The Game Outcomes Project, Part 4: Crunch Makes Games Worse 游戏专案为何成功系列之四：加班反而会把事情搞砸 网志版：http://wp.me/pBAPd-qJ 原文网址： http://gamasutra.com/blogs/PaulTozour/20150120/234443/The_Game_Outcomes_Project_Part_4_Crunch_Makes_Games_Worse.php 缩网址：http://tinyurl.com/m7kmuzf 撰文：Paul Tozour 繁体中文翻译：NDark 20150120 译按：本文是一篇统计学专业文章，若有翻译不正确的文句，请以原文为主。 This article is the fourth in a 5-part series. Part 1: The Best and the Rest is also available here: (Gamasutra) (BlogSpot) (in Chinese) Part 2: Building Effective Teams is available here: (Gamasutra) (BlogSpot) (in Chinese) Part 3: Game Development Factors is available here: (Gamasutra) (BlogSpot) (in Chinese) This article is Part 4, and a Chinese translation will soon be available. Part 5 will be published in late January 2015. For extended notes on our survey methodology, see our Methodology blog page. Our raw survey data (minus confidential info) is now available here if you'd like to verify our results or perform your own analysis. The Game Outcomes Project team includes Paul Tozour, David Wegbreit, Lucien Parsons, Zhenghua “Z” Yang, NDark Teng, Eric Byron, Julianna Pillemer, Ben Weber, and Karen Buro. 本文是五篇系列中的第四篇。 第五篇将会在一月底释出。 想要知道问卷的方法论，请参阅部落格页面 :http://intelligenceengine.blogspot.com/2014/11/game-outcomes-project-methodology-in.html 我们问卷的原始资料在此，有兴趣的朋友可迳自取用分析。 "游戏专案为何成功"团队成员包含Paul Tozour，David Wegbreit，Lucien Parsons， Zhenghua “Z” Yang，NDark Teng，Eric Byron，Julianna Pillemer，Ben Weber，及 Karen Buro。 The Game Outcomes Project, Part 4: Crunch Makes Games Worse 游戏专案为何成功系列之四：加班反而会把事情搞砸 Extended overtime (“crunch”) is a deeply controversial topic in our industry. Countless studios have undertaken crunch, sometimes extending to mandatory 80-100 hour work weeks for years at a time. If you ask anyone in the industry about crunch, you’re likely to hear opinions stated very strongly and matter-of-factly based on that person’s individual experience. And yet such opinions are almost invariably put forth with zero reference to any actual data. 延长工时（加班）在我们的产业中充满争议。无数的工作室都曾采取加班的手段，甚至一 周会工作八十到一百小时。假如我们询问业界加班的情形，我们会听到各种基於个人经验 的不同看法。 但这些意见纯粹都是主观意见，缺乏实际数据佐证。 If we truly want to analyze the impact of extended overtime in any scientific and objective way, we should start by recognizing that any individual game project must be considered meaningless by itself – it is a single data point, or anecdotal evidence. We can learn absolutely nothing from whether a single successful or unsuccessful game involved crunch or not, because we cannot know how the project might have turned out if the opposite path had been chosen – that is, if a project that crunched had not done so, or if a project that did not employ crunch had decided to use it. As the saying goes, you can’t prove (or disprove) a counterfactual – you’d need a time machine to actually know how things would have turned out if you’ d chosen differently. 假如我们真的想要用科学化的方式分析加班带来的冲击，我们应该先认知道一点也就是： 分别从各专案的特例来看都是没有意义的。我们无法清楚的知道加班对於专案是否有影响 ，因为我们从事後来看只能看到成功与失败，而不能用另一个方式再执行同一个实验，因 为我们还没发明时间机器。 Furthermore, there have undeniably been many successful and unsuccessful games created both with and without crunch. So we can’t give crunch the exclusive credit or blame for a particular outcome on a single project when much of the credit or blame is clearly owed to other aspects of the game’s development. To truly measure the effect of crunch, we would need to look at a large sample, ideally involving hundreds of game projects. 更进一步，无法否认的有很多成功或失败的专案都可能由加班或不加班的情形下完成。所 以我们不能完全用加班来指责单一专案的成败，因为造成他们的成功与失败可能来自不同 的要素。真正要测量加班的影响，我们应该用巨观，数百个专案的数据来做。 Thankfully, the Game Outcomes Project survey has given us exactly that. In previous articles, we discussed the origin of the Game Outcomes Project and our preliminary findings, and our findings related to team effectiveness and many additional factors we looked at specific to game development. We also wrote up a separate blog post describing the technical details of our methodology. In this article, we present our findings on extended overtime based directly on our survey data. 刚好"游戏专案为何成功"的问卷给了我们这个机会。在先前的系列文章中，我们已经谈论 了游戏专案为何成功这个计划的来由与初步结果，找到与团队效率之间的关系，以及游戏 制作领域的额外要素。我们也在部落格撰写了我们的方法论。 本篇文章中，我们从问卷的数据中持续寻找加班相关的线索。 Attitudes Toward Crunch Developers have surprisingly divergent attitudes toward the practice of crunch. An interview on gamesindustry.biz quoted well-known industry figures Warren Spector and Jason Rubin: 对於加班的不同态度 令人惊讶地，开发者对於加班的态度也很分歧。在Gamesindustry.biz的访问中我们可以 引述Warren Spector 与 Jason Rubin 的说法： “Crunch sucks, but if it is seen by the team members as a fair cost of participating in an otherwise fantastic employment experience, if they value ownership of the resulting creative success more than the hardship, if the team feels like long hours of collaboration with close friends is ultimately rewarding, and if they feel fairly compensated, then who are we to tell them otherwise?" asked Rubin. Rubin说："加班确实糟透了，但假如从团队成员的角度看来，那可能也是一个美妙的团队 经验，假如他们认为创作的结果高过痛苦，假如他们认为与亲密战友长时间的合作是一个 终极的满足，假如他们能够获得回馈，那麽谁又有资格跳出来阻止他们？" […] "Look, I'm sure there have been games made without crunch. I've never worked on one or led one, but I'm sure examples exist. That tells me something about myself and a lot about the business I'm in," said Spector. Spector继续说："...