Gems of Wars RNG is currently broken

Yes, of course. It’s not even much of a belief question, Occam’s Razor can answer that pretty easily. If they wanted to make it harder to win, would they rather:

a.) Insignificantly raise the AI win rate by secretly nudging some rolls slightly in favor of the AI.
b.) Significantly raise the AI win rate by adding some trivial algorithms to make the AI play a lot smarter.

Fun fact, my pet rescue times gradually went down from 15 minutes to 4 minutes for a full run. If that’s due to “Dynamic Difficulty Adjustment” it seems to be implemented the wrong way around. :smile:

2 Likes

They have done both A and B and with a simpler solution than coding AI behavior.

They manipulate summon rates and trained everyone to call it “streakiness.”

1 Like

Every LT so far this week.
Grand did the first, then black did 2, then CPT… The rewards for the last two are “identical” drop rate wise.


Just to show a break in the two images there was a good 12 hour break between the LT being collected.

The excuse in the past was “when you do a high number of LT, it’s likely a few of them will be the same since RNG will be streaky”… On purpose…we will only complete 5 LT this week. The RNG amount of streakiness will be exactly the same as if we did 300. :person_shrugging:
4 Glory Keys must be programmed to drop 90% of the time. Must be also why they don’t disclose the drop rates.

4 Likes

@awryan , at this point trust your gut feeling or experience that you have with this game. I have started to do so myself, and it really makes me come to peace with a few things. Things wont be explained/admitted, and things will most likely not change.

7 Likes

20% :ok_hand:

Edit: 35%, not 20%. Thank you for pointing it out.

1 Like

Summoning bandits is 35% chance. But I see your point.

1 Like

Hopefully this dynamic difficulty adjustment lawsuit prompts them to remove it from Gems of War, or at least admit its there. It’s a huge factor in guild wars, and basically ruins the whole week for me. You can’t just speed clear explore 5 for 3 hours and jump into guild wars, you will get absolutely annihilated by the RNG.

An observation in Emperinazar delve:

The Kobolds have Extra Turn as a chance. But the AI Kobolds always avoid taking 4 matches if they have a chance to cast which inevitably ends up in an extra turn.

Do they know something that we don’t?

I guess, depends on how its coded. It might very well be if casting is processed before the move, and if casting gives extra turn, it may default to the goblin mode. If not, then move is executed. Can be either way. We most certainly have no idea, and so may the developers if the coding is sufficiently convoluted. And yes, this is 99% correct. I have not seen kobolds move extra turn first, cast, and get extra turn in higher level kobold delve. The always do cast first.

1 Like

They don’t. If you spend some time observing the AI behavior you’ll eventually notice other situations where the AI gives spells higher priority than picking up an extra turn match, even when it can’t possibly gain an extra turn from the spell. King Silenus likes to do that a lot, the kobolds just follow the same pattern.

2 Likes

Here’s my opinion as both a retired player and a person who tried to disassemble the code.

Listen to akots. 2-3 years ago they pointed out, by statistical analysis, the GoW RNG is very prone to streaks even though if you consider those streaks in the long run it balances to match expected distributions.

AWR, at that point in the game, was very anti-akots. This was before Delves, Invasions, Raids, or a lot of current features existed. GW was the ONLY weekly event. Even though akots’ research supported the playerbase’s opinion that TDS was too powerful AND devs fixed it, at this point in time AWR’s opinion of the RNG was that it was totally fair, and anyone complaining about it was a loser who didn’t understand teambuilding.

Now that everything in GoW centers around the RNG, AWR is suddenly upset.

A game like FTL is unashamed of the fact it wants to be like a tabletop board game and only present a viable win scenario 10% of the time. You can, with skill, expand that to maybe 20%, but if the game doesn’t give you the right nodes at the right times you are going to lose. Hades is a more modern implementation of this roguelike concept, and while I’m only 65 runs in I already have a great intuition at the 25% mark as to whether I’m going to win or not. A nice thing about Hades that FTL lacks is if you think you’ll lose at the 25% mark, the EV of continuing as far as you can anyway is usually better than what you’d get if you give up.

