<?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"
	>
<channel>
	<title>Comments on: Rosetta Memory Overhead</title>
	<atom:link href="http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/feed/" rel="self" type="application/rss+xml" />
	<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/</link>
	<description>Working smarter, not harder</description>
	<pubDate>Fri, 21 Nov 2008 20:16:44 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: David</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-13</link>
		<dc:creator>David</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:54:53 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-13</guid>
		<description>I’m not sure how relevant this is:

E.g., if Safari uses an additional 140MB to store the web pages it is displaying (as mine is currently doing) then the additional memory usage is not significant. Starting programs and not doing anything with them shows only the code overhead. If I run photoshop I don’t care if it takes 20 or 60MB for code since I’m going to be using 600MB for my images. The same applies to these programs.

My Mail program is using up 56MB at the moment, so the additional 40MB overhead of running Rosetta appears to be less than 50%.

The interesting issue here is that if that code is now 5x larger it will really hurt I-cache performance, but I guess there’s not much to do about that.</description>
		<content:encoded><![CDATA[<p>I’m not sure how relevant this is:</p>
<p>E.g., if Safari uses an additional 140MB to store the web pages it is displaying (as mine is currently doing) then the additional memory usage is not significant. Starting programs and not doing anything with them shows only the code overhead. If I run photoshop I don’t care if it takes 20 or 60MB for code since I’m going to be using 600MB for my images. The same applies to these programs.</p>
<p>My Mail program is using up 56MB at the moment, so the additional 40MB overhead of running Rosetta appears to be less than 50%.</p>
<p>The interesting issue here is that if that code is now 5x larger it will really hurt I-cache performance, but I guess there’s not much to do about that.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Roland Fines</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-12</link>
		<dc:creator>Roland Fines</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:54:35 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-12</guid>
		<description>I agree that my intel iMac chokes with only 512 Mg RAM. As soon as I launch NetBeans, switching between apps slowly becomes a nightmare. 512Mg RAM is not enough.</description>
		<content:encoded><![CDATA[<p>I agree that my intel iMac chokes with only 512 Mg RAM. As soon as I launch NetBeans, switching between apps slowly becomes a nightmare. 512Mg RAM is not enough.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sean</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-11</link>
		<dc:creator>sean</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:54:15 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-11</guid>
		<description>Bubu,

My theories are:

* There is some extra debug code in there that will eventually be removed as the transition to x86 progresses.

* The Apple engineers have had many years of refining their PowerPC build chain for speed and memory consumption.

* There probably was only a few engineers tasked with keeping the secret build of OS X compiling/working on x86 and memory usage was probably not one of their key concerns.</description>
		<content:encoded><![CDATA[<p>Bubu,</p>
<p>My theories are:</p>
<p>* There is some extra debug code in there that will eventually be removed as the transition to x86 progresses.</p>
<p>* The Apple engineers have had many years of refining their PowerPC build chain for speed and memory consumption.</p>
<p>* There probably was only a few engineers tasked with keeping the secret build of OS X compiling/working on x86 and memory usage was probably not one of their key concerns.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Bubu</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-10</link>
		<dc:creator>Bubu</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:53:54 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-10</guid>
		<description>This time conclusions are OK. I am very surprised about the results; why does the intel consume more memory that the powerPC ? Tx for the test.
</description>
		<content:encoded><![CDATA[<p>This time conclusions are OK. I am very surprised about the results; why does the intel consume more memory that the powerPC ? Tx for the test.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sean</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-9</link>
		<dc:creator>sean</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:49:54 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-9</guid>
		<description>Dono,

We know that iLife ‘06 is a universal binary. We are just wondering why on an Intel iMac you don’t have the option of running it under Rosetta, like you can with OmniOutliner which is also a universal binary.</description>
		<content:encoded><![CDATA[<p>Dono,</p>
<p>We know that iLife ‘06 is a universal binary. We are just wondering why on an Intel iMac you don’t have the option of running it under Rosetta, like you can with OmniOutliner which is also a universal binary.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dono'</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-8</link>
		<dc:creator>Dono'</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:49:37 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-8</guid>
		<description>iLife ‘06 IS already a universal binary.

Dono’</description>
		<content:encoded><![CDATA[<p>iLife ‘06 IS already a universal binary.</p>
<p>Dono’</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jeff</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-7</link>
		<dc:creator>Jeff</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:49:18 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-7</guid>
		<description>iLife 06 is not compatible with G3 systems, so that’s why it won’t run under Rosetta. I don’t know how they made it not do so.</description>
		<content:encoded><![CDATA[<p>iLife 06 is not compatible with G3 systems, so that’s why it won’t run under Rosetta. I don’t know how they made it not do so.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Bubu</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-6</link>
		<dc:creator>Bubu</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:48:53 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-6</guid>
		<description>Conclusions are wrong. The tests compare memory usage of various programs on Intel Macs, with or without Rosetta. They say nothing about RAM usage on current powerPC machines.

edit: They do now.
</description>
		<content:encoded><![CDATA[<p>Conclusions are wrong. The tests compare memory usage of various programs on Intel Macs, with or without Rosetta. They say nothing about RAM usage on current powerPC machines.</p>
<p>edit: They do now.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sean</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-5</link>
		<dc:creator>sean</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:48:33 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-5</guid>
		<description>&lt;p&gt;Dis,&lt;/p&gt;
&lt;p&gt;Thanks for the tip, so I should be able to look in the App bundle and edit the plist to get say iTunes running under Rosetta.&lt;/p&gt;
&lt;p&gt;Will have to try that./p&gt;&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Dis,</p>
<p>Thanks for the tip, so I should be able to look in the App bundle and edit the plist to get say iTunes running under Rosetta.</p>
<p>Will have to try that./p></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dis</title>
		<link>http://incisivegeek.com/2006/01/22/rosetta-memory-overhead/#comment-4</link>
		<dc:creator>Dis</dc:creator>
		<pubDate>Mon, 31 Jul 2006 21:48:15 +0000</pubDate>
		<guid isPermaLink="false">http://incisivegeek.com/2006/01/21/rosetta-memory-overhead/#comment-4</guid>
		<description>&lt;p&gt;From the Universal Binary Programming Guidelines, 2nd ed.:

To prevent an application from opening using Rosetta, add the following key to the Info.plist: &lt;key&gt;LSRequiresNativeExecution&lt;/key&gt; &lt;true/&gt;

This is a possible explanation for the iLife phenomenon you saw.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>From the Universal Binary Programming Guidelines, 2nd ed.:</p>
<p>To prevent an application from opening using Rosetta, add the following key to the Info.plist: <key>LSRequiresNativeExecution</key><br />
<true />
<p>This is a possible explanation for the iLife phenomenon you saw.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