听着，我确信一定有不需要加班就产出的游戏，但我待过的开发案 从未这样，虽然我相信一定有例外。这就是我所作的工作与产业。" […] "What I'm saying is that games - I'm talking about non-sequels, non-imitative games - are inherently unknowable, unpredictable, unmanageable things. A game development process with no crunch? I'm not sure that's possible unless you're working on a rip-off of another game or a low-ambition sequel. "...我不是说那些续作专案，抄袭游戏，而是说完全原创，完全未知的产物。这种游戏开 发案子怎麽可能不加班？除非你正在抄袭或是只是作没有野心的续作。" “[…] Crunch is the result of working with a host of unknown factors in creative mediums. Since game development is always full of unknowns, crunch will always exist in studios that strive for quality […] After 30 years of making games I'm still waiting to find the wizard who can avoid crunch entirely without compromising at a level I'm unwilling to accept.” "...加班是因为我们是与未知参数的艺术媒体战斗过程的产物。只要游戏开发充满了不确 定性，为了追求品质，加班就是必然。但说起来简单，三十年的游戏制作机验後，我仍等 待某个魔术师来告诉我不需要加班就可以做出我可以接受的关卡。" On the other side of the fence is Derek Paxton of Stardock, who said in an interview with Gameranx: 在Gameranx的访问中，Stardock 的 Derek Paxton 表达了另一个角度的看法： “Crunch makes zero sense because it makes games worse. Companies crunch to push through on a specific game, but the long-term effect is that talented developers, artists, producers and designers burn out and leave the industry. "加班一点也没有意义，因为它只会把游戏搞砸。公司会用加班来压缩特定游戏专案，但 长期来看会把有才能的开发者，美术人员，制作人，设计人榨乾，逼得他们不得不离开这 个产业。" “Companies and individuals should stop wearing their time spent crunching as a badge of honor. Crunch is a symptom of broken management and process. Crunch is the sacrifice of your employees. I would ask them why crunch isn’t an issue with other industries. Why isn’t crunch an issue at all game studios? "公司与开发者应该停止把[时间花在加班]当作荣誉的象徵。加班是崩坏管理与流程的病 徵。加班是员工的牺牲品。我问其他产业为何他们不需要加班？为什麽不是每个游戏工作 室都需要加班？" “Employees should see it as a failure. Gamers should be concerned about it, because in the long term the hobby they love is losing talent because of it. Companies should do everything in their power to improve their processes to avoid these consequences.” "员工应该把这件事视为失败。游戏开发者应该认真关心此事，因为长期来看，他们对游 戏的爱会因此而遗失。公司应该要尽其可能改善流程来避免这些後果。" So who is right – Spector and Rubin, or Paxton? 所以谁才是对的？Spector 及 Rubin，还是 Paxton？ [Full disclosure: team member Paul Tozour leads Mothership Entertainment, whose flagship game is being published by Stardock.] [抢先报：Paul Tozour 率领母舰娱乐这间公司，他们的主打游戏会被 Stardock 发布。] In the Game Outcomes Project survey, we provided 3 text boxes at the end that respondents could use to tell us about their industry experiences. Where they mention crunch, they invariably mention it as a net negative. One respondent wrote: 在游戏专案为何成功的问卷中，我们设计了三个开放栏位给回答者，让他们告诉我们产业 的经验。关於提到加班的部分，不约而同地都提出负面的说法。其中一个回应这样写着： “The biggest issue we had was that the lead said ‘Overtime is part of game development’ and never TRIED to improve. As sleep was lost, motivation dropped and the staff lost hope ... everything fell apart. Hundred-hour weeks for nine months, and I'm not exaggerating. Humans can't function under these conditions ... If you want to mention my answer feel free. I'm sure it'd be familiar to many devs.” "我们最大的问题就是管理者说：[加班是游戏开发的一个部分]，而从未试着改善，当睡 眠不足，热情与希望也会随之降低与剥离。我说真的，人类不能在九个月每周上百个小时 的加班这样的环境下工作... 假如你们希望我老实讲，我敢保证其他团队状况都相同。" Another developer put it more bluntly: 另一个开发者说得更难听： “Schedule 40 hours a week and you get 38. Schedule 50 and you get 39 and everyone hates work, life, and you. Schedule 60 and you get 32 and wives start demanding you send out resumes. Schedule 80 and you’re [redacted] and get sued, jackass.” "一周四十个小时的工作，那麽工作效率差不多是三十八小时。如果排五十个小时，那麽 就会得到三十九小时外加痛恨工作，痛恨人生，及痛恨管理阶级的员工。如果排了六十个 小时，那麽会得到三十二个小时的效率外加离职潮。排八十个小时的工作，那麽只会收到 存证信函。" In this article, we will be getting a final word on the subject from the one source that has yet to be interviewed: the data. 这篇文章中，我们会访问我们的案例，也就是那些我们手中的数据资料，对这个题目给一 个总结。 The “Extraordinary Effort” Argument "超凡努力（加班，代表着热情）"理论 We’ll begin by formulating the “pro-crunch” side of the discourse into testable hypotheses. Although no one directly claims that crunch is good per se, and no one denies that it can have harmful effects, Spector and Rubin clearly make the case in the article above that crunch is often (if not usually, or even always) a necessary evil. 虽然没人直接声称加班本身就是好事，也没有人否认它有害，Spector 与 Rubin 清楚地 在前面的说法也证实通常（并非总是）加班是必要之恶。但我们先试着以"加班是好事"这 个论点来做个整理。 According to this line of thinking, ordinary development with ordinary schedules cannot produce extraordinary results. We believe an accurate characterization of this viewpoint from the gamesindustry.biz article quoted above would be: “Extraordinary results require extraordinary effort, and extraordinary effort demands long hours.” 也就是假设这样的思路下去思考，正常工期的开发方式没办法制作杰出作品。也就是相信 Gamesindustry.biz的访问中所提到论点："超凡的成果来自於超越极限的努力（超凡的努 力），而超越极限的努力需要长时间付出，也就是加班。" This position (we’ll call it the “extraordinary effort argument”) leads directly to two falsifiable hypotheses: 1. If the “extraordinary effort argument” is correct, there should be a positive correlation between crunch and game outcomes, and higher levels of crunch should show a measurable improvement in the outcomes of game projects. 2. If the “extraordinary effort argument” is correct, there should be relatively few, if any, highly successful projects without crunch. 这个论点（我们姑且称为超凡努力的论点）直接就会发现两个反证： # 假如超凡努力的论点是对的，那麽在我们的问卷中加班与游戏产出分数上会有正相关 ，越加班，就应该会产出优秀的作品。 # 假如超凡努力的论点是对的，那麽应该不可能发生没加班却高度成功的专案。 Luckily for us, we have data from hundreds of developers who took our survey with no preconceptions as to what the study was designed to test, and which we can use to verify both of these statements. We’ll agree to declare victory for the pro-crunch side if EITHER of these hypotheses remains standing after we put it in the ring with our data set. 很幸运地，我们从问卷中得到数百分开发者的回应，可以透过这些数据来证实这两件事， 尤其是在我们设计之初并没有故意去设计这样的实验。假如数据中告诉我们超凡努力论点 是对的，那麽我们也会宣告加班是胜利之道，并将其放在结论的皇冠上。 Crunching the Numbers We’ll approach our analysis in several phases, carefully determining what the data does and does not tell us. 加班数字 我们接着从数个步骤来分析，小心地看那些数据透露，或没有透露的事。 Our 2014 survey asked the following five questions related to crunch, which were randomly scattered throughout the survey: #“I worked a lot of overtime or ‘crunched’ on this project.” #“I often worked overtime because I was required or felt pressured to.” #“Our team sometimes seemed to be stuck in a cycle of never-ending crunch / overtime work.” #“If we worked overtime, I believe it was because studio leaders or producers failed to scope the project properly (e.g. insufficient manpower, deadlines that were too tight, over-promised features).” #“If I worked overtime, it was only when I volunteered to do so.” 我们在2014年的问卷中问了以下关於加班的问题，在问卷中我们还把它们都随机排列： # 我在专案中超时工作。 # 因为感受到压力，我常常超时工作。 # 我们的团队常常感觉到受阻碍，并陷入无止尽的加班。 # 需要加班的原因是领导层与制作人在时程上搞砸了。（人力不足，估计期限过短，过 度承诺） # 我加班是因为我自愿加班。 Here’s how the answers to those questions correlate with our aggregate project outcome score (described on our Methodology page). On the horizontal axis, a score of -1.0 is “disagree completely” and a score of +1.0 is “ agree completely." 这里是这些答案与总和专案产出分数的相关分数（方法论在我们的部落格已描述），水平 轴是从-1.0的完全不同意，到1.0的完全同意。 Figure 1. Correlation of each crunch-related question with that project’s actual outcome (aggregate score). Each of the 5 questions is shown, as an animated GIF with a 4-second delay. Only the horizontal axis changes.加班相关 的问题与总和产出分数的关联性，每个问题以一个四秒的周期显示出来 The correlations are as follows: -0.24, -0.30, -0.47, -0.36, +0.36 (in the same order listed in the bullet-pointed list above). All five of these correlations have statistical p-values well below 0.001, indicating that they are statistically significant. Note how all the correlations are strongly negative except for the final question, which asked whether crunch was solely voluntary. 关联性依序是-0.24，-0.30，-0.47，-0.36，0.36（顺序如问题序）。五个关联性都有少 於0.001的统计p值。也就是具有统计表徵。注意除了最後一个问题自愿加班之外，这里关 联性都是强烈的负向。 “But wait,” a proponent of crunch might say. “Surely that’s only because you’re using a combined score. That score combines the values of questions like ‘this project met its internal goals,’ which are going to give you lower values, because they're subjective fluff. Of course people who are unhappy about crunch are going to give that factor low scores – and that’s going to lower the combined score a lot. It’s a fudge factor, and it’s skewing your results. Throw it out! You should throw away the critical success, delays, and internal goals outcomes and JUST look at return on investment and I bet you’ll see a totally different picture.” 但加班的支持者可能会说："等等，这一定是因为这里是一个总合分数，包含了内部满意 度，这当然会有负分，因为那是主观意见，加班就是会让人不开心，才会导致总合分数这 样发展，应该要排除在外！我们应该要只看利润的产出分数，一定可以看到不同的结果。 " OK, let’s do that: 那麽我们也从善如流： Figure 2. Correlation of each of the 5 crunch-related questions with that project’s return on investment (ROI). As with Figure 1, each of the 5 questions is shown, as an animated GIF with a 4-second delay. Only the horizontal axis changes. Note that many of the points shown represent multiple coincident points. See our Methodology page for an explanation of the vertical axis scale.五个加班相关问题对上专案利润的关联度，如图一相同，每个 问题以一个四秒的周期显示。只有在水平轴不同。每个点都可能代表重合在一起的点。在 垂直轴的缩放方式请参照我们的部落格网页。 Notice how the lines have essentially the same slopes as in the previous figure. The correlations with ROI are as follows (in the same order): -0.18, -0.26, -0.34, -0.23, and +0.28. All of these correlations have p-values below 0.012. 