Semiconductor physics test: Neon nails transistor trap diagram, Fable 5.2 adds second-order details
xiaosun86 · x · 2026-11-20
The author stress-tests models on an n-MOSFET Id-Vg question (acceptor-like vs donor-like interface traps):
- Neon: quirky behavior — drew the schematic in ASCII first, then plotted; but got the physics right immediately
- Fable 5.2: captured the first-order effect (curve stretching as traps fill) plus second-order details — the minimal off-current floor from other mechanisms, and the kink caused by the sharp trap profile both models assumed
- The accompanying images compare three outputs: ideal no-traps, acceptor-like traps (subthreshold right shift), donor-like traps (left shift that disappears), with physical mechanism annotations
Verdict: how far Fable is ahead of Neon remains unclear on this simple domain question; the author also notes GPT-6.2 extra-effort mode didn't respond.
Related event: AI models tested on drawing MOSFET trap effect diagrams(3 posts)→
More from Models
- Leaker teases: Google's upcoming models have 'cracked the code' — iruletheworldmo · 2026-11-20
- Qwen4 Max lands on PostTrainBench at 36.7%, edging below GLM 5.3 Flash — maksym_andr · 2026-11-20
- PostTrainBench data: US-China model gap of ~8% is nothing unusual — maksym_andr · 2026-11-20
- Cost-vs-quality chart puts Gemini 4 Neon on the CursorBench Pareto frontier — zacharynado · 2026-11-20
- AI model proves optimal solutions to long-open MIPLIB optimization problems — tdhopper · 2026-11-20
- Claude 3 Sonnet spontaneously writes a full Latin hymn at 7:59 PM — repligate · 2026-11-20