Maybe not, but is the target audience that shills out $20/month really the type of people who have optimized their site to such an extent that shaving 50ms off the request latency by having your edge cache geolocated is really the type of thing that makes the difference? most of that group could probably do a lot of other optimizations that probably count for more.
The common mistake is to pick a server geographically close to yourself, only access it from low-latency connections, and then assume that everyone in the world is seeing the same thing.
Or to only visit your own site with everything already in the browser cache. If you're not seeing cold start loads, you're not seeing what every new visitor to your website is seeing.
Consider the Photopea.com website. The author explained in a comment below that he spends $60/month to host the site without a CDN. Several of us loaded the site and it took 2.5 - 5.0 seconds to load. He could sign up for a cheap Cloudflare account, reduce the size of his server (due to caching), and the load times for everyone would drop by a significant amount.
If you're hosting simple, static content like a blog for an audience that doesn't care about load times, then of course nothing matters. But for modern, content-rich websites (photos especially) it can actually be a substantial improvement to add a CDN even if you have a single fast server. You may not see it, but visitors from distant locations definitely will see a difference.
A CDN is more times than not the wrong answer to a real problem. Shave off your website and consider content-addressed protocols for big static asset download (like the textures from the article). If you run your website as a lightweight glorified Bittorrent index you'll notice your costs are suddenly a lot less, and you can still have a smaller "Download over the web" button as fallback.
This is a conclusion i am extremely doubtful of.
Ping time new york <-> tokoyo is about 180ms. So lets say as a worse case the ping time to the single server is 180ms (its probably not that bad), and lets say the latency to cloudflare edge server is 20ms.
So using cloudflare on a cache hit (best case), you save something like 160ms per roundtrip.
Which don't get me wrong is a huge savings and worth it (although this scenario is hugely exagerated).
However say you want to load the page in under 1 second instead of 5 seconds. In this scenario you would basically have to have 25 round trips to bring the site from 5 seconds to 1 second just on rtt savings of having a geo located edge server. If your site needs 25 round trips to load, something else is clearly wrong. (And this is an exagerated case, the real world the benefit would probably be much less)
To be clear i'm not saying that geo located edge caches are bad or useless. They are clearly very beneficial thing. Its just not the be all and end all of web performance, and most people in the demographic we are talking about probably have much more important things to optimize (otoh using cloudflare is cheap and doesnt require a lot of skill, so it is a very low hanging fruit)
Per packet. If you're doing a cold start, you'll pay that latency cost several times over: first the TCP handshake (3 roundtrips), and then the TLS handshake (2 more roundtrips). That's 800ms of extra latency before you even get to sending the first HTTPx request.
Cold start latency matters a lot.
You’re forgetting that the TCP protocol itself is bidirectional. High latency connections will have lower throughout, especially at the beginning of transmission, because the data isn’t literally just streaming in one direction.
[1]: https://stackoverflow.com/questions/536300/what-is-the-short...
I just think if your site takes 700ms, is there really a difference between that and 650ms?
3.44 seconds to do a search for "donkey"
