*** Loki (Dark Reign) ***

NorthernPolarity

Phantron wrote:

I use Loki for PvE just because The Hood bores me to sleep, though Loki's higher HP does matter a lot for PvE too when your enemies can drop 4 digit matches on you. For PvP will probably still have to consider the team restrictions. Loki covers purple strongly which is usually more useful than The Hood covering yellow weakly but The Hood is probably still better overall.

If Trickery isn't extremely situational they'll have to nerf Illusions by a lot. As is right now Illusions is sort of roughly equal to two top tier but balanced ability while Trickery usually has no use at all so it sort of works out.

One thing I thought might be interesting is if you have Intimidation at level 5 and do an Intimidation after the opponent has a huge cascade, as that'd immediately trigger all your Mischief tiles which is pretty much like getting that mega cascade yourself, since the AI never uses any moves after they moved a tile regardless of how much AP or how many extra turns they get. Wouldn't make sense to try to play with this strategy, but it'd be unbelievably cool when you pull it off.

At least purely considering the team composition of XF / LadyThor, Loki has definitely replaced Hood for me. Most of the time, you end up losing with XF / LadyThor because the AI gets some sort of insane match-4 cascade on you, or if the board is not cooperating with blue / red tiles. Well, Loki solves both those solutions in one go. Mischief has even more subtle value than I thought it originally did: since it turns the same colored tiles that the match-4 occurred on, it effectively drains the board of most of that one color, giving you increased chances at favorable match 4/5s. Also, it does just randomly cause cascades by itself which is really nice. Since Loki only tanks purple in this team comp as well, it's far more sustainable than XF / GT / Hood, where hood tanks yellow and blue and is a health pack drain.

The only issue is that vs XF / GT / Hood, XF / GT / Loki is worse because Hood randomly still just wins you the game in an unfavorable board. I would say that vs any other team composition, XF / GT / Loki is superior in pretty much all possible ways, but XF / GT / Hood beats out this team on defense.

Unknown

NorthernPolarity wrote:

At least purely considering the team composition of XF / LadyThor, Loki has definitely replaced Hood for me. Most of the time, you end up losing with XF / LadyThor because the AI gets some sort of insane match-4 cascade on you, or if the board is not cooperating with blue / red tiles. Well, Loki solves both those solutions in one go. Mischief has even more subtle value than I thought it originally did: since it turns the same colored tiles that the match-4 occurred on, it effectively drains the board of most of that one color, giving you increased chances at favorable match 4/5s. Also, it does just randomly cause cascades by itself which is really nice. Since Loki only tanks purple in this team comp as well, it's far more sustainable than XF / GT / Hood, where hood tanks yellow and blue and is a health pack drain.

The only issue is that vs XF / GT / Hood, XF / GT / Loki is worse because Hood randomly still just wins you the game in an unfavorable board. I would say that vs any other team composition, XF / GT / Loki is superior in pretty much all possible ways, but XF / GT / Hood beats out this team on defense.

The Hood also get purple over the 4* for having 12 in purple while no 4* has higher than 11 in their weak colors. In a Thor + X Force team, The Hood would still tank yellow, blue, and purple. On the other hand Loki would only get purple. That's a huge difference in sustainability since neither are really able to take hits (though Loki is still slightly better) but The Hood is expected to take close to 3X the damage based on his color matches. Besides, Loki is just more interesting to play than having The Hood's merciless passive grind an opponent down. Sure, Mischief is probably overpowered, but they do get one turn to spend the AP and there sure have been plenty of times where an opponent does get to unload their AP.

The only thing Loki doesn't do is that he really can't slow down boosts while The Hood can randomly slow down boosts if the board turns out to suck for your opponent in PvP. Since both characters are within the range of what can be killed by X Force (the move), this means Loki has no chance to delay an X Force but The Hood might. However with the changes to shields we might see less shield hopping in general so that should cut down the case where an opponent starts out with 5 or 6 green AP and then match green and immediately X Force. If they only have 3 green AP, denying the match 4 may very well be as good as AP drain. It is not hard to imagine you start with 3 green and a green match 4 is on board. Against The Hood you'd just take it and since that just eliminated 4 green you should be safe from Dormammu's Aid on green and the next green match will take out The Hood, but against Loki this would be an extremely dangerous move.

vudu3

NorthernPolarity wrote:

Mischief has even more subtle value than I thought it originally did: since it turns the same colored tiles that the match-4 occurred on, it effectively drains the board of most of that one color, giving you increased chances at favorable match 4/5s.

