Sign in with
Sign up | Sign in
Your question

Cedar Mill and Presler Rant - 4.2GHz+ Possible

Last response: in CPUs
Share
October 29, 2005 3:47:35 AM

Earlier this month Tom's Hardware Guide released a review of Intel's upcoming Cedar Mill and Presler processors. This generated a rash of comments that doubted the ability of these processors to perform properly, especially at high clock rates, due to thermal issues. However, the latest review by AnandTech proves otherwise.

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=257...

Anandtech found similar power savings as Tom's Hardware in the new Cedar Mill. They found that a 3.6 GHZ 660 Prescott uses 21% more power than a 3.6GHz 661 Cedar Mill. What is more interesting is that the 3.4GHz 950 Presler uses as much power as a 2.8GHz 820 Smithfield. This would mean that the 65nm process yielded a 600 MHz or 21.4% increase in processor speed and a 1 MB increase in L2 cache (which is one of the more power consuming components) without additional power usage. This is a definite increase in performance per watt.

Previously, I mentioned that Intel should release Cedar Mills clocked at 4.2 GHz. The reply from people was that the 3.8GHz 671 would never be released due power inefficiency and that a 4.2GHz Cedar Mill was "a pipedream". Once again I beg to differ. The major concern was that for every increase in clock speed the power consumption and loss increases exponentially, and in order to achieve stability Vcore increases would be needed further worsening the situation. It was also mentioned that the "Vcore has remained basically unchanged in spite of the shrink". In fact the Vcore has decreased from 1.4V in Prescott to 1.3V in Cedar Mill. In the end, a lower starting Vcore and the 21% reduction in power consumption yield ideal conditions for higher clock speeds. Anandtech proves this by overclocking a early sample of the 661 from 3.6GHz to 4.25GHz. While a 625MHz or 18.1% increase in speed may or may not sound impressive depending on your pessimism, the important thing to note is that this was down without increasing Vcore and with Intel’s Reference Cooler. Only standard overclocking, no special parlour tricks. Anandtech saids that “any serious overclocker to reach speeds greater than 4.0GHz effortlessly”. This clearly indicates that a 3.8GHz Cedar Mill is a legitimate release processor, and as such appears on Intel’s Processor Roadmap. It also indicates that a 4.2 GHz Cedar Mill is not “a pipedream”, but rather is something that Intel can reach “effortlessly”.

In fact, Cedar Mills were found to run stably at 4.5GHz on the stock cooler although the Vcore had to be increased to 1.4125V. It is also interesting to note that these overclocks were done using a 3.6GHz 661. A higher quality, higher binned part such as the 3.8GHz 671 that Intel is producing will likely still have slightly more room for speed increases.

Of course, now the concern is whether the power consumption will make such a chip unreasonable. Once again we need to look at the figures. A 950 Presler has the same power consumption as a 820 Smithfield. This is analogous to a 651 Cedar Mill consuming the same amount of power as a 521. We ask ourselves, what is the power consumption of a 3.8GHz 671? Assuming the 65nm process gives 600MHz and 1 MB of L2 cache worth of additional power consumption room, the 3.8GHz 671 would consume more power than the 3.2GHz 541, since power consumption increases non-linearly as clock speeds increase, but less than the 3.4GHz 551. Similarly a 4.2GHz 691 would consume more power than a 3.8GHz 571. Now it would appear the 4.2GHz Cedar Mill has broke the bank, so to speak, since it consumes more power than a 3.8GHz Prescott. However, several things need to be taken into account. First, Intel motherboards are designed for higher power consumption than the 3.8GHz 572 ever since the advent of the dual-core Pentium Extreme Edition. Secondly, my estimates are decidedly conservative. I mentioned above that Cedar Mill runs stably at 4.5GHz. As it turns out, Anandtech proved that the power consumption of a Cedar Mill at 4.5 GHz is less than the power consumption of a 3.6GHz 660. Therefore, a 4.2GHz 691 should have no problems remaining far below Intel’s current maximum limits.

The situation is similar in the dual core segment. If a 3.4GHz 950 can maintain the power levels of a 2.8GHz 820, Intel’s fastest dual core, the 3.46GHz 955 should remain in the power levels of the 3.0GHz 830 despite the higher 1066MHz FSB. This means that Intel still has clocking room for Presler while still remaining within the power levels of the 3.2GHz 840EE. Anandtech themselves clocked Presler to run stably at 4.25GHz.

The remaining concern is Intel’s new marketing strategy of performance per watt. In this case, significantly higher clocks can be reached while maintaining the same power usage(watt). A 4.5GHz Cedar Mill provides a 18.4% clock increase compared to the 3.8GHz 670 yet remains within the power consumption of the 3.6GHz 660. Therefore, increasing clock speeds does not necessarily mean decreasing performance per watt. From a marketing standpoint, winning is still the most important. Conroe is scheduled for H2 2006, but won’t likely be released en masse until Q4 2006 when all the kinks are worked out. That leaves Intel 1 year with no speed increases. During this time AMD is planning on releasing the X2 5000+ and the FX-60. In order for Intel to remain competitive they must release faster processors regardless of any slight decreases in performance per watt. Yonah’s weak FPU performance means it’s in no shape to fight the AMD64 despite its Digital Media Boost.

