Re:Human coders are still better than LLMs
by UncleOxidant on Thursday May 29, 2025 @03:50PM
(#44129621)
Claude and Gemini are decent at it as well. I was surprised when I asked claude (and this was several months back) to come up with a testbench for some very old, poorly documented verilog. It did a very decent job for a first-cut testbench. It even collected common, recurring code into verilog tasks (functions) which really surprised me at the time.
[ Parent
| on HN ]
Re:Human coders are still better than LLMs
by cushychicken on Friday May 30, 2025 @06:59AM
(#44134876)
Yes! It’s much better at using functional logic than I am - which I appreciate!