<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"
     xmlns:atom="http://www.w3.org/2005/Atom"
     xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>System Expanded</title>
    <link>https://system-expanded.neocities.org/</link>
    <description>Synths, Macintosh restoration, film photography and anime.</description>
    <language>en</language>
    <lastBuildDate>Fri, 28 Aug 2026 23:43:27 GMT</lastBuildDate>
    <atom:link href="https://system-expanded.neocities.org/feed.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>test post</title>
      <link>https://system-expanded.neocities.org/index.html#t</link>
      <guid isPermaLink="false">https://system-expanded.neocities.org/post/t</guid>
      <pubDate>Fri, 28 Aug 2026 04:00:00 GMT</pubDate>
      <category>Synths</category>
      <description>hello</description>
      <content:encoded><![CDATA[<p>here's the body</p>]]></content:encoded>
    </item>
    <item>
      <title>Korg Minilogue XD — Six Months In</title>
      <link>https://system-expanded.neocities.org/index.html#korg-minilogue-xd-six-months-in</link>
      <guid isPermaLink="false">https://system-expanded.neocities.org/post/korg-minilogue-xd-six-months-in</guid>
      <pubDate>Sat, 22 Aug 2026 04:00:00 GMT</pubDate>
      <category>Synths</category>
      <description>A look back at six months with the Minilogue XD and what surprised me most about the patch memory system and the digital multi-engine.</description>
      <content:encoded><![CDATA[
      <p>Six months with the Minilogue XD and I keep returning to the same conclusion: the patch memory system is a feature, not a bug.</p>
      <p>The machine encourages a different relationship with sound. You learn patches rather than saving them. The physical state of the knobs <em>is</em> the patch. When you come back the next day and the filter is somewhere else, that's not loss — it's the next starting point.</p>
      <p>The multi-engine slot is more interesting than it looks on paper. The user oscillator slot especially — the community has built oscillators that genuinely expand what the machine can do: FM operators, wavetable engines, physical modeling. All of these running through the analog filter.</p>
      <p>Six months in, this is still the synthesizer I turn on first.</p>
    ]]></content:encoded>
    </item>
    <item>
      <title>Restoring a Mac SE/30 — Part One</title>
      <link>https://system-expanded.neocities.org/index.html#restoring-a-mac-se-30-part-one</link>
      <guid isPermaLink="false">https://system-expanded.neocities.org/post/restoring-a-mac-se-30-part-one</guid>
      <pubDate>Mon, 17 Aug 2026 04:00:00 GMT</pubDate>
      <category>Mac</category>
      <description>Picked up a yellowed SE/30 at an estate sale. The caps need recapping but the board looks clean. Starting the documentation here.</description>
      <content:encoded><![CDATA[
      <p>Picked up a yellowed SE/30 at an estate sale in Pasadena. The seller didn't know if it worked. Neither did I.</p>
      <p>First thing: the caps. Almost every SE/30 board at this point has some degree of cap leakage — the original ELNA SMD capacitors have been failing for thirty years, and the electrolyte corrodes traces and pads slowly, usually without visible symptoms until something stops working. The board looked clean to my eye, but I'll rework all the caps before I power it on.</p>
      <p>Cap kit ordered. Will document the rework process in Part Two.</p>
      <p>What I'm hoping for: a clean recap, a working board, and enough of the original plastics to justify the yellowing removal process. The case is quite oxidized. Retr0bright or an H2O2 cream is the standard approach — I'll write about that separately.</p>
    ]]></content:encoded>
    </item>
    <item>
      <title>Rewatching Planetes in 2026</title>
      <link>https://system-expanded.neocities.org/index.html#rewatching-planetes-in-2026</link>
      <guid isPermaLink="false">https://system-expanded.neocities.org/post/rewatching-planetes-in-2026</guid>
      <pubDate>Mon, 10 Aug 2026 04:00:00 GMT</pubDate>
      <category>Anime</category>
      <description>The show holds up better than I expected. Hard science and softer themes about ambition and compromise — still relevant, still moving.</description>
      <content:encoded><![CDATA[
      <p>I watched Planetes when it first came out and remembered it as a solid hard-SF series. Rewatching in 2026, it reads differently — sharper, and in places more unsettling.</p>
      <p>The series is interested in work — the indignity and meaning of it, the way ambition curdles or clarifies under pressure. Hachimaki's arc is not a triumph arc. He becomes someone harder and less pleasant to be around, and the series doesn't pretend otherwise. His transformation is a cost, not a reward.</p>
      <p>The political material — terrorism, nationalism, a UN that functions as a site of compromise rather than virtue — lands differently now. The show is not naive about institutions, and it doesn't resolve the tension cleanly.</p>
      <p>Still one of the better science fiction series in the medium. The slower episodes in the middle third earn what comes after them.</p>
    ]]></content:encoded>
    </item>
    <item>
      <title>First Roll on the Canonet QL17</title>
      <link>https://system-expanded.neocities.org/index.html#first-roll-on-the-canonet-ql17</link>
      <guid isPermaLink="false">https://system-expanded.neocities.org/post/first-roll-on-the-canonet-ql17</guid>
      <pubDate>Mon, 03 Aug 2026 04:00:00 GMT</pubDate>
      <category>Photos</category>
      <description>Shot a roll of HP5+ through the Canonet I found at a thrift store. Rangefinder focus took some adjustment. Results inside.</description>
      <content:encoded><![CDATA[
      <p>Found the Canonet at a thrift store in Silver Lake for twelve dollars. The selenium meter was dead (expected) and the viewfinder had some haze, but the lens looked clean and the shutter sounded right at multiple speeds.</p>
      <p>Shot a roll of Ilford HP5+ at box speed, metering by eye. The 40mm lens is genuinely good — sharp through the center wide open, uniform across the frame by f/4. The rangefinder patch took a few frames to trust, then it was fast.</p>
      <p>The light seals need replacing. A few frames at the beginning and end of the roll show fogging. Everything in the middle came out clean.</p>
      <p>Developed in HC-110, dilution B. Scanned on a flatbed. The results are on the photos page.</p>
    ]]></content:encoded>
    </item>
    <item>
      <title>On Patch Memory and Happy Accidents</title>
      <link>https://system-expanded.neocities.org/index.html#on-patch-memory-and-happy-accidents</link>
      <guid isPermaLink="false">https://system-expanded.neocities.org/post/on-patch-memory-and-happy-accidents</guid>
      <pubDate>Tue, 28 Jul 2026 04:00:00 GMT</pubDate>
      <category>Synths</category>
      <description>Arguing for semi-modular over full patch recall — the case for impermanence in sound design.</description>
      <content:encoded><![CDATA[
      <p>The argument for full patch recall is obvious: you find something good, you save it, you come back to it. The argument against it is less obvious, but I think it's real.</p>
      <p>Full recall encourages a museum attitude toward sound. You find something good and preserve it rather than developing it. The saved patch becomes a destination rather than a starting point.</p>
      <p>Semi-modular instruments without patch memory — the Minilogue XD in its default mode, the Grandmother, the early Moogs — require you to either write down settings or accept that you're working in impermanence. That impermanence has a cost (you lose things) and a benefit (you keep moving).</p>
      <p>The happy accidents argument: when you reset a semi-modular and start from a different place, you find things you wouldn't have gone looking for. The loss of the previous patch is the condition for discovering the next one.</p>
      <p>I'm not against patch memory. I'm arguing for leaving it turned off sometimes.</p>
    ]]></content:encoded>
    </item>
  </channel>
</rss>
