Online since 2002. Over 3300 puzzles, 2600 worldwide members, and 270,000 messages.

TwistyPuzzles.com Forum

It is currently Tue Jul 22, 2014 2:51 am

All times are UTC - 5 hours



Post new topic Reply to topic  [ 66 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Threes!
PostPosted: Fri Mar 14, 2014 2:38 pm 
Offline

Joined: Mon Aug 18, 2008 10:16 pm
Location: Somewhere Else
www.threesgame.com

This just finally came out for Android. I can't put it down!

The animation on the website can explain it better than I can, but I'll try anyway: slide the pieces on the board to combine 1 and 2 to make 3. Then combine 3 with 3 to make 6, 6 with 6 to make 12, and so on. A new piece gets added every move, so you want to try to avoid running out of moves, which ends the game. The bigger your numbers get, the higher your score.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 3:24 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
The one I can't put down is 2048, inspired by Threes:

http://gabrielecirulli.github.io/2048/


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 3:34 pm 
Offline

Joined: Mon Aug 18, 2008 10:16 pm
Location: Somewhere Else
That one's good too, but I wish there was a "next" indicator. I found a hex grid version of it for Android, and since hexagons make everything better, I hope to try it soon.

https://play.google.com/store/apps/deta ... dio.hexled


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 4:29 pm 
Offline

Joined: Sun Aug 26, 2012 10:01 am
bhearn wrote:
The one I can't put down is 2048, inspired by Threes:

http://gabrielecirulli.github.io/2048/



ME TOO!! This is the furthest I've gotten:

http://prntscr.com/30tm5h

_________________
My YouTube, including a FF Siamese 2x2x2 Walkthrough


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 5:09 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
JackRTully wrote:
ME TOO!! This is the furthest I've gotten:

http://prntscr.com/30tm5h

Ah, I think you don't know the proper strategy yet.



I thought sure I would get to 2048 this game, but not quite.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 6:12 pm 
Offline
User avatar

Joined: Sat Apr 22, 2000 2:51 am
Location: New Ulm, Minnesota, USA
Is 2048 game on android at all? I hope because I like it better.

_________________
Darren & Traiphum's Dual Helicoptrahedron
X-TownCuber wrote:
Are my eyes deceiving me, or is this the coolest puzzle ever?


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 6:44 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
As a web app, it's playable on my iPhone, should work on Android as well.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 7:06 pm 
Offline
User avatar

Joined: Thu Dec 02, 2004 12:09 pm
Location: Missouri
I just played the 2048 web app on my iphone. Nice. I've never heard of this game before and I got 5208 point on my very first game. So... what is this "proper strategy"? What happens if you get to 2048? Is the game always winable? Or is it like solitaire or tetris where there is still an element of chance?

Carl

_________________
-
Image

Image


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 7:58 pm 
Offline
User avatar

Joined: Sun Mar 08, 2009 9:21 am
Location: Massachusetts, USA
My strategy in 2048 is to keep the highest tiles all in the bottom row, ideally in order from left to right or right to left. Then, combine tiles only by moving left, right, or down, while avoiding up moves as much as possible so that the bottom row is always filled with tiles. If you order the tiles, it's very easy to combine them into higher numbers in a chain.

_________________
Katniss wrote:
Only on this forum would people use a V-cube 7 as a size comparison for a cat :lol:

My Shapeways shop


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 8:09 pm 
Offline

Joined: Fri Dec 20, 2013 11:47 pm
I press up, right, down, right, and then sometimes left. I have gotten over 4500 just by spamming this :P


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 8:44 pm 
Offline
User avatar

Joined: Mon Feb 16, 2004 12:49 am
Location: Raleigh NC
4th try


Attachments:
Screen shot 2014-03-14 at 9.40.30 PM.png
Screen shot 2014-03-14 at 9.40.30 PM.png [ 40.96 KiB | Viewed 3420 times ]

_________________
LE|F
Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 9:26 pm 
Offline
User avatar

Joined: Mon Feb 16, 2004 12:49 am
Location: Raleigh NC
Same game and got 4096


Attachments:
Screen shot 2014-03-14 at 10.27.15 PM.png
Screen shot 2014-03-14 at 10.27.15 PM.png [ 34.37 KiB | Viewed 3408 times ]
Screen shot 2014-03-14 at 10.25.13 PM.png
Screen shot 2014-03-14 at 10.25.13 PM.png [ 55.32 KiB | Viewed 3409 times ]

_________________
LE|F
Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Fri Mar 14, 2014 10:17 pm 
Offline
User avatar

