WoW: The RNG (Random Number Generator) eSport? - Page 6

jimbo229 · 07/23/08, 3:04 PM

It seems to me that anyone denying that RNG has an effect on PvP is wrong. But Bliz has seeming done a good job minimizing the effects of it. RNG will effect players inside a given skill bracket, but the effect is greatly diminished on players outside of that skill bracket. For example, if a 1700 team played a 2200 team, the 1700 team would lose almost every single time, as expected. Sure there could be instances of them winning, but overall they would have a losing record. Whereas if two equal skilled 2200 teams played (assume same classes/specs/gear just to be fair) then RNG might play a huge roll.

Even in cases of same gear, specs, classes, and skill I believe the RNG determining the match is still fairly minimal. Take the sword spec warrior w/ WF for example, he has a 1% chance to get an huge damage spike. Sure that 1% is going to play a roll, but I think the chances of a player making a mistake are much higher than the 1% chance to hit for a ton of extra damage.

As one poster above said, look at poker world tours. I think it is pretty undeniable that poker is more random than WoW (betting aside, purely on card draws). Yet in the world tours you consistantly see professional players at the final tables not first time players. Even most of the people that are not profession at those final tables have been playing consistently for years prior. It becomes pretty clear that a game undeniably based on chance comes down to skill more than anything.

Someone said before that they lost because a warrior got a lucky stun on cyclone, then he resisted an intercept charge, then he dodged a disarm. Sad to say, those things happen, all the time. But if you calculate the chances of that happening they are so low that they do not effect the final outcome.

solbergb · 07/23/08, 7:10 PM

Originally Posted by Vagabond

Solb's Whole position is that in some intangible and unmeasurable way, in some indeterminate length of time before the RNG shit on you, you were actually playing badly. Had you not played like ass, God wouldn't have smote you.

Right. It starts with gear and comp. If a certain warrior+windfury combo puts you both at the mercy of the RNG, both you and the other guy have made choices that put you in that situation.

"Play badly" or "Play well" includes everthing that leads up to the crucial die roll.

A football team in control of a match doesn't "go for it" on 4th and 1. They punt, or field goal. One that is only a little behind will do the same. One that has lost control of the game may go for it, because they have put themselves in a position where they have to take that gamble. Same goes for when poker players go all-in on weak hands, they see it as the best shot to reverse the situation, when they've played badly enough and/or been unlucky enough to be put into that position, where the ante will grind them to death if they don't make a move.

There isn't any randomness when my scrub arena 5-man encounters a formerly 2000 rated team on the first day of the season. We lose. We lose so badly that we barely dent the health on any target in the opposing team. It doesn't matter our comp, or our gear, we're so bad relative to the other team that we'll lose anyway even if every single RNG event breaks our way. "Luck" only enters into it when the teams are reasonably evenly matched, even if one starts out with a significant enough advantage to "normally win" with "correct play".

I don't think it is necessarily a bad thing to have specs/comps out there that are more random than others. Random will cut both ways in the end, and your ability to exploit the good runs and survive the bad runs is the measure of player skill who has chosen a highly variable spec. It's no better or worse than picking a comp that wins against 7 of 10 comps you usually encounter, but sucks totally vs the other 3. You make those choices before you enter the arena. Your final result is based on how well you maximize advantages you gain from those choices and minimize the disadvantages.

All we're arguing about is how wide the divide has to be in rating and scissors/paper/stone comp matchups before "usually win" becomes "always win".

Mearis · 07/24/08, 4:13 AM

Originally Posted by solbergb

Right. It starts with gear and comp. If a certain warrior+windfury combo puts you both at the mercy of the RNG, both you and the other guy have made choices that put you in that situation.

I don't think you understand. There is no combo that is immune to getting rng'd by windfury+swordspec. People intentionally use swords with windfury because you will randomly win matches just through a lucky crit streak.

