stacksmashing / gb-wordle Goto Github PK

Game Boy-compatible games (like every pokemon game prior to Crystal) can use specific color palettes in a Super Game Boy or Game Boy Color.

It would be nice to use that to render yellow and green squares.

It allows me to submit certain nonsense words. Sidenote, l can't think of any word this could possibly be:

Further testing, this is practically non-functional as a game.

There's no win state. I got the word and it allowed me to keep guessing ultimately telling me I failed. Also THAMA is not an (English) word. Is this using an English dictionary?

It showed RAINK as correct.

Please specify in the readme file which exact version of GBDK you used.

A bit of auditory feedback.

If you guess two of a letter that only appears in the solution once, both will be marked as being in the word.

Hey, Great project!
I find a little hard to play in English, so I would like to make a quick translation to Spanish.

I'm finding some errors when running Make. I'm using gbdk-4.0.6 on Windows 10 x64 and tryied with Ubuntu as well.

Before sharing the specific errors, could you share the recommended environment to start from there? Thanks!

Here's an example that doesn't seem to make much sense to me. Or maybe the rules are different compared to standard Wordle?

Could you maybe also add the color mapping to the readme?

Wordle shows the correct word when you lose, while this version only says "You lost. Sorry!"

Like in the official wordle.

Much of the fun of wordle is playing the same word as someone else so you can compare solutions. I realize that multiplayer via the link cable (in order to synchronize which word is played) may be a big request. But other solutions exist, such as simply requesting a puzzle number (which could just be the word in position X of the word list) or requesting that the user input the date and then giving the same puzzle based on the date.

It would require adding some letters. For example, in Spanish we have an extra letter, "Ñ".

You could look into banking and compression.
Also don’t use strings in your word list, strings end with \0, so each word takes up 6B and 16% of the space is basically wasted.

MBC5 can have 512 banks (at least one is for your code) á 16KiB of which GBDK-2020 can address 255 with it’s functions. 256 banks are still 4MiB.

So you can fit > 1.67 million 5 char words in your list. But you also have to manage and find them. Or > 835 thousand 5 char words when you are limited to GBDK-2020’s capabilities.

TLDR: just do 0x00-0x29 for A-Z, 0x2A-0x7F for Digram Encoding, 0x80 to flag the end of it

Another trick would be to not use ASCII, it has 128 characters, but you actually just need use a-z (26 characters). You could for example use a different encoding where youcan fit two of the 15 most frequent charaters in one byte. The most frequent characters seem to be: (maybe easier to implement than digram encoding)

• 0: the other 11 characters in the second part of the byte, or <empty> if F is the second part or <null> if 0x00

• 1: E

• 2: T

• 3: A

• 4: O

• 5: N

• 6: I

• 7: H

• 8: S

• 9: R

• A: L

• B: D

• C: U

• D: C

• E: M

• F: W

• 00:

• 01: Y

• 02: F

• 03: G

• 04: P

• 05: B

• 06: V

• 07: K

• 08: J

• 09: X

• 0A: Q

• 0B: Z

This has to end with 0x00 again since this coding has a variable length. So we need some form of delimiter again. Pro of this is, that you can use the C string functions.

“WORLD” is 0xF4, 0x9A, 0xB0, 0x00
“RUGBY” is 0x9C, 0x03, 0x05, 0x01, 0x00
“FRANK” is 0x02, 0x93, 0x50, 0x07, 0x00

Digram Encoding is likely better, but that also needs some form of delimiter. Custom encoding + some form of compression is also possible.
Just Digram Encoding might always gain. With 2B references It could have 3B to 5B, because you would just use the plain string in the worst case. So it’s probably not worth it.

Another encoding is to use one bit (&0x80) for marking the end and just replacing parts of ASCII you don’t need with most common digraphs/digrams:
(er, ti, ar, ou, or, le, th as 01, 02, 03, 04, 05, 06, 07)
“OTHER” is 0x4F, 0x07, 0x81
“MOUTH” is 0x4D, 0x04, 0x87,
“ORDER” is 0x05, 0x44, 0x81
“UNCLE” is 0x55, 0x4E, 0x43, 0x86
“VOICE” is 0x56, 0x4F, 0x49, 0x43, 0xC5