M: This is going to sound silly on the surface, but you'll see what I'm talking about when you dive in. I need you to go online and try and figure out for me the actual rules, as best as you can figure out, for the current version of Marvel Puzzle Quest. I need you to try and explain the abilities of Multiple Man in practice. The character seems to be bugged or something, and I can't seem to discern a pattern in terms of what actually does or doesn't work with him.
G: TL;DR: Multiple Man is one of the least intuitive characters in MPQ because his powers don't create "tiles" in the normal sense. They create Dupes, and Dupes only behave like their tile type when they are actually removed from the board. Most of the reports I found from players who thought he was bugged turned out to be misunderstandings of exactly when a Dupe is considered to have activated. citeturn0search1turn0search2turn0search0
Here's the practical version.
Every Dupe is a special object with a color attached to it.
When you use Yellow ("My Own Best Friend"), you choose a color and spend AP of that color. For every 2 AP spent, Jamie creates Dupes of the corresponding type. citeturn0search1
The important part:
A Dupe is not necessarily that tile.
It only behaves as that tile when it is removed from the board in the appropriate way.
Yellow Dupes are trap tiles.
The text says:
when this tile leaves the board, Multiple Man is healed
Players naturally assume:
"I matched it away. Why didn't I get healed?"
The problem is that "leave the board" and "be destroyed" are not always treated identically by MPQ's engine. Multiple reports exist of players expecting healing from Yellow Dupes and not getting it because the removal method didn't count the way they expected. citeturn0search2turn0search0
In practice, the safest assumption is:
This is not unique to Multiple Man; trap tiles have had weird interactions throughout MPQ history.
Black Dupes create a repeater called "Worst Enemy."
The repeater:
every turn. citeturn0search1
This means Black Dupes are often actively harmful unless you're exploiting:
Many players accidentally generate Black Dupes expecting a benefit and then wonder why Jamie is punching himself in the face.
Green power:
The key sentence is:
Destroyed tiles deal damage, but do not generate AP. citeturn0search1
So if you're thinking:
"I just destroyed 8 red Dupes. Why didn't I get 8 red AP?"
That's not what happens.
Instead:
This makes it feel inconsistent if you're expecting normal tile destruction behavior.
His passive eventually says:
Win the match if X friendly Dupes exist.
Depending on cover level this threshold changes. citeturn0search1
The strange part is that this counts friendly Dupes, not necessarily visible special tiles you can easily count by eye.
So sometimes players think:
"I have a ton of Dupes, why didn't I win?"
while the game is counting a different number than they're estimating.
If you're seeing behavior that seems random, the culprit is probably one of three things:
Yellow and Purple Dupes only trigger under specific removal conditions.
Powers that destroy, replace, convert, or overwrite tiles often interact differently than a normal match.
Reabsorb gives AP according to its own formula, not according to standard tile destruction rules.
If you describe a specific board state ("I had 4 yellow Dupes, matched them with Polaris, and got no heal" or similar), I can probably reverse-engineer whether you found an actual bug or one of MPQ's notorious wording problems. Based on what I found, Multiple Man has generated enough confusion that there are Reddit threads literally titled things like "how does he work?" and "this power will have no effect." citeturn0search2turn0search0
M: But that's just it. It doesn't spend yellow, as far as I can tell, unless you use the ability on yellow tiles. And it feels like a lot of the related colors that aren't yellow will try and create a number of duplicates equal to the number of yellow AP you have, not half of that, which can lead to situations where, because there aren't enough tiles available, it won't fire. At least I think that's the case. And that's why I'm asking if anyone else is aware of this, or if it's just a recent thing, or what. Because when I tried looking it up, there were discussions about it being broken in so many different directions that I decided to try and get you to help filter things out. So I'd like you to go look again with an eye towards getting information just from within, like the last month.
G: I think you've stumbled onto something real, and the reason you're having trouble finding an explanation is that there seem to be three different versions of Multiple Man being discussed online simultaneously:
Looking at the current published power text, Yellow is supposed to work like this:
"Drain up to X AP in the selected color; then for each 2 AP used create 2 Dupes..." citeturn0search6
If the text is accurate, then:
What caught my eye is your observation:
"it feels like non-yellow colors are creating Dupes based on yellow AP"
That is not supported by the published description at all. If that's what you're seeing, then one of two things is happening:
The power preview may be calculating from Yellow AP while the actual resolution uses the selected color.
MPQ has had plenty of UI-preview mismatches over the years.
The creation count may genuinely be referencing Yellow AP internally.
This would explain exactly the behavior you're describing:
That would produce a very confusing "sometimes it works, sometimes it doesn't" pattern.
Historically, Multiple Man has had enough weird interactions that players reported:
So the character already has a reputation for edge-case behavior.
I spent most of the search budget trying to find reports from the last month specifically, and I honestly didn't find a recent bug thread discussing the exact Yellow-AP issue. What I found instead was a lot of fragmented discussion and old bug reports. The official forums have largely moved around and aren't nearly as searchable as they used to be. citeturn0search1turn0search16
Try this exact setup:
| Yellow AP | Red AP | Selected Color |
|---|---|---|
| 0 | 10 | Red |
| 10 | 0 | Red |
| 10 | 10 | Red |
If the number of Dupes tracks Yellow AP, then you've found a genuine bug.
If it tracks Red AP, then the bug is elsewhere.
If you tell me:
I can probably reverse-engineer the formula from the numbers. Right now your description actually sounds more like an internal AP-reference bug than a misunderstanding of the power text.
M: OK, I have to check something, because what you just said actually might make sense. Now, I'm trying to read this on a phone, and that is a very long description, so if it's true, then I don't feel bad about missing it. But I hadn't actually looked at how other AP was reacting only yellow. I was so focused on what I thought was happening that I didn't get to notice what actually might be happening. I'll be back if that actually turns out to be the case.
G: That actually sounds like the right next step.
One thing the search turned up that I had glossed over before is that the official wording is very explicit:
"Drain up to X AP in the selected color; then for each 2 AP used create 2 Dupes..." citeturn0search2turn0search3
The reason your observation caught my attention is that the power is not supposed to care about Yellow AP at all, except that Yellow is the button you're pressing. The amount drained should come from the color you selected after pressing the button. citeturn0search2turn0search3
So if you were mentally tracking:
it's very easy to accidentally conclude "Yellow AP matters" when the real test is:
The fact that you've now realized you weren't watching the non-Yellow AP pool is exactly the kind of thing that solves mysteries like this.
Also, don't feel bad about missing it. Multiple Man's text is effectively three powers stuffed into one description, with variable AP costs, special tile rules, a win condition, and a bunch of exceptions. Even veteran MPQ players spent weeks arguing over how he worked when he launched. citeturn0search3turn0search0
If your test shows that:
then the power is probably working as designed and the confusion is just the wording.
If instead you find:
then you've found something much more interesting, because that would directly contradict the published power description. citeturn0search2turn0search3
I'll be interested to hear what numbers you get when you check, because that's a testable hypothesis rather than MPQ folklore.