Comment by diath
7 hours ago
It's weird that the article does not show any benchmarks but crappy descriptions like "milliseconds to microseconds" and "tens of thousands to single digits". This is the kind of vague performance description LLMs like to give when you ask them about performance differences between solutions and don't explicitly ask for a benchmark suite.
I disagree. I think it's a nice discovery many might be unaware of and later spend a lot of time on tracking down the performance issue independently. I also disagree that a rigorous benchmark is needed for every single performance-related blog post because good benchmarks are difficult to write, you have to account for multiple variables. Here, the author just said - "trust me, it's much faster" and I trust them because they explained the reasoning behind the degradation.
> No schema changes. No new indexes. No query rewrites. Just telling Dapper the correct parameter type.
Are we automatically discarding everything that might or might not have been written or assisted by an LLM? I get it when the articles are the type of meaningless self improvement or similar kind of word soup. However, if hypothetically an author uses LLM assistance to improve their styling to their liking, I see nothing wrong with that as long as the core message stands out.