I get a 16x16 grid pattern that the circles can occupy inside the square. Some circles "overlap" the borders of the square both top and bottom and side to side, by one pixel.

I'm looking at it in Paint and magnified 800 % zoom.

Just FYI.

WOW this is a tricky one, and the different size circles doesnt help!

I think that the key at the bottom says that the pattern is as follow:

1 .Circle moves up from bottom right to top right, then top left (minus one) down to bottom left. (15 position + 14 positions)
2. Circle moves from top right minus one position to bottom left (minus one position) and then from top left to bottom right minus one position. ( 14 + 15)

There is no top left to top right movement.
There is no bottom left to bottom right movement.

Using this i got some letters which yielded either SERVER, FINDING, SECRECY or WARFARE in a decryptor. This could be a coincidence though.

I have to go training now...

Do you think the circles are supposed to be different sized? I don't. I'm going to work on the assumption that they're supposed to be the same size.

Has this always been at the bottom of the puzzle page? I never noticed it before...

I have worked out the following:

Square dimensions 25px by 25px.
Number of positions of circle in x&y dimension is 17x17 (width minus radius of circle)
Circle Diameter 8px or 9px (annoyingly mainly 8px but some are not 'proper' circles due to pixelation)


Does anyone else agree with this? Or have i worked it out wrong

Originally posted by BannedForLife
Maybe this is the way the alphabet should be?

That reminds me of this for some reason:

Originally posted by GarethAyres
Square dimensions 25px by 25px.
Number of positions of circle in x&y dimension is 17x17 (width minus radius of circle)
Circle Diameter 8px or 9px (annoyingly mainly 8px but some are not 'proper' circles due to pixelation)

You're probably right, but I get 24X24 and 16X16. I'm not worrying myself with the diameter of the circle because it changes slightly, I ASSUME that's a graphics issue and not supposed to be that way.

I'm going to work with my numbers for a while (when I get time - ACK!)

Well, this is certainly odd. I got this

C3 C5 A7 C14 A2 A6 C2
B14 C15 B6 B10 B12 B15 B7 B3
B11 B14 A12 A10 C6 B7 B13 A12 B2 C14
B10 B7 A3 B10 D2 C13 A12 A12
C13 C7 A2 C13 C5 C2 C7 C11 D3 B13
A2 B13 B12 C7 B5 C10 C10
C10 B8 B8 A12 B3 B11 A10 C3 B14 C15
A3 A1 B12 A1 C15 B3 B3

Symbols used
A 1 2 3 6 7 10 12
B 2 3 5 6 7 8 10 11 12 13 14 15
C 2 3 5 6 7 10 11 13 14 15
D 3

30 symbols used!

[edit on 14-8-2008 by Deaf Alien]

Rows A and C are identical when rotated 180 degrees.
Individually, B and D are identical when rotated 90 degrees.

Allright, assume that both rows in both sets are identical to each other when rotated or mirrored and assume that the same sets of letters are in them, I get this.

A 1 2 3 6 7 10 12
C 2 3 5 6 7 10 11 13 14 15

B 2 3 5 6 7 8 10 11 12 13 14 15
D 2

There are 24 letters used.

[edit on 14-8-2008 by Deaf Alien]

This is driving me crazy. I can't even decide how many positions there are in each direction

I'm thinking that maybe a Caesar Cipher instead of the regular alphabet follows the path taken by the circles. I'll play around with that idea and see if anything makes sense.

Hi all. Using Deaf Alien's idea that Rows A and C and rows B and D were some complement of each other, I found that there were 26 characters used.

Here's the cryptogram I came up with. I haven't even begun to check it, but I don't have a lot of hope.

ABCDEFE GHIJKLMN OGPQFRSPTU VMAVTDPP, DCEDBECWNS ESKCXQQ, QYYPXNQAGH ALKLZXX

[edit on 14-8-2008 by Benevolent Heretic]