注意到回归线仍与前一张图相同吗？利润的关联性是：-0.18，-0.25，-0.34，-0.23，及 0.28。全部关联性都有小於0.012的统计p值。 Still not convinced? Here are the same graphs again, correlated against aggregate reviews / MetaCritic scores. 不相信吗？这张图也一样，对上网页分数的关联性： Figure 3. Correlation of each of the 5 crunch-related questions with the project’s aggregate reviews / MetaCritic score (note that the vertical axis does not represent actual MetaCritic scores but is a normalized representation of the answers to this question; see our Methodology page for more info). As with Figures 1 and 2, each of the 5 questions is shown, as an animated GIF with a 4-second delay. Note that many of the points shown represent multiple coincident points. Only the horizontal axis changes.五个加 班问题对上网页分数的关联性（注意垂直轴并非表示MetaCritic真正分数，而只是一个对 问题经过正规化的数值。更多资讯，请看我们的部落格）如同图一与图二，五个答案都以 一个四秒周期的方式显示。每个点都可能代表重合在一起的点。只有在水平轴的参数是不 同的。 The results are essentially identical, and all have p-values under 0.05. 结果一样，全部都具有小於0.05的统计p值。 So if our combined score has a negative correlation with ALL our crunch questions except the one about crunch being purely voluntary (which itself does not imply any particular level of crunch), that means that we’ve disproven the first part of the “extraordinary effort argument” – the correlation is clearly negative, not positive. 总合的分数对上除了自愿加班之外的所有加班问题都是负向的关联。意思是我们能够推翻 超凡努力论点，很清楚，就是没有正相关。 Now let’s look at the second testable hypothesis of the “extraordinary effort argument.” In Figure 4 (below), we’re looking at the two most relevant questions related to overall crunch for a project. The vertical axis is the aggregate outcome score, while the horizontal axis represents the scale from “disagree completely” (-1) to “agree completely.” The black lines are trend lines. As you can see, in both cases, higher agreement with each statement corresponds to inferior project outcomes. 接着来看看我们对於超凡努力理论的第二个辩证。 在下面的图四中，我们取出两个对加班问题中最相关的问题。垂直轴是总合产出分数，同 时水平轴是从完全不同意的-1，到完全同意。黑色的线是趋势线。如你可见，在两个问题 中，越高的同意带来越低的总和分数。 Figure 4. The two most relevant questions related to crunch compared to the aggregate project outcome score.两个相关问题对上总和的产出分数。 We’ve added horizontal blue and orange lines to both images. The blue line represents a score of 80, which will be our subjective threshold for “very successful” projects. The orange line represents a score of 40, which will be our threshold for “very unsuccessful” projects. 我们接着加上了的蓝色与橘色水平线。蓝色线是80，也就是我们主观认定非常成功专案。 橘色线则代表40。也就是我们主观认定非常不成功的专案。 The dots above the blue line tell a clear story: in each case, there were more successful games made without crunch than with crunch. 在蓝线之上的点的分布清楚了代表一件事：多数成功的游戏没加班的数量比加班的多。 However, these charts don’t tell the full story by themselves; many of the data points are clustered at the exact same spot, meaning that each dot can actually represent several data points. So a statistical deep-dive is necessary. We’re particularly interested the four corners of the chart – the data points above the blue line on the extreme left and right sides of each chart (below -0.6 and above +0.6 on the horizontal axis) and below the orange line on the left and right sides. 然而，只是图并不能将细节全盘托出，很多数据点都重合在一起，分不清楚有几个。所以 我们需要再深一步的统计分析。我们对这张图的四个角落特别有兴趣。也就是蓝线以上的 左右端，以及橘线以下的左右端。（水平轴以-0.6及0.6为界线） Looking solely at the chart on the top of Figure 4 (“I worked a lot of overtime or ‘crunched’ on this project”), we observed the following pattern. Note that the percentages are given in terms of the total data points in each vertical grouping (under -0.6 or above 0.6 on the horizontal axis). 独立看图四上半（我在专案中超时工作。）我们观察到後述的模式。注意那些比例是以水 平轴已经切开（为左右两群组）的群组来计算。 We can see clearly that a higher percentage of no-crunch projects succeed than fail (17% vs 10%) and a much larger percentage of high-crunch projects fail rather than succeeding (32% vs 13%). Additionally, a higher percentage of the successful projects are no-crunch than high-crunch (17% vs 13%), while a higher percentage of the unsuccessful projects are high-crunch vs no-crunch (32% vs 10%). 我们可以很清楚地看到非加班的区块成功数量是高於失败数量（17%多过於10%），高度加 班的区块中，失败却高於成功（32%多过於13%）。成功专案中，不加班多於加班（17%多 过於13%）。失败专案中高度加班的情形多过於不加班（32%多过於10%）。 Here’s the same chart, but this time looking at the bottom question, “Our team sometimes seemed to be stuck in a cycle of never-ending crunch / overtime work.” 同样的图表中，我们看下半个问题：我们的团队常常感觉到受阻碍，并陷入无止尽的加班 。 These results are even more remarkable. The respondents that answered “ disagree strongly” or “disagree completely” were 2.5 times more likely to be working on very successful projects (23% vs 9%), while the respondents who answered “agree strongly” or “agree completely” were, incredibly, more than 10 times more likely to be on unsuccessful projects than successful ones (41% vs 4%). 结果更强烈。在不加班（回应强烈不同意与完全不同意）的案子中成功专案是超过不成功 专案的两倍半（23%对上9%）。