As such, Intel should release a 4.0GHz 681, and the 4.2GHz 691 that I’ve been arguing for all along. In order to properly combat the FX-60, a new single core Extreme Edition needs to be introduced. A 4.26GHz 966 Cedar Mill Extreme Edition could be released. Its power consumption would still be within Intel’s current maximum, and its high clock speed will mean it can finally benefit from a 1066MHz FSB. A 3.6GHz 960 Presler is already rumoured to be in the works if required and should also be released. The ultimate Intel processor, would then be a 3.73GHz 965 Presler Extreme Edition. This part may be just over the Intel’s current power comsumption limits but would certainly be worth it. Where $999 currently buys you 1 3.73GHz Extreme Edition, the same could get you 2 in dual core configuration.

Intel could easily build these processors with little cost. As Anandtech mentioned, clocks over 4.0GHz are effortless, and the current reference cooler is sufficient and does not require redesign. The 4.0GHz 681 and 4.2GHz 691 are easily reached with processors binned as 3.6GHz 661s and higher. As the 65nm process continues to improve before Cedar Mill introduction in Q1 2006, the yields will improve. A 3.6GHz 960 Presler is easily produced by slapping 2 661s together and is Intel’s current backup plan for a trump card to compete with faster AMD processors. While Cedar Mills with 1066MHz FSBs are not currently on the Roadmap, they are being produced to make the 3.46GHz 955 Presler. Producing a 3.73GHz 965 or a 4.26GHz 696 is simply a matter of improving yields and binning higher. No major retooling is necessary.

Clearly, the Netburst architecture still had some life left with the 65nm process. I know many people would disagree with me but the ability to run a 4.5GHz Cedar Mill for the same power consumption as a 3.6GHz Prescott proves that power issues can be solved or at least reduced to more acceptable measures. Sadly Intel is obsessed with their next-generation Conro, Meron, and Woodcrest architecture. What they should have done is invested more into the development of Cedar Mill and Presler. I’m not talking anything revolutionary. Just making the necessary changes to allow SpeedStep to run at lower clock speeds would have solved most of the power issues since processors are rarely at full load anyways. If Intel was afraid of a fully Enhanced SpeedStep capable Desktop processor cutting into Pentium M sales, they could introduce a SpeedStep feature into the Cedar Mill that is closer to AMD’s Cool’n’Quiet. Instead of enabling all multipliers, just enable a few at set levels such as 6, 10, and 14 to yield steps of 1.2GHz, 2GHz, and 2.8GHz. Another feature that could have improved Cedar Mill’s performance without a major redesign is fixing the L2 cache latency problem. The high L2 latency of the Prescott 2M was mainly due to Intel simply sticking an additional 1MB of L2 onto the processor. They could achieve better performance by tweaking the routine that the processor uses to address and prefetch the cache to better take into account that 2MB of L2 is available instead of 1MB. These two improvements, along with higher clock speeds, and VT implementation (the 6x3s were cancelled) would have kept the Pentium 4 alive, and profitable for the rest of the year until Conroe. It would have also given Intel some fallback room in the event that Conroe is delayed. As it is the best Intel can do right now is to continue to improve there manufacturing process to further reduce processor power consumption, and increase yield allowing for greater clock speeds.

Just my opinion on Intel's strategy.
October 29, 2005 8:44:44 AM

The point you dont seem to get. Chip size is proportional to gate length. A 90 nano chip is 30% larger than a 65 nano chip. If it's 30% larger, and 21% hotter, it's thermal density is 9% lower. In other words, the 65 nano chips will throttle at a 9% lower speed. The prescotts start to throttle at 3.4ghz. The pressler will start to throttle at about 3ghz.
Pretty basic math. You just cant break the laws of physics.
October 29, 2005 6:19:37 PM

Way to long to read it all.
Related resources
October 29, 2005 10:50:30 PM

For Wusy and HardwareBoss

I tried to warn people by calling it a rant in the title. In any case, I apologize for the length. This was originally supposed to quickly prove to the doubters the viability of higher clocked Intel processors, but it quickly became a comment on Intel's current strategy and the need for them to release higher clocked processors to compete with AMD's new X2 5000+ and the FX-60. Essentially what I'm saying is that Cedar Mills have been proven to clock "effortlessly" to 4.2GHz without increasing Vcore and using the standard Intel cooler. They can also run at up to 4.5GHz again on the standard cooler, but with an increased Vcore. However, despite concerns about extradinary power consumption levels, it turns out a 4.5GHz Cedar Mill uses less power than a 3.6GHz Prescott 2M. As such I recommended that Intel release 4.0GHz 681, 4.2GHz 691, and a 4.26GHz 696 Extreme Edition Cedar Mills. A 3.6GHz 960, and a 3.73GHz 965 Extreme Edition Presler should also be released. These could all be introduced with little effort by Intel since their current early production samples can "effortlessly" meet these clock speeds while still fitting within the power envelope of the previous Prescotts. All Intel would need to do is make new boxes.

