AI getting too many drops on console, still?

[Saltypatra]

The only time the Ai ever ‘cheats’ it does so in favour of the player at lower difficulty levels. There was a thread in March found here where our console developers ran numerous tests to determine if the AI was luckier than the player. The thread itself is a bit long and unwieldy so I’ve gone through and found all the relevant examples of this. Feel free to give it a read yourself if you have the time and inclination.

@GoldPhoenix0

the AI has no knowledge on what is going to be dropped next on the board. The AI will make a move based only on what the player can see, the board and spells it can cast. There isn’t even any code in there to line up any further moves or check for cascades, and as much as I would love to add that into the AI, I know that all it would cause is more complaints about “Cheating AI”, as much as it would then be closer to a smarter human opponent.

@GoldPhoenix0

Playing on Normal difficulty does affect the “luck” the AI will get, in which it enables what is generally called “Combo Breaker”. The AI will also be a bit more random with it’s decisions on Normal difficulty as well. These changes are always disabled in Ranked PvP, as you are unable to select a difficulty.
Gem Preferences. Players can influence their AI for gem matches in Ranked PvP. This is were the AI will now no longer be baited for Skulls if the Defending player modified the slider to give it a lower preference. This means that the AI will, in general, act closer to that human player.

@Sirrian

We define a “lucky” drop as one of the following events (relative weighting is show in brackets):

A Skull drops in to form a 4+ Skull match, when that skull was not previously visible on the board (5 pts)
A Gem drops in to form a 4+ match, when that gem was not previously visible on the board (3 pts)
A Skull drops in to a space where the opponent then immediately gets a 4/5-of-a-kind Skulls (4 pts)
A Gem drops in to a space where the opponent then immediately gets a 4/5-of-a-kind (2 pts)
A Skull Gem drops in to set up a Skull match for the opponent (3 pts)

@GoldPhoenix0

For more information on how these tests are run, here is a breakdown on how we have conducted these tests

1. These tests are run in the PvP environment. This means there is no interference with any luck/combo breaker code.
2. To keep things fair, we have the following rules for picking teams

• Teams are formed by picking random troops from the complete troop pool
• Each troop is level 15 with NO traits (As to stop any trait interference)
• Both players are given the exact same team, as to ensure that any spells cast (like mana generators) are given an equal chance to cast these spells

3. If a game takes longer than 20 minutes, the game ends and deems it to be a stalemate, and continues onto the next round

Over the past 72 hours, we have produced and analysed 2024 unique battles using the same scoring system according to @Sirrian 's post to produce the following results:

LUCK SCORES (based on weightings above)

Average Player Luck Score: 23.153
Average AI Luck Score: 22.982
Difference in Luck Scores - the human player was on average 2.62% luckier than the AI player

LUCKY GAMES (counting number of games and the associated luck score)

Games with Equal Luck: 58
Games where the Human was Luckier: 1007
Games where the AI was Luckier: 959
Difference in Lucky Games: The Human player was luckier in 1.22% more games

CASCADES (number of cascades which occurred in a match)

Average Player Cascades: 36.534
Average AI Cascades: 36.894
Difference in Cascades - the human player had on average 1.88% more games with more cascades than the AI player

4 OF A KIND MATCHES (number of 4 of a kind matches performed by each player)

Average Player 4 of a Kind Matches: 5.744
Average AI 4 of a Kind Matches: 5.514
Difference in 4 of a Kind Matches - the human player had on average 1.88% more games with more 4 of a Kind Matches than the AI player

5+ OF A KIND MATCHES (number of 5+ of a kind matches performed by each player)

Average Player 5+ of a Kind Matches: 2.217
Average AI 5+ of a Kind Matches: 2.136
Difference in 5+ of a Kind Matches - the human player had on average 6.72% more games with more 5+ of a Kind Matches than the AI player

BOARD MATCHES (number of board moves which each player has taken)

Average Player Board Moves: 26.049
Average AI Board Moves: 26.734
Difference in Board Moves - the AI player had on average 1.09% more games with more board matches than the human player

SPELLS CAST (number of spells cast by each player)

Average Player Spells Cast: 6.577
Average AI Spells Cast: 5.627
Difference in Spells Cast - the human player had on average 18.33% more games with more spells cast than the AI player

Normal Difficulty

@GoldPhoenix0

Just to confirm that our code is working correctly, we have been performing further tests using the same test conditions as the previous test, but on Normal Difficulty in a mode which is not Ranked PvP. This here would be the difference between playing the rest of the game on normal difficulty, and then playing Ranked PvP, where the Combo Breaker is no longer enabled.

