Show HN: 2048 turned 10 this year, I built an updated version to celebrate
play2048.coHi all! My name is Gabriele Cirulli, I’m the original creator of 2048. Ten years ago, someone posted 2048 on HN[1] and suddenly it seemed like everyone was playing it. Back then, I wrote a bit about my experience during those weeks[2].
Even after all these years, that experience still feels a bit surreal to me. I’m deeply grateful to everyone who connected with the game, whether in small or significant ways, and for the stories shared along the way. Some people expressed how they were going through tough times and found some comfort, however small, in playing 2048.
At the start of last year, I wanted to breathe new life into the game as it was starting to show its age. I quit my job last October to work on 2048 full time and spent a year building this new version (the original took just 5 days!). I wanted to pay tribute to what made 2048 great while modernizing and polishing the experience.
The idea of adding powerups came when Prime Gaming and I connected to see if we could create a special version of 2048 for their members, with some exclusive extras. Some of those powerups made it into the main game, though there’s still a Classic[3] mode just like the original for those who prefer a more hardcore experience. The old site is also still online[4].
2048 is now my full-time focus, and I’m excited about the ways it can keep improving. I wanted to share this update with the community where it all began, both for a bit of nostalgia and to hear your thoughts and feedback!
Thank you all so much!
— Gabriele
[1]: https://news.ycombinator.com/item?id=7373566
[2]: https://medium.com/@gabrielecirulli/2048-success-and-me-7dc6...
This may be buried in the comments and you will never see it, but thank you very much Gabriele. Your game helped me in a very weird circumstance.
I was afraid of flying, specially on the takeoff and landing (and turbulence as well, ha). So I read somewhere that if I focused on something else, it would help me. So for the past years, I played 2048 during takeoff and landing, and it worked. It helped me to focus on something else, not the airplane, and I started to enjoying more my trips.
Now I don't need to do it anymore, but just for the experience I still do it when I fly. So thank you for helping me with my fear!
That’s amazing! Such a cool story. I’m really happy that 2048 was able to help you this way :)
This is bizarre. I came into this thread to make this exact same comment - I play 2048 religiously during takeoff, turbulence, and sometimes landing. It works really well, though if you were to watch a replay of my game, you could tell when turbulence hits because I start making faster and less logical moves :)
Very bizarre, as I also found it helped with this! I was only a First Officer at the time, but even as a Captain today I find playing 2048 during take off and landing really helps.
Wait, you're a Captain, and you fear of taking off and landing? I have so many questions :). Also, as someone who is scared of taking off and landing, I'm even more scared now.
I'm afraid again now...
Always amazing to read about surprising positive impacts that a project can make in someone's life like this.
TypeScript was fairly new at that time and to learn it, I ported 2048 to TypeScript. It was fun!
Fast forward a couple of years, I was debugging an issue with a react component and glanced over the .d.ts of react. I was quite surprised when I saw that my name was in them. I never contributed to react's types myself.
It turned out that someone took some types I wrote for 2048 and used them in the very first type definitions for react: https://github.com/DefinitelyTyped/DefinitelyTyped/commit/4b...
It's still there to this date, but I've lost my TS port in the sands of time.
This doesn't make any sense. We should be celebrating 8 years, 16 years and so on.
In 8+16 years from now we can celebrate 2048 the whole year. Still time for the author to implement an Easter egg until then.
The true 10th anniversary actually happened 1000 years ago.
Congrats on the 10 year anniversary, though honestly I think having tried your new 2048 I'll go back to the classic build. It might just be the hours I've poured into the original but it feels faster without the additional animation. But still a lot of good work there and I'm wishing you the best of luck.
As for the argument about Threes!, I have to say that I've generally found 2048 to be a much more fun game; the full-screen sliding and the lack of the 1+2 mechanic makes things move much faster, which for me is a priority. That's definitely personal taste, but I hate the vitriol that comes up on the topic.
My favorite 2048 clone by far is https://ashervollmer.github.io/2048/128.html, which is just a 3x3 2048 that only spawns 2. I like it because it is possible to achieve total victory, a perfect game fills the board with a a final score of 7172.
Wow, has it already been ten years? I also wrote a clone of 2048 back then (https://github.com/nieware/gofusion), using Go and a QML-based UI, for a contest, and (to my astonishment) actually won the first prize, which consisted of a Nexus 7 tablet (which served me well for several years) and a rare original vinyl Gopher figurine (which is still sitting on my desk looking at me serenely with its googly eyes while I type this).
You know how the distribution of 2-value vs 4-value for new tiles is a weighted random function? And sometimes you get an unlikely 4 that really screw you over? Have you thought about adding a mode which always creates the worst option of the 2?
You could call it “God does play dice with the universe. They're loaded and he hates you.”
Like HATETRIS: https://qntm.org/files/hatetris/hatetris.html
I think 2048-AI had this, but I’m not sure. It’d surely be pretty evil. Imagine it turning on randomly when it detects you’re going too fast :)
Someone forked 2048-AI and added evil mode :) https://sztupy.hu/2048-Hard/ Originating in the StackOverflow comments here https://stackoverflow.com/questions/22342854/what-is-the-opt...
BTW, congrats on the whole thing; what a ride. I remember staying up all night implementing the AI after seeing it here, and the rush of seeing it win the first time, plus the added rush of seeing the AI post right next to the original post on here. Thanks for the fun!
Like a Cambodian version of minesweeper.
I've opened the game without reading the post. When I've hovered over the crown and saw the prime stuff I thought it was some kinda' joke or parody. Turns out it isn't. Which I think is even funnier in some meta/state-of-the-things-today way. As there's nothing left but laugh at the absurdity of our reality.
Same here, saw this post and thought it would be a nice way to pass time on my laptop which is tethered to 256kbps free roaming while my phone updates over the censored wifi so I don't waste my fast data (hence browsing HN as one of the few text-based websites still around).
I remembered 2048 as a nice all-javascript game that runs in memory... opened this "updated version" and was met with slow-loading fonts, Google Ads, sign in and payment features, and the game did not in fact work (UI rendered but the initial 2 tile never loaded).
What's absurd about a consumer brand having a gaming division?
For those who haven't heard of Threes, I highly recommend giving it a try. It's the original game that 1024, 2048 etc were cloned from (and I think it's still the best by far). Wikipedia has a good summary: https://en.wikipedia.org/wiki/Threes#Legacy
Thank you for your comment. I wanted to add that credit should indeed be given where it's due and that while I didn’t mention Threes in my original post, it has played a significant role in 2048’s existence. Threes is credited on the play2048.co site (and now also on GitHub), and I’ve always tried to acknowledge its influence.
That said, I think there’s a balance to strike. As I mentioned before, I created 2048 before I became aware of Threes, and while it’s important to credit inspiration, I’m not convinced that every creative project needs to trace back every indirect influence. 2048 began as a small experiment without any intent of gaining popularity, and it grew into something distinct, shaped by the viral spread and its community of its open-source variations.
I understand the value of recognizing origins, but I also believe 2048 has developed its own identity over time. I appreciate the feedback, and I’m always open to improving where needed. Thanks for sharing your thoughts.
Humans are incapable of absolutely original thought - whether our inspirations are conscious, sub-conscious or fabricated.
Wordle was just...
DropBox was just...
the iPhone was just...
Modern movies are just...
Don't allow others to dismiss your work. They are jealous they didn't do it first - which begs the question - if it was _just_.... then why didn't _they_ do it?
I was originally frustrated with your game and jealous of you for awhile, too.
I feel like this is a bit misleading. You say you weren’t aware of Threes at all when you made 2048, which may be technically true, but then you should probably clarify that 2048 was a clone of 1024 which was a clone of Threes. And this all happened within around a month of when Threes released… to the point where many people started accusing Threes of being a clone of 2048.
