<?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: Degrading qemu performance in DooM Part II	</title>
	<atom:link href="https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/feed/" rel="self" type="application/rss+xml" />
	<link>https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/</link>
	<description>Fun with Virtualization</description>
	<lastBuildDate>Fri, 26 Jul 2019 10:13:36 +0000</lastBuildDate>
	<sy:updatePeriod>
	hourly	</sy:updatePeriod>
	<sy:updateFrequency>
	1	</sy:updateFrequency>
	<generator>https://wordpress.org/?v=7.0.1</generator>
	<item>
		<title>
		By: neozeed		</title>
		<link>https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-217748</link>

		<dc:creator><![CDATA[neozeed]]></dc:creator>
		<pubDate>Fri, 26 Jul 2019 10:13:36 +0000</pubDate>
		<guid isPermaLink="false">https://virtuallyfun.com/wordpress/?p=9280#comment-217748</guid>

					<description><![CDATA[In reply to &lt;a href=&quot;https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-217659&quot;&gt;Martin Buriánek&lt;/a&gt;.

It was standard versions of qemu.  The older stuff is more responsive on Windows... That&#039;s all.

I&#039;ve been experimenting on and off here:
https://sourceforge.net/projects/qemu090b/

The most interesting thing was Mpu401 Io support... If you like DOS MIDI.

https://sourceforge.net/p/crossdjgppv1/DooM/ci/master/tree/

This is an experiment with doom, getting it to build with various compilers etc.. 

The main issue at hand is the watcom retail doom has self modifying code that kills the newer qemu dynamic translations, although the newer ones fare much better, the graphics Io seems to lag..  Maybe it&#039;s my perception I haven&#039;t measured]]></description>
			<content:encoded><![CDATA[<p>In reply to <a href="https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-217659">Martin Buriánek</a>.</p>
<p>It was standard versions of qemu.  The older stuff is more responsive on Windows&#8230; That&#8217;s all.</p>
<p>I&#8217;ve been experimenting on and off here:<br />
<a href="https://sourceforge.net/projects/qemu090b/" rel="nofollow ugc">https://sourceforge.net/projects/qemu090b/</a></p>
<p>The most interesting thing was Mpu401 Io support&#8230; If you like DOS MIDI.</p>
<p><a href="https://sourceforge.net/p/crossdjgppv1/DooM/ci/master/tree/" rel="nofollow ugc">https://sourceforge.net/p/crossdjgppv1/DooM/ci/master/tree/</a></p>
<p>This is an experiment with doom, getting it to build with various compilers etc.. </p>
<p>The main issue at hand is the watcom retail doom has self modifying code that kills the newer qemu dynamic translations, although the newer ones fare much better, the graphics Io seems to lag..  Maybe it&#8217;s my perception I haven&#8217;t measured</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		By: Martin Buriánek		</title>
		<link>https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-217659</link>

		<dc:creator><![CDATA[Martin Buriánek]]></dc:creator>
		<pubDate>Thu, 25 Jul 2019 19:30:55 +0000</pubDate>
		<guid isPermaLink="false">https://virtuallyfun.com/wordpress/?p=9280#comment-217659</guid>

					<description><![CDATA[In reply to &lt;a href=&quot;https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-209979&quot;&gt;neozeed&lt;/a&gt;.

Hi, nice projekt!!. Please can you describe it more? So you modified the source for Doom and compiled, or did you make some modifications to the new versions of QEMU, and that&#039;s why it&#039;s running well now? If you have made changes to QEMU, please describe it or post diff for version 20190218 thank you in advance]]></description>
			<content:encoded><![CDATA[<p>In reply to <a href="https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-209979">neozeed</a>.</p>
<p>Hi, nice projekt!!. Please can you describe it more? So you modified the source for Doom and compiled, or did you make some modifications to the new versions of QEMU, and that&#8217;s why it&#8217;s running well now? If you have made changes to QEMU, please describe it or post diff for version 20190218 thank you in advance</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		By: neozeed		</title>
		<link>https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-209979</link>

		<dc:creator><![CDATA[neozeed]]></dc:creator>
		<pubDate>Mon, 18 Mar 2019 23:47:48 +0000</pubDate>
		<guid isPermaLink="false">https://virtuallyfun.com/wordpress/?p=9280#comment-209979</guid>

					<description><![CDATA[In reply to &lt;a href=&quot;https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-209970&quot;&gt;Stefan Ring&lt;/a&gt;.

I&#039;m using a 13 year old machine, so I&#039;m sure you have something faster.

Also this is part 2, &lt;a href=&quot;https://virtuallyfun.com/wordpress/2019/03/05/degrading-qemu-performance-in-doom/&quot;&gt;did you see part 1&lt;/a&gt;?

The whole point was Watcom/C and the  self-modifying code vs using a GCC build with no self modifying code, along with performance timings of an ASM version of FixedDiv/FixedMul vs using a C version.]]></description>
			<content:encoded><![CDATA[<p>In reply to <a href="https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-209970">Stefan Ring</a>.</p>
<p>I&#8217;m using a 13 year old machine, so I&#8217;m sure you have something faster.</p>
<p>Also this is part 2, <a href="https://virtuallyfun.com/wordpress/2019/03/05/degrading-qemu-performance-in-doom/">did you see part 1</a>?</p>
<p>The whole point was Watcom/C and the  self-modifying code vs using a GCC build with no self modifying code, along with performance timings of an ASM version of FixedDiv/FixedMul vs using a C version.</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		By: Stefan Ring		</title>
		<link>https://virtuallyfun.com/2019/03/15/degrading-qemu-performance-in-doom-part-ii/comment-page-1/#comment-209970</link>

		<dc:creator><![CDATA[Stefan Ring]]></dc:creator>
		<pubDate>Mon, 18 Mar 2019 20:39:06 +0000</pubDate>
		<guid isPermaLink="false">https://virtuallyfun.com/wordpress/?p=9280#comment-209970</guid>

					<description><![CDATA[How is this degrading? It has been constantly improving since 0.11.0. Also, the prebuilt ones are much faster than your own builds, for whatever reason.]]></description>
			<content:encoded><![CDATA[<p>How is this degrading? It has been constantly improving since 0.11.0. Also, the prebuilt ones are much faster than your own builds, for whatever reason.</p>
]]></content:encoded>
		
			</item>
	</channel>
</rss>