Joined: Mon Feb 16, 2004 12:49 am
Location: Raleigh NC
FAIL


Attachments:
Screen shot 2014-03-14 at 11.16.15 PM.png
Screen shot 2014-03-14 at 11.16.15 PM.png [ 47.07 KiB | Viewed 3394 times ]

_________________
LE|F
Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 15, 2014 4:51 am 
Offline
User avatar

Joined: Mon Jan 01, 2007 5:28 pm
Location: CA, USA
Darn upper left 2.
So hard to keep the high numbers in the corner and not have a 2 jump in the empty slot. :(


Attachments:
2048.png
2048.png [ 40.26 KiB | Viewed 3356 times ]

_________________
...
Brushwagg
Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 15, 2014 12:18 pm 
Offline
User avatar

Joined: Thu Dec 02, 2004 12:09 pm
Location: Missouri
Finally...
Attachment:
Win.PNG
Win.PNG [ 34.92 KiB | Viewed 3317 times ]


Though I didn't make it to 4096.

Carl

_________________
-
Image

Image


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 16, 2014 11:12 am 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
I finally got 2048 as well, kept going, score was 26,936.

4096, wow... and on your 4th game? You must have a better strategy than I do. Took me dozens of tries. I can almost always get to 512, but getting beyond that is hard. Although, the last few games have gone better; I think I'm reasoning about the game better now.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 16, 2014 1:37 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
I keep almost getting there. I've watched the AI solve it and I'm amazed how many times the program can change the corner / wall that it's building the largest numbers against. Once I'm forced to move my largest number out of the corner or even any of my other large numbers from against the wall I'm usually screwed.

Attachment:
another_2048_fail.png
another_2048_fail.png [ 57.42 KiB | Viewed 3218 times ]

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 16, 2014 2:35 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
Brandon Enright wrote:
I keep almost getting there. I've watched the AI solve it and I'm amazed how many times the program can change the corner / wall that it's building the largest numbers against.

Very cool. Too bad it stops at 2048. Whatever it's doing is not much like I play it.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 16, 2014 2:42 pm 
Offline
User avatar

Joined: Mon Jun 30, 2008 8:38 pm
Location: Russia
4096 and 8192 web version available :)
http://stevenrush.github.io/
http://stevenrush.github.io/8192

and multiplayer version!
http://emils.github.io/2048-multiplayer/

2048-oficial for you android (Game's cap to win is now 8192)
https://play.google.com/store/apps/deta ... lli.puzzle

_________________
my Shop: ShapeWays, grigorusha Big Puzzle Sale: EBay
Image Image


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 16, 2014 3:38 pm 
Offline
User avatar

Joined: Thu Dec 02, 2004 12:09 pm
Location: Missouri
bhearn wrote:
Very cool. Too bad it stops at 2048. Whatever it's doing is not much like I play it.
That isn't what I'm doing either. I do everything I can to keep the largest tile in the lower left and I generally keep all the larger tiles on the left wall. Most of my moves are left or down. If the left column is full I also use up... though make sure that up move won't combine tiles in the left column. And generally you never want to use right if it at all can be avoided. Try to think a move or two ahead. Before you make a move ask yourself if you'll be able to make another left or down move after this turn. If no and your left column isn't full you may want to reconsider your options.

I was on maybe my 10th game by the time I got to 2048. And I feel I could do 4096 if I took my time. Its usually an error I make while trying to make moves too fast that kills me. 8192 I think would require a lot of luck.

Carl

_________________
-
Image

Image


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 16, 2014 5:12 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
wwwmwww wrote:
I do everything I can to keep the largest tile in the lower left and I generally keep all the larger tiles on the left wall. Most of my moves are left or down. If the left column is full I also use up... though make sure that up move won't combine tiles in the left column. And generally you never want to use right if it at all can be avoided. Try to think a move or two ahead. Before you make a move ask yourself if you'll be able to make another left or down move after this turn. If no and your left column isn't full you may want to reconsider your options.
This exactly describes my strategy except that I use the upper-right corner and keep all of the large numbers along the top wall. Basically my strategy is the mirror about the top-left to bottom-right diagonal.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Last edited by Brandon Enright on Mon Mar 17, 2014 12:47 am, edited 2 times in total.

Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 16, 2014 5:57 pm 
Offline

Joined: Mon Aug 18, 2008 10:16 pm
Location: Somewhere Else
Thanks for derailing my thread, guys... oh well. :lol:

Threes is a lot harder than 2048... I've reached the 1024 (ninth) tile a couple times in 2048 but I haven't been able to get the 384 (eighth) tile in Threes.

