Ultron feedback

DaveR4470

First, I thought it was a good event overall. The RNG was very generous to me, and I can't complain in the least. So kudos for that. Here's a little feedback on what I think worked, and what didn't...

What worked: The event overall, which was different and unique and an interesting change of pace. I think the bomb dropping mechanic was a neat twist, and the sentry nodes were challenging.

What didn't work:

-- Timers in general. If there are going to be phantom resets of nodes, put a visible timer on them. If the point reward timer on the subnodes is different than the high-level Ultron reset timer, make that clear. There were too many instances where it wasn't clear at all why a given node wasn't giving awards in some cases, and was in others. It was only during the second run that I pieced together, through observation, that apparently the POINTS were on a single 8-hour timer independent of the Ultron node, while ACCESS to the nodes (i.e. resetting the "clear" status) was controlled by the Ultron node's time AND the 8-hour "master" timer. Nowhere was that made clear.

-- Rewards for Ultron. If this is going to be a "boss" node that's going to show up repeatedly, you just can't have the 20 iso award. Even if it's just a standard token, beating Ultron should have awarded something of value, and not the booby prize, with EVERY clear, not just four of them. Much like a 10" tall Stonehenge, the fact that I was getting less for beating Big Bad Ultron than I was for beating a couple of waves of his minions with Quicksilver tended to de-emphasize the grandness of the enemy, shall we say. Especially towards the end, beating Ultron became a chore that I had to do so I could get to the Quicksilver/Scarlet Witch (as applicable) required nodes that awarded the exact same prize package as Ultron, just much more often over the course of the event. On the other hand, the generosity of the subnodes was quite appreciated. As noted, the RNG gods were very kind to me.

-- The AI ability that triggers multiple crits... please, no. Cut that back to a single crit. There were too many instances of game-changing cascades.

-- Server issues. Probably has been discussed elsewhere...

That's all I can think of right now. Thanks....

Pwuz_

Considering that PvE continued to run as normal through out this, I would have like to have been given a bit more time. It plays a bit like the Gauntlet in that there is no real rewards for placement. The Gauntlet doesn't run along side a PvE event & PvP, but Ultron does?

I like the event A LOT! But having to divide my time between PvE, PvP, & Ultron was a huge drain on my health packs.

Unknown

I like how the feedback forum has feedback complaining about feedback.

Unknown

I want to add to this.

Pros:

Excellent event overall
What @DaveR4470 said plus

I liked the waves and the different sentries was a good mix.

Ultron fight was great. (I didn't like that after I killed him there was a cascade that dropped two bombs and killed one of my guys but I digress).

Cons:

The Gunners were SUPER OP. I fought several waves where the comp was Gunner/Fighter/Fighter or Gunner/Gunner/Fighter and on turn 1 or 2 (not wave but turn) I lost a character because the fighters cacades caused more than 7 special tiles to be on the board at which point the gunners drop crit tiles that cause cascades that kills someone. That someone was always my 3* Thor who has 10k hp....

I don't know if it was intentional but only the top alliances were going to get Hulkbusters. 11 million HP on round 7 (thats where we ended in the second half) is not something even a good alliance can hit. (We are a top 100 alliance but we have real lives that don't lend to kill him every 8 hours). I think his HP should be based on the alliances ability to kill him. Everyone getting hulkbuster covers would have made you more money. In order for people who are not regularly top 1 players in events are going to have to buy covers with HP, which we have to buy from you.

The event was too monotonous. I was sick of fighting sentries by the end of it. If this is something you are interested in doing again, please change it up. Toss in Mind controlled Heroes or something. By the end of the event I couldn't wait for it to be over whether we won Hulbusters or not.

blinktag

DaveR4470 wrote:

First, I thought it was a good event overall. [...]

What didn't work: [...]

-- The AI ability that triggers multiple crits... please, no. Cut that back to a single crit. There were too many instances of game-changing cascades.

Everything that he said.

The multiple crits was such an issue that I can't help but wonder if the simulations (you do simulations, before building powers, right?) took into account that due to the other powers, the enemy specials were significantly more likely to be clustered than in other events, meaning a two crits in a single match was also more likely. Sure, there's a lot of cognitive bias involved in anecdotal observation (and not always time to note what happened), but I'd speculate that when the crits appeared, they were "doubled" (two crits in the same match, which is a massive amount of damage) at least 1/3 of the time. If not that often, certainly much more often than if the crit appeared at a random location instead of replacing a special.

DaveR4470

blink: I think, if anything, the problem is that their simulations use non-human thinking. To make a long story short, I'm sure the simulation AIs -- which probably run thousands of rounds to analyze skill power/damage -- make "human"-side matches based on "best obvious match" in the way that the game AI does. The problem is that humans do not always do that. We can think and plan ahead, and attempt to concentrate matches based on what colors we need given our battle team.

So, again long story short, in the simulator, the skills are triggering on what is, more or less, a random distribution of tiles. In real life, after several rounds of play, the board is very much NOT a random distribution of tiles -- it reflects the choices made by the player, which could (i.e. probably will) wind up with a disproportionate and non-random distribution of certain tiles that the player doesn't want.

And dropping two crits into THAT environment is going to have a significantly different impact than dropping two crits into a truly random distribution of tiles. Ultimately, I think that may have been the unanticipated consequence of the double crit skill.

blinktag

DaveR4470 wrote:

blink: I think, if anything, the problem is that their simulations use non-human thinking.

That may be part of the issue, but as I recall, the Sentries introduce their special tiles from the top ... which means their distribution is already not random.

veneretio

DaveR4470 wrote:

-- The AI ability that triggers multiple crits... please, no. Cut that back to a single crit. There were too many instances of game-changing cascades.

Never found this to be an issue except when they were paired with Ultron Prime for the rewards. That was a nightmare and wiped about 9 characters before I lucked my way into beating it fast enough. This is the only context I see them as a problem.

DaveR4470

blinktag wrote:

DaveR4470 wrote:

blink: I think, if anything, the problem is that their simulations use non-human thinking.

That may be part of the issue, but as I recall, the Sentries introduce their special tiles from the top ... which means their distribution is already not random.

The strike/attack/shield tiles were introduced from the top (they all related to a common passive skill that changed what it generated based on the type of sentry), but the active skills spawned on the board -- in a random column for the bottom-tile column clear one, and anywhere for the twin crits.

veneretio wrote:

Never found this to be an issue except when they were paired with Ultron Prime for the rewards. That was a nightmare and wiped about 9 characters before I lucked my way into beating it fast enough. This is the only context I see them as a problem.

I agree -- it was worse on the Prime nodes. It never reached the point of game-killing on the regular side. But it was confusing to track which type of sentry triggered which type of bad thing the whole event, so (to me) it seemed like you'd suddenly get crit cascades out of the blue that screwed up your board management entirely, which didn't seem like a really great mechanic. I think a single crit at maybe a slightly lower threshold would have kept the degree of difficulty factor in play without the "oh, here's a profound change in the flow of this match" element.

ErikPeter

I think lowering their 4.5x crit multiplier would help, but the real problem comes from positive feedback. If, through bad decisions or luck, you let Core Meltdown happen, it will probably blow up the board, causing a rain of new special tiles to be created, which means it may be impossible to prevent another core meltdown the next turn... And so on.

The Flyers were better balanced. They were annoying, but their ability eats special tiles which made it less likely they'd use it again right away. If they do the event again, I think that special tile generation should be paused during Core Meltdown crit cascades to prevent inescapable failure.