In 1965, Gordon Moore predicted that the number of transistors on a chip would double roughly every two years. 6 decades later, that prediction has proven to be one of the most durable trends in technology history. From Intel's 4004 processor with 2,300 transistors in 1971 to Nvidia's Blackwell GPU packing 208 billion transistors in 2024, the semiconductor industry has achieved a 90-million-fold increase in density. This dashboard visualizes the complete history — transistor counts, clock speeds, process nodes, power consumption, and performance benchmarks across 100+ chips spanning every major era of computing.
2 years remains the approximate doubling period for transistor density on leading-edge chips, though the original 18-24 month cadence has stretched. TSMC’s N2 (2nm) process and Intel 18A show the trend continuing through architectural innovations like GAA transistors and backside power delivery.
208 billion transistors are packed into Nvidia’s Blackwell B200 GPU (2024), the densest commercial chip ever produced. Apple’s M4 Ultra contains 72 billion transistors. Intel’s original 4004 processor in 1971 had just 2,300 — a 90-million-fold increase over 53 years.
2 nanometers is the smallest process node entering mass production, with TSMC’s N2 and Intel 18A (1.8nm-equivalent) both targeting 2025-2026 volume manufacturing. Below 2nm, the industry is exploring CFET transistor stacking for sub-1nm-equivalent nodes.
1965 is when Gordon Moore published his original observation that the number of components on an integrated circuit doubled roughly every year, later revised to every 2 years in 1975. The Intel 4004 (1971, 2,300 transistors) provided the earliest commercial validation.
3 major approaches are extending computing beyond traditional scaling: chiplet architectures (AMD’s MI300 uses 13 chiplets), 3D chip stacking (TSMC SoIC), and new transistor designs like CFET. Specialized AI accelerators, photonic computing, and quantum processors offer gains through architectural innovation.