Intel Readying Dual-Core Desktop Chip

sunisha.shah4eva writes "CoolTechZone is reporting that Intel is planning on introducing a dual-core Dothan chip for desktop computers. According to the article, Intel has plans to turn the performance table around with AMD. From the article: 'Finally, it looks like Intel has learned from its mistake and secretly prepping a surprise for the rest of the industry. According to the information we received, Intel is currently working on a desktop, dual-core Dothan microprocessor with SSE3 instruction set that Intel plans to launch sometime in the future. Whether the launch will take place this year or in 2006 is currently unknown.'"

Comments

2006?

[Tokerat]

...just in time for the Apple switch to Intel products?

I'm still kind of miffed about that but if they run new dual-core chips it might not be so bad.

Re:2006?

[dsginter]

...just in time for the Apple switch to Intel products?

There is something curious in the Apple deal and it is big.

Apple emphasized that they believed Intel's processor roadmap to be more impressive than any alternative. Now, Intel's current chips use boat loads of power because they haven't entered into a silicon-on-insulator deal with IBM, who owns the patent. AMD uses silicon-on-insulator to get their power consumption numbers wayy down relative to Intel's numbers.

Without getting into the details on why Intel doesn't have silicon-on-insulator (IBM wants to "trade" instead of license...), one would think that AMD would have been a *much* better choice for Apple. But Apple's emphasis on the future processors leads me to believe that Intel has something *big* up their sleeve. Probably something to compete with the Cell processor, but on a much broader scale (i.e. - not focused so much on gaming performance).

I know that Intel have been developing Ovonyx memory technology for some time now (since 2000). It is interesting to note that in the process of developing the memory, they found that it has nonbinary processing capabilities.

Is Intel going to drop a bomb?

Re:2006?

[ocelotbob]

Nah. I'll just be counting the days until they port Mac On Linux to the x86. I give it 27 days.

Re:2006?

[Tokerat]

Oh and something i forgot in my other reply: PowerPC was originally faster, much, much faster than comparable x86 technology of the time. Unfortunately, we left it in the hands of Motorola to develop it. After all, good old Moto had served us well with their 68k line of CPUs since the early days and now that it was time to bring in the new it seemed only right to give them the chance to accellerate their design, fabrication, and sales with this newfound holy grail of technology.

Instead, they just kept doing what they where doing before, and the rapid design changes in the Intel and AMD lines quickly started kicking our asses, to the point where PowerPC sucked.

IBM took over for the G5, and it was amazing. New plant, new core, new company, AMAZING throuput at much lower clock and pretty damn low power consumption. Couple that with an OS designed for it and there you go.

To this day, a Dual 2.5GHz G5 still pounds a 3+GHz P4/AMD whatev into the ground (unless you're gaming). Problem is, new plant, new core, new company. IBM is having fab problems and can't push the chips much higher for a time to come. Apple, who have learned harsh lessons about not having a backup plan, have been watching Intel for some time now and see that they're on the move with something big, real soon. Something they asked IBM for and IBM couldn't deliver, at least not for a while. By that time, x86 will have moved ahead to the point of EASILY and OBVIOUSLY eating the PowerPC 970 series for breakfast.

Motorola is still, by the way, the manufacturer of the G4, so any machine that hasn't got a G5 in it is still hindered by their asses.

All in all, it's a big mess, and to move to a unified manufacturer who is willing and able to be accomodating, can meet demand for high volume orders without problem is a godsend. Apple is too big now to say "Wait 5 years and we'll be ahead again!", they have to stay on top and moving ahead at all times because if they lose steam now they could collapse.

This is a big move for them and not to be taken lightly. I only hope they're doing what is right...I'd love to get my hands on the developer kit, that's for sure.

About time to let Netburst die

[orionpi]

I'm currently running a Dothan on my desktop (ASUS CT-479 + P4P800-VM), it makes it very clear that Netburst has been dead for some time now. Intel has been milking a dead but very expensive cow, and will continue to do so for as long as they can.

Current Intel Dual Core DOA

[the eric conspiracy]

Sounds like a good idea to me since I have already ruled out current Intel dual core designs because of their outrageous power consumption. AMD Athlon 64's are much better in this department except they are awfully expensive right now. A more economical dual core Dothan design would definitely be something I would be interested in.

Open Opportunities

[Doc Ruby]

I bet Intel's people wish that all software could be recompiled on installation, to target the specific tweaks they put into a certain chip model. Instead of waiting for the OS or app vendor to recompile for an optimized binary distribution, which rarely happens. Of course, that depends on open source...

