Hacker News new | past | comments | ask | show | jobs | submit login
Design Is Why 2048 Sucks, and Threes Is a Masterpiece (wired.com)
71 points by garycomtois on July 14, 2014 | hide | past | favorite | 80 comments

It's a nice narrative to think that the innovative product where they spent 1.5 years agonizing over the details is obviously superior to the derivative product banged out in a weekend, but that's not the case, in my opinion.

I much prefer 2048. The $2 I spent on threes was $2 well spent, but 2048 is the game I pick up when I want to play something. (although the game should have been called 4096).

2048 is a far worse game, there is literally no challenge to it. Granted I'm a big fan of threes but I can literally beat 2048 every single time I play it. I did a simple test one weekend and played 2048 to measure how many times in a row I could beat it. I got 15 and stopped out of boredom. I find it dumbfounding how anyone can compare the two and not find 2048 an extremely poor clone.

If you're winning 15 times in a row, you're not moving fast enough. The appeal of 2048 is that you move so quickly, it feels cathartic. If you're not moving fast enough to make the occasional mistake, then of course you're going to get bored. If you're not making mistakes, you're also missing out on the fun of digging yourself out of a hole.

And as I mentioned in my original post, the goal for me to get 4096 or over 50,000 points. 2048 is easy, 50,000 points isn't.

But isn't that you having to add a level of challenge to the game that it doesn't have?

House ruling is fine but shouldn't that level of challenge (either through an enforced time limit or some other means) exist in the vanilla game?

2048 is a very accessible game, which is a large part of it's appeal. Rather than adding levels of difficulty, the better variants use achievements (like my 4096 and 50,000 goals, and your time limits) to keep it fun once you've got the basics mastered.

But the point is these aren't parts of 2048, they're just things you're putting on it.

It may be, in effect, a customisable game but that feels accidental rather than anything the designer intended. I could make any game harder by drinking heavily before playing it but that external factor making the game more difficult hasn't really improved it as a game.

Worth saying, I'm not wild about either game. For me they're simply not rewarding and I found myself resenting rather than enjoying the time I spent on them.

Honestly I still disagree, I can still easily hit 50,000 points multiple times in a row. The mechanic is broken, it's a literally solved game. Even moving extremely fast I rarely have a problem hitting 4096. Left down ldldldldrldldldldr nearly always works for an infinitely high score.

I don't believe you.

shrug I don't have anything to prove here but read this: http://www.dailydot.com/gaming/2048-game-strategy-tips-cheat...

I just tried it out again and I used a slightly modified version of the technique laid out in this article (Left-Down-Left-Down until I couldn't anymore then switched to Right-Down, rinse repeat only occasionally switching directions to setup long number chains ) and I got a 4096.

> It’s not guaranteed to yield an impressive score every single time, and it’s unlikely to produce any tiles higher than 512.

Yes, I've covered that point. No it's not just left down on repeat. It's trivially different though. In any case you can choose to believe or not but it's extremely easy compared to Threes! to the point of not being worth playing IMO.

You havn't figured out the strategy yet. 50,000 is routine once you have the strategy, and the joy in the game rapidly diminishes.

>The appeal of 2048 is that you move so quickly

For you maybe, not me - not all the time anyway. Granted when I am bored and feeling lazy, or I have a quick subway ride, I'll spin up 2048, but if I want a challenge and some thinking, I'll spin up Threes. In some sense, 2048 for me now serves as a less twitchy, less exciting Tetris substitute.

Moving the goal posts back on behalf of a game makes the goal more difficult, sure, but it doesn't necessarily mean the difficulty of the actual gameplay increases at the same rate, i.e. it can lag behind the difficulty of the goal, which in my opinion is very much the case for 2048.

The game can take a while, but the version by Ketchapp (on iPhone), which was the first knockoff to hit the app store, gives goals based on what you've solved so far. Right now my display says "Your next goal is to get to the 16384 tile", whereas the Gabriele Cirulli version has the downside of interrupting me every time I hit 2048, which, as you point out, is quite common once you're used to the game mechanic.

Nonetheless, hitting 8192 remains a challenge, and I feel like I've been tantalizingly close to 16384, but have never quite managed it. So as far as I'm concerned 2048 (or, I guess, 16384) remains quite challenging.

