xkcd.com speed test — 20 Oct 2026

Measured from the wfix.shop server on 20 October 2026 at 13:41 UTC. Grade A (92/100), with the screenshots, the timings and the fixes that matter.

https://xkcd.com/

20 October 2026 at 13:41 UTC · Free test · report pt_2ed99e4d0541
Grade
A
92/100
Performance100/100
Structure79/100

11 audit(s) passed, 5 need work

First byte (TTFB)217 msmeasured from this server
Fully loaded751 msfirst byte + slowest asset
Requests10document + assets
Page weight551.2 KBcompressed transfer
HTML document2.4 KBbr
ProtocolHTTP/2nginx
Third-party536.2 KBother hosts

Where the first byte went

  • DNS lookup 58 ms 27%
  • TCP connect 30 ms 14%
  • Server wait 130 ms 60%

Measured on the connection that fetched the HTML. Every other request on the page waits for the end of this bar before it can start.

How this page compares

First bytefaster than 90% of 4,999 pages tested here217 ms against a median of 449 ms
Page weightlighter than 55% of 4,999 pages tested here551.2 KB against a median of 650.5 KB
Fully loadedfaster than 95% of 4,999 pages tested here751 ms against a median of 1.84 s
Requestssends fewer requests than 88% of 4,999 pages tested here10 against a median of 26

Every report published here is in this comparison, so it is a reference population of pages tested from one server - not a benchmark of the whole web.

What these numbers mean

  • The first byte arrived in 217 ms. That is workable, and the cheapest win left is caching or a warmer connection.
  • The page transfers 551.2 KB over 10 request(s), which is lean. Images files are 99% of it, so resize and recompress the largest images before anything else.
  • The HTML itself arrives br-compressed.
  • It runs over HTTP/2, so the assets share a connection instead of opening one each.
  • 97% of the weight comes from other hosts (analytics, fonts, embeds), which also means their slow day becomes your slow day.
  • The largest element painted at 372 ms, inside the 2.5 s Google calls good.
  • Layout stayed still (CLS 0.051), so nothing jumps under the reader's cursor.

Fix first: Image delivery and format - Serve WebP or AVIF with a JPEG fallback, size images to the space they occupy, and lazy-load anything below the fold. Images are usually more than half of a slow page.

Read on one shared server, not a visitor's phone: use it to compare pages against each other, not as a promise about every connection. We fix what these tests find.

Screenshot

What the page looks like

Loaded in headless Chromium on this server. Click an image to open it full size.

Desktop screenshot of xkcd.com
Desktop 1366 × 768 · 201.0 KB · page is 1386 px tall
Mobile screenshot of xkcd.com
Mobile 390 × 844 · 85.7 KB · page is 2121 px tall

Browser render

The same URL loaded in a real browser, so these are the readings curl cannot produce. The grade above is still built from the server measurements.

Largest contentful paint372 msgood is under 2.5 s
Cumulative layout shift0.051under 0.1 is stable
Total blocking timenone over 50 mslong tasks during load
First contentful paint288 msfirst content on screen
Browser page weight207.6 KB7 request(s), scripts included
Largest contentful paint372 ms
Goodgood under 2.5 s, poor over 4 s
Cumulative layout shift0.051
Goodgood under 0.1, poor over 0.25
Total blocking timenone over 50 ms
Goodgood under 200 ms, poor over 600 ms

Page weight by type

What the page is made of

Images 8 requests
547.7 KB 99%
Css 1 request
1.1 KB 0%

Transfer size, compressed as sent. 10 request(s) in total, 551.2 KB across all of them.

Top issues

Image delivery and format

high impact

1 image(s) over 200 KB and 3 large image(s) still served as JPEG or PNG.

Fix: Serve WebP or AVIF with a JPEG fallback, size images to the space they occupy, and lazy-load anything below the fold. Images are usually more than half of a slow page.

Cache static assets

medium impact

2 asset(s) answer without a Cache-Control max-age, so browsers revalidate them on every visit.

Fix: Version filenames on change and serve them with a long max-age plus immutable. That is what makes a returning visitor fast.

Third-party weight

medium impact

1 third-party host(s) serve 536.2 KB of this page. Each one is another DNS lookup, TLS handshake and point of failure you do not control.

Fix: Self-host the libraries you can, drop the tags you cannot justify, and preconnect the ones you keep.

Mobile viewport

medium impact

No viewport meta tag, so a phone renders the desktop layout scaled down.

Fix: Add a viewport meta tag with width=device-width and an initial scale of 1.

Strict-Transport-Security

low impact

HSTS is not set, so the first visit on a hostile network can still be downgraded to HTTP.

Fix: Send Strict-Transport-Security with a max-age of at least six months. Add includeSubDomains once every subdomain is HTTPS.

Waterfall

Slowest requests

IMAGESimgs.xkcd.com/comics/fusion_device_2x.png
534 ms
IMAGESimgs.xkcd.com/comics/fusion_device.png
506 ms
IMAGESimgs.xkcd.com/s/temperature.png
497 ms
IMAGESimgs.xkcd.com/s/a899e84.jpg
471 ms
IMAGESimgs.xkcd.com/store_news/news_back.png
461 ms
IMAGESxkcd.com/s/0b7742.png
455 ms
IMAGESimgs.xkcd.com/store_news/news_back_2x.png
417 ms
CSSxkcd.com/s/7d94e0.css
416 ms
IMAGESxkcd.com/s/919f27.ico
411 ms

Each bar is one request's own response time, scaled to the slowest. Assets were fetched in parallel, so the bars overlap in real time rather than stacking. first party third party

RequestTypeStatusTimeSize
https://imgs.xkcd.com/comics/fusion_device_2x.png 3rd partyimages200534 ms210.7 KB
https://imgs.xkcd.com/comics/fusion_device.png 3rd partyimages200506 ms100.5 KB
https://imgs.xkcd.com/s/temperature.png 3rd partyimages200497 ms27.1 KB
https://imgs.xkcd.com/s/a899e84.jpg 3rd partyimages200471 ms21.0 KB
https://imgs.xkcd.com/store_news/news_back.png 3rd partyimages200461 ms44.9 KB
https://xkcd.com/s/0b7742.pngimages200455 ms10.7 KB
https://imgs.xkcd.com/store_news/news_back_2x.png 3rd partyimages200417 ms132.1 KB
https://xkcd.com/s/7d94e0.csscss200416 ms1.1 KB
https://xkcd.com/s/919f27.icoimages200411 ms738 B
Report this page

The page is removed when the owner agrees. Sending this removes nothing by itself.

Run this test on your own site

The same engine, free, with the screenshots and the fixes. No account needed.