◆ CIRCUITS & CHIPS, DECODED FOR THE AI ERA 10 FREE CALCULATORS NO LOGIN · NO ADS · NO TRACKING BUILT FOR EE STUDENTS WORLDWIDE v0.1 LIVE
◆ CIRCUITS & CHIPS, DECODED FOR THE AI ERA 10 FREE CALCULATORS NO LOGIN · NO ADS · NO TRACKING BUILT FOR EE STUDENTS WORLDWIDE v0.1 LIVE
v0.1 — public reference · est. 2026

Circuits & chips,
decoded for the AI era.

A free, no-login reference for students learning the silicon underneath the models. Calculators that just work. A glossary built for the generation training transformers but never taught what's inside the GPU.

R1 R2

Ten calculators every EE student actually uses.

Type a value, get an answer. No accounts. No paywalls. No "click here to see the result." Built the way calculator sites should have been built ten years ago.

see all 10 calculators →

Concepts the textbook skips, explained simply.

Short, honest explanations of the hardware behind modern AI. Written for students who train models every day and have never been told what's actually inside a GPU.

featured hardware AI compute

What's actually inside a GPU?

A GPU is not a "faster CPU." It's thousands of small, dumb workers all doing the same operation at once on different data. That's why it accelerates matrix math — and that's the whole reason it accelerates AI.

Open up an NVIDIA H100 die shot and you'll see 132 Streaming Multiprocessors (SMs), each containing 128 CUDA cores plus specialized Tensor Cores for matrix-multiply-and-accumulate. That's roughly 16,000 simple arithmetic units all running simultaneously.

The key insight: a CPU optimizes for latency (finish one task fast). A GPU optimizes for throughput (finish 16,000 tasks at once, even if each one is slower). Training a neural network is mostly multiplying large matrices — exactly the workload GPUs were designed for, even before AI was the point.

The other half of a modern GPU is memory bandwidth. HBM3 stacks deliver ~3 TB/s of bandwidth to those cores. Starve them of data and the compute is useless. This is why "compute" and "memory" are both quoted in GPU specs.

why this matters
Every time you call an LLM, ~16,000 tiny arithmetic units are doing fused multiply-adds on tensors. Understanding this changes how you think about cost, latency, and what "GPU-poor" really means.
read the full entry →
↓ free download

30 circuit formulas every
EE student should memorize.

A printable one-page PDF cheat sheet. Ohm's law to op-amp configs to Karnaugh maps. Drop your email, we'll send it instantly.

✓ check your inbox — the PDF is on its way.