安装 Steam
登录
|
语言
繁體中文（繁体中文）
日本語（日语）
한국어（韩语）
ไทย（泰语）
български（保加利亚语）
Čeština（捷克语）
Dansk（丹麦语）
Deutsch（德语）
English（英语）
Español-España（西班牙语 - 西班牙）
Español - Latinoamérica（西班牙语 - 拉丁美洲）
Ελληνικά（希腊语）
Français（法语）
Italiano（意大利语）
Bahasa Indonesia（印度尼西亚语）
Magyar（匈牙利语）
Nederlands（荷兰语）
Norsk（挪威语）
Polski（波兰语）
Português（葡萄牙语 - 葡萄牙）
Português-Brasil（葡萄牙语 - 巴西）
Română（罗马尼亚语）
Русский（俄语）
Suomi（芬兰语）
Svenska（瑞典语）
Türkçe（土耳其语）
Tiếng Việt（越南语）
Українська（乌克兰语）
报告翻译问题
讨论规则及指引
举报此帖
1) RNG in computer games are pseudo RNG. However the entropy used in such applications is random enough fro this application. We are not in the esoteric realm where massive amounts of randomness is required from the pRNG like is required in things like cryptography or in data crunching massive amounts of data. Any basic computer pRNG is totally fine for gaming purpose and is 'random enough'.
2) The above example you give has little to do with an RNG, it's simply a system by which you 'smooth out' the OUTCOME. The RNG is still an RNG.
As has been said in every otehr game with an RNG
The RNG doesn't hate you
You jsut hate statistics.
RNGs in comptuers are usually based on a formula that spit out random numbers. The 'seed' is simply the initial value you use to generate the list of numbers. These are useful so you can replicate a random set of numbers in the situation where you might need a repeatable, but still 'random' data set. Some formulas are 'better' than others for randomness.
pRNG algorithms have varying levels of entropy. And ifyou're relaly really really realy into math you can argue literally your entire life about which algorithms are better and how much entropy is needed, and how to get more entropy for specific applications like encryption, etc etc etc.
For the purposes of 'to hit rolls in a RPG' any pRNG algorithm is totally fine.
And note 'random' does not mean getting 8 heads in a row is impossible. In fact it's something that SHOULD happen in a truly RNG system.
From a GAMEPLAY perpsective there is a differnce between using the RNG, and what a player's PERCEPTION of the outcome should be.
http://www.shacknews.com/article/62807/sid-meier-and-rob-pardo
Humans suck at math. They suck even more at probability. Users often do not 'like' random outcomes. ANd such, sometimes you design system to actually REMOVE randomness because at the end of the day, sometimes 'fun' is more important than 'true math'.
Really? 'cause, I think the chance is lower than that. I think you've taken that number out of thin air.
hehe, so true
Not exactly.... The Damage Resistance test is not "Dodge" persay. It is a negative modifier to the successes you got on a roll. It varies as it must be rolled as well and can fail.
Clustering Illusion.
A quick check with IDA says that the game uses whatever Unity3D ships with, and google says that Unity3D ships with xorshift128.
As far as PRNGs go, xorshift128 is fast but has a few flaws that show up in statistical testing(0). But since this is neither a numerical analysis app nor something that requies cryptographic random numbers, it's sufficient (and I've seen much worse be used).
*shrugs* I don't care enough about this to dig up the actual hit calculation code/percentile display code to see how they use the random numbers they get, but the PRNG itself is fine.
(0): F. Panneton and P. L'Ecuyer, ``On the Xorshift Random Number Generators'', 2004
So far for me this is the case you want a unforgiving RNG play Blood Bowl or Xcom. This games RNG I have hardly ever missed on normal difficulty.
yeah, who cares if the mechanics are broken! as long as the dialog is great!
we all know that game can either have one or the other, not both
At least it's not XCOM, where 100% chance to hit means a 50% chance to miss.
Praise the lord, at least someone knows what they're talking about, unfortunately I doubt you have access to the source code which may be "interpreting" the results of the PRNG.
Well if you want to talk probability then you presumably know that, as the number of isolated events increases, the chance that some will show highly improbably results approaches 1... or, to put it another way, with the number of people playing this currently a couple of people noticing an extended runs of unlikely misses is not just possible, but in fact almost certain. To put it another way; sorry, but you've probably just been on the low end of the bell curve.
This kind of discussion inevitiably comes up about every single game with visible "hit rates", to the point where some games (especially MMOs) actually have hidden systems that detect if a player is having a string of misses, and invisibly increases the chance of subsequent attacks hitting so it feels fairer... even though, in actual fact, it leads to a random probabilistic system that isn't truly probabilistic OR random. This even leads to gaming the system; there're certain games where people have figured out how many misses lead up to a 100% hit, then "farmed misses" until they've hit the golden number for a guaranteed hit before making a critical attack.