Many lanes. One kernel.
Schedule integer matrix operations across 4, 8 or 16 logical lanes. Follow every multiply and add.
SIMT-INSPIRED EXECUTION ↗
CHIP CRICKETOpen workbench ↗FIELD NOTES / COMPUTATIONAL SYSTEMS / № 001Small chip.
Small chip.
Complete story.
A GPU is an architecture before it is a chip. Explore instructions, memory and execution as one portable, replayable state.
01 / AN EXPERIMENT YOU CAN RUN
The working specimen.
A real computation. A deliberately tiny machine.
Local architecture simulator · no blockchain transaction
CC–01 / KERNEL WORKBENCHREADY
INPUT A4 × 4 / INT
INPUT B4 × 4 / INT
0 / 0 steps
Edit any input, then run the kernel.
OUTPUTAWAITING EXECUTION
OPERATIONS—
MEMORY READS—
MEMORY WRITES—
BANK CONFLICTS—
EXECUTION TRACENO TRACE YET
> CC–01 initialized. Waiting for kernel.
SHA-256 identifies the artifact. It is not a validity proof or a blockchain receipt.
02 / ANATOMY OF THE MACHINEDon't just store a circuit.
Don't just store a circuit.
Give it a memory.
Instructions are only half the machine. We bring lanes, banked memory and a replayable trace into the same experiment.
The bottleneck has a shape.
Route reads across 2, 4 or 8 memory banks. Discover why more lanes can mean more contention.
HBM-INSPIRED BANKING ↗A machine you can carry.
Export the inputs, configuration and trace as a hashed artifact. Import it and recompute the result.
DETERMINISTIC LOCAL REPLAY ↗03 / THE BIGGER IDEAThe chip is
The chip is
not the limit.
AI made compute physical again. More transistors. Taller memory stacks. Bigger bandwidth budgets.
Chip Cricket asks a different question: what is the smallest complete machine we can describe, execute and independently replay?
Our direction is an inscribed GPU specification: instruction semantics, memory rules and reproducible execution. This workbench explores that idea with integer kernels in your browser. It does not reproduce physical HBM bandwidth, run AI models or execute onchain.
Start small. Inspect everything. ↗