I hope that was an appropriately shorter summary.


For Endyen

In actuality, Prescotts do not throttle at 3.4GHz. In fact they do not even throttle at 3.6GHz. Only the early 560 Prescotts throttled. The E0 stepping brought reductions in power comsumption and heat production allowing a 3.4GHz 550 to drop from a TDP of 115W to 84W and prevented the 560J from throttling. Similarly, the 660 also does not throttle. This can be seen in the website below.

http://www.hardwareanalysis.com/content/article/1785.4/

In fact the only time throttling occurs on a 3.6GHz 560 is when people break the thermal compound seal between the heatsink and the processor and fail to apply a thermal compound of comparable quality to Intel's default one. If a 3.6GHz Prescott is installed and not touched again it should not throttle even under full load.

However, it is still important to note that according to Hardware Analysis, the 3.73GHz does throttle under full load. Although improvements in Intel's architecture does allow the 3.73GHz to throttle less than the lower clocked early 3.6GHz 560. According to Hardware Analysis, the problem is not thermal density as much "Intel's stock heatsink that is shipped with all of their processors". This is easily fixed and has already been fixed as Intel updated their reference cooler when they released the dual core Smithfield.

http://www.tomshardware.com/cpu/20050405/pentium_d-06.h...

The new coolers have a larger copper core to increase conductance while the heat dissipation area of the heatsink appears to have nearly doubled due to more thinner fins. The newer coolers should clear up most throttling issues.

Although it appears that an appropriate heat sink is the main culprit for throttling, the topic of thermal density should still be addressed. You have noted that the thermal density appears to have increased by 9%. This is of course still on an early production sample. Improvements in the production line in the next few months will probably yield still greater reductions in temperature. A good example of newer steppings doing wonders is the E0 stepping for Prescott, which allowed a 3.4GHz 550 to drop from a TDP of 115W to the lower TDP of 84W. Even an additional 4% reduction will mean that Prescott will run 25% hotter than Cedar Mill, and using the 30% larger measurement that you mentioned, thermal density will only increase 5%. I don't think a 5% increase in thermal density is catastrophic.

Thermal density increases are always hard to read. Straight forwardly it is the concentration of heat on transistors. While obviously a core shrink does increase thermal density, in this case Intel has taken measures to alleviate the thermal density problem. Instead of shrinking the whole processor proportionally, only the transistors are shrunk. The isolating space between them remains constant to what it was in the Prescott 90nm process. This means heat from one transistor still has the same amount of distance to disipate before reaching and interfering with another transistor. It is important to note that the heat production of each transistor has decreased 21% as mentioned before to further reduce issues of thermal density. From a macroscopic point of view it seems that thermal density increases, but examining the real issue behind thermal density, the ability of heat from one transistor to affect another, from an individual transistor point of view, it is clear that the situation is not as grim as it appears.

With throttling issues alleviated by better coolers, and thermal density issues reduced, Intel should release faster processors. An Intel Extreme Edition at 4.26GHz should be able to beat AMDs announced FX-60 in Office productivity and Audio/Video applications, which have traditionally favoured Intel. AMD will still be able to maintain the lead in games though. A 3.73GHZ dual core Presler Extreme Edition should likewise beat the X2 4800+ and should offer the new X2 5000+ adequate competiton. One way or another, they are required in order to keep Intel competitive with AMD for another year until Conroe appears.
October 30, 2005 1:01:56 AM

I dont understand why overclockers care about the stock speed increasing. The first thing that I would do after loading the OS on a new system would be to crank the BIOS up and see how fast it would go. I see poeple getting fired up all the time about Intel raising the speeds. The potential to run faster is there without Intel pumping them up. How far it will go depends mainly on cooling and the ability to pay the increased power bill.

Am I looking at this wrong?

ASUS P5WD2 Premium
Intel 3.73 EE @ 5.6Ghz
XMS2 DDR2 @ 1180Mhz

<A HREF="http://valid.x86-secret.com/records.php?PHPSESSID=792e8..." target="_new">#2 CPUZ</A>
SuperPI 25secs
October 30, 2005 1:36:03 AM

You really need to take off those rose colored glasses. I would love to see intel move the technology ahead. Unfortuneatly that just isn't going to happen.
Saying that a 5% increase in thermal density will allow for higher clocks, just isn't realistic.
Quote:
The isolating space between them remains constant

No, they have maintained the seperation between the gates and the substrata. This is to maintain the leakage level, since that is usually between gates and traces. Incidentally, this is their big problem. Maintaining the spacing, just prevents the problem from worsening.
As to EO stepping, it brought a real world reduction of > 10%, they just changed the way that they calculated the TDP, to make it look better.
While a pristen 3.4 may not throttle, put a couple of months of dust on the hsf, and it will. Put it in a system, with less than perfect airflow, and it will. Put it in a system with dual high powered gfx, and again, it will. Even the higher voltage caused by an oversized psu may make it throttle.
Throttling is only a symptom. All that heat is going to cause everything in the box to act a little flakey.
Then there is the scenario at work, where we have 20 computers to a room. With the new chips, the only thing people will wear, is sweat shorts, and ear protection.
October 30, 2005 6:09:15 AM

I'm a bit of a purist. I want to OC, but I want to do it without the expensive ram, or a high end hsf. The ram I use costs ~ $80 a gig. I cant see me ever including the price of a fridge, as part of a cooling system for OCing.
It's also a question of all those suckers out there getting burned. I've never been too big on caveat emptor.
November 2, 2005 6:01:02 PM

Quote:
I cant see me ever including the price of a fridge, as part of a cooling system for OCing.

