Treasure Hunt high scores 1.0.9 (screenshots)

Curse you (-2 to a random stat) @DEMONorANGEL… beat my best by 5…


@AnalSwordfish , this TH score calculation was truly impressive. I’ve been meaning to ask you, how does a mathematician even know where to start with this kind of derivation?

I recently watched Marcus du Sautoy’s BBC documentary on “The Story of Maths” (and how it evolved) and even studied 3 years of quantum physics in school, but the ability to reverse engineer a function by observing its behaviour has always been something of a mysterious back box.

For Treasure Hunt, was it sheer pattern recognition that you knew the solution had to be a generative Fibonacci power series? I don’t know anything about that. How did you know that the weighting for each type of loot would be linear?

Regarding those who think Navyguy’s score of 1386 had to be fake, I’d love to analyze the elements of the 19-vault image that he provided here, to reassure skeptics that his near-endgame score must indeed have been above 1000. But I’m afraid that I don’t know enough to unroll the summation, as you put it, to pump out the correct scoring result.

I’m sure it’s a fake, and I’m sure the final score is also calculated correctly, as the hack somehow replaces the screen contents/variables before that end calc is done…

This is one of many 1000 move scores i have seen. It is easy to call hacker when one things they are cheated by a better player. Also in reguards to the above formula it would not be hard to plug in the numbers to find the correct end score.

Why can’t devs come and confirm this? @Sirrian @Nimhain

Hey folks, well we DID take a look at those screen shots… they don’t seem to be photoshopped, and no hacking registered on our servers for those games. That was about all the time we had to spend on it with 2.0 closing in to a major deadline for us. The scores DO seem crazy high… if I had time I would attempt some math to see if they are possible, but my experience tells me you should never discount the abilities of players to achieve insane things legitimately, so we don’t ever want to be hasty in calling hacks if we don’t have hard proof in the form of server-logs!

We take Hacking/Cheating very seriously, and have lots of new countermeasures coming in 2.0.

Hacking is a constant problem in the game, and behind the scenes we have various checks running & we silently ban dozens of accounts every week - some of them are quite cheeky and contact support and pretend they HAVEN’T been hacking! :rage:

No doubt there are some out there we haven’t caught yet, but we are working on it.


To be honest i was able to secure 2 vaults in 99 moves and am not very good at the game. i can nail a vaul in as little as 30 moves easy. Knowing this and the numbers given i see no reason why it would be impossible to achieve that many vaults in that many moves.

Thanks for confirmation, and sorry for the distraction :slight_smile:
Sounds like 2.0 is getting very close :smiley:


“On the other hand, I consider myself very skilled on this kind of heuristic-based games and my inability to reach a score near to his screenshot made me feel an urge to be able to test my heuristical assumptions.”

I think you, like me and probably most people, are playing mostly on heuristics. But the guy who gets 1000 moves is doing lots more calculation of exactly what the move will do.

I’m usually trying to calculate all the chain reactions. I was talking about heuristic because the newly dropped pieces are randomly generated (actually, I’m currently trying to figure out if this assumption is correct…) and therefore, even if short term outcome is mainly due to chain reactions that can be accounted for with good precision, mid and long term effects of a specific move cannot be precisely predicted, requiring some heuristic reasoning, I’m talking about things like trying to keep more evolved pieces on the bottom of the screen, or prioritization of short term equivalent moves.

I don’t spend minutes on every move, therefore I for sure miss something and this could lead to the big score gap. What I’m trying to figure out is if the gap is due to too much little thinking about the chain reactions or if it’s due to a wrong heuristical assumption.

I use Treasure Hunt to see how fried my brain is from work. Rested >90 moves, exhausted <50.


…and finally breached the wall of 200 moves for me :slight_smile:


Grats, well done

With enough practice you can get alot of moves in and good rewards.

For me TH is still less economical than PVP for generating gold quickly… it’s pretty good for Glory though - but I just tend to target making the low tier chests and finish the game in 4–50 moves…

So I have a question regarding the authenticity of the screenshot:

How can you generate a vault on the top row?

Coins & Bags fall from the top - with occasional wooden chests. Let’s say you got 2 wooden chests to drop, and moved a third on from the second row to the top row, placing a green chest at the top-right corner. Then you could repeat that (with WILD luck) and perhaps get a red chest on the third-from-right position.

Then you would need to get some more lucky matches to shift that red chest to the right twice WITHOUT TOUCHING ANY OF THE LOWER ROWS.

Then you would need to repeat that twice more, to get a vault on the top row.

But the screenshot shows TWO vaults on the top row - and they are ADJACENT!

That’s just too much for me. The thought of how many perfect drops you would need to make that happen is… just not really in the realm of skill.

1 Like

You simply have to strain credulity a bit, et voila! Top Row Vaults.

Could do it by running out of legal moves and causing the board to shuffle. Seen strange things happen …

That being said, I really don’t think he could have gotten that many moves.


Board Shuffle when running out of valid moves as Shimrra said. So it is technically possible.


Yes! Excellent point! That would explain it neatly.