变动来自动扩张/收缩。在这方面我也试了很多方法，最后决定用一段简单的脚本来解决这个问题。简单说来，所有产业现在都会记录它们过去的规模变动，以及变动后的盈利变化。然后，它们会简单地用把这两个值相乘，计算它们的产业规模在下一年该如何变动。如果两个数值之中的任何一个是负数，那么产业规模就会缩减。如果两个都是正数或两个都是负数，那么产业规模就会扩大。

Luxury goods are highly profitable in 3.0, yet still coexist alongside regular production. Note that profitability and values are still balanced ongoingly.
在3.0中，奢侈品能够带来丰厚的利润，但省份也可以有其他的一般生产项目。注意，利润率和其他数值都还在平衡中。

Property and Infrastructure
资产与基建
To facilitate these new Trade and Production systems, we had to forgo our old system of abstract ‘Property Power’ and automatically calculated property shares. With this new model, we decided to work from a clean slate, moving away from simply having ‘Rural’ and ‘Urban’ property. Instead, there are real, discrete, buildable units of property that each are owned by a given class. Property is not focused on progression, it is instead focused on supplying real estate for industry slots to utilise, working as a primary means to distribute wealth and thus power dynamics in a province.
为了构建新的贸易与生产系统，我们必须扬弃我们旧有的抽象「资产力量」和自动计算的资产份额。在新模型下，我们希望打扫干净屋子再请客，从单纯的「农村」和「城市」资产上跳脱出来。与之前不同，现在它们是真实可见的，可以建设的资产单位，从属于各自的阶层。资产不再关注改良或增长，而是用于支持各类不动产，以供产业槽位（译注：industry slots，根据之前的3.0开发日志，每个省份都有一定量的槽位，用于形成产业）使用。它是分配财富，并由此决定省份权力动态的主要手段。

Each of these property types are used by different industries, and owned in different absolute amounts by a variety of classes, requiring unique values for construction and maintenance.
各大阶层拥有一定量的，各种各样的资产，这些资产又被不同的产业使用；各类建设和维护也需要不同数量的资产。
Progression instead falls largely upon a new Infrastructure system, entirely divorced from the old. New infrastructure is based on a ranked system of progression, each rank providing increasing bonuses while requiring dynamically sized amounts of units to be constructed for each rank. This dynamic sizing allows larger provinces to have larger costs where appropriate, while also providing a way to increase maintenance accordingly as well. In a sense, a rank is a collection of these units, providing a means for partial progression towards the next improvement in a province.
现在，大部分增长与改良都附属于基建系统，和旧的系统已经大不相同。新的基建系统现在基于改良等级，每个等级都需要建造动态数量的基建单位，提供逐级增长的加成。这一动态规模系统使得更大的省份需要更高的建设花费，也使得维护成本的增长有据可循。在某种程度上来说，基建等级就是基建单位的集合，这使得省份有办法在特定的方面得到部分改良。

Amenities is essential for those aspiring for large, powerful cities. Outside of migration, it is the only other way to reduce the inherent difficulty in maintaining a positive growth rate for urban population.
对于那些希望让城市强盛的玩家来说，舒适度不可或缺。舒适度可以让是维持城市人口正增长变得更加容易；除了移民以外，这是唯一的途径。
Trade Nodes
贸易节点
Our new trade system left a fundamental issue, which is the balanced trade nodes of 2.5 would not fundamentally work due to the limited trade node connections inherent to it. In this sense, we made the conscious choice to forsake AI balance to an extent, and instead focus on providing a larger number of merchants for a now larger, fully formed network of inter-node trade connections, providing complete functionality for trade with historical, redrawn routes and an expanded trade node set. This includes major reworks to the nodes of much of Asia, and additional work in many other areas.
我们的新贸易系统还有一个基础性的问题：由于贸易流向的限制，在经过2.5平衡性改动后的贸易节点从根本上无法（译注：在3.0中）运作。在这个层面上，我们有意识地决定放弃一部分AI平衡，转而为现在的节点内贸易提供更多的商人团，来支持现在这个规模更大，更全面的节点内贸易链接网络。我们重新根据真实历史绘制了贸易路线，并增加了新的贸易节点，使得其在新贸易系统下可以充分运作。包括亚洲的大部分区域和其他的一些区域都已重做。