You could buy the fridge, keep the PC in the frostless freezer section to OC better, and still have cold beer while you game. ;) 



:evil:  یί∫υєг ρђœŋίχ :evil: 
The <b><font color=red>Devil</font color=red></b> is in my <b><font color=red>'98 Mercury Sable</font color=red></b>!
<b>@ 201K miles!</b>
November 3, 2005 10:02:38 AM

Your claims are again as solid, robust and meaningful as a pigeon fart over Washington Square. Why don't you ever tell us stupid people what you are actually meaning?

Is it all "Intel rulez" kinda sh|t you you have been throwing at us all these years?
I know it is, so no need for reply.

*fed up and going back to the Other to have another look at Auburn*
November 3, 2005 1:07:31 PM

<pre><font color=orange> ∩_∩
Ω Ω
(=¥=)</font color=orange> - Cedrik says Era should just grow up already.<font color=orange>
_Ū˘Ū_</font color=orange></pre><p>
:evil:  یί∫υєг ρђœŋίχ :evil: 
The <b><font color=red>Devil</font color=red></b> is in my <b><font color=red>'98 Mercury Sable</font color=red></b>!
<b>@ 201K miles!</b>
November 3, 2005 2:45:26 PM

Grow up!? Look who's talking!
You yourself want to be a prima donna so bad that even people in Peking can see that.

Hail Internet!
November 3, 2005 2:55:35 PM

<pre><font color=orange> ∩_∩
Ω Ω
(=¥=)</font color=orange> - Cedrik says that Era is just too much fun! What would we<font color=orange>
_Ū˘Ū_</font color=orange>   ever do without Era? **ROFL**</pre><p>

:evil:  یί∫υєг ρђœŋίχ :evil: 
The <b><font color=red>Devil</font color=red></b> is in my <b><font color=red>'98 Mercury Sable</font color=red></b>!
<b>@ 201K miles!</b>
November 3, 2005 3:03:02 PM

Quote:
You yourself want to be a prima donna so bad that even people in Peking can see that.

A better put down would have been more like: You yourself want to be a prima donna so bad that you probably wear a pink tutu and slippers.

Come on, is that the best you can do Era? Sheesh. Tap a clue for two colorless mana already.

Cecilia, you're breaking my heart.


:evil:  یί∫υєг ρђœŋίχ :evil: 
The <b><font color=red>Devil</font color=red></b> is in my <b><font color=red>'98 Mercury Sable</font color=red></b>!
<b>@ 201K miles!</b>
November 3, 2005 3:30:04 PM

Oh my, Slvr is trying to get serious with me.
I'm trembling already.
November 3, 2005 3:33:43 PM

Clowns to the left of me, jokers to the right.
Here I am, stuck in the middle with you.
<pre><font color=orange> ∩_∩
Ω Ω
(=¥=)</font color=orange> - Cedrik wonders if SP is even capable of being serious.<font color=orange>
_Ū˘Ū_</font color=orange></pre><p> :evil:  یί∫υєг ρђœŋίχ :evil: 
The <b><font color=red>Devil</font color=red></b> is in my <b><font color=red>'98 Mercury Sable</font color=red></b>!
<b>@ 201K miles!</b>
November 3, 2005 3:51:58 PM

In your dreams, buddy.
November 3, 2005 3:59:50 PM

<font color=green>Mirrors on the ceiling
Pink champagne on ice
And she said
We are all just prisoners here
Of our own device
And in the master's chambers
They gathered for the feast
They stab it with their steely knives
But they just can't kill the beast</font color=green>
<pre><font color=orange> ∩_∩
Ω Ω
(=¥=)</font color=orange> - Cedrik wonders what it would be like to hear Arnold<font color=orange>
_Ū˘Ū_</font color=orange>   Schwarzenegger sing Hotel Cal i forn i a.</pre><p>
:evil:  یί∫υєг ρђœŋίχ :evil: 
The <b><font color=red>Devil</font color=red></b> is in my <b><font color=red>'98 Mercury Sable</font color=red></b>!
<b>@ 201K miles!</b>
November 3, 2005 5:40:04 PM

You are getting melancholy in there. Watch out for your puny hind, my boy.
You are all just prisoners here[/arrogant]
November 3, 2005 6:53:39 PM

Quote:
You are getting melancholy in there. Watch out for your puny hind, my boy.
You are all just prisoners here[/arrogant]

