|
para |
Posted: Wed Oct 22, 2008 6:46 pm
Anti-robot code? |
|
Tarn GURU
Joined: 10 Oct 2000 Posts: 873 Location: USA
|
Posted: Thu Oct 30, 2008 12:07 am |
para wrote: |
Code: |
@@@ @@ @@ @@@
@ @ @ @ @ @ @
@ @ @ @ @
@ @ @ @ @
@@@ @ @ @ @@@
@ @ @ @ @ @
@ @ @@ @ @@@@@ @ @
@ @ @ @ @ @
@@@ @@ @@ @
|
|
The 6 is obvious. If I use my imagination, I can see a 5 or a 3 next but I'm just guessing. I have no idea what the third character is supposed to be. The last is presumably an 8.
Is the user actually supposed to pass this? I would fail for both of the examples you've shown.
-Tarn |
|
|
|
umdbandit Wanderer
Joined: 31 Oct 2004 Posts: 94
|
Posted: Thu Oct 30, 2008 11:18 am |
Hey, could we have like 9 or 10 more examples? Also, could you give what the answer is "supposed" to be for each one? I'm curious to see the variance between submissions. Also, like the others said, this latest one looks like some alien font that means god knows what :P
|
|
_________________ Rufus- Winner of Stuff |
|
|
|
intoK Apprentice
Joined: 18 Feb 2007 Posts: 190
|
Posted: Thu Oct 30, 2008 1:52 pm |
upon seeing that, id just use delete command instead
|
|
|
|
Rahab Wizard
Joined: 22 Mar 2007 Posts: 2320
|
Posted: Thu Oct 30, 2008 5:10 pm |
It looks to me like the characters are deliberately distorted, with pixels shifted out of place. In fact, I wonder if the distortion is done dynamically, so that it never appears the same twice. If that is true, I think it would take full-scale OCR to translate them, and even then I wouldn't expect it to work as efficiently as the human eye, and my own human eye can't translate those. If my mud expected me to figure those things out just to keep playing, I'd find another mud.
|
|
|
|
ralgith Sorcerer
Joined: 13 Jan 2006 Posts: 715
|
Posted: Thu Oct 30, 2008 7:01 pm |
Agreed Rahab... thats nuts
|
|
_________________ CrossOver: Windows Compatibility on Mac and Linux CMUD Advocate |
|
|
|
para Beginner
Joined: 14 Sep 2008 Posts: 18
|
Posted: Wed Nov 05, 2008 12:01 am |
Its not easy to edit these, I cant just simply copy and paste those numbers here. What's more, I find it
will show differently depending on the web browser you are using. I could recognize the code 113 I posted before
using firefox, but not with microsoft IE 7.0. So weired... Anyway, thanks for all you guys looking into this.
Tarn wrote: |
para wrote: |
Code: |
@@@ @@ @@ @@@
@ @ @ @ @ @ @
@ @ @ @ @
@ @ @ @ @
@@@ @ @ @ @@@
@ @ @ @ @ @
@ @ @@ @ @@@@@ @ @
@ @ @ @ @ @
@@@ @@ @@ @
|
|
The 6 is obvious. If I use my imagination, I can see a 5 or a 3 next but I'm just guessing. I have no idea what the third character is supposed to be. The last is presumably an 8.
Is the user actually supposed to pass this? I would fail for both of the examples you've shown.
-Tarn |
|
|
|
|
Fang Xianfu GURU
Joined: 26 Jan 2004 Posts: 5155 Location: United Kingdom
|
Posted: Wed Nov 05, 2008 1:21 am |
Is it supposed to not be using a fixed-width font or something?
|
|
|
|
hkduck Newbie
Joined: 09 Feb 2009 Posts: 5
|
Posted: Mon Feb 09, 2009 4:11 am |
I am encoutering the some problem here, is it possible for cmud to read and auto answer code like this?
|
|
|
|
hkduck Newbie
Joined: 09 Feb 2009 Posts: 5
|
Posted: Mon Feb 09, 2009 5:05 am |
|
|
|
|
gmueller Apprentice
Joined: 06 Apr 2004 Posts: 173
|
Posted: Mon Feb 16, 2009 9:18 pm |
It looks like characters are 9 x 5 (rows x columns) what you can do is encode each number, and then use a "distance" function. the distance function just determines the distance between the encoded number, and the one that is being represented.
Lower distance means higher match. use an array to determine which matches are higher on the list, and then "guess" when two different numbers have the same distance.
Example:
Encoding for number 1:
X
XX
X X
X
X
X
X
X
XXXXX
pattern realized:
X
X
X
X
X
X
X
X
XXXXX
distance = 2 (two characters are different)
a more complex distance function would give you different matching potential. |
|
|
|
gmueller Apprentice
Joined: 06 Apr 2004 Posts: 173
|
Posted: Mon Feb 16, 2009 9:19 pm |
since I didn't use the "code" tag it screwed up my spacing, but you get the idea.
|
|
|
|
gmueller Apprentice
Joined: 06 Apr 2004 Posts: 173
|
Posted: Mon Feb 16, 2009 9:34 pm |
it also looks like the columns are either just a simple right shift or left shift. in that case your distance function should average or weight the right shifted, and left shifted version of each encoding.
|
|
|
|
|
|