I found this other similar game, too: https://play.google.com/store/apps/deta ... ries&hl=en

Here you have to match three elements (red, yellow, and blue) to make black, and then match black with black. It looks difficult, but I haven't tried it yet.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 16, 2014 7:37 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
bhearn wrote:
Brandon Enright wrote:
I keep almost getting there. I've watched the AI solve it and I'm amazed how many times the program can change the corner / wall that it's building the largest numbers against.

Very cool. Too bad it stops at 2048. Whatever it's doing is not much like I play it.

Check out his description on Stack Exchange. Definitely worth a read and thinking about. Others have suggested creating a snake pattern instead. I only try to keep the top row in order but I should try keeping the row below it in reverse order too. I notice I lose towards the end of the game because I have too many medium-sized numbers (16-64) on the board and it becomes very hard to collapse the small numbers that keep appearing. I need to keep more small numbers around in order so I don't run out of space. Creating a snake out of the first two rows should help do this.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Mon Mar 17, 2014 12:49 am 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Brandon Enright wrote:
wwwmwww wrote:
I do everything I can to keep the largest tile in the lower left and I generally keep all the larger tiles on the left wall. Most of my moves are left or down. If the left column is full I also use up... though make sure that up move won't combine tiles in the left column. And generally you never want to use right if it at all can be avoided. Try to think a move or two ahead. Before you make a move ask yourself if you'll be able to make another left or down move after this turn. If no and your left column isn't full you may want to reconsider your options.
This exactly describes my strategy except that I use the upper-right corner and keep all of the large numbers along the top wall. Basically my strategy is the mirror about the top-left to bottom-right diagonal.

I ended up solving it by slightly tweaking my strategy. Before the only thing I paid significant attention to was keeping all of the large number in order in the top row and aggressively building a big number in the second row on the left so that I could merge it up into the top row.

After my tweak, I paid a lot more attention to avoiding too many numbers in the bottom three rows and especially avoiding any sort of checkerboarding where large areas are useless and off limits. By focusing on keeping as much area open I had more freedom to build the larger numbers in the second row to merge them into the top. I also didn't focus so aggressively on merging just the smallest number on the left in the top row but allowed large number to be merged into other numbers in the top row while leaving the number on the left in the top row smaller. This meant less total work was needed to merge a number in the top row if I started running out of room.

I solved it in the first try using this. My guess is that it probably would work 33% or more of the time.

I sooooo didn't have the time to work on this puzzle this weekend :evil: I just can't resist problems like this, especially when I have a ton of work to do that part of me is doing its darnedest to put off.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Mon Mar 17, 2014 9:20 am 
Offline

Joined: Sun Nov 23, 2008 2:18 am
Does there exist a text-only version of this game?

As I understand it:
You start with a square grid filled with a random assortment of tiles with low powers of two.
If adjacent tiles have the same value, you can move one on top of the other to create a tile with their combined value.
The newly created space is filled with a random tile with a low power of 2 for its value.
The game continues until no more moves are possible.

Am I incorrect on any of the above or missing anything?

_________________
Just so you know, I am blind.

I pledge allegiance to the whole of humanity, and to the world in which we live: one people under the heavens, indivisible, with Liberty and Equality for all.

My Shapeways Shop


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Mon Mar 17, 2014 11:23 am 
Offline

Joined: Mon Aug 18, 2008 10:16 pm
Location: Somewhere Else
For 2048? The board, sixteen cells in a square, starts empty except for two tiles marked "2". When making a move, all tiles slide as far as they can in one of the four basic directions. Identical tiles can merge to make a tile with double the value. Then, a tile (usually a 2 but sometimes a 4, and I'm not sure if higher ones can appear?) appears on a random space. The goal is to reach the tenth level of tile, 2048.

For the original Threes, the same board is used, but tiles only move one space at a time. At the start of the game, nine tiles start on the grid, an assortment of 1s, 2s, and 3s. 1s and 2s can only match with each other to make 3s, and 3s and higher tiles can only match with twins. A new tile will only appear in a row or column where tiles moved, and will slide onto the board into an edge spot.

I don't think reader-friendly versions of these exist... sorry. :(


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Mon Mar 17, 2014 6:56 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Jeffery Mewtamer wrote:
As I understand it:[...]Am I incorrect on any of the above or missing anything?


