<?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/"
	xmlns:georss="http://www.georss.org/georss" xmlns:geo="http://www.w3.org/2003/01/geo/wgs84_pos#" xmlns:media="http://search.yahoo.com/mrss/"
		>
<channel>
	<title>Comments on: SIGGRAPH 2008: The quest for more pixels</title>
	<atom:link href="http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/feed/" rel="self" type="application/rss+xml" />
	<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/</link>
	<description>Fresh hacks every day</description>
	<lastBuildDate>Tue, 24 Nov 2009 19:13:48 +0000</lastBuildDate>
	<generator>http://wordpress.com/</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Dan Kaminsky</title>
		<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/comment-page-1/#comment-40832</link>
		<dc:creator>Dan Kaminsky</dc:creator>
		<pubDate>Fri, 22 Aug 2008 03:27:44 +0000</pubDate>
		<guid isPermaLink="false">http://hackaday.iheartcashews.com:8181/2008/08/20/siggraph-2008-the-quest-for-more-pixels/#comment-40832</guid>
		<description>Moore&#039;s Law as a measure of transistor count?  Sure, it&#039;s doing fine.&lt;br&gt;&lt;br&gt;Moore&#039;s Law as a measure of computational power?  Not so much.  Do you realize how many of those transistors are just CPU cache?  L1, L2, L3...more and more and more layers.  Sure, we&#039;re getting more cores, but lets be honest -- nobody really knows how to use &#039;em, not like we knew how to use more clock.  And as we learn to parallelize, we&#039;re abandoning x86 entirely for the 50x speedup.&lt;br&gt;&lt;br&gt;Maybe a better way to say it:  &quot;Moore&#039;s Law is dead. Long live Moore&#039;s law&quot; -- as we finally are forced to migrate off sequential computation to actually experience its advantages.</description>
		<content:encoded><![CDATA[<p>Moore&#8217;s Law as a measure of transistor count?  Sure, it&#8217;s doing fine.</p>
<p>Moore&#8217;s Law as a measure of computational power?  Not so much.  Do you realize how many of those transistors are just CPU cache?  L1, L2, L3&#8230;more and more and more layers.  Sure, we&#8217;re getting more cores, but lets be honest &#8212; nobody really knows how to use &#8216;em, not like we knew how to use more clock.  And as we learn to parallelize, we&#8217;re abandoning x86 entirely for the 50x speedup.</p>
<p>Maybe a better way to say it:  &#8220;Moore&#8217;s Law is dead. Long live Moore&#8217;s law&#8221; &#8212; as we finally are forced to migrate off sequential computation to actually experience its advantages.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Sciurus</title>
		<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/comment-page-1/#comment-40831</link>
		<dc:creator>Sciurus</dc:creator>
		<pubDate>Fri, 22 Aug 2008 00:28:21 +0000</pubDate>
		<guid isPermaLink="false">http://hackaday.iheartcashews.com:8181/2008/08/20/siggraph-2008-the-quest-for-more-pixels/#comment-40831</guid>
		<description>Wow, people are quick with the whole transistor count thing. Moore&#039;s Law is the cat&#039;s pajamas!</description>
		<content:encoded><![CDATA[<p>Wow, people are quick with the whole transistor count thing. Moore&#8217;s Law is the cat&#8217;s pajamas!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Sciurus</title>
		<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/comment-page-1/#comment-40830</link>
		<dc:creator>Sciurus</dc:creator>
		<pubDate>Fri, 22 Aug 2008 00:25:15 +0000</pubDate>
		<guid isPermaLink="false">http://hackaday.iheartcashews.com:8181/2008/08/20/siggraph-2008-the-quest-for-more-pixels/#comment-40830</guid>
		<description>pentium IV: 42,000,000 transistors&lt;br&gt;Core 2 Duo: 291,000,000 transistors&lt;br&gt;I think Moore&#039;s Law is still cookin&#039; with gas.&lt;br&gt;Old slang is the bee&#039;s knees!</description>
		<content:encoded><![CDATA[<p>pentium IV: 42,000,000 transistors<br />Core 2 Duo: 291,000,000 transistors<br />I think Moore&#8217;s Law is still cookin&#8217; with gas.<br />Old slang is the bee&#8217;s knees!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ken</title>
		<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/comment-page-1/#comment-40829</link>
		<dc:creator>Ken</dc:creator>
		<pubDate>Wed, 20 Aug 2008 22:20:00 +0000</pubDate>
		<guid isPermaLink="false">http://hackaday.iheartcashews.com:8181/2008/08/20/siggraph-2008-the-quest-for-more-pixels/#comment-40829</guid>
		<description>moore&#039;s law is, and has always been, about transistor count. NOT core clock frequency. And it&#039;s definitely not dead yet.</description>
		<content:encoded><![CDATA[<p>moore&#8217;s law is, and has always been, about transistor count. NOT core clock frequency. And it&#8217;s definitely not dead yet.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Decepticon</title>