Let it go, Matt.
How about don't support people who profit off of cheap clones and "forget" to mention the original years later.
Cirulli never profited off of 2048, as far as I know. The iOS app that actually made money was a ripoff by someone else of his MIT-licensed code.
[flagged]
It seems like you are piling on just to pile on, not simply "pointing out the origin". There's already 52 matches for "Threes" on this submission. I think we get it. The author of has added credits to Threes. You won!
[flagged]
If this was your only comment, I wouldn't have said anything. But you have 5 (now 6!) separate comment chains going on about Threes. In addition to the other 5 top-level comments, and ~20 replies that talk about Threes. All saying the same stuff.
It's boring.
I didn't post any top level comments, I'm only correcting people who I see posting misinformation. Isn't the point of Hacker News for people to talk about things they have expertise in? This happens to be a situation I know a lot about since I was also a mobile puzzle game developer at the same time this happened.
>Isn't the point of Hacker News for people to talk about things they have expertise in?
No. Although that certainly can help, it's not the main point.
From the guidelines: "If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity."
Seeing the exact same thing posted about Threes in over half of the comments here is the opposite of gratifying one's intellectual curiosity. It's boring. Especially when the comment is just a slight rephrasing of a comment that's already been posted several times (or worse, solely a link to your own comment elsewhere in the submission).
[flagged]
Please don't do this. 13 comments on this is too much, and (as is typical) they've gotten progressively more off-topic and flamewarrish.
When people start arguing about what they did or didn't say, with swipes like "Oh come on," it's clear that curious conversation was left behind quite a while ago and it's time to stop.
I get why you'd want to get rid of this thread, but surely the problem was caused the much earlier comment from doppp that said "Let it go, Matt". From what I've seen in the past, you would usually remove that kind of personalized negative response, but not in this case?
I hear you, but different people draw the what-caused-it line completely differently—it basically always feels like the other person started it and did worse (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...). It's like the old adage "The fight started when he hit me back."
The reason I replied to you is that your account was producing quite a bit more than the other accounts in this repetitive and dyspeptic discussion.
But the comments largely doesn't add anything further interesting to the discussion. Its been mentioned and spoken about ad nauseam.
As Threes was getting famous I bought it and tried a bit, but it didn't really stick.
It's with 2048 that I actually got hooked, it felt like a more natural and seamless game to play. I think it's the simplicity (no cuteness, no craftyness) that helps abstract the game, and of course dealing with powers of two makes it all the more natural. It felt a lot easier to get in and out of the game, be it for 30 sec to check the train station or 4 hours until lunch break.
I feel like Threes was the cute and whimsical game, while 2048 could probably become the classic game, in the same kind of spot as Tetris.
While it's unfortunate that the Threes team didn't receive a proper recognition, I believe the popularity of 2048 mainly comes from the fact that it can be, in some sense, "finished" unlike Threes. Yes, you can continue after 2048, but it was a good milestone that is initially hard but can be done relatively easily once you've got the hang of it. The clear milestone makes every game play much more disposable unlike Threes, even though you can surely continue even after that. (I did play Threes a bit before 2048 got popular and realized this in retrospect.) The same principle applies to more recent Suika Game and I still find it amusing that the only major change for both was the title.
Hmm, I'd guess that difference isn't significant for the vast majority of players. And for those that care, the greater depth of Threes is surely an advantage.
I think 2048 became more popular because it was a) on the web, and b) free, whereas Threes was only on the iPhone and cost a few bucks.
Oh, and c) the OP, to give him his due credit, did a really nice job with it! It had the same kind of simplicity and virality as Wordle.
I will add, from playing with this just now, three's feels so much harder.
Yeah, that was also why I stopped play Threes shortly after trying it a bit.
Suika Game has the same core mechanic (and many clones) but still represents a fundamentally different game. The gameplay is not even in the same category, as it involves some kind of physical simulation.
... or did you mean that Suika Game itself wasn't the original game with the falling-circles mechanic?
More like the latter, but not exactly (and that's my fault, sorry). As I said, 2048 is primarily distinguished from Threes by its title which implies the clear milestone. While the author of 2048 probably didn't intend this---after all, it is a clone of another clone 1024---, it seems that the Threes team also overlooked this to me. The title Suika Game also seems to share this quality probably by accident, which was what I originally meant to say.
... As effective as those titles are, I suddenly find myself imagining an alternate timeline with games and franchises like Bowser Bash, Quest for the Triforce, 25 Lines, Megalopolis, The Space Race, Shadaloo Smackdown....
> It's the original game that 1024, 2048 etc were cloned from
This argument has always been silly.
Three's didn't invent sliding tile games. Sliding tile games existed going back to the 1980s and 2048's mechanics are different enough that I don't even think they're comparable.
They've existed in wood for a lot longer than that.
https://en.wikipedia.org/wiki/Klotski
It's not different to 1024, and not giving credit where it's due is not unintentional. It's like making a falling blocks game with tetrominoes, but the board is 20 blocks wide, and you "forgot" to mention that it's inspired by Tetris.
Threes was (I believe) the first iOS game I actually paid money for. You could tell that the author was a true craftsperson who really cared about every detail of the product. It always felt unjust that the clones got all the attention, and now the nostalgia.
Threes would have worked fine as a free web game if they prioritized attention, but they chose to make it an iOS only paid app, presumably because they preferred making money over gaining attention. That's a very reasonable tradeoff in my book, I'd rather have money in pocket than stranger's upvotes.
Here's a great blog post from the Threes dev about their creation process and the impact of the rip-offs:
https://asherv.com/threes/threemails/
>2048 is a broken game. Something we noticed about this kind of system early on (that you'll see hidden in the emails below). We wanted players to be able to play Threes over many months, if not years. We both beat 2048 on our first tries. We’d wager most people that have been able to score a 768 or even a 384 in Threes would be able to do the same using the fabled “corner strategy”. You probably could too! Just try tapping “up” then “right” in alternating order until you can’t move. Then press left. You may not get to a 2048, but you might just see your highest score ever.
>When an automated script that alternates pressing up and right and left every hundreth time can beat the game, then well, that's broken.
From my experience, this greatly overstates the "exploit". In 2048 you get to maybe 128 this way typically before you can't move up/right any more, then you have to start thinking after the left press. Basically whenever you slide away from the "preferred" corner, supposing your plan is to slide back promptly, there's always a chance that a random spawn gets in your way and complicates the plan. Getting to 2048 on the first try doesn't sound like a modal experience at all. (Of course, most new 2048 players won't have had the experience of developing Threes first.)
For that matter, the developer talks about how rare it is to see a 6144, but doesn't seem to acknowledge that reaching a 4096 in 2048 is far more difficult than reaching 2048.
At any rate, it's not at all immediately clear why having the player join 1+2 first before making blocks of 3*2^n, should noticeably improve the gameplay over having only powers of two. So IMO it's not that the gameplay of 2048 is fundamentally less interesting; the implementation just sets a lower standard.
(Though for what it's worth, I've wondered how it might go with the Fibonacci sequence - allowing 1s to merge either with 2s or other 1s.)
You just have to play Threes for a few hours and it becomes obvious it’s a much more interesting and deep game than 2048. Of all the things that can be debated about this situation, that feels like the biggest stretch of all.
I don't mean to "debate" the depth of Threes (having not played it), only to say that it isn't obvious from a description. On the other side of the equation, 2048 is clearly interesting enough to have sold (and, from what I can tell, marketed by word of mouth), and its players would seem to disagree with the Threes author that 2048 is "a broken game". The corner shake might seem like a tedious but powerful strategy; but it doesn't come anywhere close to trivializing the object of the game, and this is a casual time-waster anyway.