I think the best way to understand the behavior of the game is to think of it as similar to how a billiard / pool ball rack works. Are you familiar with the triangular wooden rack that corrals the balls into a triangular shape but you can still roll them all around as a group?

If so, imagine a square rack that can hold four rows by four columns for a total of sixteen balls. To move all of the balls to the left you actually move the rack right and let them all strike the left wall of the rack. Two numbers merge when they are forced to collide with each other because they either got pushed up against a wall or they got pushed up against another number which itself was pushed up against a wall. Thinking about the game in this way allows you to understand what happens when three of the same number, say all three are four, are all in the same row. Suppose you move the numbers to the right by moving the rack to the left. The first four will hit the wall and when the second four will collide with it, they will combine to form the number eight. Then the third four will collide with the eight and will be stopped because four and eight can't combine. Seeing the moves in this way removes any ambiguity about which numbers will combine when you have situations like three in a row. Each move pushes all of the numbers against a wall or other numbers and forces all of the ones that can combine to be combined. The game is over when its impossible to combine any more numbers and there isn't any empty space left. The only other trick is that each time you do a move, in addition to all of the numbers moving and combining, a new number, either a two or a four, is introduced into an empty space. So every move adds one more number to the board so unless you spend a lot of effort to combine numbers, you run out of space.

A text version of the game could be programmed based on my description above but I don't think it's a game that could be realistically played without vision simply because reading up to 16 numbers in a grid out after every move would be very tedious.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Mon Mar 17, 2014 10:25 pm 
Offline
User avatar

Joined: Wed Jan 21, 2009 12:58 pm
grigr wrote:
[...] and multiplayer version!
http://emils.github.io/2048-multiplayer/
Thank you for that link, it was nice until some users started cheating. See below, those tiles appeared randomly, that guy made a total of like 15 moves. :(

Hopefully that won't happen often.

Skarabajo


Attachments:
cheater.jpg
cheater.jpg [ 221.98 KiB | Viewed 2446 times ]

_________________
My collection | My first mod | Making of first mod | Items for sale
Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Tue Mar 18, 2014 6:02 am 
Offline

Joined: Sun Nov 23, 2008 2:18 am
Now that I have a more accurate description of the game, the gears in my brain are already turning as to how I can write a simple C++ terminal application for this game, and even allow for grids of arbitrary size. And it would certaintly be a lot easier to program controls of than the game I described.

Well, I know what my next self-assigned programming exercise will be.

_________________
Just so you know, I am blind.

I pledge allegiance to the whole of humanity, and to the world in which we live: one people under the heavens, indivisible, with Liberty and Equality for all.

My Shapeways Shop


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Wed Mar 19, 2014 1:35 pm 
Offline
User avatar

Joined: Thu Sep 29, 2005 5:46 pm
The ante has just been upped.

http://www.csie.ntu.edu.tw/~b01902112/9 ... 740992/#_=

_________________
best single solves-
2x2x2- 7.14
3x3x3- 31.92
4x4x4- err... 330-ish
megaminx- 173.something


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Wed Mar 19, 2014 1:50 pm 
Offline

Joined: Mon Aug 18, 2008 10:16 pm
Location: Somewhere Else
I lol'd!


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Wed Mar 19, 2014 3:18 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
The really big one is just too easy. I can't really tell if it's getting harder as you have to build bigger and bigger numbers but I don't feel like it is. Using no strategy other than holding the up arrow key, and then when there are no possible moves, either hold the right key, or left key, before going back to holding up I'm here:
Attachment:
big_2n_easy.png
big_2n_easy.png [ 90.53 KiB | Viewed 2255 times ]


I think if you used this strategy and then once in a while stopped and use a bit of smarts to combine larger blocks, you could, in principle, beat it. Of course, even doing 100 moves a second it would still take a minimum of about 700,000 years so the game isn't really beatable.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Wed Mar 19, 2014 3:21 pm 
Offline
User avatar

Joined: Wed Apr 13, 2011 4:17 pm
Just played 2048 with the Fibonacci sequence rather than powers of 2.
Quite interesting twist as it is not matching pairs of the same number any more.

_________________
Only small minds want always to be right


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Wed Mar 19, 2014 7:55 pm 
Offline
User avatar

Joined: Sat Apr 22, 2000 2:51 am
Location: New Ulm, Minnesota, USA
Nebagram wrote:


Thank you very much because I love it now I wish it was for Android and I'd love to have a hexagon version to of this size. I always have my laptop hooked up to me 46 inch tv an still I can't see all of the squares at once but I do love the game a lot. Please help if someone can show me how to see all of the squares at once. Is it a quick fix I hope? Change resolution (dots per inch).