在高度加班中（强烈同意与完全同意）的案子中不成功的 案子则是成功案子的四倍（41%对上4%） Some might object to this way of measuring the responses, as it is an aggregate outcome score which takes internal achievement of the project goals into account – and this is a somewhat subjective measure. What if we looked at return on investment (ROI) alone? Surely that would paint a different picture. 有些人可能会质疑总和的分数包含了专案的内部满意度，当然就是主观意见。那假设我们 只看专案利润？会有不同的结果吗？ Here is ROI: 专案利润的图在此： Figure 5. The two most relevant questions related to crunch compared to return on investment (ROI).最相关的两个问题对上专案利润的关联性 The first question (top chart) gives us the following results: 第一个问题的结果如此： The second question (bottom chart) gives us: 第二个问题的结果如此： These results are essentially equivalent to what we got with Figure 4 -- the probabilities have shifted a little bit but the conclusions haven't changed at all. The same results hold if we look at MetaCritic scores or any of the other outcome factors we investigated. 结果跟图四本质上相同，机率稍稍有些偏移，但结论相同。假如我们只看网页分数，或其 他产出要素，结果仍然相同。 For further verification, we did a deep-dive statistical analysis of the data in figures 4 and 5, treating the left and right sides of each graph on each figure (all data points < -0.6 and all those > +0.6) as two separate populations and performing a Wilcoxon rank sum test to compare them. 更深验证图四与图五的数据，把每一张图的左右两侧都取出来（小於-0.6与大於0.6的数 据）我们对它们进行曼－惠特尼U考验法的分析。 The p-values of all of these are highly statistically significant, with the top two rows having p-values under 0.006 and the bottom two rows with p-values of 0. 全部的统计p值都有高度统计表徵，上面两列甚至小於0.006，最後一列则是0。 It should be clear that our data set contradicts both of the testable hypotheses that we derived from the “extraordinary effort argument.” But before declaring victory for Paxton and the anti-crunch side, let’s take a look at the counter-argument. 很明显我们的数据与我们从超凡努力论点而推论的两个假设都背道而驰。但在我们宣告 Paxton与非加班派的说法胜利之前，我们再来看看反面论点。 The “Crunch Salvage Hypothesis” 加班补救理论 The counter-argument goes something like this: “Your correlation is bogus, because crunch is more likely to happen on projects that are in trouble in the first place. So there’s already an underlying correlation between crunch and struggling projects, and this is skewing your results. You seem to be saying that crunch causes poorer outcomes, but the causality actually works differently – there’s a third, hidden causal factor (“project being in trouble”) that causes both crunch and lower outcomes. And although crunch helps improve the situation, it’s never quite enough to compensate for the problems in the first place, which is why you get the negative correlation.” 反面的论点意思於此： "这些关联性都是导果为因的，因为那些有问题的专案才会加班，根源就是加班与有陷入 困境的专案就必然关联。加班造成失败的因果关系不存在，而是由其他隐藏的原因所造成 （如造成专案陷入困境的原因），那些原因才造成加班与产出低落。虽然加班能弥补产出 ，但绝不足以弥补到能解决问题的情形。所以当然只得到与总合分数负相关的情形。" This position warrants further investigation. As the Spector/Rubin interview linked above makes clear, there are some developers who are willing to demand crunch even in cases where their projects are not in trouble (“crunch will always exist in studios that strive for quality,” according to Spector), so it’s clear that at least in some cases, crunch is used on projects that are not yet having problems. But the notion that crunch is more likely on struggling projects is entirely plausible. 这个看法需要进一步的调查，如同Spector与Rubin的访问中所说，有很多开发者是愿意加 班，即便他们的专案没有陷入困境。（也就是加班是为了品质，而不是为了追上进度）因 此很显然至少某些例子中，加班并非用在陷入困境的专案中。因此加班是因为专案困难的 论述有点似是而非。 Let’s test this counter-argument. Let’s assume the causation is not A -> B but C -> (A and B), where “A”=crunch, “B”=poorer project outcomes, and “C ” represents some vaguely-defined set of factors representing troubled projects. 我们试着用反证法来论述。定义Ａ是加班，Ｂ是低产出，Ｃ是其他困境专案所来自的原因 。这样的推论应是朝向：Ｃ导致了Ａ加Ｂ（某种原因导致了加班与低产出），而不是：Ａ 导致Ｂ（加班造成低产出） We’ll call this the “crunch salvage hypothesis” – the idea that crunch is more likely to be used on projects in trouble, and that this “trouble” is itself the cause of the poorer project outcomes, and that when crunch is used in this way, it leads to outcomes that are less poor than would otherwise be the case. 这边我们称这个情形是"加班补救理论"，其意指加班是用在拯救陷入困境的专案，而所谓 的困境就是造成专案低产出的原因，当加班出现时会导致产出回填一些（不那麽糟糕）。 We don’t really care about every part of this hypothesis: we’ll simply accept the first two parts (that trouble can arise on a project, and that crunch often happens as a reaction to this trouble) as self-evident truths (although whether they are correct or not isn't really relevant to this article). 我们并不特别在意假设的其中任一部分，我们简单承认两个部分：专案会有问题，加班会 因此问题而发生。这部分确实不言而喻。（虽然不管他们是对错，都与此篇文章无关） What we really care about, and what we can test, is the third part of this hypothesis – that when crunch is used in this case, it leads to outcomes that are less poor than would otherwise be the case. In other words, if a project is in trouble, is crunch an effective response? If the “crunch salvage hypothesis” is correct, then crunch should provide an improved project outcome score beyond what we would expect to see if crunch were not used, all else being equal. 