A table consisting of $5$ columns and $32$ rows, which are filled with zero and one numbers, are "varied", if no two lines are filled in the same way. On the exterior of a cylinder, a table with $32$ rows and $16$ columns is constructed. Is it possible to fill the numbers cells of the table with numbers zero and one, such that any five consecutive columns, table $32\times5$ created by these columns, is a varied one? Proposed by Morteza Saghafian
Problem
Source: Iranian TST 2019, first exam day 1, problem 1
Tags: combinatorics
math90
24.06.2019 12:18
Does it contain cyclically consecutive columns, e.g $29,30,31,32,1$? Edit: $13,14,15,16,1$.
jazzup
24.06.2019 12:42
math90 wrote: Does it contain cyclically consecutive columns, e.g $29,30,31,32,1$? The problem says a cylinder, so I think yes.
Pathological
04.07.2019 05:53
The answer is yes, here is a construction.
0000000000000000
____________________
1111111111111111
____________________
0101010101010101
1010101010101010
____________________
0111011101110111
1110111011101110
1101110111011101
1011101110111011
____________________
1000100010001000
0001000100010001
0010001000100010
0100010001000100
____________________
0011001100110011
0110011001100110
1100110011001100
1001100110011001
____________________
0110100101101001
1101001011010010
1010010110100101
0100101101001011
1001011010010110
0010110100101101
0101101001011010
1011010010110100
____________________
0011110000111100
0111100001111000
1111000011110000
1110000111100001
1100001111000011
1000011110000111
0000111100001111
0001111000011110
____________________
Let $S$ be the set of binary strings of length $5$, i.e. $S = \{0, 1\}^5.$ We will say that $s_1, s_2 \in S$ satisfy the relation $s_1 \rightarrow s_2$ if the last four digits of $s_1$ are the first four digits of $s_2.$
Now, a row of the table corresponds to $s_1, s_2, \cdots, s_{16} \in S$, satisfying $s_1 \rightarrow s_2 \rightarrow s_3 \rightarrow \cdots \rightarrow s_{16} \rightarrow s_1$.
From here, it would suffice to partition the elements of $S$ into cycles of lengths divisible by $16$. In this way, if we had a cycle $s_1 \rightarrow s_2 \rightarrow \cdots \rightarrow s_t \rightarrow s_1$ for $t | 16$, we could let our first row be such that it reads $s_1, s_2, \cdots, s_t$ each $\frac{16}{t}$ times, and cyclically shift this first row by one $t-1$ times to get $t-1$ other row.
(These groups of cyclically shifted rows are delineated in the answer with _____)
@stroller yea took me about an hour as well incl. writeup
Hamel
04.07.2019 16:46
@above Can you post your motivation?
stroller
04.07.2019 19:07
0000010001100111
1111101110011000
+ their cyclic shifts
EDIT: how long did people take to do this? I spent an hour on this P1.
Attachments:

Pathlogical
05.07.2019 00:42
The following numbers work:
000000|00000|00000
000011|00001|00001
000101|00010|00010
000110|00011|00011
001001|00100|00100
001010|00101|00101
001100|00110|00110
001111|00111|00111
010001|01000|01000
010010|01001|01001
010100|01010|01010
010111|01011|01011
011000|01100|01100
011011|01101|01101
011101|01110|01110
011110|01111|01111
100001|10000|10000
100010|10001|10001
100100|10010|10010
100111|10011|10011
101000|10100|10100
101011|10101|10101
101101|10110|10110
101110|10111|10111
110000|11000|11000
110011|11001|11001
110101|11010|11010
110110|11011|11011
111001|11100|11100
111010|11101|11101
111100|11110|11110
111111|11111|11111
This clearly works upon inspection.
Call a $2^n \times a$ table constructed on a cylinder "good" if we can fill the cells with 0 or 1 so that any $2^n \times n$ table constructed by $n$ consecutive columns is varied, where $a \ge n$, and we extend the definition of varied tables to $2^n \times n$ tables rather than just $32 \times 5$ tables. It is easy to find that if we can construct a good $2^n \times a$ table and a good $2^n \times b$ table, then by joining them together (as shown in the above solution with "|"), we can construct a good $2^n \times (a+b)$ table. Clearly, we can construct a good $2^n \times n$ table. We can show that we can also construct a good $2^n \times (n+1)$ table, by induction on n. Then, by putting together 2 $32 \times 5$ tables and 1 $32 \times 6$ table, we have a good $32 \times 16$ table, as desired.
Idio-logy
04.11.2019 17:50
0000000000000000
1111111111111111
1000100010001000 and cyclic shifts (4)
0111011101110111 and cyclic shifts (4)
1100110011001100 and cyclic shifts (4)
0101010101010101 and cyclic shifts (2)
1001011010010110 and cyclic shifts (8)
1100001111000011 and cyclic shifts (8)