Isn't this also a negative if the AI matched a color that you're trying to collect? The AI steals your red match-4 and then Mischief creates 4 red CD tiles. There's suddenly a lot less red tiles on the board than there were 2 turns ago. If you're lucky Mischief might steal back 2 or 4 red AP but it's not going to get you to the 10 needed for Smite.

NorthernPolarity

vudu3 wrote:

NorthernPolarity wrote:

Mischief has even more subtle value than I thought it originally did: since it turns the same colored tiles that the match-4 occurred on, it effectively drains the board of most of that one color, giving you increased chances at favorable match 4/5s.

Isn't this also a negative if the AI matched a color that you're trying to collect? The AI steals your red match-4 and then Mischief creates 4 red CD tiles. There's suddenly a lot less red tiles on the board than there were 2 turns ago. If you're lucky Mischief might steal back 2 or 4 red AP but it's not going to get you to the 10 needed for Smite.

Good point! It can either be a positive or negative depending on what the color is. I would say though, that since you generally control what match-4s your opponents get (i feed my opponents a lot of irrelevant match-4s with Loki), this is more positive than it is negative. In the case of the AI getting lucky cascades on you, it has the potential to be negative. In the case where you feed your opponent a match-4, you can either take the match-4 if it would destroy a color that it would need, and give it to them if it destroys a color you don't need, leading to net positive overall.

Unknown

vudu3 wrote:

NorthernPolarity wrote:

Mischief has even more subtle value than I thought it originally did: since it turns the same colored tiles that the match-4 occurred on, it effectively drains the board of most of that one color, giving you increased chances at favorable match 4/5s.

Isn't this also a negative if the AI matched a color that you're trying to collect? The AI steals your red match-4 and then Mischief creates 4 red CD tiles. There's suddenly a lot less red tiles on the board than there were 2 turns ago. If you're lucky Mischief might steal back 2 or 4 red AP but it's not going to get you to the 10 needed for Smite.

It's fairly random which color the AI ends up getting a match 4 so by the same token if it eliminated another color that makes it easier for you to collect red. Illusions benefits from the elimination of a color, though it'd likely be a negative while fighting The Hood as he benefits more from eliminating a color on the board. In general eliminating one color will lead to better cascades which is generally bad for the player because in PvE you don't really want to trade big cascades with overleveled AIs, and in PvP where you start with an advantage, you don't really want to increase the tempo that will reduce the value of your initial advantage though you do get first dibs on the new board created by Mischief.

Unknown

How are people figuring not much of a difference in terms of cascades between 3 & 5 purple? Isn't it something like 26 tiles vs. 32? That sounds like a big difference? Especially when black is a meh ability mainly used on specific PVE matchups?

simonsez

Why would 6 more switches necessarily mean more cascades? There are plenty of times I see a match-5 set up, but by the time all the switching is done, the match-5 gets switched out. And NP's simulator also says the difference is minimal.

Unknown

Switchman wrote:

How are people figuring not much of a difference in terms of cascades between 3 & 5 purple? Isn't it something like 26 tiles vs. 32? That sounds like a big difference? Especially when black is a meh ability mainly used on specific PVE matchups?

Because it's actually 52 vs 64, which is a difference of 0 if the board has 12 or more teamup tiles, and a minimal difference if it has more than 0 teamup tiles

Unknown

simonsez wrote:

Why would 6 more switches necessarily mean more cascades? There are plenty of times I see a match-5 set up, but by the time all the switching is done, the match-5 gets switched out. And NP's simulator also says the difference is minimal.

Good point! Thanks

NorthernPolarity

I've also run simulations on using illusions right after polarizing force (which is realistically the least amount of TUs on the board), and you can see those results show that the difference is like 10 vs 11 tiles destroyed on average, which is marginal as well.

turul

Some live simulations:
Level 5: http://mpqsim.0cdn.net/?commands=Magnet ... runs=10000
4: http://mpqsim.0cdn.net/?commands=Magnet ... runs=10000
3: http://mpqsim.0cdn.net/?commands=Magnet ... runs=10000

scottee

For the record, I don't think the simulations and current theories are correct.

I've been observing the swaps a lot more now that I use Loki regularly.

32 swaps doesn't mean that it selects 64 tiles at the beginning (32 pairs) and then switches them. And that if there are less than 64 colored tiles then the swaps are wasted.

At Illusions level 5, some tiles get swapped more than once (I've seen a countdown tile move 3 times in one Illusions). And also, when there are clearly less than 64 colored tiles on the board, I've seen a colored tile never get swapped.