2048 might well have won out for its simplicity (although personally I think the audio had a lot to do with it). Screenshots of Threes development (from the page linked in the post I quoted) imply that for quite a while it allowed for making numbers of any 2^x*3^y form, and earlier versions of the game must have tried even more complex rules - even larger prime numbers like 79 show up. Eventually this reduced to only numbers of the form 2^x*3 (as well as 1 and 2). To me that looks like a strange left-over irregularity, even if it does improve gameplay.
(After reading the rest of the thread, I think I regret replying at all.)
> interesting enough to have sold
Hasn't 2048 always been free? When comparing a paid game with a clone that's free (or even "free-to-play with obnoxious ads and lootbox mechanics" not that 2048 is that) the latter will usually become more popular, and that certainly happened here.
I am not sure how I feel about it. I certainly don't believe anyone should be able to legally own an idea like "sliding tile number games based on powers of 2 with or without being multiplied by 3" but I also don't have a lot of respect for those who, lacking an original idea, resort to cloning someone else's creative work (or in 2048's case, I guess, cloning another clone). So I guess I have no problem with them existing, but don't feel any desire to give them accolades or to play their game.
>Hasn't 2048 always been free?
I paid something like $2 CAD in the Nintendo eShop for a 3DS version. (If it's a clone, it hews very close to the original.) I guess I can't be sure it's supposed to work like that.
That was a clone. The creator of 2048 never distributed anything besides the free web version.
https://kotaku.com/clone-of-clone-of-clone-now-on-3ds-eshop-...
I don’t agree that it is obvious that it is more interesting or deep. I have played a lot of hours of Threes and it is mildly more complex because the tiles only move 1 square at a time, but in my opinion that doesn’t actually make it more interesting or deep.
Try getting the 32768 tile in 2048. Yes, it can be done somewhat consistently, but it requires a deep and somewhat subtle strategy. Furthermore, the “snake” formation that every player learns at the beginning stops being optimal at a certain point, and thus the late game has much more variety than the early game. I do think that Threes is more complex, but I’m not sure if this complexity is really necessary.
That's been implemented - google "Fibonacci 2048".
you're right, it's not an "exploit"; it is literally the gameplay. go watch anyone play 2048 for more than 9 seconds and then try to tell me that isn't the case...
and why is the fact that the difference between threes and 2048 "not immediately obvious" salient... at all? what is it even supposed to mean? i'm not so great at number theory... that doesn't make me think that all those people are gods among humans. same with the obverse: i am really good at geometry, so honestly are we sure that the ancient greeks were even good at math? it's not immediately obvious to me that geometry even is math. they didn't even have calculators for god's sake!
I love being able to see the prototype evolution of the game on here. Thanks for sharing!
I don't understand the peeve people have about mentioning Threes all the time. Does every discussion about a FPS game need to point out it's the same concept as Doom or Quake or whatever? The vitriol in some of the comments here are quite weird to me.
It's also dishonest to label 2048 as a clone. Personally I never cared much for Threes, and same I guess with my parents etc which all got hooked for a while on 2048. 2048 strikes a good balance on being accessible and challenging, most people don't want it more complicated or deeper.
If anything, the 2048 hype must have helped Threes tremendously. Instead, many people act as if 2048 was a slight on Threes somehow, stealing their thunder. I actually bought Threes based on all the comments back then, but didn't really like it. Too cutesy, and too challenging when I just wanted to mindlessly swipe.
Does every discussion about a FPS game need to point out it's the same concept as Doom or Quake or whatever?
It's a fair question, but Doom and Quake were both very famous and successful.
What sticks in my craw a bit with Threes is that the clones came out really fast, and 2048 in particular because much more famous and successful, so Threes never really got the chance to shine as much as it deserved (except when die-hard fans like me keep coming out of the woodwork to hype it, as you say). And I still think Threes has a much better and deeper game design than any of the clones!
If one of the many clones and variants of Wordle had been a runaway success, and the excellent original had been relatively overlooked and forgotten, I'd similarly be promoting Wordle in threads like this.
It's not that I resent the success of 2048 -- to the contrary, the OP did a great job with it and the success is deserved. But I assume that many people who have heard of 2048 have not heard of Threes, and I'd like them to try it, because it's great.
>If one of the many clones and variants of Wordle had been a runaway success, and the excellent original had been relatively overlooked and forgotten, I'd similarly be promoting Wordle in threads like this.
Why? It's not a job, or even a fun hobby, to try and ensure forgotten things get the recognition they deserve.
I'm surprised it's hard to understand! I see it as like recommending a cult movie or book to people -- that's not an unusual thing to do, right? Like when you're talking about time travel movies and there's always that one person who pops up and says "OK, but have you seen Primer? You should see Primer!"
It can get tedious if you overdo it, sure.
>Like when you're talking about time travel movies
Sure, but this wasn't a thread about games, or sliding number games, in general, it was a thread about 2048 specifically.
>It's not a job, or even a fun hobby
However, some might imagine it to be virtuous.
>If anything, the 2048 hype must have helped Threes tremendously.
This. I've never heard anyone mention Threes outside of a sub discussion about 2048. For all the people here that claim to love it, it has had very little impact outside of being discussed as being similar to, or a precursor of, 2048.
> Does every discussion about a FPS game need to point out it's the same concept as Doom or Quake or whatever?
Well the great differentiator between puzzle games is the idea of the playing mechanism. The great differentiator between FPS is implementation. If I make an FPS, I didn’t really steal from Doom because the idea is pretty obvious. But if I make a sliding game that’s very similar to 2048, you might say I stole the idea. It’s like with patents, subjectively the mechanism of a FPS shouldn’t be patentable to me, but three or 2048 might be.
> three or 2048 might be
I think there can be a different line drawn for "what should be patentable or legally protected" and "what should be celebrated as creativity."
A clone of Tetris where the line pieces are 5 blocks tall, for instance, is not notable or creative. But also, I think someone should be legally free to make it if they're not infringing Tetris's trademarks or stealing their code.
I agree: even if the descriptions of Threes (never played it) are only half true, they are completely different genres: one is a deep puzzler that aims for the kind of challenge that some absolutely do require to accept a game as entertainment but many others would consider borderline work, the other is a super casual routine builder that derives its quality only from the challenge of balancing full autopilot mode with maintaining a bit of attention. I 'd guess that not only would the latter have failed (silently, not spectacularly) if the author had tried to sell it on some app store instead of just putting it out on the web, the former would just as certainly not have gone viral like 2048 did. I absolutely do share your belief that 2048 (and the resulting discussions) has sold far more Threes than Threes ever did, or rather than Threes would have in a timeline where 2048 never happened.
(PS: and on a more meta level, people like us feeling super clever about "2048 helped Threes!" might be the secret sauce responsible for much of the longevity of the games' shared virality)
I still play Threes regularly. It’s super accessible but still remains challenging after years of play. I even occasionally still set a new high score.
It’s exactly the same for me. Threes never gets boring. I can almost finish 2048 every time. For me, that’s the definition of boring.
Came here to post a link to Threes, it does make me a bit sad whenever people mention 2048 without Threes.
[flagged]
uhhhh. well if reading their personal attestation is insufficient...
Congratulations!
"Powerups with Amazon Prime" sounds like the famous bug report that grep (I think) should search on Amazon if it doesn't find the string locally, which was submitted IIRC in protest at the Ubuntu "lens" doing just that.
Congratulations on creating such a wonderful game. My wife still plays it, I would claim, daily. I would claim it's part of her evening relaxation :) I will most certainly forward your new developments to her and also try it myself. Good luck with your new job.