Soraxis · 07/24/08, 4:40 AM

If windfury+swordspec wins that many games on a consistent basis its not really a RNG matter anymore, its pretty much just a working group comp. Thats like going and saying that the mage got a lucky crit on his shatter combo, well thats his spec, thats what he's played into and you should expect crits like that just as you should expect warriors to wtfpwn on occasion because of what is already talked about.

• Maestroquark · 07/24/08, 7:55 AM

Originally Posted by Soraxis

If windfury+swordspec wins that many games on a consistent basis its not really a RNG matter anymore, its pretty much just a working group comp.

Okay, so throw in a bunch of random numbers that's okay >90% of the time but <10% of the time just being in melee range means you lose and somehow it's not random? There's a reason Windfury Totem is changing to haste in WotLK.

Mearis · 07/24/08, 8:36 AM

Yeah this argument is retarded. Some abilities that are excessively streak are bad for game balance and Blizzard is trying to change them.

The RNG is usually such that over a long streak the best team will win, but in the context of individual games, often times you will lose just due to particularly unlucky dice rolls. Right now, every warrior I know admits that mace stun is dumb and needs to be nerfed, but right now not using it is just intentionally handicapping yourself.

Can you sometimes play in a way that RNG will not affect you? Yes - sometimes, but sometimes you cannot and you will lose even though you played flawlessly or nearly so, just like sometimes you will win through absolutely no merit but only by sheer luck. I have feared a rogue through cloak of shadows before to barely stay alive, and that's obviously just dumb luck on my side.

• Amera · 07/24/08, 7:06 PM

The beta paladin trees are a pretty good indicator of this removal. You went from talents give you a chance to resist curses, fears, disease, stuns, disorients, and dispels to just dispels, with a time reduction on the others. The likely holy spec is looking at -50% curse and disease duration and -30% to everything else, which is pretty nice.

Despite that, they still stick in Sacred Cleansing, which is a total embarrassment RNG-a-thon right now. And they haven't touched anything with dispel mechanics, which is also a little disappointing.

solbergb · 07/29/08, 5:03 PM

Originally Posted by Amera

And they haven't touched anything with dispel mechanics, which is also a little disappointing.

They did make the mage armors not dispellable, which given the new effects and talents tied to the various armors is a pretty good thing for mages. They might take that approach with other key things currently being dispelled, remove the randomness by just saying "nope, can't dispell it, you gotta live with it".

solbergb · 07/29/08, 5:11 PM

Originally Posted by Soraxis

If windfury+swordspec wins that many games on a consistent basis its not really a RNG matter anymore, its pretty much just a working group comp.

That was my basic argument, but I suspect that the complaint is not that it wins consistently, but rather that on a small but noticable basis, it turns the tide if it's an option you have, whereas the rest of the time it has no impact at all.

It sounds like the prevailing opinion on this thread is that there should not be low-probability/high impact options out there.

While it is true that removing such options will emphasize skill more, it may also detract from participation levels and interest in "e-sport" (if any such interest really exists beyond hardcore PVPers). If you consider softball versus baseball, it is possible one reason baseball is a major spectator sport and softball is not, is the home run is a reasonable way to turn around bad play elsewhere, and it's achievable with a good batter in baseball, and nearly impossible in softball.

Softball, the quality of your fielding, baserunning matter more than in baseball, although it's important in both sports. In baseball, the wider variety of pitches available allow a good pitcher to shut down a team solo, and the small but significant chance of a home run lets a good batter make baserunning and fielding irrelevant. The strikeout pitcher and the home run hitter can be thought of as "speccing into" approaches to the game that minimize overall team skill and change the emphasis to more of a pitcher vs batter duel.

To extend the baseball metaphor, you're talking about extending the baseline or changing bat or ball composition to reduce the number of games swung by a home run. (This is by the way, something that has happened a couple times in major league history. Too many home runs are bad, as are too few).