我们真正在意，且我们能够测试的是假设的第三个部分，当加班开始实行，就会导致没那 麽差的产出，至少比不加班好。换句话说，假如专案发生困难，加班是否是一个提高效率 的工具？ 假如"加班补救理论"是正确的，那麽加班就应该提高专案产出，比起没有加班，至少不会 减少。 In order to test this conjecture, we calculated a linear regression model that specifically excludes all 5 questions related to crunch/overtime. We’ ll call this model the “crunch-free model.” 为了证明这个假设，我们透过针对五个加班问题做出线性回归模型，我们称此模型叫做" 不加班模型"。 Figure 6. Correlations for the “crunch-free model” (a linear regression that excludes crunch-related questions) with aggregate game outcome scores.不 加班模型（所有加班问题的线性回归）对上产出数字的关联性。 This “crunch-free model” correlates with our overall outcome score with a correlation value of 0.811 (and a p-value under 0.001). This is, by any measure, an extremely strong correlation. 这个不加班模型对我们的产出分数有着0.811的关联性（小於0.001的p值）。也就是说， 从此来看，有着高度的关联性。 We then computed the crunch-free model’s error term – that is, we compared the actual aggregate outcome score to the predicted outcome score given by the crunch-free model for each response by subtracting the predicted score from the actual aggregate outcome score. A high value indicates that the project turned out better than the model predicted, while a negative error value indicates that the project turned out worse than it predicted. 我们接着计算不加班模型的误差项，也就是说，我们比较"实际"的总和产出以及由"不加 班模型"所算出的"预测"产出分数，实际减去预测。若是数值高，那麽就代表比预测还好 ，如果是负值就代表比实际比预测还差。 If we accept that the crunch-free predictive model is a good predictor of game outcomes (and the extremely high correlation and tiny p-value suggest that it is), then the “crunch salvage hypothesis” tells us that we should expect that it should improve the outcomes of game projects where it is used at least to some tiny, observable extent … and the more it is used, the more it should improve game project outcomes. 假如我们接受"不加班模型"是一个好的产出分数预测模型（实际上高相关与低p值也显示 如此），而加班补救理论就应该透露出加班会增强产出分数，而且是越多越好。 In other words, if crunch works, it should provide a “lift,” and for projects that involved more crunch, we should see a positive error term (that is, game projects that crunched should have turned out better than the crunch-free model predicts), while for projects that involved little or no crunch, we should see a negative error term. 换句话说，假如加班是有用的，它就会往上牵引专案，会得到一个正向的误差项（实际产 出分数应该比不加班模型预测得更好），同时不加班的专案，就应该得到一个负向的误差 项。 So according to this worldview, there should be a clear, positive correlation between more crunch and a greater positive error value for the crunch-free model. 根据这样的看法，在加班与不加班模型的正面误差项上就应该有一个清楚，正面的关联性 。 Here is the correlation for the error term with the answers to each of the two primary crunch-related questions: 下面就是误差项与两个问题的关联度。 Figure 7. The two most relevant questions related to crunch, compared to the error value of the crunch-free model. The vertical axis is the error of the crunch-free model (positive = better than model predicts; negative = worse), and the horizontal axis indicates agreement with each question (-1.0 = disagree completely, +1.0 = agree completely).两个加班相关问题对上不加班模型误 差项。垂直轴是不加班模型的误差项（正值＝超过预期；负值＝相反），水平轴指出每个 问题的同意度（-1完全不同意，+1完全同意） As you can see, there is a slight negative correlation. However, it is not statistically significant (p-value = 0.24 for the upper graph, and 0.1 for the lower one). And even if it were statistically significant, the correlations – at -0.07 and -0.1, respectively – are negative. 如你可见，有一个稍微负向的关联。然而，没有统计表徵（上面的图0.24的p值，下面的 图0.1的p值）。就算勉强要计较，那麽关联性分别是-0.07及-0.1，也就是负向的。 So where the “crunch salvage hypothesis” tells us to expect correlations that are strong, positive, and statistically significant, we see correlations that are weak, negative, and statistically insignificant. 因此"加班补救理论"所推论，应该要有强烈正面统计表徵，实际上却是微小，负面，没有 统计表徵。 Testing all of the other crunch-related questions in this way gives us similar results. 对全部的加班问题的测试的结果都类似。 If we accept the assumptions that went into calculating these correlations, then we must conclude that more crunch did not, to any extent that we can detect, help the projects in our study achieve better outcomes than they otherwise would have experienced … and in many ways appears to have actually made them worse. We are left to conclude that crunch does not in any way improve game project outcomes and cannot help a troubled game project work its way out of trouble. 假如我们接受这些关联性的推论，那麽我们就得到了：加班过多是不能帮助团队更好的结 果（我们所知的产出分数）... 很多情形似乎加班使状况更糟。 因此我们得到结论不管各种情形加班并不能加强游戏产出，也不能帮助有问题的团队解决 困境。 Voluntary Crunch 自愿性加班 But what about when crunch is voluntary? Our analysis has already indicated that a when crunch is entirely voluntary, outcomes significantly improve. Does a lack of mandatory crunch then eliminate the negative effects of the quantity of crunch? In other words, do higher levels of voluntary crunch then turn crunch from a net negative into a net positive? 但当加班是自愿的情形呢？我们的分析已经指出当加班是自愿，产出会显着提升。缺乏强 制性的加班会消除加班的负面影响吗？换句话说，是否高品质的加班就不是负面的？ In short, no. We compared the two extremes of our primary crunch question (we categorized the highest two answers to “I worked a lot of overtime …” as “High” crunch, and the lowest two as “Low” crunch) against our question about whether crunch was purely voluntary (where we condensed all 7 answers into three 3 broad categories -- the top two as “Voluntary,” the bottom two as “Mandatory,” and the middle 3 as “Mixed”). We also compared these categories using Kruskal-Wallis to prove statistical significance. 简短来说，答案并非如此。我们将两个主要的加班问题（我在专案中超时工作被标示为主 要）对上是否自愿的问题，然後将答案分随三群：自愿的，强制的，混合的。我们也把这 些类别用克-瓦二氏检定来证明有统计表徵。 Our analysis shows that although crunch seems to be significantly less harmful when it’s voluntary, low levels of crunch in each case above (voluntary, mandatory, and mixed) are consistently associated with better outcomes than high levels of crunch. 我们的分析显示了即便自愿性加班有统计表徵比较没有伤害，在三个自愿分类（自愿，强 制，混合）中，低量的加班还是比高量加班的产出分数来得好。 What Causes Crunch? The conclusions above led us to ask: what actually causes crunch? The Spector/Rubin interview above clearly illustrates the attitudes that cause at least some developers to demand extended overtime, but we were curious what the data said. 到底甚麽造成加班？ 结论让我们去询问到底甚麽导致加班？先前所说Spector及Rubin的访问中清楚的指出至少 有一些开发者是需要加班，我们很好奇数据怎麽说。 If crunch doesn’t correlate with better outcomes, what does it correlate with? Does it really derive from a desire for excellence, or is it a reaction to a project being in trouble, or do its roots lie elsewhere? 假如加班并没有造成更好的产出，那麽到底加班与甚麽东西是相关联的？是否可以从追求 卓越的慾望而推导出？或是那是团队陷入困境的连锁反应？我们想找到根源。 To find out, we analyzed the correlations of all the input factors in our survey against one another, looking specifically at how factors outside of our group of five crunch-related questions correlated with the five crunch questions. The four strongest correlations with our crunch-related questions were: +0.51: “There was a lot of turnover on this project.” +0.50: “Team members would often work for weeks at a time without receiving feedback from project leads or managers.” +0.49: “The team’s leads and managers did not have a respectful relationship with the team’s developers.” -0.49: “The development plan for the game was clear and well-communicated to the team.” (The three positive correlations indicate that they made crunch more likely; the negative correlation is the one that makes crunch less likely). 为了找出答案， 我们分析了所有问卷中所有输入要素与其他项目的关联性，特别想找出与这五个加班相关 的问题的关联性。与加班问题最强烈的四个关联项目如下： 0.51：专案人员流动率很高。 0.50：团队一段时间没有收到来自团队领导层的回应。 0.45：团队的领导层与团队的开发者没有良好的敬重关系。 -0.49：开发计画很清楚，而且与团队充分沟通。 （前三个正相关指出，越这样的团队，就越有可能加班；最後一个负向的相关则表示越强 大，加班就越少） This seems to indicate that crunch does not, in fact, derive from any sort of fundamental drive for excellence, which would have resulted in higher correlations with completely different input factors on our survey. Rather, it appears to stem from inadequate planning, disorganization, high turnover, and a basic lack of respect for developers. 事实上，这似乎指出加班并未由追求卓越而来（也就是在问卷中的其他问题。）反而是不 透明的计画，没有组织，高流动率，缺乏尊敬会导致加班。 Conclusion: We Are Not a Unique And Special Snowflake 结论：其实这些理论在产业都适用 We should be clear that we are not attempting to write an academic paper, and our results have not been peer-reviewed. Therefore, we walk a fine line between analyzing the data and interpreting it. 我们应该很明白我们并非在写一篇学术文章，这些结论也没有被审核过。因此我们仅是在 分析资料与解释它们。 However, no matter how we analyze our data, we find that it loudly and unequivocally supports the anti-crunch side. Our results are clear enough and strong enough that we believe it’s important to step over that fine line, and transition from objective analysis to open advocacy. 然而，不管我们如何分析资料，我们都发现清楚且无须质疑的反加班结论。结果清楚到能 够让我们站出仅是数据分析之线，拥护着我们的结论。 There is an extensive body of validated management research available showing that extended overtime harms health, productivity, relationships, morale, employee engagement, decision-making ability, and even increases the risk of alcohol abuse. 有更多延伸的管理研究说明加班伤害健康，产量，人际关系，士气，员工雇用，决策能力 ，甚至是酗酒的风险。 An enormous amount of validated management research demonstrates that net employee productivity turns negative after just a few weeks of overtime. Total productivity actually declines by 16-20% as we increase our work days from 8 hours to 9 hours. Even just a few weeks of working 50 hours per week reduces cumulative output below what it would have been working only 40 hours per week – those 10 extra hours of work actually have a negative impact on productivity. All of that while also increasing employee stress, straining relationships, and increasing product defect rates. 有更多的研究证明员工在几个礼拜的加班後产量就变成负的。当每日工作时数到八小时以 上，总产量实际上降低为百分之十六到二十。每周五十小时的工作量会降低效率使累计产 量变为只有每周四十个小时。那多出来的十个小时实际上对产量反而有害。这全都说明增 加员工的压力，会榨乾人际关系，降低良率。 