Interesting to see the times for speedruns of this game: https://www.speedrun.com/2048
I can't quite comprehend how a human would be able to achieve 24s
Oh, it's not obvious at all from the UI, but the left-most column is a link to the video of the run: https://www.speedrun.com/2048/runs/zxwdol8m
In the early phases of the game, just mashing two adjacent arrow keys (say, left and down) will get you pretty far without any thinking whatsoever. You only need to slow down when the board is about three quarters full.
There is a factor of luck here. I can usually get to 128 or 256 in that time, but it really depends on whether you focus on doubling up a single square or cleaning up the board.
256, sure, but 2048 would take 8 times as long
as a former record holder (1m18s, 2015) it's been fascinating to see this number go down over the years
Just to note to say thanks. Never have I played anything that gives me so much satisfaction. I found I had some weird aptitude for it.
> suddenly it seemed like everyone was playing it
That's exactly how I remember it, yes.
It started with only the tech people, but then it spread to all coworkers, friends, grandparents...
Anyway, congrats on this game but I won't fall in the same trap as 10 years prior, wasting hours upon hours of productivity!
> I won't fall in the same trap as 10 years prior, wasting hours upon hours of productivity!
One day later, did you succeed in that?
Bravo Gabriele, 2048 was an amazing piece of design and UI. Loved it. You did amazing work.
Cannot believe this is a decade old - will always remember playing 2048 on the NYC subway to school with my friends <3
Fun game. I played for the first time and got a score of 3048, with my largest tile being 256 and no powerups used. Not sure if that's a good score or not.
My one observation is that it didn't feel intellectually engaging until the board was fairly full. Until I got to ~2500 or so, I was just making random-ish moves that felt like they didn't matter much. So it took say 15min before I got to the "good part."
It would be nice if the game could be modified in some way to get to the sweet spot earlier. Maybe starting from a mostly-full "puzzle" board, with varying levels of difficulty.
Or if the early moves actually DO matter more - and my beginner brain just didn't realize it - maybe there's a way to explain that connection to newer players so that they engage their brain earlier on.
Almost feels a bit like chess opening moves, which to a beginner might not feel very impactful, but to an experienced player they can see the connection from that one early pawn move to something meaningful later on in the game.
Its more like a Rubick's cube - there is a trick to it. Getting a single tile to 2048 is fairly easy once you know it, and practically a guarantee once you've had a bit of practice.
The new powerups etc. are nice for removing a mistake, or for letting you get a score a little bit higher at the end game, but won't take you the distance unless you have good technique.
A big part of the magic is getting your head round it - like any other logic puzzle.
Unlike the Rubik's cube, I think you can get good enough at 2048 to usually get the 2048-tile based purely on intuition, though, without consciously memorizing anything. I've never looked up a strategy guide to the game and just playing purely by instinct (using heuristics I've randomly discovered myself), I can usually get there.
I'm curious what you mean by "the trick"!
All your ideas to make the game more interesting are solved: Just play Threes instead.
Thanks so much! I saw 2048 on hn back in the day (10 years?!) and loved it, and then I bought Threes! Which I still play idly today!
On a sidenote: it's been years, but anyone else slightly annoyed that the "Out of Moves" text is still not centered?
Hi gab, thanks for sharing. This was fun to play back during our Facepunch era, and it’s still entertaining now.
I thought I remembered seeing it back on there in the "What are you working on" thread. Was always interesting to scroll through.
Hello as well from a past Facepunch user, 10 years go by very quickly
@Terabytes glad to see your work, the new release is really nice
Likewise, glad to see fellow former Facepunchers here :-) Great work!
Thank you. I've thoroughly enjoyed 2048 the past 10 years. The new one isn't showing the numbers for me, though. I'm on Firefox on Android. Classic one works perfectly. Works fine on Edge for mobile (only have it for situations like this where I'm tracing browser dependencies).
This feels so much snappier than the original. I don't know if I'd use the powerups, because usually I play 2048 mindlessly and when I don't really want to think too much, but I might find myself pleasantly surprised in the future.
Random thing I noticed last week: there's a version of 2048 buried in the United Airlines iPhone app. Hamburger -> Games -> 2048.
Also in some inflight entertainment systems!
Thank you for building this game, Gabriele! I very much enjoyed playing the game - and later, building what was (at that time) the best AI for the game while procrastinating on my PhD! (https://github.com/nneonneo/2048-ai, https://stackoverflow.com/a/22498940/1204143)
I'd actually built an AI for Threes some time earlier (https://github.com/nneonneo/threes-ai), although (perhaps unsurprisingly) the 2048 AI turned out to be a lot more popular.
The timing is pretty funny too - I was on a flight just two weeks ago that featured 2048 as one of the in-flight games, so I played it myself for a bit, got to 2048, then got my AI to get me to 4096 - and promptly crashed the whole app! Your implementation, even though it took just five days, was remarkably robust and fast and has held up really well all these years - real props to the great job you did.
I guess now I'll have to update my AI for all the powerups you've added to the game :^)
Always interesting to hear the other side of the story: https://asherv.com/threes/threemails/
HN thread: https://news.ycombinator.com/item?id=7484106
This is still the only mobile game I play (besides chess, if that counts). Thanks for creating such a fun piece of software. It's snappy, beautiful, and tastefully done.
It is easy to underestimate any individual's contribution to society, especially if it can be (ridiculously) trivialized as "just a mindless game" (that hurt to type).
2048 has impacted our species. Seriously.
I can assure you from an objective, subjective, authoritative, personal, and collective position: your game has saved many people their sanity, entertained millions, saved/wasted billions, inspired untold people into math, web dev, and game design.
But most importantly, people didn't quit "mindless/thankless/frustrating/demeaning" jobs because their mind was _just_ adequately stimulated by your creation to continue to justify coming to work the next day.
Not demanding enough to command full attention (until we get a 4096, then nothing else matters), but 'choring' enough (in the most charitable sense) to keep people from finding a more or less demanding time-fill, likely compromising their jobs.
Jobs that help others, in an exponential way, sometimes.
Even if this was an isolated incident, I can tell you, you will forever be top 5 (hyper) viral games by a solo developer.
If a butterfly can cause a hurricane, 2048 could just as easily be the reason we survive the next filter, for all we know.
You helped me help others help others. And you deserve to know that.
BTW: I always blocked your ads (no offense, just cognitive stuff).
Drop a donation link somewhere, please.
Bug report: digits don’t show for me. (Interesting æsthetic, actually, unmarked tiles.) Firefox, Sway/Linux, Settings → Fonts → Advanced → “Allow pages to choose their own fonts, instead of your selections above” disabled, and web fonts blocked altogether via uBlock Origin.
That’s not really surprising with those settings.
It is, actually. Firefox's PDF renderer on text meeting certain conditions is the only thing I've encountered breaking in a similar way (not rendering text), in almost five years of using this configuration.
I expect it to just render with my chosen fonts. That's what everything else does.
Unrelated: I really wish the browser setting of "Allow pages to choose their own fonts, instead of your selections above" can be a per-origin setting. There are a small number of websites that use web fonts tastefully but for the majority I would prefer my own fonts.
I played 2048 A LOT back when it came out. I still fire it up every now and then.
Thanks Gabriele, what a fantastic game you have made.
The new power ups are a nice addition that allowed me to reach 2048 more quickly :)
I remember a coworker who said something along the lines of "I am free!" when he finally got 2048, and then never touched the game again. I hope the new power ups will make the game less addictive.
I made a version with llamas at some point, as a valentine's card. (Source code is yours, I just swapped the images!)
# of llamas = log2(original number)
https://people.csail.mit.edu/ebakke/twothousandandfortyppaca...
My favorite version is the wasm implementation written in rust.
Source: https://github.com/dev-family/wasm-2048
Playable here: https://2048.dev.family/
How time flies. I remember I wrote a zero-knowledge AI bookmarklet that did quite well...
http://stackoverflow.com/a/23853848/632039
So nice! I remember playing it back in the day, and then forking your repo to make it 128 instead of 2048.
