From 2b29a44ee0f6e6b89875b8ad24043fd46c0c60c9 Mon Sep 17 00:00:00 2001 From: Razvan Dimescu Date: Mon, 13 Apr 2026 01:02:10 +0300 Subject: [PATCH] docs: remove unfair NextDNS comparison from performance section Comparing local cache (0.8ms) vs a remote service (37ms) measures network latency, not resolver quality. Any local resolver would show the same advantage. Replaced with AdGuard Home comparison which is a fair local-to-local benchmark. --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 5c632cf..9979d46 100644 --- a/README.md +++ b/README.md @@ -124,7 +124,7 @@ From Machine B: `curl http://api.numa` → proxied to Machine A's port 8000. Ena ## Performance -0.1ms cached queries — matches Unbound, 47× faster than NextDNS. Wire-level cache stores raw bytes with in-place TTL patching. Request hedging eliminates p99 spikes: cold recursive p99 538ms vs Unbound 748ms (−28%), σ 4× tighter. [Benchmarks →](benches/) +0.1ms cached queries — matches Unbound and AdGuard Home. Wire-level cache stores raw bytes with in-place TTL patching. Request hedging eliminates p99 spikes: cold recursive p99 538ms vs Unbound 748ms (−28%), σ 4× tighter. [Benchmarks →](benches/) ## Learn More