Where the line is drawn, we can argue about. But there is a line, and there is merit to not having it be at either extreme.

You need a reason to keep watching the game after you've gotten behind. You need a reason to keep hoping and keep playing even if your team seems to be outmatched. Having a chance to win each individual game keeps you showing up, even if your odds of winning overall are slight.

heemo · 07/30/08, 5:32 PM

I think the sole reason that Blizzard's design philosophy has been to maintain random elements even in the competitive aspects of the game is marketability. Everyone but the elite is content with a slightly random playing field if it means that the little guy can occasionally beat the big guy. I think most(all) unbiased watchers were pleased to see GGE knock off SK in MLG SD. People love to cheer for the underdog, especially if they are the underdog. The biggest difference between WoW and RL sports is in WoW, the fans are the players.

I do find it somewhat objectionable that Blizzard clearly (at least to this point) has strongly encouraged the luck element - through resist talents and other random game mechanics. Hopefully over time we see these things moderated to some degree as random elements are moderated in pro sports. Ultimately, I think that's the key to this whole debacle - balance and moderation, not extremes. No, I don't think anyone wants to do away with every single resist, miss and crit, but I do think some of those things need to see some adjustments.

On the topic of the randomizing of "powerful" abilities - if they are too powerful, tone them down. Don't make them uncontrollable and/or unreliable - that just frustrates everyone on both sides. I'd be much happier with a 4 second spell lock on an RMP mage 100% of the time rather than a 5 second spell lock 80% of the time - if spell lock being overpowered is even really up for debate. It's more likely that spell resistance's effect on a crucial pet ability was initially overlooked and simply isn't on the developers' shortlist of things to address - like a lot of the game's similar flaws and imbalances.

As to the comment about pro players not complaining about RNG: seriously? Have you ever read or watched a single interview with one of them where the interviewer asked what they would change about WoW if the could change one thing? If someone took the time, they could post a pretty convincing montage of pro quotes on the subject to disqualify that statement.

Brinas · 07/31/08, 1:28 PM

I think that the reduced duration is a nice fix. A rogue getting feared for 4 seconds instead of 8 isn't game breaking to either team. I'm just curious how they could fix unbreakable will. It's a hit or miss type of ability. Making it into a reduced duration would just make the talent suck. Great so the silence lasts 1 seconds instead of 2. Who cares? my spell already got interrupted. Same thing with CS. How can they balance this talent without making it random?

Ghando · 08/18/08, 8:45 PM

Arena players are probably content to let Blizzard do the balancing based on their huge amounts of available data. But I think this basic principle is something everyone can agree on: when you take an action, you want to know what the result of that action will be. If a Druid has Innervate as his only buff and I Purge him, I should dispel it. It shouldn't take 8 GCDs and ~1800 mana to get it off. This is an extreme example as most anecdotes in this thread are, but it illustrates a crucial problem with the game if you're trying to make it into something genuinely competitive.

It's also worth noting that this is a bigger issue due to the format of tournaments. I've lost matches because a Warlock resisted Earth Shock and won them because I resisted or dodged a crucial Kidney Shot. Over the huge number of games that a serious arena team will play over the course of a season these aberrations are insignificant. However, in an offline match where only a couple games are played and those games come down to a handful of actions/details/plays the game introduces a lot of variance. It's like baseball; any given game can be swung wildly by luck, and while the 162-game regular season muffles those aberrations the playoffs are largely a crapshoot even in best-of-seven series.

solbergb · 08/19/08, 3:40 PM

If the playoffs weren't a crapshoot though, major league baseball would lose a huge amount of its revenue stream. The big advertising dollars are for the playoff games, not the individual games in the regular season, especially in brackets where the pennant race is already won.

If you can predict the outcome before the first minute of the game, there's no point to showing up. Not for the fans and not for the players. Nobody fields a little league team against the New York Yankees, except in some kind of public relations match where the little league team gets something for their participation that they value (ranging from sponsorship to just getting to associate with their heros).

