INSPIRED BY APOLLO. NOT A REPLICA.
The program
is the pattern.
A circuit can calculate an answer.
A memory can remember every answer.
A lookup table is a little logic machine.
For three binary inputs, there are only eight possible combinations. We calculate the result of each one, then encode those answers into rows. The input selects a row; that row gives you the outputs. The “compilation” is turning your expressions into this finite table.
Why does Apollo memory look like a rectangular block?
The cores were packed into grids and layers inside metal modules, with many sense wires sharing each core. Wire bundles, connectors and protective potting gave the finished memory its dense rectangular shape. It was not one isolated ring per stored bit. Our module folds addresses into a compact grid and lets you isolate a bit wire to see its route.
For readability, we show one word per core and at most 16 cores at a time. Actual Block II ropes held 12 words per core in two layers of 256 cores, with selection and sensing electronics not modeled here. This is an explanatory visualization, not manufacturing geometry.
What was actually woven for Apollo?
Wires threaded through or around magnetic cores encoded bits of fixed, read-only memory. That memory stored instructions and data; it was not the entire computer. A separate processor executed instructions, and specialized circuitry read the magnetic signals.
Our yarn model uses the same through/around idea, but its rings are not magnetic and a person reads it. Our electronic build is a woven diode ROM: diode connections store the bits. The Arduino selects and reads a row; it does not evaluate your original equations. Neither build is an Apollo core-rope replica.
Store answers—or store the instructions.
Lookup ROM stores every output combination. Switch the visualization to Executable instructions to compile the same logic into 16-bit stack-machine instructions. The browser CPU reads those words and produces outputs checked against the truth table. Apollo likewise stored instructions and data for a separate processor, but used its own instruction set and 15 data bits plus parity. Our program mode does not run Apollo flight software, and the beginner Arduino build still reads only the lookup ROM.
Small programs can make big memories.
Each additional input doubles the rows. Three inputs need 8 rows; twelve need 4,096. Explore up to 16 output bits, with 64 named intermediate signals to keep the logic readable. The 3D view and print patterns show one section of up to 16 rows at a time; the CSV contains the complete memory.
The direct-wired UNO starter circuit is still limited to 3 inputs and 4 outputs. Larger electronic memories need address decoding and additional sensing hardware; we do not generate an invalid pin map for them. Start small physically, then scale your design. Stateful programs need extra memory and a clock, which these builds do not include.
Want to explore real magnetic core rope?
Start with the restoration work linked below. A magnetic implementation requires appropriate ferrite material, pulse-drive circuitry, and sense amplification; it cannot be substituted into our DC diode reader. We intentionally do not present an untested pulse circuit as a beginner construction plan.