GoW is the opposite outside of most scenarios. It’s optimized for a game where you quit the moment your inuition tells you you’ll lose. PvP and Explore farming are where this is most evident: there’s no reason to fight a looping team for 5 minutes to lose if you know your team wins in the first 3 moves or is doomed.

That’s no good in events with sigils, and it’s why ancient GoW historians remember Sirian specifically hated stamina systems. Back then, he liked and appealed to gamers who respected a theory about Freecell that as far as I know is still unproven: that any shuffle has a solution. The idea in a game like that is you should never be backed in a corner, there should always be a move that wins.

I think GoW lacks that concept. There are too many skills that leave their effects up to randomness, and it’s not presented in a fun way.

In Hades, sometimes I start out with weak boons. When that happens I reckon my best bet for the run is to farm resources rather than try to win. But sometimes, just after I make that decision, the game gives me a lynchpin boon that alters my outlook. Then I have to watch the next few opportunities for progress and decide if it’s better for me to focus on winning than farming.

GoW completely lacks that nuance. This is why AWR is mad, and if I were him I’d quit over it. Coincidentally, I’ve already quit. Here’s the thing.

I know how AWR’s top guilds work and it’s based on performance. In a strategy game like chess, this is reliable. You can see how players compare to each other and form very objective rankings with strong probability guesses about who might beat who.

But if RNG dominates skill, this gets really weird. The people who did best last week aren’t guaranteed to be the best this week. The people who did best last month aren’t guaranteed to do well at all this month. If you can get 3 points or 20 points from an event based on RNG, it turns out any above-average player is just as good as any other!

That’s where GoW has messed up. The important matches aren’t about skill anymore. Some upper 25% of the playerbase represents the “haves” who can build good teams. 50% of those will “win” RNG this week, but “lose” RNG next week. That’s not a scenario where you can build a guild of the “best” players, because in general the people doing “best” are just on some weird local maxima of RNG and are likely to tank as soon as you recruit them.

GoW has a lot more to do with Candy Land, where a single shuffle determines the outcome, than Poker, where there are a lot of variables and understanding “what I see” vs. “what I can deduce” is a measure of skill.

That’s the real tragedy. Everything in GoW advertises itself as a skill game, but it’s a chance game.

5 Likes

I couldn’t disagree more with some of your RNG assertions. Everything you said against awr, I take no issues with lol. You’ve made some glorious posts here about awr in the past(search “bed of barf” by Slypenslyde) and you’ll enjoy it I promise!

As for the RNG I do push back a lot. You have generalized GoW but you have also missed the nuance. I used to think GW was a rng fest where going 3/2 was just as likely as 5/0. I also thought pure faction at 500 on most delves were near impossible or fully luck based unless your hoard was 300. I say used to, because I learned to take my time, come up with better strategies, listen to advice of our network players, and take my time on those modes that require it.

GW takes time and strategy. Of course RNG will still smack you down, but that’s equal for everyone in the game so we are all in the same boat. The only thing that could make this even is if all players got the same starting board and defenses to go against. The randomness is part of the charm and frustration. But after making a lot of GW adjustments I rarely go 3/2 and usually only lose 1 to 2 GW battles at most, and I play two accounts in bracket 3(I know, not as hard as B1, but still annoying). I also stream my GW battles and so often the players I’m streaming with see things I never would have caught and it’s saved me. Sometimes I see things on their games and it saves them. Sure, RNG can still crush your soul, but the odds are always in your favor to start and you don’t need all the troops in the game to succeed here just some extra patience.

As for pure faction the same applies. This past weekend players would complain it would take 7-10 of tier 7 potions and it takes 30-45 min for a final battle. Except I, and many others did it with less than 4 of tier 7 and took less than 25 minutes. We discussed the best team options, strategies and team builds to give ourselves the best chance.

