AI cascades must be addressed

the door swings both ways, obviously one no is going to complain when they get **** wrecking cascades, sometimes I get cascades so good I can end the match by turn 3-4. AI has the same chance as we do to do the same thing, but to be honest, because AI is about as smart as a brick, human players have better potential to cause big cascades since we can manage powers better, make different shaped 5+ matches and so on.

AI isn't cheating, don't get your feathers ruffled over a brick getting occasionally lucky.

Also, remember that most people tend to prioritize one or two colors. Immediately following a human turn, the board is more likely to have greater saturation of the remaining 5 colors. The result is maybe a percentage point or two in favor of AI cascades, but even a few percentage points make a difference when you're looking at the number of Turns involved in the average player's game sessions.

fmftint wrote:

NorthernPolarity wrote:

It's that time of the month again! Ahem, "No, the AI does not cheat: viewtopic.php?f=7&t=15895&start=40#p220817".

From the same developers that say Moonstone is underrated and underused? They can say whatever they want, I know what I see and experience. Either the AI "knows" what is off screen or tiles are being generated in their favor

ps I didn't say they cheat, but tiles fall in their favor far too often. I've never had 6 of the same color drop in on the vertical or horizontal but the AI gets that luck too often for my taste

I've had that happen in my favor before. It rocks. Yeah it sucks when the AI gets that lucky but as much as both sides get cascades, I don't think that kind of thing is that frequent.

OK, maybe I can help with this one.

I recorded every match I played for 2 months, (and it was a lot. I think I may have problem). After watching them back, and watching certain ones in slow motion, I have come to 2 conclusions.

1). The cascade ratio was 44% for me, and 56% for the Ai. Close enough to even to conclude it is mostly luck.

2). Watching a bunch of Ragnarok matches in slow motion and counting all the tiles, I have concluded that the display screen does not always keep up. If Rags got 8 green in a cascade, the display would only show 2 or 4 sometimes. Which made it appear the Ai did not have enough Ap to use certain powers when they actually did.

To D3, please do Not change anything. In the current PvE, Thick as Thieves, I was playing node V with Ares, Empiricist, and Bullseye. All over level 200. I made one match, then the Ai made one. I made my 2nd which caused a critical, which gave me another turn, which gave me a critical, and another turn, and so on. I got 6 cascades with critical hits 6 times in a row which gave me enough Ap to use my XF green and black and Steve Rogers red. Then it was game over. It was soooo beautiful .

Which just goes to show you, it's mostly luck.

Phantron wrote:

If it's 44% you and 56% AI then it's a huge advantage to the AI because the human always goes first, which means the human always have the first pick on the initial board and quite often the first major cascade also kills the cascade potential of the board by taking all the good stuff. Without having a further breakdown of the matches it is impossible to tell whether this is because the computer is better because it's more skilled, or the human is less skilled, or that human fatigue has a far more effect than most people think (there are a lot of times where I thought I'm supposed to get two moves in a row without stuns), or whether fights against guys like Ragnarok or Juggernaut naturally greatly favors the AI in terms of cascades, especially if they've the appropriate AP pump.

Couple things to point out here...

1). Human fatigue may very well be part of it. I can't tell you how many times I get going and don't pay as much attention as I should. I will often make a match 4 only to see a match 5 up in the corner at the last second. But it's too late at that point. Or hitting the wrong power, (MN Mags blue instead of CStorm wind storm for example, I hate when I do that). The computer is always paying attention and does not get tired. So you are definitely right about that.

2). This is based only on my games and my play style. This is was not done as close to an actual scientific study as I could have made it, but more of me being so danged curious and always over analyzing and over thinking things. That and I have way too much free time and I'm way too addicted to this game . Also, I can't stand someone trying to cheat me, so I had to know for sure if it was or not.

3). Technically, it was 44.377% me and 55.623% Ai. But less than half of 1% didn't really make a difference.

So not only does human fatigue have more of an effect than most people realize, (I didn't notice just how bad it was for me until I watched a few replays. I would see a better move while watching than the one I made while playing and wonder how I missed it), but also, players sometimes forget, (myself included), just how cheap Juggs and Rags powers are.

This is the amount of Ap I got from my first move.


And I think I have a free turn because there was a match 5.

Swings and roundabouts here. At time I am like Jesus H Christ you cheating mother smallfelines, other time I am firing off one ability after another.

For me it is fine.