<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p>I think he understands that he's talking about page fetch time to
      an audience that won't believe that latency is like "latent
      fingerprints": stuff that hasn't shown (up) yet. <br>
    </p>
    <p>Such folks annoy me (;-))<br>
    </p>
    <p>Note his quote usage in this:<br>
      <blockquote type="cite"><strong style="box-sizing: border-box;
          font-weight: 600; color: rgb(36, 41, 46); font-family:
          -apple-system, BlinkMacSystemFont, "Segoe UI",
          Helvetica, Arial, sans-serif, "Apple Color Emoji",
          "Segoe UI Emoji"; font-size: 16px; font-style:
          normal; font-variant-ligatures: normal; font-variant-caps:
          normal; letter-spacing: normal; orphans: 2; text-align: start;
          text-indent: 0px; text-transform: none; white-space: normal;
          widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px;
          background-color: rgb(255, 255, 255); text-decoration-style:
          initial; text-decoration-color: initial;">IMPORTANT NOTE</strong><span
          style="color: rgb(36, 41, 46); font-family: -apple-system,
          BlinkMacSystemFont, "Segoe UI", Helvetica, Arial,
          sans-serif, "Apple Color Emoji", "Segoe UI
          Emoji"; font-size: 16px; font-style: normal;
          font-variant-ligatures: normal; font-variant-caps: normal;
          font-weight: 400; letter-spacing: normal; orphans: 2;
          text-align: start; text-indent: 0px; text-transform: none;
          white-space: normal; widows: 2; word-spacing: 0px;
          -webkit-text-stroke-width: 0px; background-color: rgb(255,
          255, 255); text-decoration-style: initial;
          text-decoration-color: initial; display: inline !important;
          float: none;"><span> </span>about the -c {concurrency} option:
          if you ask for -c 10, each "page" will consist of 10 parallel
          fetches of URL, and the "latency" will be the amount of time
          it takes to get the last bit from the last concurrent child
          fetch.</span></blockquote>
    </p>
    <p>--dave<br>
    </p>
    <div class="moz-cite-prefix">On 2019-12-19 2:32 p.m., Dave Taht
      wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:CAA93jw6wUSEUPp2GhMHneyFgoznYe=MXPDp7c1bK5+W_Qfoqfw@mail.gmail.com">
      <pre class="moz-quote-pre" wrap="">I was not aware that jim salter had really gone to town on measuring
latency under load in the past year - notably the 4 stream 1024p + web
browsing torture test used here:

<a class="moz-txt-link-freetext" href="https://arstechnica.com/gadgets/2019/11/ars-puts-googles-new-nest-wi-fi-to-the-test/?itm_source=parsely-api">https://arstechnica.com/gadgets/2019/11/ars-puts-googles-new-nest-wi-fi-to-the-test/?itm_source=parsely-api</a>
<a class="moz-txt-link-freetext" href="https://arstechnica.com/gadgets/2019/12/amazons-inexpensive-eero-mesh-wi-fi-kit-is-shockingly-good/?comments=1">https://arstechnica.com/gadgets/2019/12/amazons-inexpensive-eero-mesh-wi-fi-kit-is-shockingly-good/?comments=1</a>

He considers under 500ms of browsing latency to be "good". Not
entirely sure how he's calculating that, I think he's measuring page
completion time rather than "latency" per se'.

The tools he uses are here:

<a class="moz-txt-link-freetext" href="https://github.com/jimsalterjrs/network-testing/blob/master/README.md">https://github.com/jimsalterjrs/network-testing/blob/master/README.md</a>

</pre>
    </blockquote>
    <pre class="moz-signature" cols="72">-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
<a class="moz-txt-link-abbreviated" href="mailto:davecb@spamcop.net">davecb@spamcop.net</a>           |                      -- Mark Twain
</pre>
  </body>
</html>