My point is GoW on it’s surface is a simple game. It’s the same game across almost all game modes with subtle tweaks. I have no illusions about what gems should or will be. It is what it is the wheel isn’t getting reinvented. But to generalize the entire game assuming one way works for all modes and that’s how it was meant to be played or is played by all, is wrong. RNG may be streaky and it may be off or even unfair in some circumstances. But there is strategy and there is skill involved. The proof is in the stats. I have months of GW data to show the players that adjust their teams, strategies and stream are more likely to score higher. Skill > RNG. Rng and skill vary by player, which is why sometimes RNG > Skill for many.

You are not wrong about pvp, explore, or events. Those are win super fast or retreat. They are built on getting through as efficiently and as quickly as humanly possible to maximize rewards. But what you don’t realize is that is fun for a lot of players. It’s a grind for others, but I enjoy the the endorphin rush I get from winning and winning fast. Pvp can sometimes be slightly more annoying than other times but if I sniff a loss in the first turn or two I’m retreating, and I see nothing wrong that. Other games lock my ability to play and don’t allow me endless attempts. My fun is tied to that speed, progression, and optimizing play. Faster does equal more fun, and if that does start to feel like a chore, I would quit. But the demands are only on the person and the guild they are in. If the game feels like a chore, it’s less likely a game issue, and more likely a guild or personal FOMO issue. GoW is a casino, players just have to learn to go in a with a set amount of chips and leave before sunrise the next morning no matter what. Gems doesn’t force you to do anything. You can complete almost all modes with some key strategies and minimal key troops(I proved this with an alt awhile ago and developed a streamlined guide for our network).

Gems have failed in a lot of respects when it has come to new updates such as arena. Whoever is developing gems does not seem to be the exact same team as a year ago. It’s sad because there was some solid hype about what 5.0 could be and beyond. Now every update I feel like the community is just bracing themselves for more disappointing things they didn’t want or ask for. So for that I fear for the future of the game. But in terms of RNG and what gems is or isn’t I disagree with a lot of your points. But hey, you didn’t like gems anymore and you quit, I really really wish many in the game community would do exactly that.

2 Likes

Quit the game and then continue to talk about the game on the game forums. Yeah… That makes total sense. :person_facepalming:
@ButtStallion you seem to fail to understand that it’s not all or nothing. People can complain about 40% of the game and still enjoy 60% of the game. But you don’t hear about the 60% because why would you if you don’t interact with that facet of the person’s involvement of the game.

As for you @Slypenslyde aka Jon Snow… Since you’ve never been me nor never been in any of my guilds. You know nothing.
So don’t fool yourself or others in thinking that you do know. Despite feeling the need to comment on an active thread for a game you have quit playing.

3 Likes

Well, if he got nothing better to do with his time, I can totally understand it. :cry:

lol, there a slew of people I could put into the “got nothing better to do with his time” pot. Do I question why he would all of a sudden jump into a discussion about the current state of the game rng when he hasn’t played actively (or at all) for months or up to a year at least, yeah I def question that. But he is also smart and knows a lot about how gems mechanics would or should work more than the average person so his input holds some weight opposed to a random drop by.

I guess where I fall on this is whether you love or hate gems, at least be playing the current state of the game to provide a more detailed opinion opposed to referencing past iterations. As we have seen, devs can make ninja changes(see gnomes in arena) so the game does change, and if you don’t keep up, you can work off outdated information. Not saying this was the case above, but that’s my argument for current players being the main posters opposed to players who quit. Idk, I’m a rambling man.

I retract my opinions, since AWR can’t get over himself long enough to accept when I agree with him without making a parting jab.

I left GoW because I felt I wasn’t in control of my fate. This both meant it demanded too much time and my success seemed to vary too much by random seed match-to-match. Note that in the original post AWR has no specific gripe other than “the RNG is bad”, so I think I can be forgiven for not understanding whatever specific niche of RNG he means.