However, the game industry is remarkably insular for such a cutting-edge and successful industry, and it seems generally unaware of this data. We tend to ignore such evidence or blithely assume it doesn't apply to us. As a broad generalization, our industry tends to value industry experience highly while undervaluing fundamental management skills. As a result, we usually promote managers from within while rarely offering the kind of management training that would enable insiders to perform their jobs adequately. 然而，游戏产业就是一个很容易赢家全拿的产业，也因此大家都不在意数据的结果。大家 都试着忽略这样的证据，然後说别人的规则不套用在我们身上。也就是说，游戏产业的团 队会倾向主张产业的经验比较重要，同时管理技能则不重要。因此我们通常很少会拔擢有 管理训练会引导其他人充分发挥的那些员工。 Is it any wonder, then, that we find ourselves completely cut off from the plethora of validated management research clearly showing that crunch is harmful? 难怪，我们会认为自己完全不受那些多如牛毛的研究所说加班有害论所规范。 The hundreds of anonymous respondents who participated in our survey answered various questions about game development factors and outcomes separately and individually, without any real clue as to the broader objectives of our study. Simply correlating their aggregate answers shows overwhelmingly that crunch is a net negative no matter how we analyze the data. It’s not even a case of small amounts of crunch being helpful and then turning harmful; we see no convincing evidence of hormesis. 参加我们问卷的几百分匿名的回应都是各自独立的回答了无数关於产业要素及产出的问题 。他们绝无可能串通知道我们的意图。不管我们如何分析，回答的关联性就直接显示出加 班是净伤害。并非仅是一点点加班就可以对专案有帮助。我们没有看到这样的毒物兴奋效 应（译按：低毒性疫苗反而对人体有益）。 It’s common knowledge that crunch leads to higher industry turnover and loss of critical talent, higher stress levels, increased health problems, and higher defect rates – and quite often, broken or deeply impaired personal relationships. Those who feel that crunch is justified freely admit to knowing this, but they don’t necessarily care about any of these harmful side-effects enough to avoid using it, as they continue to cling to the notion that “extraordinary results require extraordinary effort.” 加班会让公司流动率增加，失去有才能的员工，高压，健康问题，低良率，人际关系崩坏 。那些觉得加班是合理的人其实都知道，但他们刻意忽略这些副作用因此而加班，当他们 持续使用这种兴奋剂来加大产量时，掉落的效率让他们就必须使用更多的兴奋剂（更多的 加班）。 However, this notion appears to be a fallacy, and our analysis suggests that if the industry is to mature, we must cast it aside. 这看法是错误的，我们的分析建议：若产业想要成熟一点，我们必须不再这麽做。 Our results clearly demonstrate that crunch doesn't lead to extraordinary results. In fact, on the whole, crunch makes games LESS successful wherever it is used, and when projects try to dig themselves out of a hole by crunching, it only digs the hole deeper. 我们的结果很清楚地证实加班并不会导致绝佳的产出。事实上，整体来看，加班不管怎麽 做反而使得游戏更不成功。当专案想要透过加班来挖出宝藏时，加班只挖出自己的坟墓。 Perhaps the notion that “extraordinary results require extraordinary effort” is misguided. 也许只有超凡努力才能做出超凡结果这段话误导了我们 Perhaps “effort” – as defined by working extra hours in an attempt to accomplish more – is actually counterproductive. 努力若指的是工时，实际上是无效的。 Our study seems to reveal that what actually generates “extraordinary results ” – the factors that actually make great games great – have nothing to do with mere “effort” and everything to do with focus, team cohesion, a compelling direction, psychological safety, risk management, and a large number of other cultural factors that enhance team effectiveness. 我们的研究指出实际上透露出所谓的超凡结果，也就是让游戏更好并非纯粹只是努力。而 是跟专注，团队内聚力，明确的方向，心理安全感，风险管理，以及增强团队效率的文化 要素有关。 And we suggest that abuse of overtime makes that level of focus and team cohesion increasingly more difficult to achieve, eliminating any possible positive effects from overtime. 我们的建议是：滥用加班就像吸毒一样会让专注与团队内聚力越来越难以做到，加班会逐 渐减少团队的各种正面效果。 We welcome open discourse and debate on this subject. Anyone who wishes to double-check our results is welcome to download our data set and perform their own analysis and contact us at @GameOutcomes on Twitter with any questions. 我们以开放的态度欢迎对於这个题目的讨论。如果你想要验证我们的推论，欢迎下载我 们的资料来分析。欢迎在推特上用任何问题回覆我们。 The Game Outcomes Project team would like to thank the hundreds of current and former game developers who made this study possible through their participation in the survey. We would also like to thank IGDA Production SIG members Clinton Keith and Chuck Hoover for their assistance with survey design; Kate Edwards, Tristin Hightower, and the IGDA for assistance with promotion; and Christian Nutt and the Gamasutra editorial team for their assistance in promoting the survey. "游戏专案为何成功"团队希望能感谢数百名现任开发者及前辈，让这个问卷研究能顺利进 行。我们也同时感谢IGDA生产力同好会的成员Clinton Keith与Chuck Hoover在问题设计 方面的协助；感谢Kate Edward，Tristin Hightower及IGDA协助推广此专案；感谢 Christian Nutt及Gamasutra编辑群对我们问卷的支持。 -- "May the Balance be with U"(愿平衡与你同在) 视窗介面游戏设计教学，讨论，分享。欢迎来信。 视窗程式设计（Windows CLR Form）游戏架构设计（Game Application Framework） 游戏工具设计（Game App. Tool Design ） 电脑图学架构及研究（Computer Graphics） --

※ 发信站: 批踢踢实业坊(ptt.cc), 来自: 220.135.205.117 ※ 文章网址: https://webptt.com/cn.aspx?n=bbs/GameDesign/M.1422077786.A.9D0.html ※ 编辑: NDark (220.135.205.117), 01/24/2015 13:36:49 ※ 编辑: NDark (220.135.205.117), 01/24/2015 13:37:10 ※ 编辑: NDark (220.135.205.117), 01/24/2015 13:37:53 ※ 编辑: NDark (220.135.205.117), 01/24/2015 13:44:15