So when you come for my puny hind, will that be with steely knife or pork sword?
<pre><font color=orange> ∩_∩
Ω Ω
(=¥=)</font color=orange> - Cedrik wonders if Era has even heard of the Eagles.<font color=orange>
_Ū˘Ū_</font color=orange></pre><p>
:evil:  یί∫υєг ρђœŋίχ :evil: 
The <b><font color=red>Devil</font color=red></b> is in my <b><font color=red>'98 Mercury Sable</font color=red></b>!
<b>@ 201K miles!</b>
November 3, 2005 7:07:36 PM

With 14 tons of herrings, if you want to know.
But I'll be there also , trust me.
My revenge all belong to me.
November 8, 2005 12:28:55 PM

Quote:
Dothan/Yonah = FX killer in performance and TDP

Yonah is the X2 killer in performance and TDP

DDR2 is under $100 per gig and you can run 2GB/4GB without all the BS associated with AMD's compatabilty/timing issues.

http://valid.x86-secret.com/show_oc.php?id=45312

<A HREF="http://www.xtremesystems.org" target="_new">www.xtremesystems.org&lt;/A>

So your comparing AMD's current proc's which have been out for 2 years with intel's future procs? U honestly dont think AMD will fix the memory compatibility thing with Ref F and DDR2? I truly respect your o/cing abilities but u can be really pigheaded when it comes to your opinion's on AMD's proc's.
November 8, 2005 3:42:25 PM

:| I thought that AMD had already made great strides in improving their DDR problems...
November 8, 2005 8:41:16 PM

Quote:
Dothan/Yonah = FX killer in performance and TDP

Yonah is the X2 killer in performance and TDP

DDR2 is under $100 per gig and you can run 2GB/4GB without all the BS associated with AMD's compatabilty/timing issues.

http://valid.x86-secret.com/show_oc.php?id=45312

<A HREF="http://www.xtremesystems.org" target="_new">www.xtremesystems.org&lt;/A>

So your comparing AMD's current proc's which have been out for 2 years with intel's future procs? U honestly dont think AMD will fix the memory compatibility thing with Ref F and DDR2? I truly respect your o/cing abilities but u can be really pigheaded when it comes to your opinion's on AMD's proc's.

It should be noted there has been no significant technological change in the K8 line other than dual cores. Perhaps like myself he believes nothing radical is comming unless Intel comes out with it and AMD licences it (SSE SSE2 ect.)
November 8, 2005 9:44:46 PM

Quote:
Perhaps like myself he believes nothing radical is comming unless Intel comes out with it and AMD licences it (SSE SSE2 ect.)

It sure is a good thing Amd licenced Intel's EM64T, and the ODMC, otherwise they would be dead in the water.
Intel's last great inovation, SSE3 was also such a great step. I sure am glad we can count on Intel to keep things moving at this great pace.
November 9, 2005 11:42:40 AM

Quote:
Dothan/Yonah = FX killer in performance and TDP

Yonah is the X2 killer in performance and TDP

DDR2 is under $100 per gig and you can run 2GB/4GB without all the BS associated with AMD's compatabilty/timing issues.

http://valid.x86-secret.com/show_oc.php?id=45312

<A HREF="http://www.xtremesystems.org" target="_new">www.xtremesystems.org&lt;/A>

So your comparing AMD's current proc's which have been out for 2 years with intel's future procs? U honestly dont think AMD will fix the memory compatibility thing with Ref F and DDR2? I truly respect your o/cing abilities but u can be really pigheaded when it comes to your opinion's on AMD's proc's.

It should be noted there has been no significant technological change in the K8 line other than dual cores. Perhaps like myself he believes nothing radical is comming unless Intel comes out with it and AMD licences it (SSE SSE2 ect.)
SSE2 was a radical change designed to hide the architectural weaknesses of netburst. I wouldnt call that inovation.
November 9, 2005 11:57:51 AM

Dell knows AMD is better lol that's why there gonna start selling them.

SUCKER lol


YaY AMD
November 9, 2005 11:58:51 AM

wow the filter blocked out Ss-uh-KA(spelled properly obviously)
November 9, 2005 6:14:50 PM

Quote:
SSE2 was a radical change designed to hide the architectural weaknesses of netburst. I wouldnt call that inovation.
I don't know if I can agree with that PoV. To me it was more like SSE2 was a crutch to lean on so that they didn't have to bother fixing one of the major weakness that they introduced into the core's design when they rushed the P4 out the door to cover up their P3>1GHz failure. :lol: 
November 9, 2005 6:17:14 PM

Quote:
wow the filter blocked out Ss-uh-KA(spelled properly obviously)
Sucker? :lol:  OMG! It really is filtered. And I thought that filtering crap and xxx was funny! He he he.
November 9, 2005 11:03:20 PM

Quote:
Perhaps like myself he believes nothing radical is comming unless Intel comes out with it and AMD licences it (SSE SSE2 ect.)

It sure is a good thing Amd licenced Intel's EM64T, and the ODMC, otherwise they would be dead in the water.
Intel's last great inovation, SSE3 was also such a great step. I sure am glad we can count on Intel to keep things moving at this great pace.

Does it always have to be a pissing match with you? You understand what I am saying why do you act like I said something I didn't.

