骑马与砍杀2:霸主 开发日志——采访人工智能程序师 Ozan Gümü?

新疆35选7直播 www.jket8.cn  316e725df1d248758f358729425bda71ef3ac307.jpg

 

Title:与Ozan Gümü?的问答环节

 
原文链接:https://www.taleworlds.com/en/Games/Bannerlord/Blog/68
 
翻译:[email protected]骑马与砍杀中文站
 
Dev Blog 12/07/18
 
Greetings warriors of Calradia!
卡拉迪亚的战士,你好!
 
Before we dive into this week’s blog post, we would like to take this opportunity to introduce you to our new Steam Creator page. The Steam Creator page acts as a hub for all things TaleWorlds on Steam. Make sure to follow the page to ensure that you don’t miss out on any news or announcements regarding our company and our games!
 
在深入这周的更新日志之前,我们想借此机会向您介绍我们全新的STEAM“开发者与发行商主页”,它是TaleWorlds在Steam上发布一切信息的中心媒介。如果不想错过任何有关我们公司与游戏的新闻或通知的话,请确保持续关注以下页面。
 
https://store.steampowered.com/developer/taleworlds
 
Mount & Blade II: Bannerlord is a bit different from other sandbox games. There are some in which the story isn’t linear, so you can more or less play it in any order while you explore the world. In others, there is no such thing as a broad story that encompasses the whole game, or if there is, it is less important than the world itself and its inhabitants: you are free to roam to discover them. What we intended to do with the original Mount & Blade, and what we are working hard to do even bigger and better in Bannerlord, is to create a game where you get to create your own story from scratch, where you make your own decisions and live through the consequences. In this week’s entry of our dev blog, we talk with one of the people responsible for creating this magic: Campaign AI Programmer, Ozan Gümü?. If we think of Mount & Blade II: Bannerlord as a “box of tools” you can use to create and live your own story, he’s one of the people who help to create and shape those tools.
 
《骑马与砍杀2:霸主》与其他沙盒游戏略有不同,某些故事并非线性的,因此在探索游戏世界时,你几乎可以随意安排主次推进游戏发展。此外并没有一个宏大的故事围绕整部游戏展开,换句话说即便有的话,故事相比游戏世界本身以及NPC来讲显得并不那么重要:你可以四处漫游找到故事。在《骑马与砍杀》原版中我们打算处理的。并在《霸主》中我们致力于尽善尽美的是创造一款玩家可以白手起家、自由决定并承担后果的游戏。在本周的更新日志条目中,我们与负责创建这一复杂项目的战役人工智能程序师Ozan Gümü?对话。如果我们把《骑马与砍杀2:霸主》想成是一个玩家可以创作并主导故事的“工具箱”,那么Ozan Gümü?就是众多帮助建立并塑造工具的成员之一。
 
NAME
Ozan Gümü?
 
FROM
?zmir
 
JOINED TALEWORLDS
2009
 
EDUCATION
Computer Engineering, Bilkent Üniversitesi
 
OFFICIAL JOB DESCRIPTION
Campaign AI Programmer
 
姓名
Ozan Gümü?
 
城市
伊兹密尔
 
加入TALEWORLDS时间
2009
 
教育背景
比尔肯特大学计算机工程
 
官方工作描述
战役人工智能程序师
 
WHAT DO YOU NORMALLY DO DURING YOUR DAY?
“I can answer this question in two ways: what I do now, and what I did in the past. In the past, I was generally trying to build our living sandbox world, which is almost a simulation of the real world. There are wars, enemy and ally kingdoms, lords with different characteristics and other aspects which mirror the real-world Middle Ages... From time to time, new armies will gather and try to attack the weak points of their rivals, capture rival fiefs or raid enemy villages. At the same time, caravans are carrying goods from where they are cheap to where they are expensive, making profits for the owner of the caravan (who can be a well-known merchant in a town, or even the player themselves). Each village produces new raw materials each day and villagers carry these resources to connected towns. Each day, the residents of towns are consuming these resources and paying taxes. The collected taxes are then given to the clan which owns the fief. While villagers are carrying their goods to town, they are vulnerable to attack from bandit groups and they can be killed and their items stolen. Each night, new bandit groups are spawned around hideouts. This means that even bandits can damage the economy of a clan if they populate a hideout which is close to a village.
 
All of these elements come together to create the sandbox world and present players with activities and objectives to pass the time. Each game element is connected to other somehow and we try to make the world as realistic as possible to make players feel like a real lord in Calradia.
 
More recently, with the majority of the coding for the simulation already complete, most of my time is spent fixing bugs that are reported by our QA team.”
 
