写在前面：我计划在后面同步跟进冰霜巨人发在红迪网上的讨论话题(不过因为我只打合作，可能会对一些对战上的术语理解有误，欢迎指正)
正文：
Imagine taking the "class"-based Starcraft 2 Coop and turning it PvP. What would work, what would suck? From the last year of discussion topics, I get the feeling this is something Frost Giant has been toying with. As we wait for more information, I would like to explore this path a bit.
想象一下，将基于“职业”的
星际争霸2
:合作模式变成PvP模式。什么会有用，什么会没用?从去年的讨论话题来看，我觉得这是冰霜巨人一直在玩弄的东西。在我们等待更多信息的时候，我想对这条道路进行一点探索。

Build order complexity：
The main difference between Coop commanders and standard is the diversity of units. The depth and complexity of the tech trees and possible unit compositions for each commander is much smaller for coop than standard, which is the basis of any build order. For coop, there are 7-12 units available vs ~17 units available standard, and the choices made in coop are fewer in general. On the other hand, there are 18 commanders in coop vs 3 races in standard, creating a vastly different landscape of possible encounters in a match.
构建顺序复杂性：
合作指挥官和对战种族的主要区别在于单位的多样性。科技树的深度和复杂性以及每个指挥官可能的单位组成都比对战种族的要小得多，而对战种族是任何建造顺序的基础。对于合作模式，有7-12个可用单位对战则有 ~17个可用单位，并且在合作模式中所做的选择通常较少。另一方面，有18个指挥官在合作模式和3个种族在对战模式，在比赛中创造一个可能遭遇的截然不同的场景。

Going up against a player with inflexible scope on their build order would impact the gameplay for many players, and I would expect all non-beginner SC players would find the reduced complexity could be detrimental to the game. However, having a better understanding of what you could go up against from the class you're presented with, might make it easier for some to get a grasp of the dynamics of counters and compositions. Obviously a simpler tech tree would make the game more inclusive, but it could be possible to offset the reduced complexity entirely by making each class correspond to an early choice of build order. One class could fit the early playstyle of zergling rush with expansion to hydra or mutalisk, another could be defensive and allow for either canonrush-like tactics or defensive powering up. Even with reduced unit availability and tech trees, it would be possible to execute the locked builds very differently. At the same time, I feel like the later commanders of coop were allowing a much larger complexity through upgrades. Stetmans frantic zerg army could become a very different beast depending on the investments made, and Mengsk could adapt quickly with reactive weaponselections.
对许多玩家来说，与一个在建造顺序上不灵活的玩家对抗会影响游戏玩法，我认为所有非初级SC玩家会发现降低复杂性可能对游戏有害。然而，从你所面对的流派中更好地理解你可能面临的挑战，可能会让一些人更容易掌握动态的克制关系和部队构成(大概是这么个意思吧)。显然，更简单的科技树会让游戏更具包容性，但也有可能通过让每个流派对应于早期的建造顺序来抵消复杂性的降低。有一种流派可以适应早期的rua狗游戏风格，可以扩展到hydra或muttalisk，另一种流派可以是防御性的，可以采用类似于修地堡的战术或防御升级。即使在单位可用性和科技树降低的情况下，可以以非常不同的方式执行锁定的构建(？)。同时，我认为后出指挥官可以通过升级变得更复杂。斯太台特曼疯狂的虫族军队通过往不同方向上花费资源可以变成非常不同的野兽，而蒙斯克可以快速切换武器(刁民发枪)来适应。

Synergies and balance：
As we have seen through the years of Starcraft, balance is hard to get right, even with three races. What if the 18 commanders of SC2 coop were thrown against each other, how in the world would you balance that? Overall, the core attribute structure of Starcraft counters is transferred nicely to commanders, although the drawbacks of some races are exaggerated. For example, Kerrigans is completely reliant on Zerglings for most of a match, and even though they can receive an armor upgrade, they will melt against most AoE, especially those with bonus against light units, such as Colossus. I have found it is interesting to unpack the buildorder matching the unit composition of the AI, but I don't think the same dynamic would be as transferable to PvP.
协同效应和平衡：
正如我们在《星际争霸》中所看到的那样，即使只有三个种族也很难平衡。如果合作模式的18个指挥官互相攻击，你将如何平衡?总的来说，星际争霸中克制的核心属性结构很好地传递给了指挥官，尽管有些种族缺陷被放大了。例如，凯姐在大部分游戏中完全依赖于狗，即使他们可以获得护甲升级，他们也会在面对大多数AoE时融化，特别是那些对轻甲单位有额外伤害的AoE，如Colossus(我没去对照英文技能表，但我猜是电兵的电)。我发现解开与AI单位组成，建造相匹配的单位是很有趣的，但我不认为同样的动态可以移植到PvP中。

