LEARNING PATHS
Learn in a useful order.
Choose a path, read each lesson, take the short quiz, and keep your progress in this browser—no account required.
CPU and Memory Systems
Follow performance from the core pipeline through caches, coherence, the NoC, and coherent expansion.
- 1High-Performance Single-Core CPU Classroom: Why the Same GHz Can Be Twice as Fast✓
- 2Multi-Core Throughput Classroom: Why More Cores Do Not Guarantee Faster Work✓
- 3Cache and Coherency Comic Classroom: From the Memory Wall to CXL✓
- 4AI Inference Chip Architecture: From GPU + HBM to ASIC + CXL + DRAM + NMC✓
CPU Classroom Series
Explore the 26-lesson curriculum: single cores, parallel work, memory systems, Arm, RISC-V, and optional CPU design labs.
- 1High-Performance Single-Core CPU Classroom: Why the Same GHz Can Be Twice as Fast✓
- 2Multi-Core Throughput Classroom: Why More Cores Do Not Guarantee Faster Work✓
- 3Cache and Coherency Comic Classroom: From the Memory Wall to CXL✓
HBM and Advanced Packaging
Follow data from the CPU and accelerator through memory bandwidth, HBM stacks, repair, and packaging.
- 1High-Performance Single-Core CPU Classroom: Why the Same GHz Can Be Twice as Fast✓
- 2Semiconductor Packaging Classroom: From CoWoS to ABF✓
- 3HBM Classroom: Why GPUs Need High-Bandwidth Memory✓
- 4HBM Classroom 2: Base Die and Repair Evolution✓
- 5AI Inference Chip Architecture: From GPU + HBM to ASIC + CXL + DRAM + NMC✓
AI Technology Classroom
A growing series on LLM inference, tokens, KV cache, memory movement, and AI accelerator architecture.
- 1AI Comic Classroom #1: LLM Tokens and KV Cache✓
- 2From Tokens and Attention to the KV Cache✓
- 3AI Inference Chip Architecture: From GPU + HBM to ASIC + CXL + DRAM + NMC✓
- 4Taalas Hardwired Inference Comic Classroom: Building Model Weights into Silicon✓
- 5Cerebras Wafer-Scale AI Comic Classroom: Why Build One Processor Across a Whole Wafer?✓