<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	
	>
<channel>
	<title>
	Comments on: Of Bits &#038; Strings	</title>
	<atom:link href="https://www.taylordesign.net/code-optimization/of-bits-and-strings/feed/" rel="self" type="application/rss+xml" />
	<link>https://www.taylordesign.net/code-optimization/of-bits-and-strings/</link>
	<description>Software Development</description>
	<lastBuildDate>Tue, 04 Dec 2018 22:51:54 +0000</lastBuildDate>
	<sy:updatePeriod>
	hourly	</sy:updatePeriod>
	<sy:updateFrequency>
	1	</sy:updateFrequency>
	<generator>https://wordpress.org/?v=6.4.2</generator>
	<item>
		<title>
		By: Daniel L. Taylor		</title>
		<link>https://www.taylordesign.net/code-optimization/of-bits-and-strings/#comment-53</link>

		<dc:creator><![CDATA[Daniel L. Taylor]]></dc:creator>
		<pubDate>Tue, 04 Dec 2018 22:51:54 +0000</pubDate>
		<guid isPermaLink="false">http://www.taylordesign.net/?p=4066#comment-53</guid>

					<description><![CDATA[In reply to &lt;a href=&quot;https://www.taylordesign.net/code-optimization/of-bits-and-strings/#comment-52&quot;&gt;Cordelia Jarvis&lt;/a&gt;.

I don&#039;t have a high traffic blog by any means. But as I mentioned in this blog post, nobody really noticed the almost identical C solution I found in the forums. I think you hit the nail on the head when you said &quot;...nobody else appreciates this stuff.&quot; Most programmers today are not familiar with what&#039;s actually going on within the CPU. They don&#039;t drop down to C to optimize performance critical code. I think modern software suffers for this.]]></description>
			<content:encoded><![CDATA[<p>In reply to <a href="https://www.taylordesign.net/code-optimization/of-bits-and-strings/#comment-52">Cordelia Jarvis</a>.</p>
<p>I don&#8217;t have a high traffic blog by any means. But as I mentioned in this blog post, nobody really noticed the almost identical C solution I found in the forums. I think you hit the nail on the head when you said &#8220;&#8230;nobody else appreciates this stuff.&#8221; Most programmers today are not familiar with what&#8217;s actually going on within the CPU. They don&#8217;t drop down to C to optimize performance critical code. I think modern software suffers for this.</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		By: Cordelia Jarvis		</title>
		<link>https://www.taylordesign.net/code-optimization/of-bits-and-strings/#comment-52</link>

		<dc:creator><![CDATA[Cordelia Jarvis]]></dc:creator>
		<pubDate>Tue, 04 Dec 2018 18:27:52 +0000</pubDate>
		<guid isPermaLink="false">http://www.taylordesign.net/?p=4066#comment-52</guid>

					<description><![CDATA[Ok... nobody has commented on this? I&#039;d like to say that was a &quot;clever&quot; way of handling the problem. I&#039;m a bit twiddler and perform lots of system level bit transformations on data...so it seemed like a logical path to me.  Of course understanding the architecture of your system (ok...most everything is 64-bit these days) and optimizing your solution for it was smart.  Maybe nobody else appreciates this stuff. We had a long standing solution in our office that took about .2 seconds to run -- However, we found ourselves using this operation 50,000 times in succession.  Now we are at 10,000 seconds.  Last I checked...there&#039;s only 60 in minute. Results took HOURS. I optimized our algorithm (via a lower level language &quot;C&quot;) and got the single response time down to .001 seconds.  This allowed us to finish the workload (50,000 calls) in less than 1 minute. Indeed it can pay to be closer to the machine.]]></description>
			<content:encoded><![CDATA[<p>Ok&#8230; nobody has commented on this? I&#8217;d like to say that was a &#8220;clever&#8221; way of handling the problem. I&#8217;m a bit twiddler and perform lots of system level bit transformations on data&#8230;so it seemed like a logical path to me.  Of course understanding the architecture of your system (ok&#8230;most everything is 64-bit these days) and optimizing your solution for it was smart.  Maybe nobody else appreciates this stuff. We had a long standing solution in our office that took about .2 seconds to run &#8212; However, we found ourselves using this operation 50,000 times in succession.  Now we are at 10,000 seconds.  Last I checked&#8230;there&#8217;s only 60 in minute. Results took HOURS. I optimized our algorithm (via a lower level language &#8220;C&#8221;) and got the single response time down to .001 seconds.  This allowed us to finish the workload (50,000 calls) in less than 1 minute. Indeed it can pay to be closer to the machine.</p>
]]></content:encoded>
		
			</item>
	</channel>
</rss>
