<?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: 16 Core Chips and You!</title>
	<atom:link href="http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/</link>
	<description>General interest observations and true web-log.</description>
	<lastBuildDate>Sun, 12 Feb 2012 11:25:34 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: ECA</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1516005</link>
		<dc:creator>ECA</dc:creator>
		<pubDate>Sat, 25 Apr 2009 06:29:27 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1516005</guid>
		<description>tHE BREAK THRU ISNT going TO SERIAL ISTS LEAVING serial..
we are only doing it 1 step at a time..

WHEN you can show a MOVIE on your PC and take 0 clock cycles, I will believe what you are saying..
When your system isnt as SLOW as the slowest operating HARDWARE...I will believe you.  Turn your 300baud modem on and WATCH windows CROAK..

WHEn you need to run something that the Video/audio card can do, and NOT need to access DX or openGL that RUN under OS and OFF the CPU...Talk to me.</description>
		<content:encoded><![CDATA[<p>tHE BREAK THRU ISNT going TO SERIAL ISTS LEAVING serial..<br />
we are only doing it 1 step at a time..</p>
<p>WHEN you can show a MOVIE on your PC and take 0 clock cycles, I will believe what you are saying..<br />
When your system isnt as SLOW as the slowest operating HARDWARE&#8230;I will believe you.  Turn your 300baud modem on and WATCH windows CROAK..</p>
<p>WHEn you need to run something that the Video/audio card can do, and NOT need to access DX or openGL that RUN under OS and OFF the CPU&#8230;Talk to me.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: JoaoPT</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515616</link>
		<dc:creator>JoaoPT</dc:creator>
		<pubDate>Fri, 24 Apr 2009 10:54:11 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515616</guid>
		<description>12 Eca

We had this talk before, but... here we go:

&lt;i&gt;&quot;Around 10 years ago, INTEL wanted to go to parallel processing with the x86 chip set..
MS said NO..&quot;&lt;/i&gt;

X86 is somewhat parallel since the inception of the original Pentium, 16 years ago, the first x86cpu with a superscalar design (ie. more than one instructions dispatched in parallel in diverse execution units...)

&lt;i&gt;&quot;What would you do IF’ a card(video/audio/network/..) had access to its OWN share of ram and direct access to the HD, without dealing with the CPU?
When you DL a program off the net it goes directly to the HD, and the CPU does nothing.&quot;&lt;/i&gt;

Already does that, it&#039;s called DMA and Bus mastering. Of course CPU is minimally involved (after all the OS has to control it and the OS runs on the CPU...).

&lt;i&gt;&quot;the CPU finds a video file, SHIPS it to the VIDEO card to do the processing, insted of the CPU doing the work. The Audio and video card and setup there OWN linking directly, insted of running thru the CPU for clock cycles..FASTER video and LESS lip sync problems..&quot;&lt;/i&gt;

Things are not so linear, since tasks are complex, and evolve over time (thus negating the advantage of an all hardware solution), but OpenGl and DirectX offload 3d and video decompressing and motion estimation to the graphics card.

&lt;i&gt;&quot;Nice to see you know a few other OS’s..
Did you know that Linux can run a DVD MOVIE Quality on a P3 800mhz?
And MS requires at least a P3 1700mhz??
And do you think your DVD player has a P3 built into it??
As well as that Linux and others can be run in MANY other environments then a SERIAL processed CPU controlled environment.. Go look at the Amiga and BOs..BOTH Linux based..&quot;&lt;/i&gt;

That&#039;s not true altogether...I still remember running fullscreen video on a P3 450mhz (of course it was win98...XP is much more impacting on video performance).
Neither the Amiga nor BeOS had their roots on Unix. (and certainly not Linux, because Linus started working on what later became Linux, as a school project the summer of 91, later than BeOs and AmigaOS...)

&lt;i&gt;&quot;You mention CURRENT tech working with the MMU..IT TOOK THEM THIS LONG to fix this problem?? and the Amiga had a hardware MMU HOW LONG AGO?? Over 20 years ago.&quot;&lt;/i&gt;

MMU is old technology. The breakthrough is moving to serial, and that simply could not be done on slow hardware (do imagine the clock frequency of an HyperTransport stream? 1000Mhz for the original Athlon64, and its 3.2Ghz on the Corei7. 68040s used to run at 25Mhz doubled internally).

&lt;i&gt;&quot;So, you are agreeing that we are working with tech thats been surpassed over 20 years ago?&quot;&lt;/i&gt;

And what&#039;s wrong with that? You drive everyday on almost 150 years old tech (the internal combustion engine)...</description>
		<content:encoded><![CDATA[<p>12 Eca</p>
<p>We had this talk before, but&#8230; here we go:</p>
<p><i>&#8220;Around 10 years ago, INTEL wanted to go to parallel processing with the x86 chip set..<br />
MS said NO..&#8221;</i></p>
<p>X86 is somewhat parallel since the inception of the original Pentium, 16 years ago, the first x86cpu with a superscalar design (ie. more than one instructions dispatched in parallel in diverse execution units&#8230;)</p>
<p><i>&#8220;What would you do IF’ a card(video/audio/network/..) had access to its OWN share of ram and direct access to the HD, without dealing with the CPU?<br />
When you DL a program off the net it goes directly to the HD, and the CPU does nothing.&#8221;</i></p>
<p>Already does that, it&#8217;s called DMA and Bus mastering. Of course CPU is minimally involved (after all the OS has to control it and the OS runs on the CPU&#8230;).</p>
<p><i>&#8220;the CPU finds a video file, SHIPS it to the VIDEO card to do the processing, insted of the CPU doing the work. The Audio and video card and setup there OWN linking directly, insted of running thru the CPU for clock cycles..FASTER video and LESS lip sync problems..&#8221;</i></p>
<p>Things are not so linear, since tasks are complex, and evolve over time (thus negating the advantage of an all hardware solution), but OpenGl and DirectX offload 3d and video decompressing and motion estimation to the graphics card.</p>
<p><i>&#8220;Nice to see you know a few other OS’s..<br />
Did you know that Linux can run a DVD MOVIE Quality on a P3 800mhz?<br />
And MS requires at least a P3 1700mhz??<br />
And do you think your DVD player has a P3 built into it??<br />
As well as that Linux and others can be run in MANY other environments then a SERIAL processed CPU controlled environment.. Go look at the Amiga and BOs..BOTH Linux based..&#8221;</i></p>
<p>That&#8217;s not true altogether&#8230;I still remember running fullscreen video on a P3 450mhz (of course it was win98&#8230;XP is much more impacting on video performance).<br />
Neither the Amiga nor BeOS had their roots on Unix. (and certainly not Linux, because Linus started working on what later became Linux, as a school project the summer of 91, later than BeOs and AmigaOS&#8230;)</p>
<p><i>&#8220;You mention CURRENT tech working with the MMU..IT TOOK THEM THIS LONG to fix this problem?? and the Amiga had a hardware MMU HOW LONG AGO?? Over 20 years ago.&#8221;</i></p>
<p>MMU is old technology. The breakthrough is moving to serial, and that simply could not be done on slow hardware (do imagine the clock frequency of an HyperTransport stream? 1000Mhz for the original Athlon64, and its 3.2Ghz on the Corei7. 68040s used to run at 25Mhz doubled internally).</p>
<p><i>&#8220;So, you are agreeing that we are working with tech thats been surpassed over 20 years ago?&#8221;</i></p>
<p>And what&#8217;s wrong with that? You drive everyday on almost 150 years old tech (the internal combustion engine)&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ECA</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515568</link>
		<dc:creator>ECA</dc:creator>
		<pubDate>Fri, 24 Apr 2009 00:35:54 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515568</guid>
		<description>10,
Around 10 years ago, INTEL wanted to go to parallel processing with the x86 chip set..
MS said NO..

What would you do IF&#039; a card(video/audio/network/..) had access to its OWN share of ram and direct access to the HD, without dealing with the CPU?
When you DL a program off the net it goes directly to the HD, and the CPU does nothing.
the CPU finds a video file, SHIPS it to the VIDEO card to do the processing, insted of the CPU doing the work.  The Audio and video card and setup there OWN linking directly, insted of running thru the CPU for clock cycles..FASTER video and LESS lip sync problems..

Nice to see you know a few other OS&#039;s..
Did you know that Linux can run a DVD MOVIE Quality on a P3 800mhz?
And MS requires at least a P3 1700mhz??
And do you think your DVD player has a P3 built into it??
As well as that Linux and others can be run in MANY other environments then a SERIAL processed CPU controlled environment..  Go look at the Amiga and BOs..BOTH Linux based..

You mention CURRENT tech working with the MMU..IT TOOK THEM THIS LONG to fix this problem??  and the Amiga had a hardware MMU HOW LONG AGO??  Over 20 years ago.

So, you are agreeing that we are working with tech thats been surpassed over 20 years ago?</description>
		<content:encoded><![CDATA[<p>10,<br />
Around 10 years ago, INTEL wanted to go to parallel processing with the x86 chip set..<br />
MS said NO..</p>
<p>What would you do IF&#8217; a card(video/audio/network/..) had access to its OWN share of ram and direct access to the HD, without dealing with the CPU?<br />
When you DL a program off the net it goes directly to the HD, and the CPU does nothing.<br />
the CPU finds a video file, SHIPS it to the VIDEO card to do the processing, insted of the CPU doing the work.  The Audio and video card and setup there OWN linking directly, insted of running thru the CPU for clock cycles..FASTER video and LESS lip sync problems..</p>
<p>Nice to see you know a few other OS&#8217;s..<br />
Did you know that Linux can run a DVD MOVIE Quality on a P3 800mhz?<br />
And MS requires at least a P3 1700mhz??<br />
And do you think your DVD player has a P3 built into it??<br />
As well as that Linux and others can be run in MANY other environments then a SERIAL processed CPU controlled environment..  Go look at the Amiga and BOs..BOTH Linux based..</p>
<p>You mention CURRENT tech working with the MMU..IT TOOK THEM THIS LONG to fix this problem??  and the Amiga had a hardware MMU HOW LONG AGO??  Over 20 years ago.</p>
<p>So, you are agreeing that we are working with tech thats been surpassed over 20 years ago?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Angel H. Wong</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515556</link>
		<dc:creator>Angel H. Wong</dc:creator>
		<pubDate>Thu, 23 Apr 2009 23:21:36 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515556</guid>
		<description>Wow, ten posts and no Machead doing the &quot;That&#039;s the minimal requirement for Vista&quot; joke.</description>
		<content:encoded><![CDATA[<p>Wow, ten posts and no Machead doing the &#8220;That&#8217;s the minimal requirement for Vista&#8221; joke.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: JoaoPT</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515313</link>
		<dc:creator>JoaoPT</dc:creator>
		<pubDate>Thu, 23 Apr 2009 10:24:38 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515313</guid>
		<description>Funny ECA, that hardware is the same Mac OS builds upon, and Linux, and several flavours of UNIX (solaris, irix, BSD...) So the &quot;MS limits intel&quot; argument is moot...
Also funny, the main selling point of AMD and recently intel&#039;s core i7 is exactly memory management...they both went serial and inside chip...</description>
		<content:encoded><![CDATA[<p>Funny ECA, that hardware is the same Mac OS builds upon, and Linux, and several flavours of UNIX (solaris, irix, BSD&#8230;) So the &#8220;MS limits intel&#8221; argument is moot&#8230;<br />
Also funny, the main selling point of AMD and recently intel&#8217;s core i7 is exactly memory management&#8230;they both went serial and inside chip&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Glenn E.</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515300</link>
		<dc:creator>Glenn E.</dc:creator>
		<pubDate>Thu, 23 Apr 2009 08:34:45 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515300</guid>
		<description>&quot;Congress looking into broadband pricing...&quot;?

Yeah, just like the way they &quot;looked into&quot; cable tv rates. NOT!  I guess Congress just realized they were running short of lobbyist dollars. So once more they&#039;ve threatened to &quot;look into&quot; some monopoly&#039;s price scheme. With the implied threat that they might regulate it. Nothing will bring the lobbyists running with their attache cases stuffed with presents (and cash) and vacation getaways to tropical resorts, like the sound of a regulatory bill in the wind. But don&#039;t expect Congress to actually do anything effected for the general public. This in only a scheme to elicit funds and favors. As Congress almost always does. And if they work the bill right, it could even raise broadband rates. It will most certainly kill any chance of &quot;free broadband&quot; existing anywhere.</description>
		<content:encoded><![CDATA[<p>&#8220;Congress looking into broadband pricing&#8230;&#8221;?</p>
<p>Yeah, just like the way they &#8220;looked into&#8221; cable tv rates. NOT!  I guess Congress just realized they were running short of lobbyist dollars. So once more they&#8217;ve threatened to &#8220;look into&#8221; some monopoly&#8217;s price scheme. With the implied threat that they might regulate it. Nothing will bring the lobbyists running with their attache cases stuffed with presents (and cash) and vacation getaways to tropical resorts, like the sound of a regulatory bill in the wind. But don&#8217;t expect Congress to actually do anything effected for the general public. This in only a scheme to elicit funds and favors. As Congress almost always does. And if they work the bill right, it could even raise broadband rates. It will most certainly kill any chance of &#8220;free broadband&#8221; existing anywhere.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Uncle Patso</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515297</link>
		<dc:creator>Uncle Patso</dc:creator>
		<pubDate>Thu, 23 Apr 2009 08:01:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515297</guid>
		<description>Off topic alert!

In no agenda 090, about that $ 51.50 Johnny Green sent in: wasn&#039;t IBM&#039;s very first PC model # 5150?</description>
		<content:encoded><![CDATA[<p>Off topic alert!</p>
<p>In no agenda 090, about that $ 51.50 Johnny Green sent in: wasn&#8217;t IBM&#8217;s very first PC model # 5150?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ECA</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515296</link>
		<dc:creator>ECA</dc:creator>
		<pubDate>Thu, 23 Apr 2009 07:52:58 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515296</guid>
		<description>PS,
and 90% of the problem comes with the software MMU that cant handle ALL this without getting confused..</description>
		<content:encoded><![CDATA[<p>PS,<br />
and 90% of the problem comes with the software MMU that cant handle ALL this without getting confused..</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ECA</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515295</link>
		<dc:creator>ECA</dc:creator>
		<pubDate>Thu, 23 Apr 2009 07:49:39 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515295</guid>
		<description>5,
So, the BASE OS and programming created by MS, is partly at fault...
THEN the hardware WHICH MS requires INTEL to build</description>
		<content:encoded><![CDATA[<p>5,<br />
So, the BASE OS and programming created by MS, is partly at fault&#8230;<br />
THEN the hardware WHICH MS requires INTEL to build</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: joaoPT</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515291</link>
		<dc:creator>joaoPT</dc:creator>
		<pubDate>Thu, 23 Apr 2009 07:37:35 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515291</guid>
		<description>Well, basically the troubles with multicore computers is code and bus. The code has to be written to be parallel (ie. every task is subdivided into concurrent tasks) that is challenging for a number of reasons, but notably because there are tasks that are intrinsically sequential. So there&#039;s so much you can gain by parallelizing. 
The other bottleneck is access to memory, resources and interprocess comunication. As more cores you put to the task, the more they step on each other&#039;s toes...</description>
		<content:encoded><![CDATA[<p>Well, basically the troubles with multicore computers is code and bus. The code has to be written to be parallel (ie. every task is subdivided into concurrent tasks) that is challenging for a number of reasons, but notably because there are tasks that are intrinsically sequential. So there&#8217;s so much you can gain by parallelizing.<br />
The other bottleneck is access to memory, resources and interprocess comunication. As more cores you put to the task, the more they step on each other&#8217;s toes&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ECA</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515289</link>
		<dc:creator>ECA</dc:creator>
		<pubDate>Thu, 23 Apr 2009 07:15:58 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515289</guid>
		<description>as to computer math and such...
using 0,1..in 8 bit
1,2,4,8,16,32,64,128..
Values and colors..
Adding each to those before it..
8+4+2+1=15 which leads to the next number 16
Adding 8bits makes 1 byte, with a value 128+ all the lower numbers 255..OR (128x2)-1..which is REALLY 0-255 which is 256 numerals..equal to 256 colors..(8bit color)
same with the rest of it..
16bit color..

You are correct about the groupings you mentioned.. 4x5 QUAD core.. 80 cores.
IF each is run consequentially..as in sharing processing in a Beowulf config the power is enormous..BUT MS dont do that..ITS COPYRIGHTED..