_________________
Darren & Traiphum's Dual Helicoptrahedron
X-TownCuber wrote:
Are my eyes deceiving me, or is this the coolest puzzle ever?


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Wed Mar 19, 2014 8:49 pm 
Offline
User avatar

Joined: Wed Jan 21, 2009 12:58 pm
Darren, you can resize the content of an html page with [Ctrl]+[-] and [Ctrl]+[+] on most Windows browsers. Or [Command]+[-] and [Command]+[+] on most Mac browsers. You can set it back to default with [Ctrl]+[0] or [Command]+[0].

Skarabajo

_________________
My collection | My first mod | Making of first mod | Items for sale


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Wed Mar 19, 2014 10:34 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
Nebagram wrote:

No -- NOW the ante has been upped. :)

http://ehzhang.github.io/2048-hardcore/


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Thu Mar 20, 2014 8:51 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Jared wrote:
Threes is a lot harder than 2048... I've reached the 1024 (ninth) tile a couple times in 2048 but I haven't been able to get the 384 (eighth) tile in Threes.
Threes is a lot harder. 2/3rds of the cards are blue or red and because they're only matchable with each other you have to be much more aggressive about clearing area and planning ahead so you can keep clearing area no matter what comes up.

I'm somewhat consistently getting to 768 now. Probably once every 5 games or so. I get to 384 most games. I still haven't gotten two 768s in one game, much less been able to combine them to 1536. I'm amazed by the insane scores people are getting in the game center.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 22, 2014 2:11 am 
Offline
User avatar

Joined: Wed Jan 21, 2009 12:58 pm
There are some variations that haven't been mentioned yet...

2048: Doctor Who Edition
http://games.usvsth3m.com/2048-doctor-who-edition/

2048 Cubed
http://cs.brown.edu/people/awstlaur/2048-cubed/

Hexagonal 2048
http://baiqiang.github.io/2048-hexagon/

Enjoy,

Skarabajo.

_________________
My collection | My first mod | Making of first mod | Items for sale


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 22, 2014 5:10 am 
Offline
User avatar

Joined: Thu Sep 29, 2005 5:46 pm
The hexagonal one was a lot easier than I was expecting, I nailed it on my first try. Then again, they did use the phrase 'use the arraw keys' (sic) in the description.

_________________
best single solves-
2x2x2- 7.14
3x3x3- 31.92
4x4x4- err... 330-ish
megaminx- 173.something


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 22, 2014 7:56 am 
Offline
User avatar

Joined: Sun Aug 04, 2013 4:50 am
Location: The Netherlands
http://ehzhang.github.io/4/
Join the numbers and get to the 4 tile! After that, try to get the maximum score!

_________________
- Eric


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 23, 2014 7:18 pm 
Offline
User avatar

Joined: Wed Jan 21, 2009 12:58 pm
It would be nice to see who the opponent is in the multiplayer version. I just played a very close game and would like to say "great game!" to whoever that was.

I wonder what would had happen if I had the time to make the two extra moves required to be in the exact configuration as the winner. Would the system say "Tied game" or something?

Skarabajo.


Attachments:
closegame.jpg
closegame.jpg [ 205.37 KiB | Viewed 1893 times ]

_________________
My collection | My first mod | Making of first mod | Items for sale
Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sun Mar 23, 2014 8:09 pm 
Offline

Joined: Sun Nov 23, 2008 2:18 am
Well, I wrote my own version, and thought I would share.

My version is called Z!(I pronounce it Zed with emphasis despite being American) and it has an alphabetic theme instead of a numeric one. Combine two of the same tile to get the next letter of the alphabet, and if you manage to combine a pair of Z tiles, they annihilate! Allows you to play on square grids up to 10*10. Empty spaces are represented by 0s and movement is controlled with WASD.

My source code is attached(rename to .cpp) as I have no access to a Windows machine and known nothing about cross-compiling. Please let me know if you find any bugs or can help me make improvements(such as how to dynamically allocate memory for 2-dimensional arrays instead of wasting space by allocating for the maximum size or implementing arrow key controls).

I have only played Z! on a 3*3 grid thus far, and my high score on that grid size is F. Not sure what the optimal grid size for the game(i.e. grid size where making Z tiles is possible without perfect play and rigging the RNG, but not so easy as to make maximizing the number of annihilations a matter of time rather than luck/skill.

