Intel 45nm Fab Process Launched And Penryn Preview

NinjaKicks writes "Intel has decided to make public details of their new 45nm manufacturing process and also has broken news that next-gen Penryn core processors are running various versions of Windows and Vista successfully. Penryn will offer a host of core tweaks over Conroe, larger cache sizes, and SSE4 support. Also, although clock speeds will be increased, processors based on Penryn should fall within the same thermal power range as Conroe. Word is Penryn will also be compatible with some of the existing motherboards on the market while others will need either a BIOS update or perhaps other board-level changes."

Comments

Test Bench

[BajaTech]

But does it run linux?

Re:Test Bench

[Harmonious Botch]

It will when I get one.

Re:Test Bench

[Guy Harris]

But does it run linux?

Yes. (Just search for the "L" word.)

Can we see some clock speed advances?

[Anonymous Coward]

Dear Intel,

Can I see the clock speed boosted? Not everything can be parallelized and besides I don't think anyone at Microsoft knows how to.

Thanks,
A Windows User Not Looking forward to badly syncing apps

Re:Can we see some clock speed advances?

[cp.tar]

Can I see the clock speed boosted? Not everything can be parallelized and besides I don't think anyone at Microsoft knows how to.

Who knows, maybe we're looking at the days of assembly programming again... you know, the days of well-written, optimized software?

The ever-growing processor speed had all but removed the need for optimization; maybe the long-forgotten art is facing a revival...

Re:Can we see some clock speed advances?

[shankarunni]

Can I see the clock speed boosted? Not everything can be parallelized and besides I don't think anyone at Microsoft knows how to. Another blurb in TFA talks about this:
HK + MG Combined:

• Drive current increased >20%, (>20% higher performance) OR
• source-drain leakage reduced >5x

What this means is that you can get higher performance (~20%) at the cost of higher power consumption (on the order of today's processors), OR you can get the same performance at substantially (not 1/5x, though) reduced power. The first few Penryn processors are apparently targeted at the Mobile market, so we can see where they are going with this in the short term.

Re:Can we see some clock speed advances?

[Joe The Dragon]

In the sever market the power consumption of the FB-DIMM's are alot higher then ECC DDR 2

Re:Can we see some clock speed advances?

[Surt]

Most significant software projects already have a performance optimization phase. You run a profiler, find out where your application spends 90% of its time, and you fix that. The fact that software today in some cases seems slow mostly has to do with it doing more, not a lack of optimization. There are exceptions, but by and large most software cannot be made even twice as fast without loss of functionality.

Re:Can we see some clock speed advances?

[Anonymous Coward]

Care to elaborate? I find it very difficult to believe this. How do you explain that software with very similar feature sets often have huge differences in speed (compare OpenOffice to MS Office, for example... OpenOffice often seems glacial in comparison).

Re:Can we see some clock speed advances?

[trimbo]

Not everything can be parallelized

Most things that desperately need to be parallelized can be when someone takes the time to do it. What's the most CPU intensive thing the average user does on their computer? Media encoding? Gaming? Both of these are ripe for parallelization if someone spends the time to do it. Even high end computational uses like fluid dynamics and 3D rendering are parallelizable.

The multi-core strategy has been superb for those who already have parallelized needs and a dire lack of rackspace, A/C and power. Intel is selling these chips like gangbusters to those people who see the price/performance in it. A 3.0ghz dual-core Woodcrest might cost $200 more than a 2.66ghz version and gives you a 12% speed bump on an serial algorithm. A slower 1.86ghz quad core Clovertown costs $100 less than a 2.66ghz dual core Woodcrest, but can still give you upwards a 40% bump if your algorithm is perfectly parallelized. That math is why Intel's strategy is working for businesses that need it.

Note: prices taken from Dell's main, non-discounted site.

Re:Can we see some clock speed advances?

[Anonymous Coward]

Not everything can be parallelized

True. Fortuanately these things ofter appear in batches of thousands or millions and each can be run in parallel.

Re:Can we see some clock speed advances?

[Surt]

OpenOffice and MS Office, while superficially similar do significantly different work. OpenOffice is multiplatform, which means its presentation layer and file system layers at least have to be very generalized, and that will account for most of what you experience as slow. MS Office gets to use native APIs, and unpublished APIs that allow it to be faster than any competitive application on the windows platform. Compare that to MS Office for Mac, which everyone experienced as horrible compared to the windows version. Finally, OpenOffice doesn't have quite the funding for optimization behind it that MS office does. So I could well believe that with work, open office might get twice as fast.

Is this a major breakthrough?

[kestasjk]

• ~2x improvement in transistor density, for either smaller chip size or increased transistor count
• ~30% reduction in transistor switching power
• >20%improvement in transistor switching speed or >5x reduction in source-drain leakage power
• >10x reduction in gate oxide leakage power

As a layman this sounds like a pretty massive improvement. Is this a major breakthrough or is this progress as usual?

Re:Is this a major breakthrough?

[marcosdumay]

I'd classify it as a minor breaktrough, such things happen every few years but are a huge gain at once.

But feature size isn't everything.

Re:Is this a major breakthrough?

[Ant P.]

If they can build a high-end chip that produces less waste heat than the 386, _that_ would be a breakthrough.

Re:Is this a major breakthrough?

[shankarunni]

As a layman this sounds like a pretty massive improvement. Is this a major breakthrough or is this progress as usual? Depends on your definition of "breakthrough". Breakthroughs aren't what they were 30 years ago. This one gives us maybe 10 additional years on the Moore's law curve. Sounds like a middling leap at best.

Re:Is this a major breakthrough?

[brejc8]

Yes and No.
There are two improvements here which are happening at the same time. Process shrink (which happens all the time) and the use of High-K dielectric (which is something reather new in the mass manufacture feild anyway)

These 2 are just due to process shrink and nothing special:
  * ~2x improvement in transistor density, for either smaller chip size or increased transistor count
  * ~30% reduction in transistor switching power

This one is interesting and the OR should be regaded as an XOR.
  * >20%improvement in transistor switching speed or >5x reduction in source-drain leakage power
Basicly the individual transistors become tunable to decide if they should be fast or low power. Critical path ones will be fast and others will be low power.

And this one is a breakthrough:
  * >10x reduction in gate oxide leakage power
With static power now accounting for up to 50% of all power this is excelent.

Re:Is this a major breakthrough?