The WOW arena system, however, has "major league teams" warming up in the little leage brackets and any team can have a major league "ringer" on it that can warp outcomes. That was ok when you were still rewarded with something you valued for failure. S4 shows what happens when that's no longer the case. I doubt inclusion or exclusion of random elements would have the kind of impact the rating system had, but my guess would be that the trend away from randomness would in fact lower participation some among weaker players, and inclusion of more "upset" powers would increase participation of weaker players.

Whether that's good or bad for the idea of an Esport I have no idea.

Herrera · 08/22/08, 7:56 AM

So to recap:

- Almost all "chance to resist fear" talents are now "reduces the fear duration"
- Fear/Roots will break after X amount of damage done to target based on target's % of health
- "Chance to resist stun" changed to "reduced damage while stunned" or stun duration
- Spell pushback completely revamped, not much RNG related, but leads to
- "Chance to resist spell pushback" changed to "reduces spell pushback (duration)"
- *drumroll* Mace spec: stuns - removed. (insert million thank you's) Changed to +rage for warriors and to something else for rogues

I think that with the addition of new talents and new skills, classes will become more versatile, more things can be done, more decisions, more choices and all this would lead to generally more interesting and challenging pvp fights.

• Calantus · 08/22/08, 11:25 AM

I'm very happy with the way things are going. Right now the only things they've missed on my wishlist are some of the proc stun/root talents and making interrupts immune to dodge/parry. I'm not holding my breath on the second but hopefully they get around to blackout/impact/imp hamstring/etc and give them the mace spec treatment.

• Maestroquark · 08/22/08, 1:40 PM

Originally Posted by Calantus

and making interrupts immune to dodge/parry

Nothing can dodge/parry/block while casting. If an interrupt is dodged, it's because it was attempted after the cast finished. Yes, lag can make this deceptive. Don't wait until the last moment.

• Calantus · 08/22/08, 10:56 PM

True. I had forgotten. :P

nife · 08/26/08, 3:07 PM

Along the same lines of questioning the rng, I would love for Blizzard to post the algorith mthey use for their psuedo random number generator. There have been a few mishaps in the rng field lately(read online casinos) and since blizzard is trying to make this an e-sport with 100's of thousands of dollars on the line it seems reasonable to expect transparency.

• mutagen · 08/26/08, 3:14 PM

Blizzard has stated that they use the IBAA PRNG.

nife · 08/26/08, 4:43 PM

Originally Posted by mutagen

Blizzard has stated that they use the IBAA PRNG.

Blah Cookie expire too quickly. Haha had a page long post but it got killed. Here is the toned down version:

IBAA is a decent prng from what I have read(Haven't looked at it specifically.) Hoever the devil is in the details. Several sites(casinos) have had trouble generating a number in the range. Its a pretty common mistake since its how most programmers were taught.

NUMBER = RANDON() % 100

The above code results in a bias. There are not an equal number of numbers that are divisible.

Here is some c# code and results showing the bias:

        static void Main(string[] args)
        {
            int[] count = new int[100];
            int accum=0;
            const Int64 MAX = int.MaxValue;
            for (Int64 i = 0; i < MAX; i++ )
            {
                count[i % 100] += 1;
            }
            System.Console.WriteLine("Index\tNumber\tDiff");
            for (int x = 0; x < count.Length; x++)
            {
                int diff = count[x] - count[count.Length-1];
                accum += diff;
                System.Console.WriteLine("{0:D}\t{1:D}\t{2:D}", x, count[x], diff);
            }
            Double per = (((Double)accum/(MAX)*100));
            System.Console.WriteLine("Bias  = {0:D}", accum);
            System.Console.WriteLine("Bias  = {0:F15}%", per);
            System.Console.ReadLine();
        }

I've only run the tests on 32 bit. It seems likely that blizzard is on 32 bits since their servers were built before 64 bit machines were common.