But screw it you honestly think the ODMC and AMD64 are some sort of revolution in computing power? Both technologies have been around in one form or another for a quite a few years before AMD packaged them in a great processor, but hey you read what you want to.

Frankly don’t care too much either your opinion is just that but my statement that nothing new has come out for quite some time does hold true. Unless your a wizard and know of some remarkable technological change in the K8 that I was unaware of.

With considerations at least Intel is trying to recover but no one gives em props for that too easy to bash and stare at your fag e-penis scores, when simply put you recommend/like/purchased the better platform. Everyone knows its AMD's ball right now quite assuming I don’t think otherwise cause all you do is piss me off further with your lackluster macho responses.

Quote:
SSE2 was a radical change designed to hide the architectural weaknesses of netburst. I wouldnt call that inovation.

If you have the slightest understanding about code or execution resources or really anything in regards to this topic Harlan you wouldn't be saying that. SSE2 isn’t anything special though just helping extend the junk known as x86, lost as to why any of you defend such a terrible ISA but whatever.

Quote:
I don't know if I can agree with that PoV. To me it was more like SSE2 was a crutch to lean on so that they didn't have to bother fixing one of the major weakness that they introduced into the core's design when they rushed the P4 out the door to cover up their P3>1GHz failure.


More of the Intel rushed the P4 wow I can’t believe people that claim they are so educated believe such a farse. The project was in the works for nearly 7 years possibly longer how is that a rush job, oh wait Intel’s platform design teams work as one group and come out with a uber processor, No they don’t do they?

It was a evolution that’s all the net burst project was and you know what it might be sucking hardcore now but it sure made Intel a crap load of cash didn’t it? Seem like the failed technology ended up paying for itself and that in itself is impressive, but I forgot where I am right now, you guys live in whatever the fad is, and the common word even if its grossly opinioned and unsupported, is the word of the land.
November 10, 2005 3:40:16 AM

Quote:
If you have the slightest understanding about code or execution resources or really anything in regards to this topic Harlan you wouldn't be saying that. SSE2 isn’t anything special though just helping extend the junk known as x86, lost as to why any of you defend such a terrible ISA but whatever.

So the line of chips from Intel and AMD that have driven the modern day computer industry for the past 20 years are junk? As someone who always retreats and hides behind Intel's inability to get a product that actually matters to the general consumer to make money and their whole buisness principle in general, you should be able to understand that while X86 might not be the strongest architecture in technical terms, It IS the strongest processor architecture because of its market penetration. I2 will remain a niche product as long as AMD and X86 have anything so say about it. I'm sorry if that dissapoints you. I remeber the days when you were telling me about how netburst would turn into IA64 and that would be the way the industry would roll. You were wrong in the end because itanium is a dead horse for now and will not mean anything to any normal consumer within the next 10 years. The software is not there. The product is simply too expensive to produce and continually R&D. Intel has pretty much stated that netburst is dead. Netburst was their big thing for years. Can't you possibly see history repeating itself with IA64 as well? Reasonable doubt is all i'm looking for here.
November 10, 2005 1:33:34 PM

Quote:
More of the Intel rushed the P4 wow I can’t believe people that claim they are so educated believe such a farse. The project was in the works for nearly 7 years possibly longer how is that a rush job, oh wait Intel’s platform design teams work as one group and come out with a uber processor, No they don’t do they?

You seriously need to wake up. The release of the P4 was rushed. There's no way of even remotely denying that if you read anything at that time.

Yes the P4 was in development for a long time. It was in the back, away from the limelight, an engineering project just evolving and waiting for technology to make it possible to fill the engineer's dreams.

If you'd looked at the original design specs, you'd have seen that the P4 was supposed to have had much more transistors for things like a second independant FPU. Imagine what a P4 with much better x87 performance would have been like!

But to get the P4 out the door to cover up the sorry state of the P3, Intel management forced the engineers to hack useful things like that out of the core to save space, and directly to that, to save costs. They even used compression technology on their instruction cache to reduce die space. (I mean look how big Willy was even after that!)

Had Intel waited on releasing the P4 until the next process shrink, instead of rushing something to market like they did (against their own engineer's advice/complaints) then CPU industry today would be a very different place indeed.

But of course I wouldn't know anything about that because I never read those articles years ago since I'm not educated. No, I'm compeltely clueless and just making up this farce.
November 10, 2005 11:17:54 PM

Quote:
If you have the slightest understanding about code or execution resources or really anything in regards to this topic Harlan you wouldn't be saying that. SSE2 isn’t anything special though just helping extend the junk known as x86, lost as to why any of you defend such a terrible ISA but whatever.

So the line of chips from Intel and AMD that have driven the modern day computer industry for the past 20 years are junk? As someone who always retreats and hides behind Intel's inability to get a product that actually matters to the general consumer to make money and their whole buisness principle in general, you should be able to understand that while X86 might not be the strongest architecture in technical terms, It IS the strongest processor architecture because of its market penetration. I2 will remain a niche product as long as AMD and X86 have anything so say about it. I'm sorry if that dissapoints you. I remeber the days when you were telling me about how netburst would turn into IA64 and that would be the way the industry would roll. You were wrong in the end because itanium is a dead horse for now and will not mean anything to any normal consumer within the next 10 years. The software is not there. The product is simply too expensive to produce and continually R&D. Intel has pretty much stated that netburst is dead. Netburst was their big thing for years. Can't you possibly see history repeating itself with IA64 as well? Reasonable doubt is all i'm looking for here.