Key splits have occurred in China, SEA and Central Asia, to provide rational trade range, rational limits to inter-provincial trade, reasonable competition and better flow between nodes.
在中国、东南雅和中亚，我们做了一些关键的分割，以提供合理的贸易距离、更合理的省份内贸易、更合理的竞争和更好的节点间流动。
Classes
阶层
The bedrock of the 3.0 rework originally was a class system, providing discrete units of population focused on socioeconomic differentiation, including within Rural and Urban. Rurals were split into Slaves, Peasants and Freeholders, while Urbans were split into Slaves, Residents and Artisans. Through the last two years of development however, we’ve found that the greatest value of classes was in discrete economic units for the ‘Elites’ of society. Consolidating Rural and Urban classes back into united classes provided better results in our systems, improved performance and understanding.
3.0重做的基石原本是阶层系统：一个更加细化的，专注于社会经济分化的，包括农村和城市的阶层系统。在原计划中，农村人口被细分为奴隶、农民和自由农；城市人口则被细分为奴隶、居民和工匠。然而，在过去两年的开发过程中，我们发现，阶层系统最大的价值在于在细分的经济单位造成了社会的「精英」阶层。对于我们的系统而言，将农村人口和城市人口合并为统一的各大阶层会得到更好的结果，可以优化性能，更便于理解。
The fundamental divisions however are now reflected through the systems developed in the last few months. For example, we’ve moved from a Serfs class to a more multi-factored approach to distinguishing the wellbeing and freedom of Peasants. Aspects like their property share, their ‘Freedom’ value or the rights applying to them all have strong effects on their fulfilment and wealth, representing the intent of the divisions without the need for separate classes.
现在，基本的人口细分由我们在最近几个月里开发的系统呈现。比如，我们不再有农奴阶层，而是用一种更加多维度的方法来区分农民的生活质量与自由度，这些维度包括他们的资产份额，他们的「自由度」值，或他们所保有的权利；这些都对他们的满意度和财富有着密切的关系，于是我们就不需要一个单独的（农奴）阶层来表示农民阶层所包含的意味。

Also demonstrates the new approach to modifiers and localisation. Note that 'Dole' is related to an interaction recently implemented, another example of the WIP of 3.0.
在这张图中，我们也展示了各种修正和本地化的实现方法。注意：失业补助 Dole 和一个最近才添加的功能相关，毕竟3.0还在制作途中。

~完~
一些其他情报：
首先解释一下最后一张截图。经M&T制作组确认，这张图是开局时间（1356）之前刚刚发生的黑死病造成的荒废度，地图模式就是原版荒废度。回过头仔细看，是不是每张省份面板上都有荒废度？至于为什么遮掉中西非就不清楚了。
所谓7个槽位都用上了，是因为3.0中移除了以前占用一个槽位的，根据文化/文化组改变的家庭结构理念。

M&T 3.0 Dev Diary #10 - InstitutionsM&T
3.0 开发日志 #10 - 思潮
wogudwkd12, March 6, 2020
I must confess that institution is one of my favorite parts of EU4. It’s a fascinating system that provides an organic and compelling way of making some regions to be more technologically advanced than others. It’s also a nice source of some alternate history vibe, such as embracing renaissance as an east asian country. It provides good gameplay, and is overall very fun.
我得坦白，思潮系统是我最喜欢的EU4机制之一。这个迷人的系统用一种有机，有理又有据的方式让某些地区在科技上比其他地区更发达。它同时也可以为游戏增添一把架空历史的风味，比如以东亚国家接纳文艺复兴思潮。在游戏性方面它也居功甚伟。总而言之，这是个非常有趣的系统
However, as much as I like it, I still had some issues with it. My main complaint was that once someone spawns some institution, others were barred from reinventing it in their own lands. Therefore, when it became time to work on institutions in 3.0, I had one clear object in my mind, which was to make institution reinvention be a thing. All other changes made on institutions that I will also talk about are stemmed from that goal.
然而，虽然我非常钟意思潮系统，于我而言它还是存在着一些问题。我时常抱怨，只要有一个国家的国土上产生了某个思潮，其他人就没法再重新发明这个思潮了。所以，当M&T制作团队开始制作3.0的思潮机制时，我脑海中就出现了一个明确的目标：让思潮可以重新发明。其他关于思潮的改动也都是基于这个目的发展而来的。