It seems to me, that one player would sometimes be locked into the only counter against a certain composition, caused entirely by the reduced unit types. It might be possible to find a balance by reducing the general bonus of attribute counters or tinkering with the composition of attributes, but this might remove too much flavour of the units themselves. Another avenue to offset this potential imbalance would be by through team play, as discussed by Frost Giant. Players could pick a set of classes which complement each other, effectively doubling the unit type availability. On the other hand, this would also increase the complexity of balance, and the metagame developer-vs-players whackamole would go bananas. I wouldn't personally be a fan of the time investment required to keep up with the "current meta" all the time, and much prefer the gradual drift of nerfs and boons we've seen in Starcraft so far.
在我看来，玩家有时会被锁定在对抗特定组合的唯一反击中(经典雷诺飘大和)，这完全是兵种减少造成的。也许可以通过减少属性克制的加成或修改组合达到平衡，但这可能会过多地削弱单位本身的特色。正如Frost Giant所讨论的那样，另一种消除这种潜在失衡的方法便是团队合作。玩家可以选择一组相互补充的流派，有效地增加单位类型的可用性。另一方面，这也会增加平衡的复杂性，并且游戏开发者与玩家的矛盾也会增加。我个人并不喜欢花时间去保持“当前元”(救命，有人知道这是什么吗)，我更喜欢我们在《星际争霸》中所看到的逐渐削弱和增强。

Dynamic or fixed roster：
Most of the balance issues could be alleviated with a smaller set of classes, say five to eight. But it seems to me, that a major appeal of Coop has been the (as of the creation of Frost Giant abandoned) expanding roster of commanders, and how would you even set a limit on it then? As any RTS would rely heavily on the competetive scene, the aforementioned balance issues should be a top priority, and to me it would just make so much more sense to start with a small subset of classes, and explore the balance from there. In most games with classes, like DOTA or Apex Legends, knowing the abilities of the available classes can make or break the experience of play. With an RTS, each different class adds so many different units and abilities, that it would take a considerable amount of time to understand even one. This may be fun and challenging, and the constant suprises would keep the game fresh for longer, but for a competetive game it might be alienating that there is so much you need to learn and so many opportunities for mechanical griefing in the unknowns you might encounter. I still hate the canon rush, and it would be a serious pain to learn counters to even more cheesy strategies, which I am sure would become much more prevalent in a diversified class ecosystem.
动态或固定名册：
大多数的平衡问题可以通过一组较小的组来缓解，比如说5到8个。但在我看来，合作模式的一个主要吸引力在于(在创建Frost Giant Disserved时)不断扩大的指挥官名册，那么你会如何对其进行限制呢？由于任何RTS都会严重依赖竞争场景，上述平衡问题应该是头等大事，对我来说，从一小部分课程开始，并从那里探索平衡会更有意义。在大多数有职业的游戏中，如DOTA或Apex Legends，了解可用职业的能力可以决定游戏体验的成败。对于RTS，每一个不同的流派都会增加很多不同的单位和能力，甚至需要相当长的时间才能理解其中的一个。这可能很有趣，也很有挑战性，不断的惊喜会让游戏保持更长时间的新鲜感，但是对于一个竞争性的游戏来说，你需要学习的东西太多了，在你可能遇到的未知事物中有太多的抱怨的机会，这可能会让人疏远。我仍然讨厌修地堡，学习更低俗的策略(大概是指狗招吧)将是一种极大的痛苦，我相信在一个多样化的阶级生态系统中，这种策略将变得更加普遍。

So, would it even make sense to explore the Coop model of multiple, simplified commanders, or would it be too volatile or boring? I for one love the mechanical exploration the Coop commanders have provided, and would be really interested in seeing it as a PvP game, but it might be a risky move that would only reveal its true form after years of development. Coop works because it is safe and whimsical, while PvP is brutal and pure. Do you believe it would be possible to create a hybrid of the two?
所以，探索多人或者简化的合作模式是否有意义，或者它是否太不稳定或无聊?我很喜欢合作指挥官所提供的机制探索，也很有兴趣将其视为一款PvP游戏，但这可能是一个冒险的举动，因为在经过数年的开发后它才会呈现出真正的形式。合作模式之所以可行，是因为它安全且异想天开，而PvP模式则是残酷且纯粹的。你认为有可能创造出两者的混合体吗?

（垃圾贴吧怎么一层楼只能发这么点字符）

冰霜巨人现在在干嘛

pve不用讲平衡，搞成pvp会丢失很多有趣的机制

冰巨。。等吧，先看看能不能成

但愿是好事。

md，就离不开pvp这条路了是吧

他们没有星际的版权，不怕被暴雪给告了吗？

双打/单刷比赛也有很多玩的地方