30+ years and any programmer with half a brain will agree x86 is junk the silicon isn’t the IA is. Retreats? Intel's inability to produce a product that matters? Market penetration? Do you hear yourself?

I don’t post here enough for you to generalize I retreat, nor will I ever. A product that sells recovers initial development costs is by definition is a successful product. As well as the several truck loads of cash net burst just produced out of no where.

IA-64? Where did I mention anything to do with it? Are we trying to get me to go yaya IA-64 roxorz your boxorz? Granted I do believe IA-64 is superior to x86 it simply does not work with the industry conditions. Nor will it ever work at least in the mainstream. But whatever feasibility and market demand go figure it cant/wont get a foot hold.

Wrong in the end? I don’t follow, ah wait I'm sure you’re attempting to rehash old conversations or something to that extent. Vague as they are too me I would have to assume I said IA-64 will become mainstream or something retarded.

Reasonable doubt? You don’t even have any real facts to support anything said since it was all opinion so you can reasonably doubt that Intel cant compete you can reasonably doubt Intel has anything in store for the industry and you can reasonably doubt I care as well. No doubt about it.
November 10, 2005 11:38:53 PM

Quote:
More of the Intel rushed the P4 wow I can’t believe people that claim they are so educated believe such a farse. The project was in the works for nearly 7 years possibly longer how is that a rush job, oh wait Intel’s platform design teams work as one group and come out with a uber processor, No they don’t do they?

You seriously need to wake up. The release of the P4 was rushed. There's no way of even remotely denying that if you read anything at that time.

Yes the P4 was in development for a long time. It was in the back, away from the limelight, an engineering project just evolving and waiting for technology to make it possible to fill the engineer's dreams.

If you'd looked at the original design specs, you'd have seen that the P4 was supposed to have had much more transistors for things like a second independant FPU. Imagine what a P4 with much better x87 performance would have been like!

But to get the P4 out the door to cover up the sorry state of the P3, Intel management forced the engineers to hack useful things like that out of the core to save space, and directly to that, to save costs. They even used compression technology on their instruction cache to reduce die space. (I mean look how big Willy was even after that!)

Had Intel waited on releasing the P4 until the next process shrink, instead of rushing something to market like they did (against their own engineer's advice/complaints) then CPU industry today would be a very different place indeed.

But of course I wouldn't know anything about that because I never read those articles years ago since I'm not educated. No, I'm compeltely clueless and just making up this farce.

So you know this because you work there Silver insider trade information? You can say the same for the I1 oh wait no one does because the Willy and the I1 had the same problem they both sucked.

Unless Intel has no real direction of intelligent staff which all couldn’t see the P3 scaling much further, which then lead to the P4 development team going hey we have a half asses POS processor that doesn’t really improve on anything but the other team has a feasible replacement at 0.13.

As well for such a rush job I do enjoy the marginal 15% core difference (cache) the Northwood had over the Willy. Just screams to be rushed.

Ya I know what they wanted and what they got which are 2 different things, doesn’t change the fact that that’s that and this is this.

Hmm another FPU ya I can imagine cooking my dinner on a 3.0 instead of using a heat sink. But I am pretty sure the second FPU would have dragged the clocking down further than its shameful 3.8 or 3.6 I don’t know how far they scale anymore since it doesn’t matter.

Hmm more theories about Willy and its terrible performance, but alas they are yours and I have no interest in discussing the finer points of technology that neither or us can claim we honestly know since neither of us are working at Intel to confirm.

Hmm if they waited to release ya I can see how that would make it AMD's market very easily.

Oh ya I figured such a comment would come from you in regards to your education and ya whatever you say we all know I called you a idiot or whatever retarded thing you want to think of doesn’t bother me your self-esteem not mine.
November 11, 2005 1:31:11 PM

Quote:
So you know this because you work there Silver insider trade information?
...
Hmm more theories about Willy and its terrible performance, but alas they are yours and I have no interest in discussing the finer points of technology that neither or us can claim we honestly know since neither of us are working at Intel to confirm.
Right. That's how I know it, I have "Silver insider trade information". Not because it wasn't widely published by sources as trustworthy as the eetimes. Not because Darrell Boggs, Intel's principal engineer for the desktop platform group, openly admitted it to press. Not because anyone could have easily found the information and read it, espeically when the admission from Intel first came out and flooded the tech sites. No, I know it because I have privileged "insider trade information" about this "farce" that "people who claim they are so educated" made up.

Xeon, maybe you'll want to shut up now before you get your head so far up your arse that you resemble a human donut. Obviously one of us actually is educated, and one of us is being schooled.