In Threes, 768 is the furthest I've gotten. I've found that when things get tight in Threes, you can easily be stymied not just by new tile placement (the only real wildcard in 2048), but also by the "+", "1", and "2"'s that can result in the position being unplayable, thus adding in my mind a greater degree of chance to the game.

I also had a lot of wins by pushing at random. So there's that.

I didn't know it was that easy to beat. I've never done it. Do you mind sharing your tactic?

To expound a bit more on faet's reply, pick either the top or bottom row as your workspace. I'll assume bottom.

Generally speaking the beginning doesn't really matter all that much so just start building up numbers however you see fit (some people go left right left right multiple times before crunching downwards). But start building and pinning your largest number on either the left or right corner. So, for example, try to make a 32, and pin it in the bottom right. Then start working on a 16 to its left, and an 8 to the 16's left, and a 4 to the 8's left. Generally, I try to make sure the entire workspace row is locked as often as possible by making sure it's always filled. This minimizes the chance that you'll need to unpin the largest number from its corner.

Once you've got your workspace row locked, there are a few things you can do. I haven't decided which I really prefer yet, but I basically just start creating another workspace row above the first workspace row. I'll usually try to focus on creating row #2's largest number on the opposite side of row #1's pinned corner number so that I can start crunching downwards as numbers begin to vertically match.

That's the very general gist of it. There are a few minor special case tactics you can use, but I'll leave that for you. Oh, and as faet said, don't swipe up (if your workspace is at the bottom). It's a recoverable mistake, but can make things much more difficult if you get a 2 stuck below your workspace (but it does kind of add some more spark to the game trying to recover from it). Personally, once you start using this strategy, the biggest thing you need to look out for is to not create a scenario where the game will generate 3 locked rows, because that will force you to swipe up. You'll see what I mean. Good luck.

Generally you want to favor 1 corner. Try to only use down/left. Use right only once your bottom row has 4 so your "Main" block stays in the bottom left.

I just tried and only got to 64 in two tries. Then tried the tumbler where you just press left-up-right-down in a loop and got to 256.

As far as I can remember 2048 isn't even always beatable by bots due to its random nature, so I suspect your literal 15 beats in a row aren't that literal.

2048 isn't always beatable by the best known bot. I don't know if anyone has proven either way whether 2048 is guaranteed solvable by a perfect bot for all possible random spawns. At the least, no one has found a counterexample like how Tetris has a guaranteed loss with a long enough string of S and Z pieces.

To be honest you're right, I'm just based on the best known bot and intuition.

Seems like a good math problem.

[EDIT] http://stackoverflow.com/a/22498940 won 100 out of 100 games using expectimax. 100 are not a lot of games, but hey.

There is empirical evidence that your experience is an outlier.

I personally find Threes to involve quite a bit more strategy than 2048. The game is also remarkably polished with legitimately excellent music and a ton of personality. It's been a while since I checked up on the myriad of 2048 clones, so maybe something's changed, but none of them that I tried seemed to have either the same degree of strategy or polish or charm that Threes has.

My introduction to this "genre" was through the first web-based 2048 clone posted here on HN, so if anything I should be biased towards that. But when I picked up Threes I was just flat-out impressed with the cohesiveness of the experience, which went above and beyond 2048 in basically all ways--strategy, art, etc.

At the end of the day, I imagine it comes down to a few simple things. First, Threes is not free. Second, difficulty calibration. Threes is probably just too difficult, which prevented it from hitting critical mass as fast as 2048 did (which is, IMHO, trivial in comparison). Third, sound. Since it's a mobile game a fair share (if not majority) of players will play on mute and hence won't really get to enjoy the music of Threes.

I must disagree.

Threes wins in every category. In 2048 once you know the strategy you don't really need to do anything. The rules are too simplistic. In threes everything is wonderful, from the audio, to the animations, to the fun little sounds every block makes on combining, even the little story about each block's personality, and of course a very good set of rules to keep the game predictable(ish) and challenging. It's actually really fun and difficult.

