Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
2048 game to the Atari 2600 VCS (github.com/chesterbr)
43 points by postit on March 25, 2014 | hide | past | favorite | 22 comments


Wow, programming in Assembly is so much easier now than it was then. Imagine the luxury of having source files that you could modify in a text editor. Joy.

I remember my first LDA's on the Commodore 64, directly in the console, having to assign a memory location to each and every line as you went along, and woe unto you if you didn't NOP yourself out enough extra space from time to time to be able to insert code (or at least JMPs) as you added features. I'm not complaining, mind you. At least we had 3 letter codes we could use for instructions instead of having to memorize hex codes for everything.

But yeah, even a simple assembler that handled the line numbers for you would have made me the happiest 13 year old around.


I get your point, but I must say that the Apple II had LISA (http://en.wikipedia.org/wiki/Lazer%27s_Interactive_Symbolic_...), which wasn't half bad for the machine - guess I was a luckier 13 year old, because I happened to have stumbled upon a copy :-)

Also, developers who had a bit more of resources (e.g., minicomputers) even used them to emulate 8-bit systems (unlikely to have been done for the Atari, due to the complexity of TIA, but a textbook case is how Bill Gates and Paul Allen developed the Altair Basic using Allen's 8080 emulator for the PDP-10: http://en.wikipedia.org/wiki/Altair_BASIC).


Proper macro assemblers (covering at least the functionality used in this source code) have been available for the C64 since 1982, so it wasn't necessarily as hard as punching the instructions straight into RAM with a monitor cartridge.


Very impressive, I'd been thinking of something similar.

In playing it in Stella, I frequently get into a "no move" situation where the grid looks like

  2   8   16    2
  4 128    2    4
 16  64    8    8
  4  16    4    _
Based on this config, I should be able to move right to collapse the two 8s, but no input it being allowed. I suspect it's a bug in the "NEW RANDOM TILE" code not allowing that last cell to be selected for a new item, since in the Stella debugger, memory at A8 is stuck with value 3.

Update: https://github.com/chesterbr/2048-2600/pull/2


Awesome, thanks for the (impressively quick) fix! Already merged and built into the repo binary.


For anyone who wants to try it out in their browser: http://jamesfriend.com.au/2600/2048/

(give it time to load, there's about 7mb of JS)


A js game, ported to assembly, emulated in js. All for fun.

I don't know why this makes me feel happy.


It's abstractions all the way down.


Is this by any chance Chrome/webkit-only? It's been loading for over 10 minutes here (using Aurora).


Hello, JSMESS developer here.

It should work on Chrome/Firefox/Opera/IE 10+/Safari/just about anything with typed arrays.

Chrome and Firefox have USB joypad support, but you need to activate it (by pushing a button on the joypad when the page loads) before the emulator starts.


Works for Firefox on Fedora


Love the sprite work to fit each of the numbers into 8-pixel-wide sprites. (The sample screenshot lacking what would be a rather chunky 256 may not be a coincidence.) The 2600 has no text or character capabilities, so the only way to display numbers is to use its very few hardware sprites, and there aren't enough to write out each digit in its own sprite.


Glad you liked it! I spent some (fun) time to squeeze the different width tiles in ways that would work well on different sizes, some working better than the others. If someone wants to improve upon it (the tiles are human-editable here: https://github.com/chesterbr/2048-2600/blob/master/tilegen.r...), I'd be happy to take a look!


The mobile version of the site hides the screenshot, so here's a direct link:

https://camo.githubusercontent.com/14454ebb4fa0b959a20b15f8a...


I'm waiting for the Kinect version of the game, so I can play it Minority-Report style.


I'd love Kinect + Oculus Rift puzzles: you'd have the board surround you, and manipulate the tiles with your hands. This would convince me that the future arrived! :-)



This is excellent! Hav you considered using hardware the hardware acceleration available in the Pitfall II mapper to get enough raster time for colors?


Heh, my current struggle is precisely trying to have tiles with different colors for each value (my last attempt fell 2 cycles short of achieving it). David Crane's DPC chip is indeed impressive (like pretty much everything he did on the Atari), and would at least free one register and shave enough to allow it, but I'd like to do that with "standard" Atari hardware. Thanks for the suggestion anyway!


It would have been wonderful if this had been released back in the day.


This is better than an interactive fiction version for me.


When's the Apple Lisa port coming?




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

Search: