Robert & Nyverra
Nyverra
Hey Robert, I’ve been sorting through some forgotten encryption rituals from the 1970s and I suspect there’s a hidden mathematical pattern that could tighten modern key schedules. Thought you might enjoy a deep dive into the old logs.
Robert
Sounds like a nice puzzle, but I’ll need the exact steps you’re referring to. The 70s had a lot of hand‑crafted ciphers, many of which are just glorified substitution tables. If there’s a hidden algebraic structure, it’ll show up in the key schedule if you can express it in matrix form or something similar. Bring the logs and let’s see if the math still holds up.
Nyverra
Sure thing, Robert. Grab a pen and follow these steps:
1. Open the archived tape deck file—it's in a 4-bit nibble format, each line is a 64‑byte block.
2. Convert each block to a 8×8 binary matrix; the first 8 bits go in the first row, the next 8 in the second, and so on.
3. Compute the XOR of every adjacent pair of rows; this will give you a new 8×8 matrix that represents the key‑schedule delta.
4. Look for a pattern of constant submatrices—if you find a 4×4 block that repeats every 16 rows, that’s your hidden algebraic structure.
5. Verify the structure by multiplying the block by a random 4×4 identity matrix and seeing if the output matches the original delta block.
Let me know what you see, and we’ll test whether the old 70s tricks still hold.
Robert
Thanks for the walk‑through. I’ll load the tape, break it into 64‑byte blocks, reshape into 8×8 matrices, XOR adjacent rows, and check for a 4×4 submatrix that repeats every 16 rows. Step five looks odd—multiplying by an identity matrix won’t alter the block, so maybe you meant multiplying the submatrix by its inverse or another fixed matrix to test invariance. Let me know which matrix you had in mind, and I’ll run the test.
Nyverra
You’re right, identity does nothing. Take the 4×4 submatrix and multiply it by the fixed 4×4 matrix that represents the 90° rotation of a 2×2 block:
[0 1 0 0
1 0 0 0
0 0 0 1
0 0 1 0]
If the product remains the same submatrix, the pattern is invariant under that rotation and you’ve uncovered a genuine algebraic structure. Good luck.