Over the past 90 hours, we have produced and analysed 2870 unique battles using the same scoring system according to @Sirrian’s initial post to produce the following results:

LUCK SCORES (based on weightings in the initial post)

Average Player Luck Score: 40.107 (vs 23.153)
Average AI Luck Score: 19.989 (vs 22.982)
Difference in Luck Scores - the human player was on average over 100% luckier than the AI player

LUCKY GAMES (counting number of games and the associated luck score)

Games with Equal Luck: 1
Games where the Human was Luckier: 2870
Games where the AI was Luckier: 0
Difference in Lucky Games: The Human player was equal or luckier in every game

CASCADES (number of cascades which occurred in a match)

Average Player Cascades: 39.532 (vs 36.534)
Average AI Cascades: 30.655 (vs 36.894)
Difference in Cascades - the Human Player had on average 38.40% more games with more cascades than the AI player

4 OF A KIND MATCHES (number of 4 of a kind matches performed by each player)

Average Player 4 of a Kind Matches: 5.784 (vs 5.744)
Average AI 4 of a Kind Matches: 2.893 (vs 5.514)
Difference in 4 of a Kind Matches - the human player had on average 38.40% more games with more 4 of a Kind Matches than the AI player

5+ OF A KIND MATCHES (number of 5+ of a kind matches performed by each player)

Average Player 5+ of a Kind Matches: 2.252 (vs 2.217)
Average AI 5+ of a Kind Matches: 1.000 (vs 2.136)
Difference in 5+ of a Kind Matches - the human player had on average 35.61% more games with more 5+ of a Kind Matches than the AI player

BOARD MATCHES (number of board moves which each player has taken)

Average Player Board Moves: 24.580 (vs 26.49)
Average AI Board Moves: 24.332 (vs 26.734)
Difference in Board Moves - the Human player had on average 8.48% more games with more board matches than the human player

SPELLS CAST (number of spells cast by each player)

Average Player Spells Cast: 7.262 (vs 6.577)
Average AI Spells Cast: 3.737 (vs 5.627)
Difference in Spells Cast - the human player had on average 44.11% more games with more spells cast than the AI player

[jeno]

oh no!!! some facts! the internet worst enemy!

[beanie42]

@Saltypatra

The only time the Ai ever ‘cheats’ it does so in favour of the player at lower difficulty levels. There was a thread in March found here where our console developers ran numerous tests to determine if the AI was luckier than the player. The thread itself is a bit long and unwieldy so I’ve gone through and found all the relevant examples of this. Feel free to give it a read yourself if you have the time and inclination.

Nice link, but you’re assuming they care about proof compared to opinion. Back to your throne

[Vangor]

@HKdirewolf

Edit: I changed the title to more appropriately reflect the issue inside. Shame on whoever changed it before. Also you should change it back to the original

What was it before you changed it?

[HKdirewolf]

When will the players stop insulting the Devs’ intelligence or something as such.

I was expecting to read some well thought-out post defending the devs’ stance on something and got the complete opposite.

[Vangor]

Heh thats what it was originally

[Jackalopi]

I think originally I saw it as the opposite, something like “When will the devs stop insulting the players intelligence”

[HKdirewolf]

Yeah that sounds more like what it should have been. Someone definitely changed it. This sounds like a case for Sherlock Holmes.

[beanie42]

One good thing came from this thread though. It caused @HKdirewolf to make an appearance

[HKdirewolf]

If only you could tell Ogunther to come back, we could get the band back together.

[turintuor]

I like these stats aside that I don’t agree with the conditions (no trait, why? lvl 15, why? same troops, why?). There is a huge biais which is the troops.

Is it not possible to simulate only gem drops and see how much cascade AI or humans have? And do it on PC/mobile and console separately to compare?

[Lyya]

Sure, it’s possible, but frankly, if people aren’t going to believe the devs saying “the same method is used to replace holes in the board, no matter who made the holes” then they won’t believe the dev stats, either. The only “remedy,” as it were, would be data collection by an unbiased party. (Though, as usual, I’m baffled as to what possible motivation the devs would have for deliberately frustrating the player base and then denying it when questioned.)

[EmsDad15]

That’s what I thought too. As amf that’s what made me come back to it after I was finished tending to my daughter. And when I got back the title was changed.

[Saltypatra]

Not much point in testing it separately on PC/Mobile and console when we are moving over to the console AI in the near future.

[Hoop]

I play on both. Seems the console ai acts even dumber than the pc ai, if that’s possible

[Vangor]

[Sidousai]

[Saltypatra]

You get a like because this is one of my favourite memes.

Let’s keep the memes there though, okay everyone? (@Vangor)

[Maxx]

@Dan_ozzzy189

Today I had a pvp game that lasted 25 minutes.
Opposition had one troop - celestasia.

Same thing happened to me yesterday I was fighting an all dragon team and was down to just Celestasia who was 1 skull match away from death but she managed to get a cast of and just looped for a while and every time I broke the barrier she would just cast again and get it straight back. After a while I just retreated and I had more fun losing that battle than I did winning against constant Famines.

[turintuor]

Some “voices” in this forum said that there is an issue with cascade on Console and it could be nice to check if is there really a difference between Console and PC/Mobile on this feature.
It should be a shame to jump on another huge issue that will make “suffer” this game…