Hacker News Reader

Comment by kevingadd | Hacker News Reader

I think the link is meant to be https://pagespeed.web.dev/report?url=https%3A%2F%2Fgoogle.co... - a perf report on the Google homepage. When I click the link from HN it pulls up a performance report on the performance report service (which is also bad.)

It's fascinating that according to the report, someone on a Moto G4 would have to wait over 6 seconds for the google search page to be responsive so they could type in a query.

View on HN

I wonder how accurate are the results really, or maybe I wasn't interpreting the numbers correctly. Accessing Google.com from my Moto G7 Power (a phone about 30% faster than G4 in terms of CPU performance) shows no seconds long delay. In face it feels almost instant.

On the other hand it seems to give a fair critique to other sites that are known to be bloated such as the one below.

https://pagespeed.web.dev/report?url=https%3A%2F%2Fnzherald....

They simulate slow wireless data, which I suspect is the cause of the gap. It's certainly the case that a big % of the world's population has bad mobile data even if they do have a relatively fast smartphone.

Wonder if it's due to a warm cache. I never use Google (hence it won't have google.com cached) and accessing it from Firefox on my high-end Android S21 takes a few seconds to load via my NZ fibre connection. Certainty not instant. DuckDuckGo on the other hand loads nearly instantly (although it's probably a warm cache on my part). To be more fair I tried Bing (which I also never use) which seems to be on par with Google, I suspect due to Bing loading a photogenic but ultimately useless background image.

Tried it again on desktop chrome with mobile simulation turned on and cache turned off. The Google.com front page still loads in about 1 second and it honest feels identical on an actual phone.

Despite what other comments seems to suggest here, the Google landing pages is actually pretty clean and not very heavy on the JavaScript.

dang informed me that HN stripped off the second half of the URL. It routinely distills them down to the canonical version. I feel slightly better about myself and dang restored it.

Appreciate that. Not sure how I managed to screw that up so royally.

Update: it wasn't me! https://news.ycombinator.com/item?id=29670584

Reply on news.ycombinator.com