![]() |
![]() |
#1 |
Aug 2002
216D16 Posts |
![]() |
![]() |
![]() |
![]() |
#2 | |
"Marv"
May 2009
near the Tannhรคuser Gate
22×3×67 Posts |
![]() Quote:
An interesting game. Kinda like "Mastermind" with letters. Do you use a word list that was posted somewhere or just "wing it"? TIA |
|
![]() |
![]() |
![]() |
#3 | |
Bamboozled!
"๐บ๐๐ท๐ท๐ญ"
May 2003
Down not across
2D8E16 Posts |
![]() Quote:
So far, I have just winged it. I do have a list of 15,949 five-letter words left over from when I was password cracking and playing similar games. Note that number can be represented in 14 bits and you have 6 trials. In other words, you need to make your choices so as to reveal at least 7/3 bits per trial. I also have 24 year old software which plays a mean game of Jotto. That code was written no later than 1997-01-18 according to the source's timestamps. Tempted to modify it to play Wordle. |
|
![]() |
![]() |
![]() |
#4 |
Bamboozled!
"๐บ๐๐ท๐ท๐ญ"
May 2003
Down not across
1166210 Posts |
![]()
Hmm, according to Wikipedia, the Jotto I played is a more complex version than the ones given there.
There are six hidden 6-letter words. A probe word is given six scores, one per hidden word. The score returned is the number of exact matches but not their locations. The first three probes are special, in that they are all given simultaneously, as are their scores --- so you can't learn anything from the first before submitting the second and third, likewise learning about the second before submitting the third. Thereafter probes and their scores are given singly. You may state that you know all six words and give them. If all are correct you win; if even one is wrong you lose. It used to take me about 8-10 probes. Wordle is a much simpler game and so the six allowed should be easily sufficient. |
![]() |
![]() |
![]() |
#5 | |
3·23·127 Posts |
![]() Quote:
These days, thinking about AI and materials science advances along with the ubiquity of those things in "The Matrix" made me smile when thinking about Jotto, one time pads.. |
|
![]() |
![]() |
#6 | |
6809 > 6502
"""""""""""""""""""
Aug 2003
101ร103 Posts
5·2,179 Posts |
![]() Quote:
|
|
![]() |
![]() |
![]() |
#7 |
"Ed Hall"
Dec 2009
Adirondack Mtns
5×1,051 Posts |
![]()
I got lucky with my first try and then wrote a program that helped me solve it after one more try. I do consider that cheating, but I exercised my C++ skills. The dictionary file I use for all such programming only has 3574 five letter words. The program returned 13 based on my first try, of which seven could be eliminated by further inspection. I could have programmed elimination of the seven, but left it for later.
|
![]() |
![]() |
![]() |
#8 |
Dec 2017
22×19 Posts |
![]()
Question: Is there a strategy which can identify any target word within a fixed number of guesses? Say 3. In that case, which word should be selected as the first guess.
|
![]() |
![]() |
![]() |
#9 | |
Bamboozled!
"๐บ๐๐ท๐ท๐ญ"
May 2003
Down not across
1166210 Posts |
![]() Quote:
My Jotto program maximised the entropy of the 4th and subsequent guesses. The first 3 were left up to me. I could have found an optimum triplet but never implemented it. |
|
![]() |
![]() |
![]() |
#10 |
6809 > 6502
"""""""""""""""""""
Aug 2003
101ร103 Posts
5×2,179 Posts |
![]()
Ideally you guess 3 words with unique letters. That would get you info on 15 letters out of 26. That would narrow the field of possible words quite drastically.
|
![]() |
![]() |
![]() |
#11 | |
Bamboozled!
"๐บ๐๐ท๐ท๐ญ"
May 2003
Down not across
2·73·17 Posts |
![]() Quote:
I did teach Alec Muffett of a way to store dictionaries which is particularly efficient for both storage and accessibility. If you don't know who Alec is, a search on his name and "crack" might prove informative. |
|
![]() |
![]() |