[RicktheBrick]

If one has a penny and doubles it one has $.02. If one has a million dollars and doubles it one has 2 million dollars. Most people would consider the latter to be a more important improvement. It is the same for microprocessors, doubling now is that much greater than it was 30 years ago.

Re:Is this a major breakthrough?

[Bender_]

This is the first time since 1969 that a major modification to the MOSFET gate stack occured. In fact it is fairly major. I should remind you that this is a structure that is replicated around 1e19 times each year and is responsible for the biggest part of the 270 Billion US$ semiconductor market.

At least ten years of work in academia and industry and billions of dollars were poured into this. Intel is the first company to make the move and introduce high-k.

(Yes, there were a few minor modifications to the SiO2/Poly stack in between: Plasma nitridation and numerous improvements on SiO2 growth)

Re:Is this a major breakthrough?

[Bender_]

That is not entirely correct. Intel had basically maxed out SiON previously, I doubt they could have gained that much in performance without high-k.

Re:Is this a major breakthrough?

[wtarreau]

* ~2x improvement in transistor density, for either smaller chip size or increased transistor count
* ~30% reduction in transistor switching power
* >20%improvement in transistor switching speed or >5x reduction in source-drain leakage power
* >10x reduction in gate oxide leakage power
As a layman this sounds like a pretty massive improvement. Is this a major breakthrough or is this progress as usual?

While not a major breakthrough per se, it demonstrates intel's commitment to work on architectural evolutions again. They work both to improve the IPC (instructions per cycle) and the transistors quality, and that is a good thing. It's also written in the article that the layer are now deposited one atomic layer at a time. I believe that by demonstrating their skills in this area, they're improving our global knowledge and capabilities in microelectronics. The stupid GHz race has ended and smart people can express themselves again. Now that is a major breakthrough.

Willy

Re:Is this a major breakthrough?

[Anonymous Coward]

Yes, but not for any of those reasons.

This is the single most important factor:

...larger cache sizes...

On chip caches are great, but only if of an adequate size. I can't count how many P4s with 16K L1 and 128K L2 caches I've seen that perform worse than PIIIs or even older processors.

My 1000MHz PIII Tualatin is adequate at 32K L1 and 256K L2 but I think double that would be better. As it is, it outperforms the majority of P4s up to 2GHz crippled by inadequate caches.

Non-technical people buy machines with such processors in them because of the cheaper price, unaware that they are buying a horribly crippled machine.

It is long overdue for Intel to stop making such crap. I can't imagine why they did in the first place... well perhaps to compete with AMD pricing. Though I suspect that they will continue to sell such crippled processors and that they will continue to be the most popular because they are comparatively ~$20 cheaper than a real processor.

Re:Is this a major breakthrough?

[dreddnott]

Slow down, cowboy, although you are spot on with the L1 cache size, Intel's Pentium 4 never had a cache size of less than 262,144 bytes, and from the more efficient Northwood on (January 2002), cache size ranged from 524,288 to 2,049,152 bytes (trying to avoid a mebi vs. mega debate).

The reason your Pentium III handily outpaces Pentium 4s of higher clockspeed has more to do with the inefficient Netburst core having an excessively deep pipeline that allowed higher clocking but lowered IPC and penalised branch prediction misses than thread size being too large for the cache.

This comment would have been more appropriate in late 2001 than early 2007, although the Pentium 4 was more expensive than the Pentium III at release even though it was less efficient. In case you just stepped out of your time machine to post a troll on Slashdot, I need to inform you that Intel recently stopped selling the "horribly crippled" Pentium 4 in favor of an architecture based on (surprise!) the P6 architecture, which powered the Pentium Pro, II and III CPUs.

I hope you feel at least somewhat vindicated.

Re:Is this a major breakthrough?

[acidrain]

If all you have is $20 in your pocket and no idea how you are going to earn money, getting another $20 means a lot more than handing a second billion to a billionaire. Processing speed is becoming meaningless on the desktop for most users. This is mostly relevant to servers now that we are moving to thin clients in the form of web-browsers, and expecting more an more processing from webservers.

Re:Is this a major breakthrough?

[dabraun]

And this one is a breakthrough:
    * >10x reduction in gate oxide leakage power
With static power now accounting for up to 50% of all power this is excelent.

I hate to be pedantic, but you can't have a "10x reduction" in anything, a 1x reduction is zero. What does this really mean? 90% reduction (1/10th remaining)?

Re:Is this a major breakthrough?

[x2A]

"I hate to be pedantic"

no you don't, you love it :-p

More (Better?) Coverage

[Anonymous Coward]

http://dailytech.com/Life+With+Penryn/article5869. htm
http://www.msnbc.msn.com/id/16839253/
http://money.cnn.com/2007/01/27/technology/bc.micr ochips.reut/index.htm?cnn=yes
http://www.nytimes.com/2007/01/27/technology/27chi p.html?em&ex=1170046800&en=59a4d10473c4a8c8&ei=508 7%0A
http://news.com.com/Chip+companies+entering+their+ metal+period/2100-1006_3-6153962.html
http://anandtech.com/cpuchipsets/showdoc.aspx?i=29 15

Still on the FSB

[Joe The Dragon]

If you are going to the make the chips smaller how hard is it to come out with a true quad-core?
Havening 2 duel-cores linked by a fsb bus will get in the way even faster as the speed of the cpu gets higher.
And a 4 cpu quad-core sever will likely choke up at the chipset to ram link as well as the chipset to chipset link.

Also if your duel quad-core workstation only have has the pci-e lanes for 1 x16 slot and the 8 other ones are used for the chipset to chipset link amd based ones will blow it away even more so with KL8 cpus. Right now an 2 cpu amd board has 4 pci-e x16 slots running at x16 x8 x16 x8 with 2 x4 lanes left over + each cpu can have a HTX slot or other HT based chip hook up to it.

Re:Still on the FSB

[RightSaidFred99]

Unfortunately (for your point) this has been proven wrong. The FSB works fine up to 2P machines. Intel will soon be quad FSB machines which should work very well for machines up to and including 4P/16Core. The proof is in the pudding, benchmarks show that the FSB limitation is only meaningful in a very few instances, and in most cases the superior uArch of Core2 more than makes up for it.

As for K8L, looks interesting - we'll have to see. If you think it's going to have a broad 40% improvement over Core2, though, I've got a bridge in Hanoi to sell you.

Re:Still on the FSB

[Joe The Dragon]

quad FSB machines may still choke up at the chipset with all of the cpus needing the use the same link to get to ram.

