News for hackers. Stuff that matters.
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by OskarS on Tuesday April 10, 2018 @08:39AM (#16800877)
I feel compelled to write every time 2048 comes up: 2048 is a shameless rip-off of another game called "Threes" which came out just before it. "Threes" is more or less a game-design masterpiece, far superior in every way (graphics, sound, design, gameplay, you name it) to 2048. The reason 2048 is more popular is that it was a free clone of Threes!, and people refuse to pay even a little bit of money on mobile platforms for games.

This is nothing against this article, of course, which is an excellent write-up. But: if you like 2048, you'll like Threes better, and the creators of Threes deserves to be rewarded for coming up with the design.

Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by k3d on Tuesday April 10, 2018 @09:13AM (#16801072)
For folks who are interested in the what went into designing it, the devs published their emails over the course of development. It's a pretty neat glimpse into how a game's design evolves, as well as the effort it takes to come up with a clean, "simple" mechanic: http://asherv.com/threes/threemails/
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by thomasahle on Tuesday April 10, 2018 @09:59AM (#16801381)
It's interesting he says that only 6 people in the world have ever seen a 6144 tile. Automatic controllers can reach it pretty often, https://arxiv.org/pdf/1606.07374.pdf says 7.83% of the tries.
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by ryanianian on Tuesday April 10, 2018 @10:45AM (#16801724)
Maybe inspired or stole the core concept, but 2048 is, to me, much more fun than Threes (I've logged a couple hours on Threes and many dozens of hours on 2048; I paid for Threes and kinda hate the 2048 app I use but it works - it's not the "free" quality of 2048 that keeps me playing it).

It's fair to say 2048 is a shameless rip-off, but that doesn't mean it's not better in ways that matter.

Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by scrooched_moose on Tuesday April 10, 2018 @09:52AM (#16801323)
The "fanboyism" around these two games is crazy. 2048 might have been a knockoff, but it's a very different and unique game. I actually really dislike Threes because it's a "play until you get screwed by the RNG" type game. I gave up on Threes after about an hour (I didn't "refuse to even pay a little") and went back to 2048.

In 2048 you can keep playing far longer and are much more in control. Sometimes you get a number in a bad spot, but some strategy and additional luck can get you out of many of them. The sliding mechanism is also behaves differently which significantly changes the gameplay.

Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by KirinDave on Tuesday April 10, 2018 @10:42AM (#16801698)
It's pretty unfair, in my opinion, to gate your acceptance of 2048's theft on which game you like more.
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by scrooched_moose on Tuesday April 10, 2018 @10:53AM (#16801784)
Why is it "theft" with these two games though?

Mario "stole" from Snokie which "stole" from Quest for Tires which "stole" from Jump Bug which "stole" from Wanted: Monty Mole which "stole" from Donkey Kong which "stole" from Space Panic.

The entire history of video games is taking an existing concept and adding small tweaks to create something new. Threes/2048 is no different but for some reason and incredible amount of emotion and vitriol is tied up in this one.

Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by Cyph0n on Tuesday April 10, 2018 @11:51AM (#16802265)
I'm not sure how Mario "stole" from Donkey Kong if both games were made by Nintendo...
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by mclehman on Tuesday April 10, 2018 @05:14PM (#16805503)
I wouldn't characterize "stole from" as a transitive relation.
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by Retric on Tuesday April 10, 2018 @11:38AM (#16802137)
The games are no where near identical which means it's not theft.
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by zmk_ on Tuesday April 10, 2018 @09:22AM (#16801124)
If I remember correctly Threes is free now as well.
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by criddell on Tuesday April 10, 2018 @10:35AM (#16801652)
You're right, both are free and if somebody hasn't tried either yet, I would say download both.

I had played 2048 first so when I tried Threes, it didn't feel right to me at all. If I had tried Threes first, I probably would have had the opposite opinion.

Some of the rules are slightly different (I don't remember how), but the two games are pretty similar. Other than rule differences, the sounds in Threes were very grating to me. I ended up muting all of the sound effects and music. One huge plus to 2048 (IMHO) is that it was released as open source.

Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by Froyoh on Tuesday April 10, 2018 @08:45AM (#16800908)
It's not a "shameless rip-off".
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by Froyoh on Tuesday April 10, 2018 @08:53AM (#16800955)
"Created by Gabriele Cirulli. Based on 1024 by Veewo Studio and conceptually similar to Threes by Asher Vollmer."
Re:The Mathematics of 2048: Optimal Play with Markov Decisio... by Retric on Tuesday April 10, 2018 @11:39AM (#16802159)
Conceptually similar is a not theft. It's a significant improvement on a very simple game.

It's like comparing Chess and Checkers. Similar sure 8x8, move one piece at a time, promote on other side of board etc, but they play as very different games.