What I'd like to see: Every paid game should have a 3-day free trial. Once you really like it after 3 days, boom purchase. I hate buying games only to realize I wouldn't play it past first sitting. This is where Threes lost, not because its better or paid, but because free is easier to approach.

What, exactly, do you prefer about 2048? Why do you think it's the superior game?

Let's ignore price for the purposes of this.

(I'm comparing threes on a Samsung Captivate Glide and 2048, Evaldo Rossi version, on an ipad mini)

* The interface of threes frustrates me. It takes a relatively long time to start compared to 2048 almost starting instantly and is feels awkward to interact with. Much to convoluted for most settings to me.

* The counting up of score at the end takes too long. I either want to move on to the next game or walk away from it. I don't want to sit there and watch numbers being added.

As to Game play:

* Three's requires more strategy yes, but it is trickier to play in a casual manner, when I have a few free moments here and there. I can pick up a 2048 game, and get somewhere reasonably in a short period of time where threes seems to take more concentration and time (Even beyond the time it takes to get the app going).

The first time I saw threes the performance, theming, and just general styling of the game felt like a knock off of something else. It was bad enough that I didn't think I'd found the original and looked again. The edition of 2048 I came across had a much cleaner interface, smoother game play and works well in time constrained or pickup-play settings while still being interesting enough to be worth playing longer. Threes just doesn't scratch that itch. It's a little too much for casual play but not quite enough for something more serious to me.

They're very different games. With 2048, you make several moves per second. With Threes, you take many seconds per move. I just prefer that style of play more in my casual games.

Threes occupies that awkward middle ground between casual and 'real' games. If I want to play a real game, I'll pick up a 4X or something. If I want a casual game, I play 2048 or sudoku or something similar.

So it's the speed? I've heard that argument before, just curious. People don't seem to argue design or rules or something else like that, just that the tiles move faster.

I would argue the speed of gameplay is a fundamental part of a game's design. The fact that 2048 plays faster and the games are short is better design in a mobile game in my opinion.

Parent is talking about play speed (thinking consciously vs intuitinally/reflexively), not animation speed.

I understand.

As far as I'm concerned the game should have been called 1024. That way I would actually have won a game by now ;)

The success of 2048 is some good evidence that minimalism in design and minimalism in aesthetics are both good ideas.

Simply put, 2048 is more elegant than Threes. It might have flaws that the developers of Threes have solved. It turns out that those flaws don't really matter.

How do you figure this at all? Threes has been incredibly successful, top 5 in the Apple app store, currently 21st, and an Apple design award. This thread assumes 2048 is more popular but we have no real proof of that. Beyond that 2048 was available as an App in the store and it has not even close to the kind of sales/buzz as Threes on iOS or Android. In fact right now Threes is 28th on Android paid games list and the top 2048 clone is 83rd in free. On iOS it's similar with Threes at 21 in paid and the top 2048 at 57th on the free list. I know 2048 is primarily a web app but the best comparison we have shows Threes in a pretty positive light.

The free web app has loads of players that haven't bought a native app

Just because Wired is reblogging Threes' own piece on 2048[1] doesn't make it any more legit.

[1] http://asherv.com/threes/threemails/#letter

I remember reading this when 2048 came out and it came out as a huge whine from the developers. It completely turned me off of Threes.

Someone succeeded using a similar idea, and in many people's opinions, they did it better and it was free. The next step shouldn't be to write a smug blog post about how the other product is crap and people should be paying you instead.

Did you read the threemail post? It wasn't how you describe.

I thought this looked less like an article and more like a paid shill piece. Seems that impression was correct.

Do they ever admit to an influence from Zepto's Pudding Monsters and/or Triple Town? Seems a bit hypocritical if they don't.

Considering Threes! is a failure compared to 2048's usage amounts, it would be more interesting to hear what they did wrong. E.g. wasted over a year working on something they should have pushed out immediately. Spent money on design that wasn't needed. Etc..

How does Threes! get considered as a failure though? There are a lot of assumptions in this thread but little actual proof. What we can say for sure: - Threes is currently higher on the paid app store lists than 2048 is on the free lists, and as best as I can tell it's hight in the combined free and paid lists. - Threes won an Apple design award - Threes continues to sell fairly well (top 30 on iOS and Android)

We have no real numbers on 2048 and if you view it via the HN echo chamber I'm sure it looks like 2048 is a more popular game but I'm not sure that's true in the larger outside market.