你平时做些什么?
“我可以从两方面回答这个问题:我现在做什么,以及我以前做了什么。以前我一般在尽力构建鲜活的沙盒世界,这几乎是对现实世界的模拟:战争、敌人、盟国、不同性格的领主以及其他反映中世纪现实世界的方方面面••••••有时新的军队会聚集在一起,试图攻击对手的弱点、占领其领地或者劫掠敌人的村庄。与此同时,商队也将货物四处运送贱买贵卖,为商队所有者谋取利润(商队所有者可能是镇上有名的商人,或者是玩家本人)。每天村庄都生产新的原材料,村民将这些资源运送到相关的城镇。城镇居民每天都在消耗这些资源并缴纳税款,征收的税款交给拥有封地的封臣家族。当村民运送货物进城时,他们很容易受到匪徒的袭击并被劫掠一空、性命难保。每天晚上新的匪徒会出现在藏身处周围,这意味着如果他们在靠近村庄的藏身处定居的话,即使是土匪也能破坏封臣家族的经济。
 
这些元素共同构建了沙盒世界,并向玩家提供活动与目标来打发时间。各个游戏元素都以某种方式与其他元素相互联系,我们试图让游戏世界尽可能的真实,以便玩家觉得自己是卡拉迪亚一位真正的领主。
 
最近由于模拟的大多数代码已经完成,我将大部分时间用在修复我们的问答团队报告的游戏错误上。”
 
WHAT DO YOU LIKE THE MOST ABOUT BANNERLORD?
“In Bannerlord, the power of a kingdom doesn’t come from just its total manpower count, but also from the economic power of each clan within the faction. Through taking different actions, players and kingdoms can win wars just by restricting the income of their rivals. In Warband, this side of the game was quite weak when compared to Bannerlord because NPC lords never suffered from a shortage of funds. However, achieving this was quite difficult because we needed to create an economically balanced game world (the total amount of money in circulation shouldn’t increase too much over time).
 
In addition to this, fiefs can be developed through projects. We have lots of different projects which affect different things positively, and managing these issues helps to make the player feel like the lord of this fief.”
 
你最喜欢《霸主》的哪些方面?
“在《霸主》中,王国的实力不仅来自其人力总量,还来自派系中各个家族的经济实力。通过采取不同行动,玩家与王国可以通过限制对手收入赢的战争。与《霸主》相比,《战团》这一特点显得微乎其微,因为NPC领主从未遭受过资金短缺的困扰。然而实现这一目标相当困难,因为我们需要创建一个经济平衡的游戏世界(流通中的货币总量不应该随着时间的推移而增加太多)。
 
除此之外,可以通过工程开发封地。我们有很多工程可以积极地影响不同的事情,管理这些事务有助于玩家感觉自己是这个封地的主人。”
 
WHAT'S THE MOST DIFFICULT THING THAT YOU SOLVED SO FAR, DURING THE PRODUCTION OF BANNERLORD?
“The first rule for creating a realistic world for our game was to make NPC lord and kingdom armies make rational decisions. We had to ask ourselves questions such as: after one army has gathered and laid siege to a town, what should the besieged town’s faction do in response? The answer may seem quite easy, especially if their army is patrolling in the area, however, what happens if the army is involved in an offensive siege of its own? Should they abandon their siege to head home and defend their own lands?
 
When making these kinds of decisions a lot of factors come into play. Distance, army power, garrison size, nearby parties, other armies, the strategical importance of fiefs... Ideally, the most logical decision should be found by the AI and their army should do this. If parties give irrational decisions in critical moments like this then players will notice this and start to interrogate our game world. Also, if players find situations in which the AI cannot react to or solve cleverly then they could misuse it. For example, if raiding a village alerted all AI lords and most of them travelled to the village, then other parts of their kingdom would be exposed to attacks. Players would notice this and the end result would be that they would be forced into using this strategy every time, which would negatively affect their gameplay experience. Solving these kinds of issues and not leaving things open to abuse by the player is a difficult problem for us, but it is something we work hard to achieve.”
 
在《霸主》的制作过程中,迄今为止你解决的最大困难是什么?
“为游戏创造真实游戏世界的第一个原则是让NPC领主和王国军队做出合理的决定。我们必须问自己这样的问题:当一支军队聚集起来围攻城镇后,这个被包围的派系该如何回应?答案看起来可能很简单,特别是当被包围派系军队在该地区附近巡逻时。然而如果被包围派系军队正卷入针对敌人的攻城战中会发生什么呢,他们是否应该放弃围攻敌人的城市,转而赶回去保卫自己的土地呢?
 
众多元素作用于这类决定:距离、军队实力、驻军规模、附近友军、其他军队、封地的战略重要性等等。最理想的状况是:最合理的决定应该由人工智能发觉,电脑的军队应该做到这些。如果电脑在关键时刻做出不理智的决定,玩家就会注意到这点,并开始质疑我们的游戏世界。另外,如果玩家发现人工智能不能反映或机智地解决问题,玩家可能会滥用AI缺陷。例如,如果劫掠村庄的行动引起了AI领主的警觉,大部分AI领主赶往村庄的话,他们王国的其他部分就会因为缺少防卫容易受到攻击。玩家会注意到这点,最后他们会被迫每次使用这个策略,这将对其游戏体验带来负面影响。解决这类问题并杜绝玩家滥用AI失误,对我们而言是一个困难的问题,但这正是我们努力实现的。”
 
WHAT DO YOU CURRENTLY WORK ON?
“Currently, I am working on income and expense issues. In Bannerlord, clan wealth is mostly controlled by the clan leader, and all taxes are sent directly to them. However, the clan leader is also responsible for nearly all of the expense issues too (except for some minor personal expenses like recruiting troops and purchasing food). Each clan member is expected to deal with these minor expenses themselves, however, clan leaders can choose to gift money to their clan members. There are actually a lot of additional mechanics surrounding clans but it would be too much to explain right now!
 