Legalism spread. It's an institution that distinguishes those who inherited the heritages of classical civilizations from those who did not.
法制主义思潮的传播地图。是否传承古典时代文明遗产是有无这个思潮的区分点。
The idea was quite simple. Basically, there’d be a set of conditions which when satisfied would give some institution progress gain to some provinces. The gameplay would be in setting up your nation to satisfy those conditions.
想法很简单：总地来说就是，如果满足一系列条件，某一思潮就会在某些省份增加一些进度。玩法上，你需要将你的国家导向满足这些条件的道路。
However, my idea quickly faced one major problem, which was that no progress could be given to an institution that is not yet enabled. If an institution was to be enabled, then it incurred a tech cost penalty on every country that did not embrace it. Obviously, having something like a +1000% tech cost penalty from game start was not very desirable to say the least.
然而，我的想法很快就遇到了一个大问题：如果一个思潮没有出现，那也就没办法给这个思潮增加进度。如果一个思潮已经出现了，那它就会给所有没有接纳它的国家提高科技成本。显然，如果开局的时候就有+1000%科技成本之类的东西，那说得客气点，这就有点难顶了。
To remedy that, I removed the tech cost penalty from all institutions and instead greatly increased the tech cost malus from tech advancement. That made it so that as you advance through tech, advancing to the next level becomes more and more costly. Then, as a final touch, I made all institutions give tech cost reduction when embraced.
为了解决这一问题，我移除了所有思潮的科技成本惩罚，并反其道而行之，提高了科技随科技等级增加的成本。这样一来，你会在科技发展的道路上越来越难以取得突破。最后一步则是给所有思潮在接纳时提供一个科技成本减免。

Commercialization spread. It represents the beginning of new type of civilizations, of commercial revolution and its embracement.
商业化思潮的传播地图。它代表国家接纳了一种新的文化：商业革命。
This had many advantages. One was that I could now enable all institutions from the start without having to worry about giving countries a ridiculous amount of tech cost penalty. That basically enabled the whole institution system for me.
这样做有很多好处。首先一点就是，我可以在开局的时候就启用所有思潮，而不用担心它会给所有国家带来巨难顶的科技成本惩罚。如此一来，整个思潮系统就由我掌控了。
Another advantage is its fluidity. It makes it so that low level techs can be accessed even by countries that lack some vital institution. More institutions become necessary only as you reach higher techs.
另一个好处是它不再生硬。比如，就算是缺少某些关键思潮的国家也可以很容易地研究低级科技。只有在你的科技等级达到一定程度时，才需要接纳更多的思潮。

Various conditions for institutions are stored within their respective descriptions.
思潮的各种条件都放在了它们各自的描述中。
Once this new model was implemented, it didn’t take much more to implement the institution invention system. It works like this
在实现了这个新模型后，加入思潮发明系统就易如反掌了。它的工作机制如下所述。
- Every institution has a global condition, which must be true before any country can start inventing it.
- 每个思潮都有一系列全局条件，只有在满足全局条件之后，任意国家才能发明它。
- Every institution also has a country condition, where only the countries that satisfy them can invent it.
- 每个思潮还有一系列国家条件，只有满足这个条件的国家才能发明它。
- Lastly, every institution has a province condition, which the province must satisfy.
- 最后，每个思潮还有一系列省份条件，省份必须满足这些条件。
- For every institution whose global condition is true, every province that satisfies the province condition whose owner also satisfies the country condition will begin to gain institution progress.
- 对于任何满足全局条件的思潮，思潮进度会在任何满足了省份条件，且拥有者满足了国家条件的省份中增长。
- When the institution progress reaches 100%, the institution becomes present in the province, like how it’s done in vanilla.
- 和原版一样，当省份思潮进度达到了100%，思潮就出现了。
- Provinces ability to reinvent what is already invented is not in any way hampered by other provinces having invented it first. The progress does not go away, and the province can still continue to invent it.
- 就算任何省份首先发明了思潮，也不会阻止其他省份重新发明之。任何进度都不会消失，而是继续增长。
- Every province that meets the ‘spread’ conditions for an institution can receive the institution through an enhanced vanilla spread model, using border-to-border spread, trade center based spread (in any direction, if a merchant is present or its a home-node), and overlord to colonial spread.
- 任何满足了一个思潮的「传播」条件的省份，都可以从其他省份那里接受思潮传播。这是一个改良后的原版传播模型，包括接壤传播、贸易中心传播、宗主国向殖民地传播等等。
That’s about it for today. Peace.
今天差不多就这样了，祝你平安。

~完~
一些其他情报：
图中可以看到，3.0暂定的思潮有：法制主义、商业化、大众教育、哥伦布大交换（殖民地货物输入）、科学方法、全球贸易、学术和民族主义。
3.0没有艺术中心了。

中西非可能是缺乏资料，像2.0做人口的时候，非洲资料就很难收集