Quote:
Under the original plan, the Pentium 4 was to have one slow ALU, two fast ALUs, two arithmetic address-generation units, two fully functional floating-point units, 16 kbytes of L1 cache, 12,000 instructions of execution trace cache, 128 kbytes of L2 cache, 1 Mbyte of external L3 cache, an allocator/register renamer and a bus architecture.

But with fabs costing more than $2 billion, even the world's largest, most profitable semiconductor company had to reconsider its plans when it became apparent that the chip size was growing too big.
November 11, 2005 8:27:53 PM

Hmm what they wanted they did not get I don’t see the issue at all, other than your smart ass comments of intellectual superiority or whatever your angle is today.

With regards to what you think and I think I will look at it objectively and look how the entire project has panned out and to me it doesn’t appear that a rush was in order unless you honestly believe that the P3 out of no where stopped scaling for Intel completely flooring them since they only test for 20 min and don’t test anything that isn’t being sold.

Then there is simply the supporting fact that the P4 thermal density is currently too great let alone getting all the goodies initially slated. Cut out 1/2 of the IC and drop what’s left to the foundries and do all this in less than 6 months.

I really really really find that whole situation improbably, what’s more probable is 12-18 months before the P3 hit its clocking limits the company realized holy moly the P3 is not gonna clock any further without modification and we have the P4 guys claiming they can have something workable in 12-18 months regardless if they are gutting the original specs. 12-18 months is very much feasible for a core redesign, now silver if you call 12-18 months a rush job then OK I agree it’s a rush job.

But looking at the actual guestimated time it would take to completely change direction of the P4 and have it work correctly out of the box would/did require some time points to a more realistic option.

Now if 12-18 months to Intel is rush work then ya I would have to agree with you but 12-18 months to me is a long time difference of opinion which doesn’t need your grade school comments coming to fruition but whatever acting like adults was lost when children and morons got access to the internet
November 11, 2005 10:11:48 PM

Quote:
Mirrors on the ceiling
Pink champagne on ice
And she said
We are all just prisoners here
Of our own device
And in the master's chambers
They gathered for the feast
They stab it with their steely knives
But they just can't kill the beast</font color=green>
<pre><font color=orange> ∩_∩
Ω Ω
(=¥=)</font color=orange> - Cedrik wonders what it would be like to hear Arnold<font color=orange>
_Ū˘Ū_</font color=orange>   Schwarzenegger sing Hotel Cal i forn i a.</pre><p>
:evil:  یί∫υєг ρђœŋίχ :evil: 
The <b><font color=red>Devil</font color=red></b> is in my <b><font color=red>'98 Mercury Sable</font color=red></b>!
<b>@ 201K miles!</b>


Numero Uno - I like the sig!!!!!!!!!!!!!!! LMAO!!!!!!!!

Now to the important stuff.... Im curious as to what your interpretation of the Hotel Cali would be especially the lines above. [I dont suppose we can go off topic anymore?]
November 11, 2005 10:14:33 PM

Woops, I didnt realize how serious this thread had gotten [/hides under bar stool]
November 12, 2005 12:36:46 AM

Quote:
the entire project has panned out and to me it doesn’t appear that a rush was in order unless you honestly believe that the P3 out of no where stopped scaling for Intel completely flooring them

Yup, it's not like Intel new ahead, the limits of thier architecture.

It doesn't matter what your particular concept of reality is, when it's just wrong. Truth is, that Intel had trouble scaling the copermine core. Nor did they have a crystal ball to see Tuulitan coming.
I know you are still young, but most people can remember that far back.
If you want a history lesson, start here
November 12, 2005 4:06:57 AM

Saying they didn’t know is asinine, as well no logical assumptions can say they didn’t see the scaling issues, simply put they have been selling clock speed for a long time they would need to know those envelopes way before the end of product. Now Intel does shady things as we all know, its far more feasible that released the 1.13 in hopes no one would notice to stretch the timetable for P4 delivery from the foundries.

We are talking about big business here saying they didn’t know is completely ridiculas, saying they were buying time by dropping sub par parts on the market and then acting like they didn’t know is more Intel than they just didn’t see it coming.

But like I said it’s a difference in opinion and how many links you have which doesn’t change the facts that if you take a moment and think about it, it really doesn’t look the way it used to.
November 12, 2005 7:06:47 AM

Of course they saw it coming. The thing they really saw though, was that the Athlon would scale fast, and without putting out something, and fast, they would be behind the eightball. All the more so because of people like Thomas Pabst.
They new they needed 6 months or so, so they dumped Williamette on the market. It was a throw away, but it got people's attention. Sure, it was a pos, but it was a fast clocked pos, and it gave them a little more time to work on northwood.
Problem was, they couldn't get northwood to work on the 133 fsb. A major step backward, on a chip that really needed the band width, so they dumped the northwoodA on the market. Again with the damage control. It was a pos, but it scaled better, and it gave them time to bring out the P4b. Problem then was the quad pumped bus. They wanted to work on a 166 fsb, but quad pumped, that would be 666, the number of the beast. Cant do that in god's country. But they could scale, and so they did.
It was only the ability of the P4b to scale, that got Intel back the crown.
!