		<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/comment-page-1/#comment-40828</link>
		<dc:creator>Decepticon</dc:creator>
		<pubDate>Wed, 20 Aug 2008 19:29:58 +0000</pubDate>
		<guid isPermaLink="false">http://hackaday.iheartcashews.com:8181/2008/08/20/siggraph-2008-the-quest-for-more-pixels/#comment-40828</guid>
		<description>Wow, these demos are amazing.  Can&#039;t wait to see if they ever get implemented in big box office productions.</description>
		<content:encoded><![CDATA[<p>Wow, these demos are amazing.  Can&#8217;t wait to see if they ever get implemented in big box office productions.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: underdog5004</title>
		<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/comment-page-1/#comment-40827</link>
		<dc:creator>underdog5004</dc:creator>
		<pubDate>Wed, 20 Aug 2008 19:20:48 +0000</pubDate>
		<guid isPermaLink="false">http://hackaday.iheartcashews.com:8181/2008/08/20/siggraph-2008-the-quest-for-more-pixels/#comment-40827</guid>
		<description>Wow, I was really impressed by the &quot;Photo Tourism&quot; section, although I have a feeling it&#039;s not the same as actually being there :)&lt;br&gt;</description>
		<content:encoded><![CDATA[<p>Wow, I was really impressed by the &#8220;Photo Tourism&#8221; section, although I have a feeling it&#8217;s not the same as actually being there :)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mr. Mediocre</title>
		<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/comment-page-1/#comment-40826</link>
		<dc:creator>Mr. Mediocre</dc:creator>
		<pubDate>Wed, 20 Aug 2008 17:31:04 +0000</pubDate>
		<guid isPermaLink="false">http://hackaday.iheartcashews.com:8181/2008/08/20/siggraph-2008-the-quest-for-more-pixels/#comment-40826</guid>
		<description>So many sweet things this year. So many...</description>
		<content:encoded><![CDATA[<p>So many sweet things this year. So many&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: hawkeye18</title>
		<link>http://hackaday.com/2008/08/20/siggraph-2008-the-quest-for-more-pixels/comment-page-1/#comment-40825</link>
		<dc:creator>hawkeye18</dc:creator>
		<pubDate>Wed, 20 Aug 2008 15:19:41 +0000</pubDate>
		<guid isPermaLink="false">http://hackaday.iheartcashews.com:8181/2008/08/20/siggraph-2008-the-quest-for-more-pixels/#comment-40825</guid>
		<description>Great article! One massive, glaring error, though, and it&#039;s one that no tech writer should ever make. It&#039;s regarding your statement about Moore&#039;s law, which I get the distinct impression that you&#039;ve never actually read. Moore didn&#039;t state that processor SPEED would double, but that the number of transistors that can be inexpensively placed on a chip would double about every two years.&lt;br&gt;&lt;br&gt;And in case you haven&#039;t noticed, processors have been growing wider, not taller. Take, say, a Pentium 4 531... 3 ghz, 125 million transistors. It was hot shit about two years ago, the best money could buy.&lt;br&gt;&lt;br&gt;Now, let&#039;s take a Core 2 Extreme, an X6900. Not even the best money can buy nowadays; there are better, but it has a clock speed of 3.2ghz, still better than the 531... and 291 million transistors. That&#039;s 2.3 times more transistors.&lt;br&gt;&lt;br&gt;Seems Moore&#039;s law is still alive and quite well.&lt;br&gt;&lt;br&gt;I do look forward to the 3D exploded view diagrams, though. Those are a bitch to try to figure out in 2D!</description>
		<content:encoded><![CDATA[<p>Great article! One massive, glaring error, though, and it&#8217;s one that no tech writer should ever make. It&#8217;s regarding your statement about Moore&#8217;s law, which I get the distinct impression that you&#8217;ve never actually read. Moore didn&#8217;t state that processor SPEED would double, but that the number of transistors that can be inexpensively placed on a chip would double about every two years.</p>
<p>And in case you haven&#8217;t noticed, processors have been growing wider, not taller. Take, say, a Pentium 4 531&#8230; 3 ghz, 125 million transistors. It was hot shit about two years ago, the best money could buy.</p>
<p>Now, let&#8217;s take a Core 2 Extreme, an X6900. Not even the best money can buy nowadays; there are better, but it has a clock speed of 3.2ghz, still better than the 531&#8230; and 291 million transistors. That&#8217;s 2.3 times more transistors.</p>
<p>Seems Moore&#8217;s law is still alive and quite well.</p>
<p>I do look forward to the 3D exploded view diagrams, though. Those are a bitch to try to figure out in 2D!</p>
]]></content:encoded>
	</item>
</channel>
</rss>