This article shows how mobile gaming may be profitable on the short term, but in the end of the day, AAA games aren't going anywhere. It takes so much few time and expertise to clone mobile games, investing on a obile company is just insanity.

Comparing this to Tetris is just hubris. I bet there won't be anyone people playing Threes in 30 years instead of the latest fad.

Mobile games have been around for 6 years ( and that's really pushing it, I'd argue that we only have 4-5 years of actual mobile game history to look at ) compared to 30+ years of AAA studio type of game dev. We are still learning what works and how to build games. More and more indie devs are figuring it out. Those are the people that are worth investing in. One of the underlining points of the article (granted it's not pointed out) is that a lot of the same people are finding repeated success in mobile games (Ridiculous fishing, canabalt, Threes!), both devs involved in Threes! and all of the advisors mentioned in the article had already succeeded in the App Store.

The comparison to Tetris is also valid. It's a very popular, simple to play, hard to master game on a wildly successful platform. A game that draws in large numbers of non traditional gamers.

30 years ago it was 1984, and at the time game development was probably more akin to embedded software engineering today, having to program at a very low level of abstraction (if any) from the hardware.

It doesn't really seems right comparing game development now and then (or even 15/20 years ago), the amount of advanced tools, public knowledge, step-by-step tutorials available today is something developers of the time could never dream of, and had to learn/build for themselves.

My point was less about the difference between now and then and more about the built up knowledge base that exists from that much collective experience. Especially about game UI design. Super Mario Brothers (or Quake/Doom or warcraft etc...) has more in common with a more modern AAA game than a touch screen based app when it comes to Player controls and design of the Player interaction. We are only lately really starting to figure out how to make really good serious touch first games. Leo's Fortune, Fates Forever, Thress!, Monument Valley, and Wayword Souls are all great examples of this process. In game play none of them are completely revolutionary but when you combine the touch controls and player interaction with the type of game play they offer, then IMO each of those is a major step forward in game design for touch devices.

Yes I got it and I agree with you, my point was that Mobile Game Development, regardless of its relative youth, is evolving at a much faster pace that traditional games have, so comparing them in terms of years is kind of pointless.

Genres like real-time strategy, first person shooters, MMOs took years to be "invented", while we see new (and successful) concepts of mobile games basically every month.

> while we see new (and successful) concepts of mobile games basically every month.

With the rise of Free2Pay, and the blight of IAP, I would argue otherwise.

The app store is certainly starting to move in a different direction. A lot of new games with new models are finding success in the App Store recently. See Fates Forever, wayword souls, leo's fortune, Threes!, Civ Rev 2, Monster Hunter 4, and monument valley. All are either paid apps or an "ethical" Free to play model. Vainglory and Revolution 60 are example's of games in the pipeline also.

Ah, sorry, I misunderstood your point, Carry on I completely agree then.

Tetris is a masterpiece and I don't think anyone is suggesting Threes is, that's not the level of success that's being suggested.

But that's not to say that you can't make money (and therefore successful investments) out of mobile games. Perhaps some of the valuations have been out (and massively out) in some cases but it's a market where there is money to be made.

The designers wailed and gnashed their teeth, "but the design!" while the rest of the population went back to drinking their coffee.

-went back to drinking their coffee and playing 2048, the free and open source game available on almost any platform.

Yup, when I ask the random guys at my coffee shop why they're playing 2048 instead of Threes, they just won't shut up about the awesome Linux support and open-sourced code!

Right, and the designers/developers of threes! are making a living, even a pretty good living off of Threes! and no one is making anything off of 2048.

Threes! is a better experience across the board and it is cross platform in it's intended form factors, it's available for iOS and Android.

My point is, 2048 took off like wildfire because it's free and the majority of clones were forked from it.

The last time the Threes vs 2048 debate was posted here, half the comments were "I never heard of Threes. I didn't even realize 2048 was based on another game."

