News
 

Bookmark and Share

(7) 

Intel Corp. will describe in details its next-generation micro-architecture for x86 microprocessors at the upcoming Intel Developer Forum (IDF) that kicks off next week, according to the company’s plans. While not many sessions will be dedicated to the new technology, the world’s largest chipmaker has already said that the new chips will be 20% faster their counterparts from Advanced Micro Devices.

Claims reportedly made by Mooly Eden, who is vice president and general manager of Intel’s mobile platforms group, suggest that Intel’s code-named Conroe and Merom processors would deliver performance advantage of at least 20% when compared to speed of AMD processors planned to be released in the second half of 2006.

“We believe we’ll be able to open a major gap with the new processors,” Mr. Eden is reported to have said about performance difference with AMD processors. “It will take at least a year and a half to two years to close such a gap.”

All the peculiarities of the next-generation micro-architecture (NGMA) are not clear at the moment. Though, from various official statements and unofficial information leaks, it can be summarized that the new micro-architecture will utilize shorter pipeline and high performance per clock ratio. The new processors will feature 14-stages pipeline, down from 31 or more stages found in current Intel Pentium (Prescott) designs, 4-issue out-of-order execution engine as well as improved performance of the floating-point unit (FPU). Initially, Intel will offer desktop processors clocked from 1.86GHz to 2.93GHz, but later on the clock-speed should improve. Also, the new chips and platforms on their base will also feature capabilities like virtualization, LaGrande technology, x86-64 in addition to EDB, EIST and iAMT2.

According to Intel, the combination of the architectural changes, large cache of up to 4MB and fast processor system bus of up to 1333MHz will allow the world’s largest chipmaker to claim victory over the smaller rival, the chipmaker believes. But one thing that the new processors from Intel will lack is built-in memory controller, something that conditions AMD’s consistent wins in benchmarks.

An executive from AMD, who is going to introduce processors with dual-channel DDR2 memory controller later this year, said he did not expect Intel’s processors to be 15% - 20% faster compared to the Athlon 64 and Turion 64 products.

Discussion

Comments currently: 7
Discussion started: 03/02/06 04:03:13 AM
Latest comment: 03/03/06 12:47:35 PM
Expand all threads | Collapse all threads

[1-1]

1. 
A lot of the comments here ignore that the silicon is already out there, and from all indications it will be conservatively 20% faster than the Athlon 64, while using less power. It should be more.

Why do people get offended by stuff like this anyway? Intel has failed a lot, particularly with the Pentium 4, but this isn't a Pentium 4. The Pentium M has been consistently excellent, and this is more a derivation of that chip. The Yonah is already an extremely useful processor.

So, I know it hurts people that consider AMD part of their family, but it will be a better processor than the mediocre Athlon 64, and a Hell of a lot better than the awful Pentium 4. It's good for everyone, except for AMD, because it allows us to have better choices. AMD will still be around probably selling them for less since they will be inferior to the Intel product, like they were with the K6. They survived, and kept Intel prices in check because while AMD will have an inferior product, it won't be vastly inferior and most people don't need the fastest processors anyway. Power use will be a big advantage for the Intel processors though, and that is useful to everyone. Still, the Athlon 64 will survive as a value processor.

While the Athlon 64 is a mediocre processor, keep in mind that AMD is certainly working on their next generation processor too. Will that be better than the Merom? It easily could be. I don't know anything about it, but certainly AMD has something in the works. The AM2 that all some idiots were talking like it was a huge improvement appears to be a major dud, with the first benchmarks showing inferior performance to the current DDR solution. The final product will be better, but clearly this isn't a major improvement.

So, accept it. Intel will pass AMD and the planet will survive. It happened with the Pentium 4 too for a while against the Athlon. AMD survived. This will be a bit different, because Intel will have a faster and more efficient processor, but AMD will survive, come out with their new processor at some point, and it'll go on that way for a while. My major concern is AMD's lagging manufacturing capability. I don't think anyone outside of Intel wants AMD to fall by the wayside and become irrelevant. I don't think they'll ever fall too far behind in design, because at this point the designing part is pretty mature, but manufacturing is a big problem right now.

0 0 [Posted by:  | Date: 03/02/06 11:47:37 AM]
Reply
- collapse thread

 
Hurts people? Haha, that's funny, I highly doubt that, it takes alot to hurt my feelings and I'm pretty laid back as they come... Iit's just most people like myself have a problem with when I smell BS, it's kinda like smelling marketing schemes, they both usually stink.

Okay, I find this laughable:" the Athlon 64 is an inferior processor"??? If that is such the case then answer me these questions... Why doesn't Intel incorporate a on-die memory controller on the "Cores" that would give Intel's a major boost and totally overcome AMD in performance? Intel's complete marketing mentality of trying to keep thier bottom line on chipsets completely keeps them from doing it that's why! Why don't they update that relic of a memory system/FSB to the free, bandwidth superior solution and open design of Hypertransport??? Because again, thier marketing department RULES that company, they will never accept Hypertransport, conceding to thier arch-rival's solutions! They will go down in flames to produce thier own first!(as what has already happened once with CSI, it was originally planned for Tigerton but it got cancelled along with CSI and was pushed back to 2008!) Thier aging FSB can no longer hold a candle against the Hypertransport's bandwidth in Opterons setups in 4-way and especially 8-way. Why did Intel do a dual package processor instead of designing a true dual core processor from the start like AMD did with the K8? Becuase they got caught with their pants down again with the coming X2's and quickly sandwiched two Prescott cores into a flaming Pentium D so they can say ME TOO, ME TOO, WE'RE FIRST, WE'RE FIRST! There is a reason AMD is making HUGE strides in the server markets, desktop markets, eating into Intel's market share, gaining in stock price on the market and it's because of these FACTS.

While the Pentium M is a huge step forward for Intel compared to the dismal effort that was the Pentium 4, dressing up a Pentium 3(which was admitted by Intel's own people mind you) and give it some low power tweaks here and there with some performance enhancements does not equal what I consider a new "technology" or a "innovative" processor. Granted the same could be said of the Athlon 64 in some ways, but it's evolution is a huge step compared to it's predecessor the K7 thanks to the intergrated memory controller and hypertransport bus. IMHO Intel needs to start over from the ground up with a new microarchitecture if they want to regain thier previous "glory" period, but they know they don't have time to do it now, as AMD is on hot on thier heels marketwise. Perhaps in 2007 or 2008 they will finally get thier act together.

Contrary to yours and others opinions, AM2 will be far from a dud. The first benches were surprisingly done on Tom's Hardware(a well known lacky of Intel's Marketing department basically) with Tom admitting the motherboard was very cripped and with a engineering processor sample with obvious things still left to fix on it before production. Hardly a respectable measuring stick much less useful information reguarding it's performance at release date. FWIW, most benchmark programs are hardly anywhere relative of real world performance,(much less the roof it's done under)it's something I do actually care about when it comes to any kind of benchmark, anything else not independently done by a unbiased review site is worthless IMHO. So far we have yet to see the real numbers. Also, if you think AMD is just releasing this as a "stopgap" for DDR2 will also be horridly mistaken, alot can be tweaked between now and the date of release to gain performance, the 90nm shrinked Venice/San Diego core socket 939/754 processors are a testament to that. So if you think for two seconds AMD is going to let out a chip that will be inferior to it's predecessors, much less it's competition, I have some lovely beachfront property to sell you in Nevada.

About AMD's manufacturing capacity, FAB 36 has gone online as of this year producing 90nm and later on at 65nm. FAB 30 is still online cranking along with a deal with Charter that AMD signed last year to help keep up any unexpected peak in demand. Something most people don't realize either about AMD's fabs, they are some of the most advanced in the world with far higher rate of working silicon in comparison to most FABs due to the methods of thier tooling and very advanced machinery. This allows them to shift a process (like what happened with 130 to 90 nm) without shutting down the FAB and allowing for continued production even with high yields, something that surprisingly Intel has yet to learn how to do with thier many fabs. This is a big reason for alot of the increased Dell rumors this year and another reason AMD is making alot of big wins and deals with retailers/corporations since last year.

So in short, yes, AMD and both Intel will survive, that is without a doubt. But Intel's dubious claims as is your fool hearty views are yet to be proven in the real world experience. Thanks to Intel's continuing enviroment they promote within thier company's politics of letting thier marketing lead them, as opposed to AMD's ideals of letting the engineering team lead them, I surmise that things will be no different from this year than they were last year. Expect to see Intel loose more market share again this year and some people get surprised when the "Cores" aren't all that and a bag of chips when the AM2/F1/S1 sockets are released and in the wild. Thanks to Intel's long history of doing all it can to establish it's own monopoly within the semi-conductor industry and seeking to destroy all it's opposition in the market, along with it's legions of rabid fanboys touting them as the 2nd coming of Christ over the years, I can't say Intel didn't have it coming to them this time. They've been king of the hill too long and had it far too good at the expense of their peers in the market, people are now waking up and realizing this.(the recent anti-trust lawsuit against Intel has really made this aware too) Times are changing. IMHO it's time Intel stepped aside, end thier fruitless escapade of crooked monopoly practices and lackluster technology so more scrupulous, inventive, ingenious people in innovative companies can lead the way in the market, not just AMD.
0 0 [Posted by:  | Date: 03/03/06 09:39:17 AM]
Reply

[1-1]

Add your Comment




Related news

Latest News

Tuesday, July 22, 2014

7:38 pm | AMD Vows to Introduce 20nm Products Next Year. AMD’s 20nm APUs, GPUs and Embedded Chips to Arrive in 2015

4:08 am | Microsoft to Unify All Windows Operating Systems for Client PCs. One Windows OS will Power PCs, Tablets and Smartphones

Monday, July 21, 2014

10:32 pm | PQI Debuts Flash Drive with Lightning and USB Connectors. PQI Offers Easy Way to Boost iPhone or iPad Storage

10:08 pm | Japan Display Begins to Mass Produce IPS-NEO Displays. JDI Begins to Mass Produce Rival for AMOLED Panels

12:56 pm | Microsoft to Fire 18,000 Employees to Boost Efficiency. Microsoft to Perform Massive Job Cut Ever Following Acquisition of Nokia