Re:Didn't AMD shoot down Intel's "dual core" claim

[Anonymous Coward]

Well... uhh, no. The Pentium D is an ugly hackjob, they are technically dual core but don't really have any design features that actually take advantage of that fact. This is why they are readying a *NEW* dual core chip.

I didn't think this was actually news, info about Dothan being released in mid/late 2006 has been floating around for a while now.

Can Apple Contribute to Intel?

[camperslo]

If Apple gained rights to some technology when Motorola and IBM didn't deliver, perhaps they could bring Altivec to Intel?

Benchmarks

[akuma(x86)]

Tom's Hardware has some interesting benchmarks with a Dothan in a desktop system with a halfway decent memory system.

http://www.tomshardware.com/cpu/20050525/pentium4- 21.html

Why Apple Couldn't Consider AMD?

[nazgul000]

A curious fact about Apple's choice of Intel over AMD, as I learned over on the Ars Technica forums -- AMD's CEO, Hector de Ruiz, was formerly the director of semiconductor products at Motorola.

I think this is one big reason why Steve Jobs and Apple could not / did not consider AMD -- they notoriously burned their bridges with Motorola/FreeScale over the G4's lackluster performance and slow development. Thus, Jobs and de Ruiz probably don't have a particularly good relationship.

There's a chip on your shoulder (snicker)

[5n3ak3rp1mp]

Actually, for anyone who cared about such things (chip geeks), the popular consensus WAS that PPC's WERE better than anything in the x86 camp. That is, during the G4 era. The instruction set was much saner (even Intel fans will complain about the bass-ackwards quirky x86 instruction set), it pushed more numbers with far less power, AltiVec showed a ton of promise (if you were willing to either wait for a good compiler or use the vector unit by hand).

With the introduction of the G5 and the failure on the promise to break through the 3GHz barrier without incurring much larger power requirements, the situation probably began to change, but us Apple Believers, admittedly, chose to ignore this slowly-dawning information. Just like those stubborn Windoze users who only surf the web and check email, and reinstall Windoze every year and spend 2 hours a week disinfecting and have pieces of apps lying around on their hard drive that failed to successfully "uninstall" (a concept foreign to OS X users), and STILL believe that their choice is cheaper and/or more effective than getting even a used Mac for the job. Their time must not be worth a damn thing. I know, because I tried to convince just such a non-technical person to buy a used Mac (since I put equal time on Macs and PC's and knew what was best for this person), but they insisted on a crappy PC laptop, and then had the nerve to call me over for free tech support... Objectivity is hard to come by all around.

Maybe we liked having a different processor because it was a different TAKE on things. It was outside the box. And certainly, every last one of us understood that it added COMPETITION to the market. Competition is good for everyone. Something else you Wintel fans seem to not care about or understand, as you freely throw your money at an industry with a leader who is a convicted monopolist. You should be kissing AMD's ass that they lit a fire under Intel's butt, because around the year 2000, it certainly did look like PPC was going to hand Intel's ass to it. (And of course, if Microsoft didn't consider open source a "threat", it would have zero incentive to change, either. Why improve when you can market instead and charge as much as the [exorbitant fee just under what would force people to buy elsewhere because it's the only game in town]?

I can't believe I even devoted this much thought to your jerkitude.

What this REALLY means

[Anonymous Coward]

Nobody is surprised Intel is making a dual core Pentium M with SSE3. This is old news called Yanoha.

What is big is that Intel will finally support the platform on the desktop. That means offering an Intel branded motherboard and Pentium M chips clocked for the desktop thermal envelope. It also means much better prices on boxed Pentium M processors.

Pentium M chips can already be overclocked 50% and in fact need to be if you want them to compete with real desktop processors. For many people overclocking is simply not an option. Even at 50% overclock the chips are barely warm meaning Pentium M has the potential to make a huge leap in performance with some minor tweaking.

Also motherboard support is nonexistent at best. Motherboards for Pentium M are outrageously priced (~$250) and come from unreliable manufacturers. No IT department is going to fill the office with AOpen hacks, they want Intel brand mobo in their white boxes. Finally, the chips themselves are also outrageously priced and barely a deal after overclocking. Most boxed chips are more for development applications so this is to be expected. Bringing Pentium M to the consumer desktop market will thankfully drive prices down closer to Pentium 4 prices.

On a side note, Pentium M has yet to make the transition to 64-bit. With Microsoft moving forward on Windows x64 and Apple moving OS X to Intel's mobile chips first, Intel needs Pentium M 64-bit support NOW. Apple can't go on supporting 4-way fat binaries with 32/64 PPC and 32/64 x86.