Threes is a wildly successful game in it's own right, an Apple design award winner, currently 21 on the top paid Overall app store list, and at one point was in the top 5 in the top paid list. A lot of those users that half bought likely have never heard of 2048. This a good example of selection basis in affect. 2048 was wildly popular on HN but I have rarely seen it mentioned on other media sites, even mobile gaming focused outlets.

For yet another possibly biased data point just like your own: some weeks ago I found myself sitting between two strangers playing 2048 in the subway.

FWIW, Threes is(/was) the App of the Week at Starbucks Canada last week. Go pick up a card if you want to get Threes for free (no purchase necessary, I guess. Just grab one off the counter and go).

I for one feel like this hasn't been discussed enough on hn.

We need a couple hundred more 2048 clones before we can discuss design merits.

Threes is actually pretty badly designed as a mobile game, imo. It has too many potential control points on a small screen, is audio intensive, and has a long startup time (I mean, wtf. It takes 5-10 seconds to start up!)

Threes might be a better "game", in that there's an easy strategy to do well in 2048, but Threes is a bad mobile game. The ideal mobile game would be something which could be played for 30-90 seconds, paused or looked away from without lasting negative consequence, but with timing as a key factor, and 4-8 possible touch points on the screen, IMO.

"Rez" would probably do well for mobile (the playstation vibrator game).

I've played and enjoyed both games.

Threes is beautiful, but to be honest it sometimes feels so...heavy. The animations, the sounds, the repetitive music...sometimes it grows tiresome. Meanwhile, 2048 is something I can throw up on a browser on any device (something I can't do with Threes) and just hash through a few games. This is not including the time I spent on the various 2048 clones, made possible because (again unlike Threes) the source code for 2048 is freely available online.

It may be less challenging, but when has challenge ever been a criteria for a game to be fun, or even a "masterpiece"?

A great success by the threes team for sure.

But I have to wonder what if that final version had been banged out in the first weekend, would it have been recognized as the winner? Would a team then stop innovating and switch to build it (and saved 1.2 years for who knows what else)?

From a product dev standpoint i'm torn. There's a lot to be learned through iterating through complexities, challenging everything and ending with a simpler solution. But there's plenty to be lost by the argyle distractions as well.

HN, how do you do know when to stop brainstorming and ship?

I'm not sure myself, but I'd wager it's when the brainstorming and iterations are getting less significant. When you're changing less and less through each iteration, then you know it's getting close to being done.

That being said, at some point someone (preferably someone wise and smart) makes the decision to ship. And they do need to make that decision.

We all know the Doge2048 is the superior game to both (I'm only half joking...)

I think the problem with Threes is that it is slightly too hard, which leads to too much frustration. We all love a challenge, but at least its possible to eventually beat 2048.

then why is it Flappy bird was a hit!?

3s > 2048! But 2048 is more popular?! How did it happen? I've heard this story before.

I think the author of 3s screwed up by going iOS first. There are millions of hungry Android devs out there. If you release a game that's iOS only, and it turns out to be really popular, and it has fairly simple game logic... there will be an Android clone in under 24 hours.

It's a cautionary tale. Release on iOS first at your own risk.

Yeah, the Threes devs released iOS first and made a ton of money, and basically no one made money on 2048. I can see why releasing on a platform where people actually pay for apps is a mistake.

I disagree 2048 sucks. It's a totally different game from Threes. But I do think a little design will make 2048 better -- that's why I created "Edge Up 3072". Try it out: http://3072.clingmarks.com/

The article fails to explain in explicit terms what exactly makes "threes" so much better than 2048. Time in development is irrelevant to the user, who only ever sees the end result.

Is there an end to Threes? The highest tile I reached is 768 which isn't that high (but it was quite hard to achieve).

Maybe I could interest you in my game called Thirties!

Exactly the same, but with far more impressive numbers.

Everyones talking about 2048, but how do Threes free clones affect the real paid version?

This is interesting, because I think the most likely effect will have been to increase sales of Threes, especially since the author of 2048 provided a link to Threes on the page.

I thought we'd all added "Threes! vs 2048" to the ever increasing list of useless topics of argument.

Skimming over the comments here, I could easily replace Threes! and 2048 with Mac and PC, or XBox and Playstation, or Beer and Wine......

I don't buy it.

Overdesign is why Threes sucks and 2048 is a masterpiece.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact