<?xml version="1.0" encoding="ISO-8859-1"?>
<!-- generator="HardwareAnalysis.Com" -->
<rdf:RDF
    xmlns="http://purl.org/rss/1.0/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
    xmlns:dc="http://purl.org/dc/elements/1.1/">
    <channel rdf:about="">
        <title>Hardware Analysis - Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <description>Hardware Analysis Community Forums</description>
        <link>http://www.hardwareanalysis.com/content/topic/42087/</link>
        <image rdf:resource="http://media.hardwareanalysis.com/halogo.gif" />
       <dc:date>2008-11-22T16:25:55-05:00</dc:date>
        <items>
            <rdf:Seq>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#268941"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#268938"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#265484"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#264970"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#262266"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#262257"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#262253"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#262159"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#262117"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#262112"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#262049"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#262044"/>
                <rdf:li rdf:resource="http://www.hardwareanalysis.com/content/topic/42087/#0"/>
            </rdf:Seq>
        </items>
    </channel>
    <image rdf:about="http://media.hardwareanalysis.com/halogo.gif">
        <title>Hardware Analysis</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/</link>
        <url>http://media.hardwareanalysis.com/halogo.gif</url>
    </image>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#268941">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-25T10:43:55-05:00</dc:date>
        <dc:creator>tom</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#268941</link>
        <description>sizzling hot snake oil?????&lt;br /&gt;
&lt;br /&gt;
WFT does that mean???</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#268938">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-25T10:29:19-05:00</dc:date>
        <dc:creator>Sander Sassen</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#268938</link>
        <description>Alright, after some delay Intel shipped us a 3.2GHz (840) dual core processor and 955 chipset motherboard, before I let her go on a standard set of benchmarks that'll not do much for dual core, I'd like for you to tell me what you'd like to see. Keep in mind that we don't have time for elaborate testing schemes, as we're out to see if dual core can prove its worth on the desktop.&lt;br /&gt;
&lt;br /&gt;
If you have any suggestions feel free to post them in the below thread, or email me directly at &lt;a class=&quot;ext&quot; href=&quot;mailto:ssassen@hardwareanalysis.com&quot;&gt;ssassen@hardwareanalysis.com&lt;/a&gt;.&lt;br /&gt;
&lt;br /&gt;
Dual core benchmarking, we'd like your input&lt;br /&gt;
&lt;a class=&quot;ext&quot; href=&quot;/content/topic/42925/&quot; target=&quot;_blank&quot;&gt;http://www.hardwareanalysis.com/content/topic/42925/&lt;/a&gt;&lt;br /&gt;
&lt;br /&gt;
Best regards,</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#265484">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-15T09:38:14-05:00</dc:date>
        <dc:creator>THOMAS STEWART VON DRASHEK</dc:creator>
        <title>Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#265484</link>
        <description>I think I have it figured out. As member of Symour Crays'. XMP24 team at George Mason University in mid 1980s' we threw out bit identification bits, leaving 24 bits which ran straight thru, we assumed all was perfect &amp;amp; needed no start/stop identifiers. That made 24 bit game platform cime qalive, with nintendo &amp;amp; playstation still using theory proven there.&lt;br /&gt;
Now amd using 48 bits active within 64 bit enviorment, while Intel uses only 40. Then it &amp;quot;hit&amp;quot; me, ohh,wwwooow, AMD is doubling 24 bit sample into 48 bits &amp;amp; placing identifiers at each end. It will be super with gaming code. Although just slightly slower than &amp;quot;true&amp;quot; 24 bit gaming, in potential.&lt;br /&gt;
Whats Intel up to? Seems to me if you took 40 bits times 3, you have 120 bits, add 8 bits of identifiers and you have full 128 bit sample. Look to Intel model when &amp;quot;core&amp;quot; puts out its 128 bit &amp;quot;core&amp;quot; array boxes, or start of such bit length pc software. it will take time to get to desk top, took 24 bit 20 years &amp;amp; now its screaming ability is being somewhat utilized, (so I precieve), yet when (&amp;amp; if) world ever goes 128 bit, Intels 40 bits built with 64 bit will have lead way, not only in gaming arrays in near future, yet with in all computing, when world is ready for ipv 7 (or 8?). Snake oil today, entire market tomorrow.+&lt;br /&gt;
sIGNED&lt;img src=&quot;http://media.hardwareanalysis.com/smilies/smile5.gif&quot; width=&quot;14&quot; height=&quot;14&quot; border=&quot;0&quot; alt=&quot;:P&quot; title=&quot;:P&quot;&gt;HYSICIAN THOMAS STEWART VON DRASHEK M.D.</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#264970">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-13T22:50:24-05:00</dc:date>
        <dc:creator>Wolfen</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#264970</link>
        <description>Encoding question(s) ...&lt;br /&gt;
&lt;br /&gt;
I am a beginner with understanding the interactions that happen when encoding media and obviously I want to get the smallest and cleanest content when I am done.  Any advise out there with not only Motherboards, but support boards, TV cards, encoder cards, etc. etc. ... I assume I know squat on this and lay it out.&lt;br /&gt;
&lt;br /&gt;
As for the dual processor game my experience deals with weapons technology, or processing an image on the fly to figure out where the &amp;quot;target&amp;quot; is going or is likely to go and to connect the dots, quickly!!  As for a home system I have a server I will retire soon that is Intel 2.5GHz HT tech on an ASUS board.  Works fine, but just not enough power.  I will upgrade to a server with OK graphics that will serve some outside concerns and my development concerns.  I will have background tasks of IIS, and a Database running, at least three or four IE windows open, and a Firefox window open with at least half a dozen tabs.  MS Visual .NET running, maybe a word processor sitting in the background (spell checking!!) Outlook running 24x7 etc.  So I guess I have a lot of crap running all the time.  How about the soft firewall, virus checker, adware stuff running, etc.  I NEED more processor power and I LOVE multiple threads.&lt;br /&gt;
&lt;br /&gt;
Now will these two processors be HT so I can have 4 processes happening at the &amp;quot;same&amp;quot; time??&lt;br /&gt;
&lt;br /&gt;
DBD.&lt;br /&gt;
...</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#262266">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T16:11:06-05:00</dc:date>
        <dc:creator>Brian Stewart</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#262266</link>
        <description>Encoding?&lt;br /&gt;
&lt;br /&gt;
AMD is actually better if you use better settings (slowest in divx, q=0 in LAME, for example)&lt;br /&gt;
That's because AMD is much more efficient, and higher quality settings (for media encodes) employ more complex math, which use more registers and instructions per calculation.&lt;br /&gt;
But why would this hurt Intel?  It's not like AMD gets extra clock cycles to complete the extra number of calculations.  Intel doesn't have a larger increase in clock cycles per calculation than AMD does - all those registers are still there, they're just all being used now.&lt;br /&gt;
&lt;br /&gt;
The answer - Hyperthreading.&lt;br /&gt;
When encoding media, there's always instructions waiting to be processed, and when you use typical settings, theres a lot of unused registers.  Hyperthreading really helps in this case.  But if you switch to the slower (higher quality) settings, more registers are used to store values during a calculation, and hyperthreading cant help (no room to stick in a new instruction).&lt;br /&gt;
&lt;br /&gt;
I for one always use the highest quality settings possible (q=0 in LAME, &amp;quot;Insane mode&amp;quot; in the latest divx beta, etc) because I LOVE geting my stuff at great quality with small file sizes.&lt;br /&gt;
&lt;br /&gt;
I've been saying this for years.</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#262257">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T15:48:34-05:00</dc:date>
        <dc:creator>Blood Brother</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#262257</link>
        <description>caps lock!!!!&lt;br /&gt;
&lt;br /&gt;
anyways Intel has always been one step ahead of AMD in encoding tasks. because of higher clocks the same thing is bound to happen in the dual core line too</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#262253">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T15:36:53-05:00</dc:date>
        <dc:creator>THOMAS STEWART VON DRASHEK</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil? MORE SPEED=MORE SPEED!</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#262253</link>
        <description>GETTING MORE RIGHT NOW MAY MEAN CHIPSET 945/955. WHY WAIT FOR 965, ITS' INCREASED SPEED OF ALL PERIPHERAL UNITS THAT INTEL SUPPORTS. PLUS THERE ALMOST AS BENCHMARK 5 EQUAL AS CAN BE. I APPAUD SNAKE OIL, WHEN ITS ON SOMEONE ELSE, IN THIS CASE, MAYBE I'M CANNIBAL, DONATE YOUR FLEASH HERE, WE NEED EVERY GIMICK POSSIBLE TO GET SEVER LEVEL THAT CAN SUPPORT STRAMING PAY PER VIEW AND BREAK EVEN, THAT MAY BE BIG, WHO KNOWS, EQUIPMENT IS JUST BEING DEVELOPED RIGHT NOW THAT DOES THAT IN MASS &amp;amp; GAMING ABILTY &amp;quot;EL COMPLEXO&amp;quot; ON HDTV LARGE SCALE FORMAT. MORE SPEED DUDE, INTERLACED SUPPORT FROM SDOFTWARE, SIGNAL &amp;amp; TV DISPLAY &amp;quot;PROGRAMABILITY&amp;quot; AHOY!!!&lt;br /&gt;
SIGNED&lt;img src=&quot;http://media.hardwareanalysis.com/smilies/smile5.gif&quot; width=&quot;14&quot; height=&quot;14&quot; border=&quot;0&quot; alt=&quot;:P&quot; title=&quot;:P&quot;&gt;HYSICIAN THOMAS STEWART VON DRASHEK M.D.. </description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#262159">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T09:54:56-05:00</dc:date>
        <dc:creator>Frozen FX</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#262159</link>
        <description>The Pentium EE 840 will have an unlocked multiplyer, though I don't see the point when it gives 130W heat. I have seen benchmarks of the EE (Which by the way uses an 800MHz FSB, not 1066MHz as previously thought), and in applications, generally it is faster (Than an FX55), such as:&lt;br /&gt;
&lt;br /&gt;
Windows Media Encoder 9&lt;br /&gt;
3DMark 05 CPU&lt;br /&gt;
PCMark 04&lt;br /&gt;
PCMark04 CPU&lt;br /&gt;
SiSoft Sandra 2005, Dhrystone ALU&lt;br /&gt;
SiSoft Sandra 2005, Whetstone FPU&lt;br /&gt;
SiSoft Sandra 2005, Integer x8 SSE2&lt;br /&gt;
SiSoft Sandra 2005, Floating-point x8 SSE2&lt;br /&gt;
SiSoft Sandra 2005, RAM Bandwidth Floating-point Buffered&lt;br /&gt;
&lt;br /&gt;
Though it is slower in:&lt;br /&gt;
&lt;br /&gt;
Doom 3&lt;br /&gt;
Half-Life 2&lt;br /&gt;
3DMark 05&lt;br /&gt;
PCMark 04 Memory&lt;br /&gt;
SiSoft Sandra 2005, RAM Bandwidth Integer Buffered&lt;br /&gt;
SiSoft Sandra 2005, RAM Bandwidth Floating-point Buffered&lt;br /&gt;
&lt;br /&gt;
The Pentium EE 840 is also faster at multi-tasking, to use a benchmarked example, importing a 130MB .pst file into Outlook while using Firefox.&lt;br /&gt;
&lt;br /&gt;
It is thought that it will be late 2006 when games are finished being re-written to take advantage of dual core, so in my opinion, a better option is to buy teh FX57, which I have heard should be available around the same time as the Pentium EE 840. After all, anyone spending this kind of money can probably afford to upgrade again in a year and a half.</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#262117">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T04:27:15-05:00</dc:date>
        <dc:creator>Brian Stewart</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#262117</link>
        <description>Yes, but overclocking aside, will there be any tangible benefits? I assumed there would be some kind of on-chip load balancing so both cores would be used pretty much equally. If there isn't then there doesn't really seem to be any point other than giving the marketing guys something else to hype up.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
I assumed this too, it wouldn't be hard to implement in hardware, but apparantly neither Intel nor AMD is looking into this - they expect software to handle it.  Saves them money, more reliable, sure, but hardware would be faster and would allow all programs to take advantage of it.&lt;br /&gt;
&lt;br /&gt;
A simple method to handle load balancing on the hardware level would be to have a small (very small) flip flop, that just alternated every clock, and on clock 1 it would send the instruction to processor 0, on clock 2, it would go to 1, clock 3 would be 0 again, etc.&lt;br /&gt;
&lt;br /&gt;
It would be perfect load balancing, a very small hardware solution (a single D-flip flop!!).&lt;br /&gt;
The problem - it would have to operate at a little over twice the speed of each core.&lt;br /&gt;
Pipelining isnt even a problem, since it would be done on an instruction level, and all instructions are per clock, with assemblers automatically putting in null ops if needed.&lt;br /&gt;
&lt;br /&gt;
I'm sure it's more complicated than what I'm thinking of, but not much more.&lt;br /&gt;
Hell, I'll ask a guy who build the damn things.</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#262112">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T04:11:29-05:00</dc:date>
        <dc:creator>t@kis !</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#262112</link>
        <description>anandtech.com has real good benchmarked reviews of intels dual core chips (pentium d, no HT, and pentium xtreme, w/ HT) vs. single core p4 xtreme and amd fx-55.  sweet!!</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#262049">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T01:00:59-05:00</dc:date>
        <dc:creator>Chris McGee</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#262049</link>
        <description>Yes, but overclocking aside, will there be any tangible benefits?  I assumed there would be some kind of on-chip load balancing so both cores would be used pretty much equally.  If there isn't then there doesn't really seem to be any point other than giving the marketing guys something else to hype up.</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#262044">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T00:36:42-05:00</dc:date>
        <dc:creator>PCGEEK</dc:creator>
        <title>Re: Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#262044</link>
        <description>thats why AMD and INTEL has said they will be locking the duel core down so it can not be overclocked.</description>
    </item>
    <item rdf:about="http://www.hardwareanalysis.com/content/topic/42087/#0">
        <dc:format>text/html</dc:format>
        <dc:date>2005-04-05T00:33:02-05:00</dc:date>
        <dc:creator>Chris McGee</dc:creator>
        <title>Re: Intel’s dual core processors, sizzling hot snake oil?</title>
        <link>http://www.hardwareanalysis.com/content/topic/42087/#0</link>
        <description>I presume you'll be saying the same thing about the AMD dual core processors then?</description>
    </item>
</rdf:RDF>