Turns out the fork is still up. It's been literally 10 years.
https://128.arthurhess.com
One of my finals during my masters degree required us to implement a 2048 engine using OCaml. I had a bit of practice with OCaml but this specific exercise really made something click in me. It made me understand how to leverage a type system to represent a state, and make impossible states impossible to represent. I don’t use OCaml anymore but this exercise has definitely made me a better developer. That’s in part thanks to OCaml, and in part thanks to 2048 so thank you for inventing the game :)
2048 is only 10 years old, I am shocked.
I found it a while ago when looking for the source of xjumpjump and instantly liked it. Thanks for creating 2048.
Every so often I try and see if xjumpjump is available on cell phones, but none yet.
Sometimes I have trouble listening without my thoughts drifting elsewhere. I found out that playing a Tetris clone that never increases in speed and never ends is a good way to keep a clear focus on listening. After changing phones recently my perfect Tetris clone stopped working as it was 32bit and the support was dropped. I've been looking for an alternative and finally found one in a 2048 clone with an Undo button. I thank you for that. It has saved me a lot of trouble staying focused!
uBlock Origin blocks 330 items on that page, which I think is a personal record for the sites I have been to.
Consent banner says data sent to "901 partners"...
I like the game though, but I use the PWA made by Opera team at https://2048-opera-pwa.surge.sh/
There are a lot of comments mentioning Threes! and holding it up as superior. I play a lot of games, and I greatly prefer Threes myself -- but I also remember very well when 2048 came out; I was in college, and 75% or more of my dorm had no interest in Threes and preferred playing and remixing 2048. Threes is very stressful and intense, while 2048 goes straight for your brain's pleasure centers, like many other mobile games nowadays.
There's value in that. There's massive value in 2048 that Threes! does not capture. Just because someone spends a lot of time designing something, it does not entitle them to attention or praise. Many great artists get famous for works they absolutely hate. Gabriele has been consistently courteous about all of this for 10 years, and I'd suggest people here could be a bit more courteous to him too.
There is nothing wrong with pointing out the origins of ideas and concepts, especially on a site like Hacker News. 2048 was much more popular than Threes, often to the point where many people are unaware of Threes at all, so I think it’s important in these kinds of forums to make sure people know where the game idea came from, and that the originators get credit.
This manifestation of the threes trope below everything 2048 related reminds me of a comment I recently read below a Stephen Wolfram related post:
https://news.ycombinator.com/item?id=41406645
All: please don't repeat the usual Wolfram trope. (If you don't know what I mean by that, a decade's worth of explanation can be found via https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que....) The issue is not that it's wrong, it's that it's extremely repetitive and we want fresh discussion on HN, preferably about the specific content of an article.
I saw that comment but didn't follow the algolia link.
I didn't realize Stephen Wolfram was dang's personal We Do Not Mention the Orangutan. I suppose that given the inevitable prevalence of clichéd back-and-forth tropes and exchanges on HN, there's little choice but to draw a line in the sand on specific versions and hope it serves as a stand-in for all of them.
I don't know who the orangutan is, but that's just one of a thwack of tropes that make me wince on HN. A multithwack. A plenithwack.
They can't be stopped but one can maybe dampen them a bit sometimes. Or not.
TIL, and, on the off chance you didn't go look this up, it's extremely worth it:
https://www.reddit.com/r/EdgarAllanPoe/comments/17cehq8/we_d...
(no matter whether it's true or not)
Wasn't that just because Threes was a paid iPhone-only app and 2048 was a free near-clone?
no, brah, 2048 had better game mechanics, brah
To me, Threes poorly straddles the middle ground between a "real game" and a "time filler". If I want a "real game" I'll play Factorio or something. When I need a break from mental work I can pull 2048. I rarely want anything in between.
Lots of people do want that “middle ground” game, especially people who exclusively play games on mobile rather than PC or console.
[flagged]
[flagged]
An opinion was stated. An opinion that will no doubt be agreed with by many, me for one. But absolutely no claim was made that it is correct for all people. Or even most people.
To answer your sort-of question:
> Ok? But lots of people do want that “middle ground” game…
Those lots of people can play threes, or other games that exist in such a middle ground.
Not every game needs to support every play style, nor does every person need to like every game.
Or to be facetious and answer you in your own style:
> Ok? But lots of people do want that “middle ground” game…
OK? But lots of people don't want the “middle ground” game, especially people who like more complex games on desktop/console and only simpler time-fillers on mobile.
Or even more so:
> Ok? But lots of people do want that “middle ground” game…
Yes… and?
Alright I'll admit the "Ok? But" was unnecessary, and I've taken it out.
[flagged]
there's also value in creating the "thinking man's" progenitor to 2048. both are extremely valuable. and if we're doing a weighing of the souls, 2048 probably wins. but you can't blame people on this site for gravitating toward the underlying ideas rather than their instantiation-- they're (ostensibly) makers, or at least moreso than the general population.
Totally agree. I prefer Threes! myself but ultimately, the market determines what's fun. Grabriele has been so humble about his success and while I feel for the Threes! developers, they came across as exceptionally sore.
Was a fun game, I still remember trying it back then!
383,392 points on the first try, using the new powerups. At that point I had a 16384, 8192, 4096, 2048, 1024, 256 and two 128's, but getting that extra 512 and then successfully folding the tiles into a 32768 likely wouldn't have been feasible, even if I had had more luck. I think this was enough for me for a while =D
Your game has inspired me back then to create a game on quickly finding isomorphic graphs. Since then, I haven't created any other game in 10 years. So thank you for a powerful stimulus!
