Limits to the energy efficiency of CMOS microprocessors | Epoch AI
epoch.ai · 945 words · saved by 1 readers
How far can the energy efficiency of CMOS microprocessors be pushed before hitting physical limits? We find room for a further 50 to 1000x improvement.
Summary Compared to the ENIAC in 1946, modern microprocessors can perform roughly a quadrillion times more computations for every unit of dissipated energy. This illustrates one of the most important trends in the history of computing – Koomey’s Law – under which the energy efficiency of computers has increased drastically over the last eight decades. But how much longer can this progress persist before hitting physical limits? The answer to this question has important implications for forecasts of future progress in hardware: if we are close to the fundamental limits, the returns to…
saved by
related reading
- When will computer hardware match the human brain? by Hans Moravecjetpress.org
- Transformer Inference Arithmetic | kipply's blogkipp.ly
- Could a Neuroscientist Understand a Microprocessor? | PLOS Computational Biologyjournals.plos.org
- All About Rooflines | How To Scale Your Modeljax-ml.github.io
- pp-beyondCMOSece.cmu.edu
- Can AI scaling continue through 2030? | Epoch AIepoch.ai
- Ushering in the Thermodynamic Future | Extropicextropic.ai
- Transformer Math 101 | EleutherAI Blogblog.eleuther.ai
- Strangely, Matrix Multiplications on GPUs Run Faster When Given "Predictable" Data! [short]thonking.ai
- AI Chip Architecturesjacobpeake.com
- An efficient probabilistic hardware architecture for diffusion-like modelsarxiv.org
- Slowing Moore’s Law: How It Could Happen · Gwern.netgwern.net