I sort of thought I might come back after a little break. Maybe the thrill of missing some key troops would’ve been fun again. But alas. AWR is still dismissing my points ad hominem because he’d rather dunk on me than agree with himself, and a valued mentor suggests that since I’m a whole maybe 10 troops out of date the game has changed so dramatically I don’t have the cognitive capacity to understand it.

Good luck everybody.

I might be a bit of an outlier, but I don’t really think ‘Dynamic Difficulty Adjustment’ is a thing in GoW. Mostly because:

  • They said it’s not (aside from making it easier for new players)
  • It would be a pain to code, and the devs seem typically go for easy/simple solutions
  • None of the posts claiming it’s a thing have imo sufficiently refuted or accounted for the option that it could be confirmation bias, and just assumed that it was a thing/fact and that everyone agreed, or went for, “either you believe it or you don’t”

I’m still not necessarily happy with Skull drop rates in high level Delves and Arena (not that I’ve been playing it since the 2nd event), and wonder if the drop rate was made to be higher in Arena to make up for the slowness (not sure about the Delves, could just be the high consequences), but I tend to think those are different issues.

In terms of repeat LTs, I’m open to the possibility of duplicated rolls, the devs have made drop rate stuff ups in the past, and not communicated them particularly well, so it doesn’t seem impossible. I kind of agree with what akots was saying about it being a bit of a black box, though, and I don’t really see the devs opening it up in the near future (so I don’t fixate on that possibility).

In terms of the 4x Glory Keys, I had a feeling it was potentially a reward from a couple of the different LT reward tables, which I think turned out to be correct:

Drawn right to left, so 90% chance an LT won’t have 50 Glory Keys, 30% chance it will have 4, 0% chance of the same reward type in the following column, then 90% * 70% (chance of not getting the keys in Table 2) * 30% (getting them in T3).

(0.9 * 0.3) + (0.9 * 0.7 * 0.3)
= 0.27 + 0.189
= 0.459

I.e. a 45.9% chance of getting 4x Glory Keys as a reward in any LT, which is pretty lame – but I think we already knew LTs are usually disappointing.

1 Like

The drop rates are server side… And can be changed on the fly… absolutely no reason to think…or to not think… that the 5.2 version of GoW has the exact same drop rate as the 3.1 version of GoW in terms of LT.

Even if it did…I understand the abundance of them… The consistent back to back to back streaking is simply unprecedented. Unlike what I’ve saw…or even bitched about… in the past.
I would see a “repeat”… Then 3-4 non repeats… followed up by another “repeat”.
Back then…I thought those streaks were bad. But was told those streaks were just normal RNG… Now these streaks are Megazord Level in comparison. And still told the same lame duck shit “normal RNG streaks”.

We get two types of answers in gems of war anymore. Either “everything is fine” or “no answer at all”.
If the devs see anything is actually wrong… they’d rather just make us feel like we are dumb, crazy or argue with those who think otherwise (even if they agree with us) while they work on it behind the scenes to make it right. 30-90 days later a ninja fix will roll out and players will post the changes for other players to learn about them.
God forbid they just say “hey, yeah that doesn’t look right, we’ll look into it and see what we can do.” :man_shrugging:

And announce POSITIVE IMPROVEMENTS TO THE GAMES EVERY TIME THEY MAKE THEM.

I watched the crown the other day, I don’t know if that Flanagan guy actually said it irl, but on the show he said “I don’t have mental issues… I’m just poor”.
Well in my case, "I’m not paranoid when it comes to issues in the game… I’ve just been playing Gems of War for 4 years.

2 Likes

Yes, same thing here. Something is wrong, I also feel it. However, being a scientist, I cannot simply accept it and need to look for a proof instead. I also don’t think it is limited to streaks, gnomes, or general manipulation by the developers. So, no, you are not paranoid, at least not the only one. :wink: Whether it is objectively justified is another story though.

2 Likes

Agreed. Silenus has always failed us in a good defense that we set. So does Thrall, am I right?