legacy_site/cave-update-ore-analysis/index.html

<a
  class="link"
  style="left: 1%; top: 1%; position: absolute"
  href="https://peter-tanner.github.io"
  >🔗 Back to home page</a
><br /><br /><br />

These graphs are interactive - scroll controls the zoom (alternatively, hover
over the bottom-left corner and use the magnifying glass to select an area). Use
the home icon to reset the view.

<div style="text-align: center">
  <h1>Absolute frequency</h1>
  <iframe
    id="inlineFrameExample"
    title="Absolute frequency"
    src="abs_freq.html"
    onload="this.style.height=this.contentWindow.document.body.scrollHeight;this.style.width=this.contentWindow.document.body.scrollWidth;"
  >
  </iframe
  ><br />
  *Note that unfortunately the ore per chunk layer doesn't appear on the right
  y-axis. You can add the right axis by using the formula: y/(2*512/16)**2<br /><br />
  <hr />
  <h1>Relative frequency</h1>
  <iframe
    id="inlineFrameExample"
    title="Relative frequency"
    src="rel_freq.html"
    onload="this.style.height=this.contentWindow.document.body.scrollHeight;this.style.width=this.contentWindow.document.body.scrollWidth;"
  >
  </iframe>
</div>
<br />
<hr />
<br />
This text was copy-pasted from the github repo at
https://github.com/peter-tanner/minecraft-ore-analysis
<h1>Minecraft ore analysis</h1>
<h2>
  Let&#39;s see how the new cave update ore generation compares to the old!
</h2>
<h2>
  Go to
  <a href="https://peter-tanner.github.io/legacy_site/cave-update-ore-analysis"
    >https://peter-tanner.github.io/cave-update-ore-analysis</a
  >
  for interactive graphs!
</h2>
<h3>
  Both tests performed with seed 0 and single biome mode enabled (Extreme hills
  biome only, to include emerald ore statistics). Size of 1024*1024 scanned.
</h3>
<h3>
  Original world files used for research are located at
  <a
    href="https://drive.google.com/drive/folders/1RsmfDp4nl5KaFWfCpL5sanK6cGUp45TO?usp=sharing"
    >https://drive.google.com/drive/folders/1RsmfDp4nl5KaFWfCpL5sanK6cGUp45TO?usp=sharing</a
  >
</h3>
<p>#</p>
<p>
  The scripts I&#39;ve created take advantage of a slightly modified
  <a href="https://github.com/DonoA/PyAnvilEditor">PyAnvilEditor</a> to parse
  the region files.
</p>
<p>
  I&#39;ve replaced the code in the <code>276</code> (and everything else under
  <code>def close(self):</code>) of the <code>world</code> class of
  PyAnvilEditor with <code>true</code> - this is because I&#39;m not writing to
  the world file, I only intend to read from it. This saves a large amount of
  time spent writing that would go to waste.
</p>
<p>
  To generate the graphs I&#39;ve used Matplotlib and I use mpld3 to convert
  these graphs into interactive html files for use on my website.
</p>
<p>
  We load a large section of the world into memory - ideally, you want to set
  the block size to the maximum as this is quicker than loading small sections
  of the world repeatedly. Then we iterate through each coordinate in the block
  - this takes about 90 seconds with a block radius of <code>128</code>. At each
  coordinate we iterate each count for a tile in a particular layer. Once a
  whole block has been processed, we add it to the total.
</p>
<p>
  The &#39;sum&#39; represents the total amount of a particular ore block within
  the whole sample (A 1024*1024 square meter area). The relative frequency
  represents the proportion of the total amount of a particular ore that occurs
  at a particular y-level.
</p>
<p>
  This sample size can be improved on but it would obviously take longer given
  that my RAM is limited.
</p>