My main area of responsibility is the simulation of the game world. The game is quite huge in scale and the simulation is open to many different kinds of bugs, many of which will only occur is certain combinations happen, so half of my time is dedicated to fixing these bugs. I actually prefer bug-fixing periods over implementing new things!”
 
你目前在做些什么?
“目前我正在处理收支问题。在《霸主》中,封臣家族财富主要由家族首领控制,所有的税收都直接发给他们。” 然而,家族首领也负责了几乎所有的费用问题(除了一些如征募军队和购买食物之类的个人开销)。每个家族成员都要自己处理这些细小开支,家族首领也可以选择给他们的家族成员送钱。事实上围绕着家族还有很多其他的机制,但是现在要解释的话实在太多了!
 
我的主要职责是模拟游戏世界。游戏的规模相当宏大,并且模拟有出现各种游戏错误的可能性,许多只会在某些组合时发生,所以一半的时间我在致力于修复这些游戏错误。事实上,比起实现新事物,我更喜欢修复游戏错误这段时光。”
 
 
WHAT FACTION DO YOU LIKE THE MOST IN BANNERLORD?
“I think Battania is so characteristic when compared to the other factions. Their architecture and equipment draw me closer to them. Also, their culture is more closely connected to wildlife, which I also like.”
 
你最喜欢什么派系?
“我认为相比其他派系巴旦尼亚更具特色,他们的建筑与装备深深吸引了我。同时他们的文化与野生动物紧密相连,这也是我所喜欢的。”
 
MORALE IN BATTLES: WILL THERE BE A CHANCE FOR A BATTLE THAT DOES NOT END IN A TOTAL BLOODBATH ON EITHER SIDE?
“Party morale is a critical variable. It decreases massively if the leader does not pay wages regularly or if the soldiers cannot find food to eat. Leaders have two main responsibilities: wages and food. Other than this, travelling long distances without entering a town makes soldiers tired and decreases their morale. Some strategies, like leaving soldiers behind to run away from strong parties also make men unhappy. When you enter a battle with troops who have low morale, after taking some casualties your men can lose their courage and start to flee. This means that a battle can be lost even if most of your troops are still alive. These deserting troops go on to form a new deserted party on the campaign map which the party leader can no longer control. Thankfully, there are lots of different ways to keep party morale high!”
 
战役中的士气表现:双方是否有机会不以血战结束战役?
“部队士气是一个关键变量。如果领主不定期发放军饷,或者士兵缺乏食物补充,士气则会大幅下降。领导者有两个主要职责:发放军饷、提供食物。除此之外,长途跋涉缺少在城镇休息会使士兵们感到疲惫并降低他们的士气。某些诸如留下小部分士兵掩护其他人逃离敌方大部队的策略也会使己方部队士气消沉。当你的部队士气低落时你参与战役,在遭受伤亡后你的部队可能会胆怯地逃命。这意味着即使有大量士兵存活,战役依然会失败。这些逃兵之后在战役地图上组成无人领导的无主部队。值得庆幸的是,有很多方式保持士气高昂。”
blog_post_48_taleworldswebsite_02.png

相关新闻

评论

  • 男子480万元彩票奖金被冒领调查 官司打赢仍难追回 2018-12-14
  • 【理上网来·辉煌十九大】加强产业融合互动 壮大现代产业体系 2018-12-13
  • 青年创新前沿论坛在海淀创业园成功举办 2018-12-13
  • 合工大今年计划招8200人 合肥学院整体列入一本招生 2018-12-13
  • 共产主义劳动不再是谋生手段就是劳动不再是看着就想笑鼓吹私有制下的责权利所谓平滑对接下的为生存而做资本的雇佣劳动奴隶劳动,而是在共产主义公有制里劳动由于一辈子做 2018-12-12
  • 山西新闻网校园小记者站小记者报名表 2018-12-12
  • 漫游宝澄清“无证”传闻 2018-12-11
  • 愿潇湘玉竹版主“(农历)五月平安”、“端午节安康”![微笑]这样的祝福是最确切的[哈哈] 2018-12-11
  • 【奥吉通丰瑞车型报价】奥吉通丰瑞4S店车型价格 2018-12-11
  • 春节我在岗:铁路人的两面人生 2018-12-10
  • 电视剧抱团出海 又有哪些作品走出了国门 2018-12-10
  • 《陕西省县域经济社会发展监测考评办法》解读 2018-12-10
  • 回复@看着就想笑:同时,市场不具有配置全局性长期性战略性资源的功能。 2018-12-10
  • 武警驻滇某部坚持以考促训 现场评判现场亮分 2018-12-09
  • 河砂出厂价飙涨近6倍 涉及房地产、基建等行业 2018-12-09
  • 267| 925| 197| 23| 310| 346| 189| 158| 737| 418|