<?xml version="1.0" encoding="UTF-8"?>
	<rss version="2.0"
		xmlns:content="http://purl.org/rss/1.0/modules/content/"
		xmlns:wfw="http://wellformedweb.org/CommentAPI/"
		xmlns:dc="http://purl.org/dc/elements/1.1/"
		xmlns:atom="http://www.w3.org/2005/Atom"

			>

	<channel>
		<title>Cycling 74  &#187;  Topic: cpu: who&#039;s the killer?</title>
		<atom:link href="http://cycling74.com/forums/topic/cpu-whos-the-killer/feed" rel="self" type="application/rss+xml" />
		<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/feed</link>
		<description></description>
		<pubDate>Tue, 18 Jun 2013 23:42:17 +0000</pubDate>
		<generator>http://bbpress.org/?v=2.2.4</generator>
		<language></language>

		
														
					
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-48592</guid>
					<title><![CDATA[cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-48592</link>
					<pubDate>Sat, 20 Feb 2010 17:15:32 +0000</pubDate>
					<dc:creator>personal_username</dc:creator>

					<description>
						<![CDATA[
						<p>hi all, </p>
<p>i&#8217;m trying to assemble three different patches in one, and i want to use just one main subpatch at once.<br />
You can imagine how important it is to pay attention to CPU process limiting.<br />
Right now i came up with two solutions:<br />
- [mute~] to stop all unwanted dsp activity<br />
- a [toggle] which stops alll unwanted other Max data activity</p>
<p>To be onest &#8211; even if in probing mode i see that msp patchcords are actually freezed (they send static 0 messages) &#8211; I don&#8217;t really see this better improvement.<br />
could it be too many subpatches? should i use more abstractions(i don&#8217;t know if there&#8217;ll be an improvement)?<br />
should i buy a new computer? ^_^ </p>
<p>PS1 cpu utilisation is around 30% BUT with many, many jumps to 80-90%.<br />
PS2 the patch isn&#8217;t that complicated (even if i use 4-5 reverberators and resonators)<br />
PS3 i have a two-year old macbook 2.2 Ghz Core duo2, 3 gb ram 667 Mhz</p>
<p>thanks!</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174741</guid>
					<title><![CDATA[Re: cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174741</link>
					<pubDate>Sat, 20 Feb 2010 17:45:49 +0000</pubDate>
					<dc:creator>Ch</dc:creator>

					<description>
						<![CDATA[
						<p>Hi,</p>
<p>Just in case. if there&#8217;s subpatches in the patcher you&#8217;re muting, be sure you send the list 1 1 or 0 1 to [mute~], as stated in the help file.</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174743</guid>
					<title><![CDATA[Re: cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174743</link>
					<pubDate>Sat, 20 Feb 2010 19:44:02 +0000</pubDate>
					<dc:creator>personal_username</dc:creator>

					<description>
						<![CDATA[
						<p>thanks, already done that!<br />
I just realized I could use matrix!<br />
for istance, for the resonators&#8230; now i have 4.<br />
i could simply put only one, as a matrix element, and route incoming and outgoing signals, no?<br />
what do you think?</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174744</guid>
					<title><![CDATA[Re: cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174744</link>
					<pubDate>Sat, 20 Feb 2010 20:25:49 +0000</pubDate>
					<dc:creator>Chris Muir</dc:creator>

					<description>
						<![CDATA[
						<p>You might also make a copy of your patch, and get rid of as many graphic objects as you can, to see if that helps. The anit-aliased, alpha-composited graphics world of Max 5 is great visually, but expensive. Every time a graphic object has to update, a fair amount of stuff happens.</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174745</guid>
					<title><![CDATA[Re: cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174745</link>
					<pubDate>Sat, 20 Feb 2010 21:30:38 +0000</pubDate>
					<dc:creator>personal_username</dc:creator>

					<description>
						<![CDATA[
						<p><a href='http://cycling74.com/forums/users/chris/' rel='nofollow' class='bbp-mention-link chris'>@chris</a><br />
yes, good tip!<br />
i&#8217;ll do that as last thing after optimizing everything, but i&#8217;m sure it&#8217;ll help!<br />
thx</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174746</guid>
					<title><![CDATA[Re: cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174746</link>
					<pubDate>Wed, 03 Mar 2010 12:48:53 +0000</pubDate>
					<dc:creator>lewis edwards</dc:creator>

					<description>
						<![CDATA[
						<p>using [poly~] to mute unwanted effects or signals is great. i use it for sound effects, and works very well. especially when you get to using more than 4 or 5 at once.</p>
<p>and as chris said. get rid of anything that you dont need. such as expensive graphics. leave those until last&#8230;</p>
<p>&#8212;&#8212;<br />
lewis<br />
&#8212;&#8212;<br />
smokingbunny.co.uk<br />
&#8212;&#8212;</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174747</guid>
					<title><![CDATA[Re: cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174747</link>
					<pubDate>Wed, 03 Mar 2010 16:30:16 +0000</pubDate>
					<dc:creator>Hans Höglund</dc:creator>

					<description>
						<![CDATA[
						<p>The <code>defer</code> object moves activity from the scheduler thread to the main (GUI) thread. If you have an expensive activity that does not require scheduler precision, try to defer it to the main thread.</p>
<p>Overdrive + Scheduler in audio interrupt + UI activity is a known killer as well.</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174748</guid>
					<title><![CDATA[Re: cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174748</link>
					<pubDate>Wed, 03 Mar 2010 16:33:22 +0000</pubDate>
					<dc:creator>Hans Höglund</dc:creator>

					<description>
						<![CDATA[
						<p>Btw, the golden rule of optimization is to <em>focus on the bottlenecks</em>. Try to figure out which objects are causing most trouble.</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174749</guid>
					<title><![CDATA[Re: cpu: who&#039;s the killer?]]></title>
					<link>http://cycling74.com/forums/topic/cpu-whos-the-killer/#post-174749</link>
					<pubDate>Wed, 03 Mar 2010 21:02:02 +0000</pubDate>
					<dc:creator>Joshua Kit Clayton</dc:creator>

					<description>
						<![CDATA[
						<p>Without seeing your patch, it is hard to know what to suggest. However, your description of using resonators and reverberators with CPU performance spikes makes it sound like it may be an issue with denormalized floating point numbers and these feedback based systems on Intel.</p>
<p>Unless an object deals with this in the object code (as many C74 objects do, but perhaps you are using third party objects), you can avoid this issue by feeding a small amount of noise rather than silence to your feedback portion of the patch (resonator or reverberator).</p>
<p>Besides that, Shark can be a useful profiling tool to try and track down tricky performance issues. Once you isolate a clear example of performance problems, you can post here for better help.</p>
						]]>
					</description>

					
					
				</item>

					
		
	</channel>
	</rss>

