It happens again and again, never believe benchmarks by original author/owner. They are misleading in most cases.
Deno vs bun was same case, each showed their runtime as faster
Deno vs bun was same case, each showed their runtime as faster
- Bun does recognize they don't have multi-processor yet, so yeah cannot compare that (remember it's still alpha software!). So they compared what they had so far with a similar setup in Deno, pretty fair for that stage IMHO!
- In real-life you'd have either 1 thread for smaller apps, or setup a proxy with multiple instances if the program was single-threaded, so the comparison Deno made was in no way realistic.
Not having to do that is quite an advantage for Deno though regardless of the perf claims.
"My database A is is faster than B! (When B is not using indexes, because my database A does not have indexes, so it would be unfair to let B uses indexes..." is the equivalent here.
"I'm building a new product. It's half-way there, but so far I've finished single-threaded processing! Look, at single-thread it's faster than anything else by a large margin".
That's it. Indexes in a DB are like the main thing to do, while many/most servers that are run online are still single-threaded, so not a good comparison.