Even if the K8L is %1-%5 faster it still kills Intel in sever / workstation chip sets in terms of number of pci-e lanes and Intel has no to way to better a high-end duel cpu quad-core amd KL8 system.
as there 2 cpu system uses FB-DIMMs and only has 20-22 pci-e lanes with no sli.

Re:Still on the FSB

[Anonymous Coward]

This is AMD fanboys grasping at straws. Intel is blowing AMD out of the water, as is evident by major switches by large manufacturers (see http://hardware.slashdot.org/article.pl?sid=07/01/ 22/1449257). All AMD fanboys have is wishfull thinking as to what will happen with the next generation of chips (KL8 or whatever). Keep dreaming fanboys.

Re:Still on the FSB

[antifoidulus]

2 duel-cores

Spelling Nazi time: It's dual! Dual! It's only "duel" if your processors are firing pistols at each other from 10 paces at dawn!

Re:Still on the FSB

[dreddnott]

Actually, IBM and AMD, as partners, will also be using this new 0.45 process in their new chips, but it may take them longer to get there than Intel.

As an AMD fanboy, I was actually very happy to see the release of the Core and Core 2 chips - Intel was willing to admit that Netburst was a mistake, and they forced AMD to lower their dual-core CPU prices by 40%-60% in one fell swoop.

Re:Still on the FSB

[Aadain2001]

If you are going to the make the chips smaller how hard is it to come out with a true quad-core?

Not very. It just takes time to go through all the design -> verification -> layout -> production steps that all new processors require. Yes, the two dual-core processors in a single package is a minor cheat, but it's not meant to be the only quad-core Intel produces. It simply let the consumer have a quad-core chip now which will help push software manufacturers produce programs that utilize the processing resources made available by quad-core. Its an easy bet that Intel already has a true quad-core design in the pipeline and it will be out to consumers in due time. So stop bashing on Intel because they couldn't snap their fingers and have a working quad-core appear on shelves. After all, Intel's solution is a hell of lot better and less expensive than AMD's quad-core solution (2 dual-core chips in the same system).

Re:Still on the FSB

[Joe The Dragon]

I am talking about amd's up coming single cpu quad-cores

Re:Still on the FSB

[Aadain2001]

And Intel has up coming signal chip quad-core CPUs. To think they don't have one in their pipeline would be ignorant.

Re:Still on the FSB

[coredog64]

It doesn't have to be pistols -- it could be mortars.

Re:Still on the FSB

[networkBoy]

I think positrons would be more app. in this instance, no?

Re:Still on the FSB

[networkBoy]

you know something funny?
NetBurst was not a mistake...
I have two high end machines. A Prescott 3.6 GHz machine and a core2duo extreme edition machine. the core2duo is faster in all apps, yes. In games and such it kicks ass, yes. In the one app where I *really* care, repetitive mathematical transforms the core2duo is faster, by only 5% and it has double the ram of the Prescott, interesting eh?

Now while I will gladly concede that if you're a gamer the Prescott was abysmal; for scientific, encryption, and encoding applications it really kicked ass.

Now since that is fundamentally the only thing I've used the P4 for since I've built it I have to come to the conclusion that the only reason to buy a core2duo was heatloading.
(though being winter I kinda appreciate the wattage, at least I'm doing something with the damn electrons on their way to becoming heat, unlike my spaceheater)
-nB

Re:Still on the FSB

[fitten]

Alternatively, you could have one core computing as fast as it can and the other one "uncomputing" what the first one did... ;)

I wonder if they have tiny banjos.

Windows and Vista?

[solevita]

news that next-gen Penryn core processors are running various versions of Windows and Vista successfully

I know Microsoft told us that Vista was new and all, but I didn't know it was this new!

Re:Windows and Vista?

[markdavis]

news that next-gen Penryn core processors are running various versions of Windows and Vista successfully

Actually, the article does not say "various versions of Windows and Vista successfully". It says: "Intel said it had already manufactured prototype microprocessor chips in the new 45-nanometer process that run on three major operating systems: Windows, Mac OS X and Linux."

Yes, you read that right.... they actually said "Linux". :) Of course, it is not a surprise it can run Linux (that is a major "duh"), but it is nice to see it in print, next to MS-Windows and MacOS 10.

FTFA

[Aaron England]

This is a show of strength if you will, and an impressive one at that. How impressive? We'll wrap-up here with a few quotes from Gordon Moore (Intel), Professor Dimitri Antoniadis (MIT), and Yoshio Nishi (Stanford) telling you what they think of Intel's achievements.

"The implementation of high-k and metal gate materials marks the biggest change in transistor technology since the introduction of polysilicongate MOS transistors in the late 1960s" - Gordon Moore

"The Intel 45-nm CMOS technology marks a historic milestone for the semiconductor industry. Similar to the transition from single metal (Al) gate to polysilicon gate that has allowed optimal nFET and pFET design, the introduction of dual metal with high-k-insulator gate-stack opens the path for optimal design of both types of FETs, at insulator thicknesses necessary for continuing device scaling that are impossible to reach with the industry-standard silicon-dioxide-based insulators. Many options of high-k gate-stacks have been the target of intense industry and academic research for many years now, but Intel's demonstration of a manufacturable dual-metal/high-k solution is a remarkable first." - Prof. Dimitri Antoniadis

"It is a huge break through to replace more than three decade's long successful polysilicon gate technology with a new high-k+metal gate technology. Though the combination of high-k dielectrics and metal gate electrode for advanced CMOS has been extensively studied by many researchers around the world as the ideal MOS gate structure, the technical hurdle to bring the technology to manufacturing floor has been believed still too high for the 45nm node. As a researcher in this field, I am pleasantly surprised by the announcement and would like to congratulate Intel researchers for their success that Intel has demonstrated 45nm microprocessors with their high-k and metal gate technology. Even though specific metal and high-k material have not been disclosed at this moment, this is a revolutionary step toward the world of sub-50nm CMOS integrated circuits, as this new technology will drastically improve transistor performance in all fronts of electrical specifications, resulting in significant improvement of IC performance." - Yoshio Nishi

Re:FTFA

[stevesliva]

"The implementation of high-k and metal gate materials marks the biggest change in transistor technology since the introduction of polysilicon MOS transistors in the late 1960s" - Gordon Moore

I don't know, I may have to disagree with his eminence on this one... but, parsing his statement a little more finely, he does specifically mention transistor technology. I still view this as an evolutionary refinement of CMOS, and not as big as the transition from bipolar to CMOS for mainstream processors. It is a bigger deal than strained silicon. And by specifying "transistor" he of course sidesteps the comparision to the similar chemical reengineering of the back-end metal stack that came with the introduction of copper and low-k dielectrics. I speculate that like those changes, though, the implementation of the process changes will be relatively transparent to the circuit designers. Yes, the models change, hopefully the subthreshold leakage goes way down, but you've still got the G,D,S,B paradigm... right?

Re:FTFA

[WaZiX]

Yes, the models change, hopefully the subthreshold leakage goes way down, but you've still got the G,D,S,B paradigm... right?
riiiiight!

Re:FTFA

[warrior]

I think this is just a lot of Intel FUD. There are other semiconductor companies that will have similar improvements in their designs as well. One that I think is most interesting that Intel will not have is called HOT - Hybrid Orientation Technology. This is only possible for the vendors that use an SOI process. Basically, HOT allows for the Si crystal to be oriented in two different ways, one way for PFETs and one for NFETs. What this means is that we no longer have to choose which FET is fastest in a given process, P and NFETs have the same gain, NOR gates are as fast as NAND, etc. Now you don't have to make PFETs more than 2x the size of your NFETs to get the needed drive and your design gets more dense - smaller, cheaper die or more on-die cache. You also get to keep one FET type in SOI, and the other in bulk to get the best of both worlds (fast SOI fets and predictable bulk FETs with fixed body voltages). Intel may be able to create some form of HOT for their fin-FET, until then they won't get HOT because they don't use SOI due to NIH (Not Invented Here).

Re:FTFA

[Bender_]

I can tell you this is not FUD. If you follow the literature about this subject closely you will notice that there is no report about a metal usable for a pMOS transistosr. Many companies are not significantly ahead of that. I expect some pretty interesting publications later this year. But it still takes several years to get a new material into production. Intel has already made big strides in putting high-k into production while other companies may still be in the screening process. I believe they have a head start of at least one year, if not more. Note that IBM announced the usage of High-k for 2008, at least a year later than intel.

Re:FTFA

[Bender_]

also: HOT and SOI are quite expensive, it saves a lot of money (at least now) to engineer around those solutions.

Re:FTFA

[warrior]

The cost of a 12-inch bulk wafer is about $400, an SOI wafer $600. The back-end processing for each is $2000+ ( slightly more for bulk for additional junction engineering to get the additional performance ). The cost difference is about a wash. I guess SOI does take a cost hit when it comes to yield. I'm not trying to take anything away from Intel's accomplishment, but they're not the only ones working on exciting new technologies. I think they're just shouting about this because they're the first ones to metal gate and they want to make up for getting beat to SOI ( and then being stubborn about it and not using it, I'm pretty sure Intel had some great SOI tech but won't use it out of spite ).

Re:FTFA

[Bender_]

AFAIR Intel presented some interesting technologies studies on SOI a couple of years ago.

Your wafer cost figures are off, but I will rather not comment due to lack of public sources. Still, in most parts of the semiconductor business 10% cost difference is not a wash but determines whether you make profit or not. It may work for a short while if you have a very high margin product (like a technologically superior CPU), but then you are suddenly in a price war...

Re:FTFA

[Mycroft_VIII]

I doubt it's out of spite, if it would make them more money they'd do it. They are a publicly traded company and as such the focus is profit.
          They may have adopted the N.I.H. attitude in for pr reasons e.g.: "We're Intell, we lead not follow!".
            Or perhaps they felt the cost at the time was prohibitive relative to the advantages and by the time the tech was cheap enough to implement they already had something more promising (perhaps this tech?) on the near horrizon.
            Not saying they made the right or wrong choice, just that I doubt the board members sat around saying "they invented it so I don't want it no matter how good it is."

Mycroft

The biggest advance

[Anonymous Coward]

I don't know why everyone is going gaga over this processor technology, when it's clear that the biggest advance is that they've made Windows run successfully.

Must have taken an army of late night patch coding wizards.

Re:The biggest advance

[rbanffy]

Wish I had modpoints. +1 funny for you.

I mainly find it funny because I usually don't run Windows. If I did, I would feel something a little bitter added to the funny mix, kind of a Dilbert strip.

Increased Processor Speed Needed?

[Anonymous Coward]

The need for an ever increasing processor speed is needed either if you video edit, work on graphic intensive apps and gaming.

I do not why the majority of computers need to be *upgraded* to faster chips just because computers need to be *faster*.

In terms of speed, office app productivity hasn't changed and will be the same for years to come.

Re:Increased Processor Speed Needed?

[eebra82]

I am not sure what you're trying to tell us? I don't think anyone is in doubt that newer CPU:s are typically for high-end users, but Intel is also targeting power efficiency among other things. Plus, some people also want to upgrade because using old hardware isn't very reliable.

Obviously, Intel must advance with more powerful processors to compete. If not, there would never be any vast software development either. Surely you must admit that running Office 2007 on a six-year-old computer isn't a very good idea.

Re:I would like to be the first to say

[Aristocat_Faust]

is that you, Jeff K?

Re:Who cares about clock speed, just overclock

[ratbag]

If you're a company or uni then overclocking your servers (definitely) or desktops (probably) really isn't an option. Even as a consumer, surely you'd be wondering if it's worth the potential warranty issues, especially if you've got to stump up for "higher grade memory and a non-crappy motherboard"?

Re:Who cares about clock speed, just overclock

[Talrinys]

That might apply to desktops but laptops don't have cooling systems even close to modern desktops. Besides these chips will run a lot cooler no matter what, so for the overclockers it will have a lot of extra potential too.

Re:Who cares about clock speed, just overclock

[Anonymous Coward]

All of that "free speed" comes at the cost of a higher grade of memory (adding $100+), a "good" motherboard instead of the ones you can buy cheaply bundled with a CPU (adding $100+), a higher quality heatsink and fan than the stock one (~$50), and a high end PSU ($60+ more than the standard type). Now you have to ask yourself, after you've spent all this money on higher quality components, wouldn't it just be easier to buy a higher-rated cpu in the first place, and have a *guaranteed* higher clock? After all, even though you can overclock some E4300+ to >3GHz, that doesn't mean that _every_ one will be able to do it. Also, as another poster mentioned, if something breaks, good luck with your warranties.

Re:Who cares about clock speed, just overclock

[MoxFulder]

A production E4300 running at 1.8 GHz can be run at 3 GHz on stock air cooling.

Didn't you get the news? The clock speed race is sooooo 2005. For everything but the most CPU-bound number-crunching applications, increasing clock speed is no longer very desirable.

Today it's all about PERFORMANCE PER WATT (crucial for server farms and portables) and on-chip parallelism/SMP (useful for everything from desktop GUIs to web serving to RTOS embedded systems).

Re:I would like to be the first to say

[dagamer34]

O_o

amd watching?

[Anonymous Coward]

where is amd now?

they still have to catch up with core duo, now this...

high capacitance

[dlenmn]

In the article, they keep on talking about high capacitance as if it's a good thing, but I was under the impression that you want to minimize the capacitance to let the transistor switch faster. Am I wrong? Is the article wrong? Is this a different capacitance that they're talking about?

Re:high capacitance

[NMerriam]

In the article, they keep on talking about high capacitance as if it's a good thing, but I was under the impression that you want to minimize the capacitance to let the transistor switch faster. Am I wrong? Is the article wrong? Is this a different capacitance that they're talking about?

I interpreted it to mean that the high capacitance referred to the low level of leakage -- ie, more energy stays inside the path as productive power rather than radiating as heat or creating static interference with nearby paths. Resistance should be the same or lower, allowing the same charge/discharge speeds. Although even if it were higher and the switching was slower, the increase in performance per watt (since you're wasting massively fewer watts) might be worth the tradeoff.

Re:high capacitance

[WhiplashII]

In a CMOS-like technology, you want overall capacitance as small as possible - because you have to charge and discharge that capacitor once per cycle. A first order approximation of power used is 0.5*Capacitance*(voltage squared). A first order approximation of the speed of the device is voltage*current/capacitance (where current is an exponential function of voltage). This means that from a circuit perspective, capacitance is the root of all evil (it both slows you down and uses up power).

When you look at it from inside the device, things change. Transistors work by having the gate-channel capacitor charge up and discharge. You need a certain amount of capacitance for it to work. An easier way to look at it might be to consider the electric fields instead of capacitance. A high K dielectric concentrates the fields in the channel instead of inside the gate oxide. This way, the fields are applied where they are useful (the channel) instead of where they are unneeded (the gate oxide). Obviously, this is somewhat of a simplification, but hopefully you get the gist of it - the capacitance is necessary inside the device, and by using a high-K dielectric you can concentrate that capacitance where you need it.

Re:high capacitance

[Anonymous Coward]

For CMOS devices, *some* capacitance is needed between the gate and the substrate so that a charge carrier channel can be opened when the appropriate voltage is applied to the gate.

Let's say you have a regular NMOS device, to open the charge carrier channel and make the NMOS conduct some current, you apply a positive voltage at the gate. The positive voltage applied at the gate attracts negative charge carriers on the substrate just beneath the gate, due to the capacitance between the gate and the substrate. The negative charge carrier here opens a conducting channel between the drain and the source and the NMOS is turned "on".

One of the problems with the current sub-90 nm processes is that, to work these increasingly shrinking transistors, you need increasingly lower gate voltages all around the transistor or else you'd burn the transistor. But you still have to attract a certain amount of charge carriers just below the gate to open a conducting channel. So what do you do? You increase the capacitance between the gate and the substrate. Capacitance can be increased in a few ways:

1. You increase the area - this does not make sense since we're talking about shrinking the transistors in the first place
2. You decrease the distance between the gate and the substrate - this is what's being done, with the side effect of huge leakage currents
3. You use a high-k material to separate the gate and the substrate - this is what Intel is talking about

So what Intel's plan is this - they'll use a thicker layer of high-k material between the gate and the substrate, so that the appropriate amount of capacitance is achieved between the gate and the substrate without making the distance between the two even shorter. A think layer of high-k material stops leakage currents without affecting other switching characteristics of the transistor.

Re:high capacitance

[Sebastopol]

Actually, drop the 0.5 if you are talking about a cycle, you have to assume the energy enters and leaves in a complete cycle.

Re:Who cares about clock speed, just overclock

[Anonymous Coward]

Didn't you get the news? The clock speed race is sooooo 2005. For everything but the most CPU-bound number-crunching applications, increasing clock speed is no longer very desirable.

Today it's all about PERFORMANCE PER WATT

How about putting it this way: Overclock a $175 E4300 to 3.0 GHz and you end up with the performance of a $500 E6700, at similar performance per watt.

But your performance/watt/$$$ is much greater. Simple enough for you?

So how many GHz?

[rbarreira]

So does this mean that the "future improvements will be on number of cores, not on individual core speeds" state of things in CPUs isn't true anymore? Anyone have any quotes on how much raw performance these 45nm CPUs will attain?

Re:So how many GHz?

[Wesley Felter]

It's still mostly true. Going from a 3.0 GHz Conroe to ~3.7 GHz Penryn is only a 30-40% performance increase, but we want a 100% increase. So we can expect that the number of cores and frequency will both increase.

Both

[try_anything]

Moore's Law says that massive improvements ARE progress as usual, but people have been so pessimistic about the future of Moore's Law that giving it a new lease on life counts as a major breakthrough.

Re:Both

[cnettel]

Actually, for a long time, the refinements were highly in the field of litography, decreasing the wavelength and so on. Performance made breakthroughs, the actually technology "just" progressed. The last few years, that's been coupled with a lot more materials research. True, that's been going on for decades as well, but lots of the supposed advancements during the 80s never made it. Now, it seems like we actually need to reevaluate every piece of the process to do those things.

Tour the Intel 45nm fab on video

[Anonymous Coward]

Scoble has video of the Intel 45 nm fab. Really great tour and interviews with senior technical fellows from Intel:

What low K will the Penryn use?

[cyfer2000]

TFA didn't mention anything about the Low K part, anybody has any information? Porous MSSQ? Porous SiLK? Porous Black Diamond (if it exists)?

Re:What low K will the Penryn use?

[Anonymous Coward]

Low-K material has been around for long time to build interconnect structure. AMD, ATI (TSMC), NVIDIA (TSMC) and Intel all use it now. High-K is used for faster and lower leakage transistors.

RTFA

Re:What low K will the Penryn use?

[Bender_]

This article offers some hints:

http://www.fabtech.org/content/view/2079

Re:What low K will the Penryn use?

[cyfer2000]

It looks like a new generation of Aurora. I guess they must introduced kind of pores in the Aurora. Or something else, I can't imagine the doped silicon oxide alone can reach such low K. But I also can't imagine how to make pores in Aurora too. Or maybe they simply figured out a way to keep the dielectric value low at small scale. Comparing to the porous MSSQ used in AMD/IBM 45nm product, I bet Aurora is better in reliability.

Re:What low K will the Penryn use?

[cyfer2000]

This article http://www.eetimes.eu/production/196701258 implies that the Black Diamond can reach K value of 2.4, but the mechanical properties is even worse than porous SSQ filem. Thus Aurora should be able to reach 2.4 too.

2.13 Ghz dual versus 4.5 Ghz

[Anonymous Coward]

would you rather a dual 2.13 GHz or a 4.5 GHz ?

By the way, by now we should be at 12 GHz, given we were at 1 Ghz in the year 2000.

1990 33 Mhz (486 DX 33mhz)
1995 200 Mhz (Pentium Pro)
2000 1 Ghz (Pentium III Coppermine and PII Xeon) (ie, from 1990 to 2000 clock speed went up 33 times)

2007 3 Ghz
2010 ?? May be 4 GHz .. if lucky 5?

By 2010 we should be at 33 GHz .. but it aint happening!

They may give us 8 cores at 4 Ghz instead .. but that's cheap crap, you can bet the compilers and apps for it will be donkey inefficient. I hope a competitor realizes the importance of instructions per second.

Re:2.13 Ghz dual versus 4.5 Ghz

[Ghoser777]

I hope you realize that clock speed is only one measure of the performance of a machine, and that increasing clock speed can trade off with other important factors.

Re:2.13 Ghz dual versus 4.5 Ghz

[dreddnott]

While Intel's ridiculously inefficient Netburst core might be the impetus for your post, that's not Business as Usual.

When Intel moved from the 80386 to the 80486, the FPU was integrated for the first time, as was onboard cache, and instruction execution rates nearly doubled on a clock-for-clock basis. The eventual clock speed of the 80486 was 100MHz, three times the speed of the fastest Intel 80386 at 33MHz (AMD released 133MHz and 40MHz versions of the 80486 and 80386, respectively).

Going from the 80486 to the P5 once again brought great improvements, such as superior FPU performance, dual pipelining, and doubled memory bus width, which gave the P5 once again nearly double the performance per clock over the 80486. The P5 (P54CS) topped out at 200MHz, twice the Intel 80486's top speed.

Things weren't looking so good with the Pentium II - it was basically a Pentium Pro, a very nice chip with superior performance per MHz than the P5, with its largest advantage cut in half (the large, full-speed L2 cache). Still, the Pentium III Coppermine brought the cache speed back up to full, and the P6 architecture eventually topped out at 1.4GHz, with far superior performance efficiency to the Pentium 4's Netburst core. In fact, Intel had to return to the basic design of the P6 to produce chips capable of defeating AMD's Athlon 64, the Intel Core and Core 2, the latter of which seems to squeeze more performance out of a MHz than any other desktop chip.

Re:2.13 Ghz dual versus 4.5 Ghz

[init100]

They may give us 8 cores at 4 Ghz instead .. but that's cheap crap, you can bet the compilers and apps for it will be donkey inefficient. I hope a competitor realizes the importance of instructions per second.

I hope you realize that there are some physical limits and constraints that cannot be broken, or can be broken at a massive disadvantage for other parameters. If I don't remember wrong, power consumption and therefore heat emission is proportional to the square or even the cube (at least nonlinear) of the clock speed at a certain manufacturing process. So you might have a 33 GHz processor (10x speed increase compared to what we have today) but the power consumption would increase from about 100 watts to 10 kilowatts. Not only would this bring a hefty increase in your power bill, imagine the cooling system required and the noise output.

An additional problem would be leakage current, since it increases with clock speed (or I'd rather guess it increases with temperature, which is a function of the clock speed). Too much leakage current, and the processor will cease to function. Leakage means that charges will not only travel along conductors, but jump between nearby conductors. Leakage not only increases with clock speed and temperature, but also with line density on the chip. The smaller the components, the larger the leakage at the same clock speed and temperature output.

I might be wrong in the details, but I think most of it is right.

Re:I would like to be the first to say

[Anonymous Coward]

mom? is that you?

Well then...

[Wicko]

Sounds to me like they went to reach as far ahead of AMD-ATI as possible, to keep their lead. God I love competition.

I have been convinced...

[Belial6]

I have been convinced for a long time that software bloat is not a problem. You touch on the reason. For the last decade, it has been cheaper to throw more hardware at a problem than it has been to optimize code. At some point in time, there will likely be a stall in speeding up hardware. When that happens we have a many years of continuing our computer speed ups via software optimizations. Heck, I know that I write inefficient code all the time. It is a simple cost/benefit choice. My clients do not want to pay tens of thousands of dollars to solve a problem that can be solved with $1000 worth of hardware. It's not that I couldn't optimize my code, and it's not that I wouldn't love to optimize my code. It's just the most companies don't want to pay for it.

Re:I have been convinced...

[cp.tar]

It's not that I couldn't optimize my code, and it's not that I wouldn't love to optimize my code. It's just the most companies don't want to pay for it.

I have encountered the same problem regarding good translations. I could do them, and I'd love to do them, but no-one in my country would pay me their worth.

Re:I have been convinced...

[Raenex]

At some point in time, there will likely be a stall in speeding up hardware.

Yes, that's been the case for at least 4 years now. That's why there's been the push for multiple cores. The Free Lunch is Over

Re:Who cares about clock speed, just overclock

[julesh]

that "free speed" comes at the cost of a higher grade of memory (adding $100+)

To be fair, you probably get a greater performance improvement from increasing the FSB speed than you do from increasing the processor speed, so this may well be worth it.

Re:Who cares about clock speed, just overclock

[julesh]

A production E4300 running at 1.8 GHz can be run at 3 GHz on stock air cooling

And if that's possible with a 65nm chip, even just the normal benefits of scaling suggest that 4GHz should be attainable from an equivalent 45nm chip. And it sounds like this generation is going to give intel better than normal improvements, so I'd expect to see 5-6GHz from Penryn cores with air cooling, assuming the data paths can switch fast enough (I expect Intel have designed it with some headroom, so this is likely to be the case). Plus that chip will have larger cache, so is likely to actually be able to make use of some of the extra cycles, rather than just sitting there waiting for the RAM to catch up like a 65nm model would at that speed.

Intel's more informative site

[Abtin]

The link in the main article is paraphrased from http://www.intel.com/technology/silicon/45nm_techn ology.htm

IBM and Intel both a anounced major breakthrough

[ysegalov]

You are missing the point here. IBM and intel, on the same day (Friday), independently announced a breakthrough in transistor design. Now isn't this strange? The biggest advance in transistors in the last 40 years or so - and two different companies announce it on the same day?!?!? Fishy.

Error in TFA

[dreddnott]

The article linked above refers to "Halfnium", with is both an element that does not exist and a gross misspelling of Hafnium , which is the new High-K replacement for silicon dioxide. It's also worth pointing out that both IBM and Intel announced this breakthough almost simultaneously, and AMD will reap the windfall benefits through its own partnership with IBM (they will move to the 0.45 process some time in 2008). AMD has also announced a low-K breakthrough that they will be implementing in their 0.65 process as well.

To give Intel sole credit for this breakthrough might be a little inaccurate.

Re:Error in TFA

[cattlebruiser]

> The article linked above refers to "Halfnium" I think that's a typo. The actual substance is called "Unobtanium" and was originally used to drill down to the center of the Earth.

Re:Who cares about clock speed, just overclock

[Anonymous Coward]

If an E4300 conservatively clocked at 1.8 GHz can be run at 3 GHz, then speedup is also possible with the Penryn. Overclocking is possible because chip manufacturers under-report a chip's performace just to play it safe and cover the performace variation introduced by the manufacturing process. No chip can be truly clocked above its critical temperature. The reported clock speed is a conservative, statistically valid figure, but many individual processors can be run above this. Also, chip manufacturers sometimes include higher-performace chips in a lower-performace batch just to decrease the supply of high-performace chips and raise their price. So basically, take the advertised clock speed with a grain of salt.

As this process is refined, I would expect the same sort of under-reporting to occur. If anything, a newer process would be easier - and riskier - to overclock because the process variation would be wider. I am not aware of the number of prototype Penryns Intel has, so I cannot comment on how tight their performance is. But this is a major change in transistor technology, while over-clocking is something of a gimmick.

Re:IBM and Intel both a anounced major breakthroug

[dreddnott]

I posted links to similar articles as a reply to the FP (an anonymous troll), but I hadn't read your article.

Initially I interpreted it as Intel and IBM cooperating on researching the new hafnium-based technology (this is the interpretation that Wikipedia's Hafnium article currently uses), but on further reading I realised that they were doing this quite independently, and AMD, Sony, and Toshiba were partnered with IBM on this research.

IBM may be basing the 4 to 6 GHz clock speeds of its new POWER chips on the implementation of High-K hafnium-based replacements for silicon dioxide. If they are, Intel may not be as far ahead as I thought. If they're not already using hafnium in the POWER6 architecture, this could result in some VERY scary chips from International Bull Moose.

they'll be friendly with existing boards?

[ystar]

so eventually can I drop a 45nm shrink into my c2d macbook?

Re:they'll be friendly with existing boards?

[maztuhblastah]

Nope. The Macbook's proc, like most laptops, isn't socketed.

Re:45nm ?

[Anonymous Coward]

Well, you should know since you've been sucking on it.

Re:IBM and Intel both a anounced major breakthroug

[edxwelch]

AMD will have a process with the low k thingy one year after Intel. I suspect IBM delibrately made their announcement on the same day to take the wind out of Intels sails, even though they're a year behind.

Re:Who cares about clock speed, just overclock

[eknagy]

Is there any chance that they will produce 1-Watt (silent), low-transistor count (cheap) Pentium 2 / Pentium 3 / SDRAM133 replacements with this new technology? Combined with Solid State Disks, plenty much of old hardware could be saved and lots of users (like my fiance or my sister) would be happy with a P2 or P3 class refurbished machine, provided there is at least 256 RAM to run a browser and some office applications.
Not to mention thin clients or second/third world countries.

Re:Who cares about clock speed, just overclock

[Doppler00]

Not a very good chance. They do that in the embedded space sometimes but usually they don't use brand new process technology for it. I mean, the Pentium 2, and Pentium 3 architectures are not as advanced. Why waste new technology on old architectures? It's more likely they will have embedded versions of Core 2 or more likely they will have an ultra low power 1GHz or 500 MHz processor. And honestly, for Intel there is no market for third world countries, so unless there is a billionaire philanthropist willing to invest in it (like the $100 computer) it won't happen.

Re:Who cares about clock speed, just overclock

[Doppler00]

Yeah, it's funny they are still pushing performance/per/watt. I certainly don't hope they are doing that on the desktop space. Consumers don't give a crap about power efficiency of their products. You can't even get people to replace 2, 60 watt light bulbs in their house, why would they care about performance per watt? It only matters for servers.

Also, there are several classes of applications and problems that cannot be handled well with multiple cores, no matter how much you wish it would. You could have a 3Ghz single core vs. a 2GHz 128 core, and the 3GHz machine may be faster. I think the GHz race will need to continue someday, just not with silicon. Quantum computers here we come!

Re:Who cares about clock speed, just overclock

[MoxFulder]

Yeah, it's funny they are still pushing performance/per/watt. I certainly don't hope they are doing that on the desktop space. Consumers don't give a crap about power efficiency of their products. You can't even get people to replace 2, 60 watt light bulbs in their house, why would they care about performance per watt? It only matters for servers.

Well, *I* for one care about power efficiency of my desktop. I've gone over to compact fluorescent bulbs (12 W, 60 W equiv) exclusively. I imagine anyone who cares about their electric bill will sit down and realize that their always-on computers consume several hundred watts (several times that when the AC's on and must remove all that waste heat).

Besides, performance per watt is about much more than the power bill. Higher performance per watt means more performance for less HEAT. And getting rid of heat requires larger desktop cases, with more noisy fans... greater performance-per-watt will allow us to have smaller, quieter cases. On a microscopic level, heat actually LIMITS the attainable clock speeds! So if you want to see higher clock speeds, then you'd better be pushing for better performance-per-watt as well.

Also, there are several classes of applications and problems that cannot be handled well with multiple cores, no matter how much you wish it would. You could have a 3Ghz single core vs. a 2GHz 128 core, and the 3GHz machine may be faster. I think the GHz race will need to continue someday, just not with silicon. Quantum computers here we come!

Hmmm... I can't think of a single REAL WORLD task that is so massively un-parallelizable that I would prefer a single 3 GHz core over many 2 GHz cores. Of course there is software that hasn't yet been optimized for parallelism, but there are few real-world problems that resist it entirely.

I do agree that the GHz race will continue at some point in the future, with silicon even. But for now performance-per-watt is a more significant bottleneck in terms of using computers efficiently. This is trivially obvious in the case of server systems. And even on my desktop I can't think of many tasks that I do which are CPU-bound... except for some heavy matrix crunching with Octave. And that, of course, is easily parallelizable. I'd much rather have a few more cores than a few more GHz.

Re:Who cares about clock speed, just overclock

[eknagy]

It's not a waste - from one "new technology" processor you can make 4..16(?) P3-compatible processors, as they are more simple, consist less transistors and even -maybe- you can re-design them a bit better. So, think about 50 USD processors and 2 G potential customers and not about 400 USD processors and 10 M potential customers.

Plus, all those users who *like their machines as they are* and *do not want to migrate to new computer/OS/technology* and *want to keep their system working* might pay for it - to keep their legal Win98 with their legal accounting program running on a Pentum2, as my friend (sure, he also has a P4 laptop), but without noise (plus more RAM, a solid state disk, a CD=>HD replacement?, a passive cooler and a backup in one 200 USD refurbishment package). Maybe a new TFT and mouse, altough he would prefer VNC, I guess.

Yes, they could migrate from 933 Celeron to 1200 Non-Celeron, but then they would need bigger fans and it would cost a lot (you see, they have to pay somebody to replace it, because they can't change it for themself now). And maybe a PSU replace was necessary.
And of course, you need to replace your motherboard 3-4 times to migrate from low-end P3 to high-end P4 (crazy, isn't it?).

Yes, some of us Java programmers would not buy it, but maybe lots of Chinese mothers will (would?).
And enviromentally educated people like my Fiance (translator) and my sister (lawyer). They don't need anything fast, they said "this is better, I can work 5-10% faster" when I replaced their 200 MHz and 400 MHz CPUs with a 900 MHz and a 1800 MHz. Because - guess what - even the 200 MHz CPU was usually waiting for them or for the disk.
And you can develop your Java on servers from a P1 with ssh/Remote Desktop/X/VNC fairly well.

And please imagine a classroom with 24 fan-triplets less than now. And please calculate the difference between 24*200 and 24*1000.

Let's face it, Pentium 3s and ThunderBirds are strong enough to run even Vista with a browser and some office applications, it is enough in itself to code SQL, C or even some Java - we don't really need stronger PC CPUs. Everyone knowns that, that's why they are moving towards Mobile, Media and Multi-core - of course, Vista SP2 might help it to increase the CPU intake.
From 4 MHz, we got to 4 GHz - I believe that we use about 75% of the difference to waste.

And yes, I would like to upgrade my old Web server's P3 processors to some faster stuff that is cheaper than gold (in weight) and I don't want to dump the whole server just to upgrade processors (and it's okay for the given traffic). And yes, there are old, slow 9 GB disks inside, and because they are in a RAID and they are SCSI, and the web pages are cached in memory, they will remain there for another 10 years, unless something brakes.

So, it is not a question of "waste of technology" - rather the question of "waste of nature" and "waste of CPU power / electric power / fresh air".
And maybe a question of "why we let them create so many incompatible stuff and why we have not rejected all this RAM, CPU and GPU sockets we don't need"?
And if we did, why don't we ask them to produce compatible parts with more modern and enviromentally friendly technology?
Why do we dump our computers every year instead of using it for 10 years, replacing and upgrading it every two years as it was in 89? Do we remember, when an IBM-compatible PC was modular?
And now? We replace the CPU, thus we must replace the motherboard, and then (AT=>ATX=>BTX) the case and the PSU, and yes, the RAM. The PCI cards and the disk can stay.

And whose children will clean all this up?

Yep, it's demagogue and it's off the mark.

Men, I need a blog. And a beer. Or a Bed. By.

Re:Who cares about clock speed, just overclock

[snarfbot]

well actually the e4300 has a 9x multiplier, so you would only need ddr2 800, or pc6400 ram to run it at 3600 mhz. which currently runs about 20-30 dollars more than pc5300 for 2 gigs, which is the minimum required for an e6600 for example which itself costs roughly 150 dollars more than the e4300, and only operates at 2.4ghz.

now say you bought the cheapest lga775 mobo you could find that supported ddr2 memory and a 266mhz fsb. maybe 50 bucks for a pos with integrated video and no expandability. fine. but for 90 dollars you can get a decently overclockable board, that will exceed the overclocking ability of your chip, and you can run any conroe at up to 3ghz on the stock cooler. so ill be conservative and say for less than 100 dollars more, you can get performance that exceeds that of the 6800 extreme edition.

so to reiterate

cheap pos mobo $50
e6600 $315
2gb pc 5300 $175
all else equal running at 2.4ghz no expandability for video and whatnot 540 dollars

asus p5b $130
e4300 $190
2gb pc 6400 $220
tuniq tower heatsink $60
all else being equal, with a pci-e 16 graphics slot. running at 3.6ghz at lower temps than a stock e6600, idle and loaded. for only 600, or 60 dollars more.

now you tell me what is the better value?

Re:Who cares about clock speed, just overclock

[Raenex]

I can't think of a single REAL WORLD task that is so massively un-parallelizable that I would prefer a single 3 GHz core over many 2 GHz cores.

Games. Go read some Carmack quotes. He wants a single CPU that goes faster, not many cores. Also, for my desktop I can't think of too many applications that will make use of more cores. I'm not confident that parallel optimization for your average application will occur. I'd definitely take a single 3GHz CPU over two 2GHz cores.

Re:Who cares about clock speed, just overclock

[Raenex]

Also, there are several classes of applications and problems that cannot be handled well with multiple cores, no matter how much you wish it would. You could have a 3Ghz single core vs. a 2GHz 128 core, and the 3GHz machine may be faster. I think the GHz race will need to continue someday, just not with silicon. Quantum computers here we come!

So, umm, what are these applications that can't benefit from parallel computers but that can benefit from quantum computers? It sounds like you think a quantum computer is just a really fast CPU. Not even close. Quantum computing is not useful for general computing.

Re:Who cares about clock speed, just overclock

[Anonymous Coward]

lame. you dont think that there'd be massive improvements in ai and physics if there were cores dedicated just to those tasks? carmack is a schmuck, whats the last game he wrote that didnt suck? quake3? quake?

I would like to be the first to say

[Anonymous Coward]

G0 FUCKING K1LL Y0URSELF FUCKTARD!!