here are the results:

Index   Number  Diff
0       21474837        1
1       21474837        1
2       21474837        1
3       21474837        1
4       21474837        1
5       21474837        1
6       21474837        1
7       21474837        1
8       21474837        1
9       21474837        1
10      21474837        1
11      21474837        1
12      21474837        1
13      21474837        1
14      21474837        1
15      21474837        1
16      21474837        1
17      21474837        1
18      21474837        1
19      21474837        1
20      21474837        1
21      21474837        1
22      21474837        1
23      21474837        1
24      21474837        1
25      21474837        1
26      21474837        1
27      21474837        1
28      21474837        1
29      21474837        1
30      21474837        1
31      21474837        1
32      21474837        1
33      21474837        1
34      21474837        1
35      21474837        1
36      21474837        1
37      21474837        1
38      21474837        1
39      21474837        1
40      21474837        1
41      21474837        1
42      21474837        1
43      21474837        1
44      21474837        1
45      21474837        1
46      21474837        1
47      21474836        0
48      21474836        0
49      21474836        0
50      21474836        0
51      21474836        0
52      21474836        0
53      21474836        0
54      21474836        0
55      21474836        0
56      21474836        0
57      21474836        0
58      21474836        0
59      21474836        0
60      21474836        0
61      21474836        0
62      21474836        0
63      21474836        0
64      21474836        0
65      21474836        0
66      21474836        0
67      21474836        0
68      21474836        0
69      21474836        0
70      21474836        0
71      21474836        0
72      21474836        0
73      21474836        0
74      21474836        0
75      21474836        0
76      21474836        0
77      21474836        0
78      21474836        0
79      21474836        0
80      21474836        0
81      21474836        0
82      21474836        0
83      21474836        0
84      21474836        0
85      21474836        0
86      21474836        0
87      21474836        0
88      21474836        0
89      21474836        0
90      21474836        0
91      21474836        0
92      21474836        0
93      21474836        0
94      21474836        0
95      21474836        0
96      21474836        0
97      21474836        0
98      21474836        0
99      21474836        0
Bias  = 47
Bias  = 0.000002188608051%

For PVE:
So assuming that you have 14 dps classes doing 500 swings/spells per boss fight. (Proably a little low number for rogues and high for mages/locks but just an extimate) There is a ~.0774% chance that your raid should have done more damage.
Not all that high but it starts to add up.

PVP:
Apply this to all procs, dispells, and damge spells done, and you can see that there is some posibility that its affected the outcome of an arena.

Big caveat is that this is all naptkin math and not really sure they aren't doing it right.

Aditionally after reading more it looks like ibaa is prone to small cycles(at least the realier versions) so streaks are possible.

Prinsesa · 08/27/08, 4:44 AM

Originally Posted by Herrera

- *drumroll* Mace spec: stuns - removed. (insert million thank you's) Changed to +rage for warriors and to something else for rogues

One step forward, two steps back, methinks.

+ Warriors lost RNG mace stuns

- Warriors gained 30% chance to Overpower on a Rend tick
- Warriors gained 30% chance to Execute on a melee crit
- Warriors gained 30% chance to refresh MS cooldown on an Enraged Assault (which is itself dependent on having Wrecking Crew's Enrage available for consumption, and Wrecking Crew is procced by a melee crit)

While I do not doubt that they're aware of the problem, some of the newer talents appear problematic.

Herrera · 08/27/08, 4:48 AM

IBAA though seems unbiased.

C code for IBAA

/*
 * ^ means XOR, & means bitwise AND, a<<b means shift a by b.
 * barrel(a) shifts a 19 bits to the left, and bits wrap around
 * ind(x) is (x AND 255), or (x mod 256)
 */
typedef  unsigned int  u4;   /* unsigned four bytes, 32 bits */
#define ALPHA      (8)
#define SIZE       (1<<ALPHA)
#define ind(x)