Im not up on current windows, but windows has ALWAYS had a problem with sharing, ubtil Win NT..THEn they changed the configuration of RAM and HOW multiple CORES were used and run with Duo and QUAD cores...windows is still having a few problems..even with an environment THEY CREATED..</description>
		<content:encoded><![CDATA[<p>as to computer math and such&#8230;<br />
using 0,1..in 8 bit<br />
1,2,4,8,16,32,64,128..<br />
Values and colors..<br />
Adding each to those before it..<br />
8+4+2+1=15 which leads to the next number 16<br />
Adding 8bits makes 1 byte, with a value 128+ all the lower numbers 255..OR (128&#215;2)-1..which is REALLY 0-255 which is 256 numerals..equal to 256 colors..(8bit color)<br />
same with the rest of it..<br />
16bit color..</p>
<p>You are correct about the groupings you mentioned.. 4&#215;5 QUAD core.. 80 cores.<br />
IF each is run consequentially..as in sharing processing in a Beowulf config the power is enormous..BUT MS dont do that..ITS COPYRIGHTED..</p>
<p>Im not up on current windows, but windows has ALWAYS had a problem with sharing, ubtil Win NT..THEn they changed the configuration of RAM and HOW multiple CORES were used and run with Duo and QUAD cores&#8230;windows is still having a few problems..even with an environment THEY CREATED..</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mark T.</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515225</link>
		<dc:creator>Mark T.</dc:creator>
		<pubDate>Thu, 23 Apr 2009 01:37:48 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515225</guid>
		<description>What is the logical progression for putting more and more cores on a chip?  So far we have had two, then four, then eight, and now 16.  What is next, 32?  Will it keep doubling or is there some other method to the arrangement of cores on the chip?

I read that Intel had shown an experimental 80 core chip so that blows my theory on how the core count would double as it does with the 8, 16, 32, 64 bit addressing.  Would an 80 core chip be 8x10 or quad 4x5 config or what?

BTW, is 128 bit processing decades away for consumer computers?  I would only think that 128 bit processing would be necessary when CPUs need to address more than 16.8 million terrabytes of RAM (phew!).  Say hello to HAL 9000.</description>
		<content:encoded><![CDATA[<p>What is the logical progression for putting more and more cores on a chip?  So far we have had two, then four, then eight, and now 16.  What is next, 32?  Will it keep doubling or is there some other method to the arrangement of cores on the chip?</p>
<p>I read that Intel had shown an experimental 80 core chip so that blows my theory on how the core count would double as it does with the 8, 16, 32, 64 bit addressing.  Would an 80 core chip be 8&#215;10 or quad 4&#215;5 config or what?</p>
<p>BTW, is 128 bit processing decades away for consumer computers?  I would only think that 128 bit processing would be necessary when CPUs need to address more than 16.8 million terrabytes of RAM (phew!).  Say hello to HAL 9000.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ECA</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515217</link>
		<dc:creator>ECA</dc:creator>
		<pubDate>Thu, 23 Apr 2009 01:20:15 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515217</guid>
		<description>IF&#039;
you could designate CORE use...
Video to THAT core,
GAMES to that core
EACH window gets a CORE
EACH program GETS A CORE..
It would be nice..
16 Cores is to many otherwise, as the CORES dont have enough to even get warm..
12 would be nice..</description>
		<content:encoded><![CDATA[<p>IF&#8217;<br />
you could designate CORE use&#8230;<br />
Video to THAT core,<br />
GAMES to that core<br />
EACH window gets a CORE<br />
EACH program GETS A CORE..<br />
It would be nice..<br />
16 Cores is to many otherwise, as the CORES dont have enough to even get warm..<br />
12 would be nice..</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: deowll</title>
		<link>http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/comment-page-1/#comment-1515204</link>
		<dc:creator>deowll</dc:creator>
		<pubDate>Thu, 23 Apr 2009 00:47:43 +0000</pubDate>
		<guid isPermaLink="false">http://www.dvorak.org/blog/2009/04/22/16-core-chips-and-you/#comment-1515204</guid>
		<description>The number of cores that can be used effectively depends on the OS, the chip design, and how much you need to get done.

I doubt if I personally need one right now.

I doubt if soy toner is an issue other than price and how well it works.

Biden is on the take from the crappy wood and any thing for a buck music companies just like Bush and his boys were. No change there.

If they can&#039;t get the security issues under control we are all bleep.</description>
		<content:encoded><![CDATA[<p>The number of cores that can be used effectively depends on the OS, the chip design, and how much you need to get done.</p>
<p>I doubt if I personally need one right now.</p>
<p>I doubt if soy toner is an issue other than price and how well it works.</p>
<p>Biden is on the take from the crappy wood and any thing for a buck music companies just like Bush and his boys were. No change there.</p>
<p>If they can&#8217;t get the security issues under control we are all bleep.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

