<?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: udp in flash</title>
	<atom:link href="http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/feed/" rel="self" type="application/rss+xml" />
	<link>http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/</link>
	<description>and still for good reason.</description>
	<lastBuildDate>Mon, 05 Jul 2010 21:26:02 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.2</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Ammon</title>
		<link>http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/comment-page-1/#comment-166268</link>
		<dc:creator>Ammon</dc:creator>
		<pubDate>Wed, 22 Apr 2009 01:13:36 +0000</pubDate>
		<guid isPermaLink="false">http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/#comment-166268</guid>
		<description>Unfortunately, Flash 10&#039;s new &quot;UDP&quot; support is even more restricted than TCP. It is only for VoIP. :(</description>
		<content:encoded><![CDATA[<p>Unfortunately, Flash 10&#8217;s new &#8220;UDP&#8221; support is even more restricted than TCP. It is only for VoIP. <img src='http://ammonlauritzen.com/blog/wp-includes/images/smilies/icon_sad.gif' alt=':(' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Norgg</title>
		<link>http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/comment-page-1/#comment-165772</link>
		<dc:creator>Norgg</dc:creator>
		<pubDate>Fri, 17 Apr 2009 15:54:22 +0000</pubDate>
		<guid isPermaLink="false">http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/#comment-165772</guid>
		<description>With a signed (even self-signed) java applet, you can get full privileges to do whatever you want, including opening a socket to anywhere, but means your user has to accept this.

You can communicate with both java applets and flash with javascript, you could maybe use this as the communication bridge.

Looks like there&#039;s be UDP in Flash 10 anyway though.</description>
		<content:encoded><![CDATA[<p>With a signed (even self-signed) java applet, you can get full privileges to do whatever you want, including opening a socket to anywhere, but means your user has to accept this.</p>
<p>You can communicate with both java applets and flash with javascript, you could maybe use this as the communication bridge.</p>
<p>Looks like there&#8217;s be UDP in Flash 10 anyway though.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ammon</title>
		<link>http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/comment-page-1/#comment-34918</link>
		<dc:creator>Ammon</dc:creator>
		<pubDate>Sat, 04 Aug 2007 10:56:27 +0000</pubDate>
		<guid isPermaLink="false">http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/#comment-34918</guid>
		<description>Asynchronous TCP sockets could probably give at least &lt;i&gt;different&lt;/i&gt; performance than a single one - but I don&#039;t know if that difference would be measurable.

Part of the bottleneck is probably just how TCP stacks are implemented. Most machines don&#039;t send your packet out quite instantly, they let it sit in a buffer first. Add TCP&#039;s error control elements on top of things, and you&#039;ve got a lot of potential for minute delays in excess of what it actually takes to send the data between client and server.

Opening multiple sockets might give some sort of strange performance increase, but it also complicates things a great deal and ties up more system resources. Plus, there&#039;s always the chance that going to all of that extra effort provides no real performance gain at all, since the packets quite possibly just sit in the same TCP buffers.

Also wrt using Zinc for the proxy... it&#039;s not even as good as it sounds initially. To get any decent coverage, you&#039;d have to buy what amounts to essentially two separate licenses if you want to compile projectors for both Windows and OSX. There &lt;u&gt;is&lt;/u&gt; no Linux option, and heaven forbid you wanted to run something like that on an embedded system ;)</description>
		<content:encoded><![CDATA[<p>Asynchronous TCP sockets could probably give at least <i>different</i> performance than a single one &#8211; but I don&#8217;t know if that difference would be measurable.</p>
<p>Part of the bottleneck is probably just how TCP stacks are implemented. Most machines don&#8217;t send your packet out quite instantly, they let it sit in a buffer first. Add TCP&#8217;s error control elements on top of things, and you&#8217;ve got a lot of potential for minute delays in excess of what it actually takes to send the data between client and server.</p>
<p>Opening multiple sockets might give some sort of strange performance increase, but it also complicates things a great deal and ties up more system resources. Plus, there&#8217;s always the chance that going to all of that extra effort provides no real performance gain at all, since the packets quite possibly just sit in the same TCP buffers.</p>
<p>Also wrt using Zinc for the proxy&#8230; it&#8217;s not even as good as it sounds initially. To get any decent coverage, you&#8217;d have to buy what amounts to essentially two separate licenses if you want to compile projectors for both Windows and OSX. There <u>is</u> no Linux option, and heaven forbid you wanted to run something like that on an embedded system <img src='http://ammonlauritzen.com/blog/wp-includes/images/smilies/icon_wink.gif' alt=';)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jlGauthier</title>
		<link>http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/comment-page-1/#comment-34838</link>
		<dc:creator>jlGauthier</dc:creator>
		<pubDate>Fri, 03 Aug 2007 16:25:11 +0000</pubDate>
		<guid isPermaLink="false">http://ammonlauritzen.com/blog/2007/07/16/udp-in-flash/#comment-34838</guid>
		<description>hrmf.
I want to use an UDP server, and have run into the same set of problems.. and came up with many of the same not so great solutions.

The zinc solution you have there might be the best... but ich. 

Resigning I am thinking for speed over tcp you could open multiple tcps and use them asyncronously? do you know if that would work?

james</description>
		<content:encoded><![CDATA[<p>hrmf.<br />
I want to use an UDP server, and have run into the same set of problems.. and came up with many of the same not so great solutions.</p>
<p>The zinc solution you have there might be the best&#8230; but ich. </p>
<p>Resigning I am thinking for speed over tcp you could open multiple tcps and use them asyncronously? do you know if that would work?</p>
<p>james</p>
]]></content:encoded>
	</item>
</channel>
</rss>