Also, assuming a fair RNG, A tiles spawn with a 4-in-8 chance, B tiles with a 3-in-8 chance, and C tiles with a 1-in-8 chance. All larger tiles must be created.

I also saw on another forum(I won't link as doing so might violate TP.com rules) where someone had written a version of 2048 for the Sega Genesis including graphics and music of the coder's creation.


Attachments:
File comment: Z! source code
Z.txt [4.68 KiB]
Downloaded 30 times

_________________
Just so you know, I am blind.

I pledge allegiance to the whole of humanity, and to the world in which we live: one people under the heavens, indivisible, with Liberty and Equality for all.

My Shapeways Shop
Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Tue Mar 25, 2014 12:00 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 29, 2014 1:33 pm 
Offline

Joined: Mon Aug 18, 2008 10:16 pm
Location: Somewhere Else
The authors of Threes wrote an article about how they felt about 2048 and other similar games, and they included a LOT of background design process stuff. Almost all of it, actually.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 29, 2014 2:42 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
Huh. I still haven't tried Threes, but I disagree with them about 2048, and that does come across as sour grapes. There's a reason everyone is playing it.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 29, 2014 3:03 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
bhearn wrote:
Huh. I still haven't tried Threes, but I disagree with them about 2048, and that does come across as sour grapes. There's a reason everyone is playing it.

I read everything up to the point where they showed the creation process. Their response is rather schizophrenic. It's like they can't decide if they're mad, sad, angry, or happy about all of the various games inspired by Threes. If anything, it seems they're trying to prove that they put a lot of thought and hard work into Threes. I don't think anyone is disputing that though and all of the puzzles inspired by Threes seem to acknowledge Threes.

The way I see it:
  • Just about everyone seems to like Threes
  • The popular puzzles like 2048 inspired by Threes aren't complete rip-offs
  • "Imitation is the greatest form of flattery"
  • They're making a ton of money from the success and popularity of Threes

So I don't understand the tone of their post. It's really neat to see how much the game evolved though...

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 29, 2014 4:09 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
Jared wrote:
The authors of Threes wrote an article ...

"When an automated script that alternates pressing up and right and left every hundreth time can beat the game, then well, that's broken. "

This is just not true. I've been playing with it; I cloned the github repository. When I automate this policy, generally the biggest tile made is 128 or 256. Once, out of a few hundred runs, it made a 1024 tile.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 29, 2014 6:55 pm 
Offline

Joined: Mon Aug 18, 2008 10:16 pm
Location: Somewhere Else
Even if it's not literally true - they admit as such further down - 2048 is still an easier game overall. Whether or not that makes it "worse" is up to debate, but I can't blame these guys for being biased towards their own game.

Here's a couple more games in the same vein:

http://hgentry.github.io/81/ - From the long list of 2048 forks, match three numbers instead of 2. A great deal harder than 2048.
http://jatekos101.github.io/20euros/ - Another 2048 fork, Join 1 and 1 to make 2, but then a 1, 1, 1, and 2 or 1, 2, and 2 to make 5. Then two 5s to make 10, and go up the line of Euro denominations. Also difficult.
http://bartbonte.com/25/ - Not a fork! More similar to Threes, with single space moves, but 2048's matching rules (i.e. no initial pair to combine like in Threes). To make up for that, red blocks appear that have to be aligned in rows of 3+ to eliminate.


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 29, 2014 7:43 pm 
Offline
User avatar

Joined: Tue Aug 11, 2009 2:44 pm
I just wrote a very simple Monte-Carlo based player for 2048. On its first run, it scored 33,892, ending with a 2048 tile, a 1024 tile, and a 512 tile. :)

This one is in some ways more fun to watch than the existing AI solver -- that mostly follows a comprehensible strategy. This one doesn't ever bother to keep the big tiles in the corner! It looks like it's always living on the edge.

I'll post it to github after I've made some improvements. Actually I was planning on writing a Monte-Carlo Tree Search, but the dead-simple MC code works pretty well!


Top
 Profile  
 
 Post subject: Re: Threes!
PostPosted: Sat Mar 29, 2014 9:37 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
bhearn wrote:
I just wrote a very simple Monte-Carlo based player for 2048. On its first run, it scored 33,892, ending with a 2048 tile, a 1024 tile, and a 512 tile. :)
What exactly do you mean by "Monte-Carlo"? It does moves at random? I can't imagine that working very well. Are you using some sort of random move input plus a outcome score / heuristic to choose a random move?

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 66 posts ]  Go to page 1, 2  Next

All times are UTC - 5 hours


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  

Forum powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group