https://vpavlenko.github.io/fluffy-graph/
Here are the threads I could find with 20 or more comments. Smaller threads are listed in a collapsed reply. These are in chronological order for a change:
2048 - https://news.ycombinator.com/item?id=7373566 - March 2014 (410 comments)
2048 AI - https://news.ycombinator.com/item?id=7379821 - March 2014 (189 comments)
2048 – multiplayer - https://news.ycombinator.com/item?id=7384974 - March 2014 (113 comments)
2048 x 2 = 4096 - https://news.ycombinator.com/item?id=7396134 - March 2014 (24 comments)
2048 in the terminal - https://news.ycombinator.com/item?id=7398011 - March 2014 (88 comments)
2048 with Leaderboard and achievements, with Kivy (Python/OpenGL) - https://news.ycombinator.com/item?id=7404515 - March 2014 (25 comments)
Show HN: 2048 in 2048 bytes - https://news.ycombinator.com/item?id=7406605 - March 2014 (54 comments)
2048 in 3D - https://news.ycombinator.com/item?id=7416777 - March 2014 (66 comments)
2048 in 4D - https://news.ycombinator.com/item?id=7418219 - March 2014 (113 comments)
Flappy 2048 - https://news.ycombinator.com/item?id=7431047 - March 2014 (121 comments)
Show HN: Logarithmic Flappy 2048 - https://news.ycombinator.com/item?id=7432448 - March 2014 (64 comments)
HN Plays 2048 - https://news.ycombinator.com/item?id=7433524 - March 2014 (81 comments)
Show HN: 2048 Tetris - https://news.ycombinator.com/item?id=7435569 - March 2014 (60 comments)
2048 for physicists - https://news.ycombinator.com/item?id=7438567 - March 2014 (33 comments)
2048 Numberwang - https://news.ycombinator.com/item?id=7439444 - March 2014 (110 comments)
Show HN: 2048 without numbers - https://news.ycombinator.com/item?id=7443166 - March 2014 (46 comments)
Dropbox 2048 - https://news.ycombinator.com/item?id=7443379 - March 2014 (2 comments)
8402: 2048 from the other side - https://news.ycombinator.com/item?id=7446139 - March 2014 (53 comments)
2(048) - https://news.ycombinator.com/item?id=7453543 - March 2014 (46 comments)
2048 in sed - https://news.ycombinator.com/item?id=7464294 - March 2014 (35 comments)
2048 game to the Atari 2600 VCS - https://news.ycombinator.com/item?id=7466097 - March 2014 (22 comments)
2048 Solver - https://news.ycombinator.com/item?id=7473486 - March 2014 (38 comments)
Threes: The Rip-offs and Making Our Original Game - https://news.ycombinator.com/item?id=7484106 - March 2014 (208 comments)
2048 As A Service - https://news.ycombinator.com/item?id=7510670 - April 2014 (52 comments)
2048 implemented in 487 bytes of C - https://news.ycombinator.com/item?id=7535666 - April 2014 (45 comments)
2048 in 3D - https://news.ycombinator.com/item?id=7543483 - April 2014 (51 comments)
Flappy 2048 - https://news.ycombinator.com/item?id=7637009 - April 2014 (32 comments)
2048 in Famo.us - https://news.ycombinator.com/item?id=7660206 - April 2014 (21 comments)
2048, success and me - https://news.ycombinator.com/item?id=7704800 - May 2014 (222 comments)
Show HN: 2048 in Swift - https://news.ycombinator.com/item?id=7845441 - June 2014 (34 comments)
Implementing 2048 in 90 lines of Haskell - https://news.ycombinator.com/item?id=7896187 - June 2014 (24 comments)
243 Game – inspired by 2048 - https://news.ycombinator.com/item?id=7991773 - July 2014 (36 comments)
Design Is Why 2048 Sucks, and Threes Is a Masterpiece - https://news.ycombinator.com/item?id=8030413 - July 2014 (80 comments)
The Mathematics of 2048: Counting States with Combinatorics - https://news.ycombinator.com/item?id=15327837 - Sept 2017 (46 comments)
The Mathematics of 2048: Counting States by Exhaustive Enumeration - https://news.ycombinator.com/item?id=15894126 - Dec 2017 (22 comments)
The Mathematics of 2048: Optimal Play with Markov Decision Processes - https://news.ycombinator.com/item?id=16790338 - April 2018 (43 comments)
Show HN: 2048.cpp – Play 2048 in directly your terminal - https://news.ycombinator.com/item?id=17897283 - Sept 2018 (45 comments)
The Mathematics of 2048: Optimal Play with Markov Decision Processes (2018) - https://news.ycombinator.com/item?id=28310842 - Aug 2021 (50 comments)
Show HN: 1024, a 2048 Puzzle Game - https://news.ycombinator.com/item?id=32405510 - Aug 2022 (48 comments)
Show HN: Exponentile – A match 3 game mixed with 2048 - https://news.ycombinator.com/item?id=39897112 - April 2024 (40 comments)
Show HN: King Thirteen: 2048 with chess pieces, in under 13 KB - https://news.ycombinator.com/item?id=41623814 - Sept 2024 (80 comments)
This corroborate well with my memory of 2048 being absolutely all over HN for a while, heh. It was so fun with all the variants popping up, really a cultural phenomenon for a while.
But also made me think: what other "fads" have we seen?
Some that come to mind is a wave of "{x}, a {y} written in rust", or more recently the wave of things being thin wrappers/prompts on top of some llm/transformer. Was also a time it felt like it daily was a new js library on the front page. It would be cool with a history of hn of sorts, about what's been in vogue at times.
We've gotten a lot more aggressive (er, proactive) about downweighting follow-ups in the succeeding years: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que....
Makes me think we could maybe ease up on that when it comes to ongoing work.
Here are the threads I could find with fewer than 20 comments—again in chronological order:
Ask HN: Why is the 2048 post so popular? - https://news.ycombinator.com/item?id=7378022 - March 2014 (15 comments)
2048 was too easy : welcome 4096 - https://news.ycombinator.com/item?id=7386557 - March 2014 (9 comments)
Simple trick to beat 2048: left up, left down, repeat - https://news.ycombinator.com/item?id=7388699 - March 2014 (3 comments)
2048 with undo - https://news.ycombinator.com/item?id=7389112 - March 2014 (5 comments)
2048 in Augmented Reality - https://news.ycombinator.com/item?id=7391322 - March 2014 (2 comments)
JS1k demo details: 2048 - https://news.ycombinator.com/item?id=7407055 - March 2014 (14 comments)
Show HN: 2048 with shareable replays - https://news.ycombinator.com/item?id=7408918 - March 2014 (4 comments)
2048 Benchmark - https://news.ycombinator.com/item?id=7408952 - March 2014 (4 comments)
How to win 2048 everytime - https://news.ycombinator.com/item?id=7410582 - March 2014 (6 comments)
Invisible 2048 - https://news.ycombinator.com/item?id=7418310 - March 2014 (4 comments)
Why Do I Find 2048 So Damn Addictive? - https://news.ycombinator.com/item?id=7418806 - March 2014 (2 comments)
Words oh so great: 2048 meets Scrabble - https://news.ycombinator.com/item?id=7419264 - March 2014 (5 comments)
Xkcd comic about 2048 game - https://news.ycombinator.com/item?id=7427895 - March 2014 (3 comments)
Flappy 2048 in HTML5 Canvas - https://news.ycombinator.com/item?id=7428175 - March 2014 (17 comments)
Flappy 2048 AI - https://news.ycombinator.com/item?id=7432754 - March 2014 (2 comments)
HN Plays 2048 (democracy) - https://news.ycombinator.com/item?id=7434051 - March 2014 (14 comments)
Show HN: Gameboy port of 2048 - https://news.ycombinator.com/item?id=7434529 - March 2014 (2 comments)
2048 in bash - https://news.ycombinator.com/item?id=7435174 - March 2014 (18 comments)
Flappy + 2048 side-by-side - https://news.ycombinator.com/item?id=7436813 - March 2014 (4 comments)
All 2048 games - https://news.ycombinator.com/item?id=7438716 - March 2014 (4 comments)
2048 Directory - https://news.ycombinator.com/item?id=7439663 - March 2014 (4 comments)
WebRTC 2-player 2048 on same board with matchmaking - https://news.ycombinator.com/item?id=7445906 - March 2014 (8 comments)
2048.py - https://news.ycombinator.com/item?id=7446524 - March 2014 (2 comments)
2048 with Stats - https://news.ycombinator.com/item?id=7446608 - March 2014 (4 comments)
2048 game – How it started? Interview with the author - https://news.ycombinator.com/item?id=7448745 - March 2014 (3 comments)
1125899906842624 - https://news.ycombinator.com/item?id=7448931 - March 2014 (3 comments)
Popular versions of 2048 with previews - https://news.ycombinator.com/item?id=7449160 - March 2014 (2 comments)
What is the optimal algorithm for the game 2048? - https://news.ycombinator.com/item?id=7450244 - March 2014 (13 comments)
2048 showcase - https://news.ycombinator.com/item?id=7459448 - March 2014 (2 comments)
Myo armband controlling 2048 [video] - https://news.ycombinator.com/item?id=7463808 - March 2014 (4 comments)
Ask HN: Why doesn't 2048 have a '1' tile? - https://news.ycombinator.com/item?id=7464533 - March 2014 (2 comments)
Ask HN: how do you commercialize something like 2048? - https://news.ycombinator.com/item?id=7473770 - March 2014 (4 comments)
2048 Golf - https://news.ycombinator.com/item?id=7474824 - March 2014 (2 comments)
Code a 2048 bot on hackerrank - https://news.ycombinator.com/item?id=7481542 - March 2014 (8 comments)
Show HN: I found a way to 'cheat'/solve 2048 (With live demo) - https://news.ycombinator.com/item?id=7486762 - March 2014 (10 comments)
Free Open Source Software and the 2048 Problem - https://news.ycombinator.com/item?id=7490316 - March 2014 (9 comments)
Udacity Course: Make Your Own 2048 - https://news.ycombinator.com/item?id=7536123 - April 2014 (2 comments)
Building 2048 in AngularJS - https://news.ycombinator.com/item?id=7554348 - April 2014 (14 comments)
2048 for stars - https://news.ycombinator.com/item?id=7603732 - April 2014 (2 comments)
A 2048 spinoff to raise funds for cancer - https://news.ycombinator.com/item?id=7606252 - April 2014 (3 comments)
A physics version of 2048 – stellar nucleosynthesis - https://news.ycombinator.com/item?id=7618293 - April 2014 (2 comments)
2048 on 8-bit Sega Systems - https://news.ycombinator.com/item?id=7620914 - April 2014 (4 comments)
Show HN: 2048 Game of Thrones Edition - https://news.ycombinator.com/item?id=7656436 - April 2014 (2 comments)
2048 in Idris - https://news.ycombinator.com/item?id=7668540 - April 2014 (11 comments)
2048 in Emacs - https://news.ycombinator.com/item?id=7705947 - May 2014 (15 comments)
2048 in bash - https://news.ycombinator.com/item?id=7708183 - May 2014 (6 comments)
(Yet another) parody of 2048, where you try to lose as fast as possible - https://news.ycombinator.com/item?id=7713249 - May 2014 (5 comments)
2048, Wolfram Style - https://news.ycombinator.com/item?id=7721685 - May 2014 (6 comments)
2048 gestures - https://news.ycombinator.com/item?id=7780359 - May 2014 (2 comments)
2048 in Erlang - https://news.ycombinator.com/item?id=7869179 - June 2014 (14 comments)
Cubic 2048 - https://news.ycombinator.com/item?id=8022498 - July 2014 (4 comments)
Show HN: Gunoki – Trainyard-and-2048-inspired Puzzle Game for iOS and Android - https://news.ycombinator.com/item?id=8329527 - Sept 2014 (4 comments)
Twilio Plays 2048 – Multiplayer Without Internet - https://news.ycombinator.com/item?id=8498398 - Oct 2014 (3 comments)
Can AI beat 2048? - https://news.ycombinator.com/item?id=8724754 - Dec 2014 (12 comments)
Play 2048, get paid in Bitcoins - https://news.ycombinator.com/item?id=9195167 - March 2015 (3 comments)
Zer0: addictive Web number-game like 2048 between two news from HN ^^ - https://news.ycombinator.com/item?id=11202264 - March 2016 (5 comments)
Show HN: WebRTC Serverless 2-player 2048 Game with Annotated Source - https://news.ycombinator.com/item?id=12788147 - Oct 2016 (13 comments)
Show HN: Hex 2048 - https://news.ycombinator.com/item?id=15890753 - Dec 2017 (3 comments)
Ask HN: What are other awesome games like 2048? - https://news.ycombinator.com/item?id=16471374 - Feb 2018 (3 comments)
Show HN: Adversarial 2048 - https://news.ycombinator.com/item?id=31922891 - June 2022 (2 comments)
x86-64 Operating System to Play 2048 - https://news.ycombinator.com/item?id=36516496 - June 2023 (3 comments)
I tried it and it is very nice, but I completely ignored the extra features :)
My brain is just too hardwired to use cursor keys and doesn’t even think to look outside the box and click the buttons with a mouse, I guess.
I still remember how quickly this spread and people made their own versions, right about the same time as Twitch Plays Pokemon came out! I'm glad it's had such a positive impact both globally among your players and to your career. Praise helix!
20k run on the first try, enough for next 10 years https://cdn.jpg.wtf/futurico/8d/36/1729783571-8d36e171f9deb9...
My favorite 2048 clone is 2048 4D https://huonw.github.io/2048-4D/
What's with the premium version that's free with Amazon Prime?
I remember seeing it back then, and playing it for quite awhile. Great game, thank you!
It seems like there's a little pay to win even in a simple game like 2048 nowadays :)
Still very fun!
I didn't notice anything different from how it was in my memory. That seems like a successful modernization :)
I didn't notice that there were power ups until reading your post.
Just downloaded the new classic version with ads and wanted to pass along a suggestion that Thor makes.
While I understand that you’ve gotta make a living, ads are far more palatable if the user can consent to them.
And the way to make the user a part of the process is to offer a trade for the ad, rather than obtrusively running them.
“Would you like 500 gold to watch an ad?”
Yes? Ad for in-game currency that provides powerups? That’s a great deal and it’s win/win.
No? Okay, no ad, but no power up.
Statistically, you almost always get a yes, but the players are 100x happier than the slimey games that force ads in unscrupulous ways.
This is interesting feedback. I always thought that ads being unobtrusive and not part of gameplay would be the saner choice. In my mind, when games ask you to trade watching an add for some kind of in-game benefit or currency, that feels like an unreasonable demand. The ads impinge on the gameplay. The way I've set up the site, ads are tucked in their own corner and should never affect gameplay (though there have been instances of ads spilling over the powerup bar, I'm working on that). Thanks for sharing!
>And the way to make the user a part of the process is to offer a trade for the ad
I think that is a worse pattern overall. It's essentially converting from the idea of watching ads to support the developer to watching ads as a pay to win mechanism. Pay to win mechanisms are always worse than any other alternative.
Im curious on the real cost of the impact of this game had on big co. I remember everybody was playing non stop at work.
I have to ask, did the game make you rich?
I used to play this in high school. Great work you did. And I hope things turn out well for you.
What do you say to people when they accuse you of ripping off "Threes"? Is it true?
I liked Threes a lot, and I bought it.
It was fun. Later I discovered 2048, and it was also fun, and felt like a bit of a different game to me.
The author's reaction to 2048 rubbed me the wrong way a bit. https://asherv.com/threes/threemails/
Thankfully fans of 2048 pushed back and most decided that it's a simple enough game that you can't really expect there not to be clones, or clones of clones. I like this one: https://mdjorge.github.io/doge2048/
---
Edit: From the post about Threes:
> We do believe imitation is the greatest form of flattery, but ideally the imitation happens after we’ve had time to descend slowly from the peak -- not the moment we plant the flag.
Fair enough. Things moved too fast for that back then and they move even faster now. A coding AI could help create a ripoff.
I do feel a bit bad for them. Putting so much work into a project only to have it cloned a week or two after they released it to much bigger fanfare (I'd never even heard of Threes up until this point but everyone I knew was goofing around on 2048 when it came out) has got to be soul crushing.
But at the same time, this should really be a learning lesson to everyone- if your game or idea is simple enough to clone that someone can "rip it off" with a week of effort, you'd better damn well make sure your original excels in something that can't be cloned quite so quickly. Or not spend so much time working on it. Having only played 2048 and seeing how relatively polished and pleasing the presentation of it was for a so-called week of effort, I find it a bit baffling that it took the Threes team a year of work.
Threes never really left the top 100 charts in the App Store. I think they are fine.
Threes is the much better game IMO - 2048 gets boring after a while. I still play threes after all this time and it’s still satisfying.
Thank you for the entertainment 2048 gave me years ago, hope your new version has the reach you want!
I play it evertime when I watch a tv show. I played it probably 10 thousands time.
2048 is fantastic, and going full time on your project is a huge milestone. Congrats!
I spent so many hours on this back in Campus, thank you for building 2048
Question: why does it start at 2 instead of at 1?
How did you do animations for the menus and stuff?
I love 2048 but the page linked in OP is a disgrace: it filled with ad trackers, and the game itself doesn't even work if they're blocked.
2048 is a brilliant game, I love it. I understand that it's been used widely (I even saw it on a plane IFE system last month) and that likely didn't generate much ot any income, but the first impression when opening this was a bit dodgy. There are over 700 'partners' you want me to consent to sending data to? Come on, you know that nobody could make an informed decision on that.
Meh, wake me up when you release 16384. Just kidding, glad this can be a full time venture for you. Mixed feelings about this orbiting back into my field of view again. So far, having fun.
I'm sorry, but that ad placement is really obnoxious. Blocking these cool, new power ups after someone is halfway into a game comes off as a dark pattern, whether it's intentional or not.
Just another scam Internet surveillance.
Why do you need to share tracking data with many hundreds of shady organisations.
> 2048 is now my full-time focus
How? How does that game pay anywhere nearly enough to allow that?
Store and/or access information on a device ( 723 partners included )
Consent + Develop and improve services ( 379 partners included )
Legitimate Interest Consent + Use limited data to select content ( 101 partners included )
Legitimate Interest Consent + Use limited data to select advertising ( 472 partners included )
Legitimate Interest Consent + Create profiles for personalised advertising ( 529 partners included )
Use precise geolocation data ( 547 partners included )
Consent + Actively scan device characteristics for identification ( 551 partners included )
If this doesn't describe the modern internet I don't know what else does.
Plus a requirement to subscribe to Amazon Prime if you want to unlock more DLC? chef's kiss
FWIW, on desktop Firefox, between uBlock Origin, EFF's Privacy Badger, and Ghostery "Never Consent" cookie dismissal, I got no trackers, no cookies, no ads, and full site functionality.
Many have the "legitimate interest" checkbox selected. Does anyone know what does it mean in practice in this case? I roughly understand the general idea behind legitimate interest but I wouldn't know how it would apply here.
AND no way to conest the "legitimate interest" of these 9 trillion partners in one click, disgusting
Show some tolerance the guy list his job in October
Then don’t use it. People who keen on whining about this clearly don’t know anything about building a business.
You do not need what was posted above to build a business.
Buuuuuuurn
That you thought this was a post worth making, I'm embarrassed for you.
[flagged]
> Then don’t use it. People who keen on whining about this clearly don’t know anything about building a business.
You could heed your own advice and just not use the site, yet you're keen to stay and whine. Curious!
I’m fighting on a meta level. for a world where people don’t pretend we have a nanny state telling us what cookies we should use or not. Just use the site, if you don’t fancy the cookies just log off and keep going
I don't think its reasonable for every site I visit for the rest of my life to badger me for the right to have a legion of companies spy on my every move.
Crazy, I know.
I'm disappointed in HN for stomping this person for the game that they made. They had no idea Threes existed, made this game as an experiment, and it went viral. Despite this, they still acknowledge Threes on their site.
But they post here and are dogpiled on by people saying they ripped off another game and should feel like a scumbag. That's wrong, let's be kind to each other instead of exhibiting rude behavior that discourages people from pursuing their projects because "it's a ripoff of xyz". This started as an experiment, and is now this person's full time career. Can we instead encourage others to follow?
We're not talking about a 1:1 ripoff, we're talking about a game whose gameplay is adjacent to another game. Is Call of Duty a rip-off of Medal of Honor because they're both WWII FPS games?
HN isn't a person—it's a statistical cloud. There is always a spectrum of opinion. Most of the comments here agree with you.
There are some bilious and excessive responses but that's inevitable when a thread is large. We do our best to moderate the site so that they don't dominate.
People commonly overgeneralize-then-anthropomorphize the data points that rub them the wrong way. There are cognitive biases that we all have which make this an easy trap to fall into ([1], [2], [3]), but it's important that we all try to avoid it.
If you (I don't mean you personally, of course, but all of us) allow the negative tail of the spectrum to form your picture of the community, that's bad—because your picture of the community influences how you participate in it and this will cause harmful secondary effects that can, in the worst case, turn into doom loops.
I don't think the primary harmful effects (e.g. bilious comments and upvotes for indignation) of running a large internet forum like HN are avoidable, but I do think the secondary ones are. At least I hope so!
[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
[2] https://en.wikipedia.org/wiki/Hostile_media_effect
[3] https://en.wikipedia.org/wiki/Fundamental_attribution_error
"They had no idea Threes existed" This makes it even worse, because it means they ripped off 1024 instead (the game that ripped off Threes). That means they put even less effort into their clone, even the name is a rip off of that game.
I genuinely don't get how people here can say "Oh it's just like Call of Duty and Medal of Honor" or say that Triple Town and Threes were similar. Obviously there's a spectrum from inspiration to cloning, but 2048 is so far on the cloning end of that spectrum that the whole comparison seems like a joke.
I think 2048 is a better game and I like it more.
Especially when other merge games (Triple Town) existed for years.
They're all in the same general genre.
>Especially when other merge games (Triple Town) existed for years.
That matt guy that's posted like 100 times in this thread doesn't want to hear it, but I've literally only ever heard Threes brought up in comparison to 2048 and Triple Town. No one brings up Threes on their own organically.
where is the stomping happening? i think everyone here knows the score, and both teams have points.
[flagged]
2048 is a better game, which is why it's 100x or more popular than Threes.
I'm sure that you, a connoisseur, greatly prefer Threes, which is great. It can be fun to have a taste which is that of a very small minority.
But, to state the nearly tautological, the difference in popularity is because 2048 is better, in the view of the vast majority. It's more fun to play, and that's what makes a game good.
Why is everybody saying this when the game mechanic of 2048 is fundamentally different. The difference in play between swiping over one tile at a time VS swiping and moving all tiles to one side is a big difference.
Furthermore, the author didn't know Threes existed when they made it, so calling it a "blatant rip" is rudely false. Let's lift each other up instead of pushing each other down.
And Threes is a blatant rip off of other games before it as much as 2048 is of Threes. Let's go down that rabbit hole.
[flagged]
Forget it, Jack, it's Tripletown.
Triple Town was great in its own right, but the games are very different and have lots of different mechanics. Nobody would argue that Threes is as similar to Triple Town as 2048 is to Threes.
[flagged]
[flagged]
I wanted to address this, because I understand where you come from. Threes is actually credited in the about page[1] on play2048.co.
The frustration is completely understandable, as Threes is definitely where the concept originated. Due to the chain of derivative games that led to the creation of 2048, I had not seen or heard of Threes until after it went viral. I think the inspiration Threes provided, although valid, is somewhat indirect in the case of 2048. I did try my best to reflect this in the about page and when I talked about my experience here[2].
I noticed there's no mention of it on GitHub, so I'll add it there. I am happy to do better given the feedback, but I'm not sure what kind of improvement I could make here. Let me know
[1]: https://play2048.co/about
[2]: https://medium.com/@gabrielecirulli/2048-success-and-me-7dc6...
[flagged]
Again, I understand where you’re coming from, but I don’t fully agree. While Threes was an definitely inspiration along the chain, by your logic, any creative work would need a detailed credit trail, even when it grows and evolves into something different. 2048 was not intended as anything other than an experiment, yet it took on its own life, driven by the unexpected virality and maybe by its simplicity. I’ve always tried to acknowledge Threes’ role, but I don't know that I should attach this label on every mention of 2048. I appreciate the perspective you shared.
Don't be a cunt.
He mentions Threes! prominently in both link [1] and link [2].
[flagged]
What’s the problem, exactly? Is there some stated HN rule that requires a developer to clearly state the inspiration for their work?
To add to the clone topic, 2048 was cloned from 1024, so any claims of being only inspired by Threes fall apart quickly. IMHO, the way you're trying to sell 2048 is revisionist history and just plain shit. You can't turn back time, and let the Threes creator have a moment to shine and profit from the design, but you can at least give credit where due, and not when 50 people call you out on it.