What I think is happening is, at the beginning of the cast, it selects 32 tiles only. Then it processes through each individual tile. For the first tile, it finds another colored tile and switches the pair. Then it moves on to the 2nd tile, and finds a tile to swap with, regardless of whether either tile has moved yet. This would explain how some tiles can move three times, while others move none.

turul

scottee wrote:

For the record, I don't think the simulations and current theories are correct.

I've been observing the swaps a lot more now that I use Loki regularly.

32 swaps doesn't mean that it selects 64 tiles at the beginning (32 pairs) and then switches them. And that if there are less than 64 colored tiles then the swaps are wasted.

At Illusions level 5, some tiles get swapped more than once (I've seen a countdown tile move 3 times in one Illusions). And also, when there are clearly less than 64 colored tiles on the board, I've seen a colored tile never get swapped.

What I think is happening is, at the beginning of the cast, it selects 32 tiles only. Then it processes through each individual tile. For the first tile, it finds another colored tile and switches the pair. Then it moves on to the 2nd tile, and finds a tile to swap with, regardless of whether either tile has moved yet. This would explain how some tiles can move three times, while others move none.

Actually, i did your way of shuffling erlier for simulation. It does not make much of a difference.

Unknown

I'm definitely seeing the same CD that gets swapped more than once in Illusions.

However in general it doesn't matter whether you're moving a tile around the world or moving the world around the tile, so that shouldn't have a significant impact on the cascade. It will have a bigger impact on the likelihood of getting a particularly troublesome tile out of a bad spot (like the bottom row). If a tile can be swapped more than once, this makes it harder to reach any particular tile in general because some are swapped twice. Note that you get no additional value for swapping a troublesome tile more than once so it's a strict loss. This would mean having more swaps matters more to dig out a particularly bad spot, since moving the world around usually does very little to help you reach a CD that's sitting at the bottom left/right corner and you really want to move that tile to a more manageable spot.

scottee

Yeah, I think for cascades, the increased chance is minimal, though there is an increase.

Before, the thought was, if there's only 52 colored tiles, level 4 and 5 make no difference.

I think it's more accurate to say, if there's only 52 colored tiles, level 5 does make a difference, but it's small.

I'm planning to keep level 5 though, as I see no need for level 5 in black.

MarcusGraves

scottee wrote:

I'm planning to keep level 5 though, as I see no need for level 5 in black.

you want 5 in black if you're using him with patch (who is arguably his best partner still) to make sure the enemy gets no strike tiles with rank 5 BR.

NorthernPolarity

scottee wrote:

Yeah, I think for cascades, the increased chance is minimal, though there is an increase.

Before, the thought was, if there's only 52 colored tiles, level 4 and 5 make no difference.

I think it's more accurate to say, if there's only 52 colored tiles, level 5 does make a difference, but it's small.

I'm planning to keep level 5 though, as I see no need for level 5 in black.

I currently have 5/3/5 Loki, and I've found a new, slightly more important justification for it: animation speed. Level 3 illusions takes almost half the time to cast animation wise because it swaps half the number of tiles. This means that you're probably saving a couple of seconds every time you cast illusions, which adds up over time when you're shield hopping and rushing for a PvE/PvP, not to mention that its just annoying waiting for that many tiles to be swapped. Those seconds saved + making 5 black usable (which I still haven't used) more than justifies the marginal increase in cascade potential.

scottee

MarcusGraves wrote:

scottee wrote:

I'm planning to keep level 5 though, as I see no need for level 5 in black.

you want 5 in black if you're using him with patch (who is arguably his best partner still) to make sure the enemy gets no strike tiles with rank 5 BR.

I have level 5 berserker rage. Level 3 black from Loki changes 5 of 6 tiles. They still do no damage.

Unknown

The only time I think the level on the black would matter is if you did 2 Berserker Rages or you're fighting Muscles that cannot move the board. Even then, if you're fighting something that requires 2 Berserker Rage to win you should have a lot of time to get the AP to setup for contingency plans, while the latter is a fairly trivial situation (if the enemy can move the board, you'd already be dead if even a couple Threaten tiles are on the board)

simonsez

scottee wrote:

I have level 5 berserker rage. Level 3 black from Loki changes 5 of 6 tiles. They still do no damage.

But if you're against an opponent that also spams strike tiles (Daken, Blade, whoever), they're going to get in the way of level 3 black, and there's a good chance you'll leave more purple than you want.