Hacker News Reader

Obviously the sample sizes are too small to be concluding much in this direction but for fun, it looks like the HTTP/3 numbers have more outliers than the HTTP/2 numbers. Is that to be expected due to the nature of the new protocol or the experiment?

View on HN

That was my initial takeaway. I suspect that the outlier is the initial connection / handshake, and that all subsequent requests were much faster thanks to the QUIC / 0-RTT session ticket. But I can't see anywhere this is mentioned explicitly, and those outliers are a fair bit worse than HTTP/2.

Reply on news.ycombinator.com