Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Upcoming 45nm Xeon Processors from Intel

Wow, I just need to wait another 32 years to get my MacBook Pro with the 2.5 watts 64 core chip and no fans!!!!!!!!!!!!!!!!!!!!

Now, where the heck I left that (hope) stool?
Wonder if I live to see it.:eek:
 
CalBoy said:
I don't think this will happen soon. Notebooks have to be designed with power in mind. A quad-core would drain a lot more power than a dual-core.
Click to expand...
Not necessarily. Depending on the clock speeds and other capabilities, it might not. Keep in mind that Intel already has tech that can power-down idle cores and even functional units within a core. These configurations could all be controlled by firmware, configured by the Energy Saver preference panel.

I distinctly remember reading about Intel working on a "Core 2 Quad" for mobile/desktop use. It won't be a Xeon, but nobody is expecting to put one of those in a laptop.
MrCrowbar said:
6 GHz = badass power consumption and heat. Apple did good.
Click to expand...
POWER is not PowerPC. Nobody is making PPC chips at those speeds.

And for the market where POWER is being sold (high-end workstations, mainframes, and other big-iron), the power consumption and heat is likely acceptable, especially when compared against the previous generation of chips used for those kinds of products.
mickeymikey said:
Couldn't agree more. I love my Apple gear, but I feel like I'm caught in Goldie Locks and the Three Bears. The MacPro is "too big" and the iMac is "too small". For my needs, an "in between" product would be the way to go...just make sure it has enough video power to drive the 30".
Click to expand...
That's one way to look at it. Another is to say "I've budgeted $x for a new computer" and look at how much computer you can get for your budget, without regard for today's need (working with the assumption that your needs will grow over time anyway.)

When I bought my PowerMac G4, I spent about $3500. If I needed to replace it today, that same $3500 can get me a nicely equipped quad-core Mac Pro. If I don't need to replace it until next year, it will buy even more of a computer. Do I need that much power? Probably not, but if I'm within my expected budget, then it's not really a problem. It just means it'll be a long time before I need to upgrade again.
 
mahonmeister said:
I wonder if AMD will have any processors that can compete with intel in the high end by the time Penryn hits the market.
Click to expand...

Expect to see headlines about how wonderful Barcelona is, with great benchmarks on scientific floating point apps.

Don't expect to see great numbers for dual socket Barcelona systems running multi-threaded or IO intensive applications. AMD's architecture of using two memory controllers connected with a serial bus has some issues when the program runs on one socket and the memory happens to be connected to the other socket.

Those great scientific benchmarks are achieved by running 8 separate copies of the program, with each copy and its memory forced to a single CPU. (Which is not unreasonable for scientific cluster computing, by the way, it's just not representative of lots of other interesting applications.)
 
RBilRamZ said:
Honestly, how would this merit a negative rating?
Click to expand...

My guess: A bunch of Anti-Intel, pro PPC nutcases. I still run across one every so often in the flesh. It it quite remarkable that this species of Mac fan still survives. I mean the proof is in the enhancements. IBM or freescale would never be moving at this pace, nor would the switchers who are virtualizing Windows to allow them to go both ways. Apple would never be where they are right now without the transition to x86. And yet there are still people who think it was a bad idea. :rolleyes:

AidenShaw said:
Expect to see headlines about how wonderful Barcelona is, with great benchmarks on scientific floating point apps.

Don't expect to see great numbers for dual socket Barcelona systems running multi-threaded or IO intensive applications. AMD's architecture of using two memory controllers connected with a serial bus has some issues when the program runs on one socket and the memory happens to be connected to the other socket.

Those great scientific benchmarks are achieved by running 8 separate copies of the program, with each copy and its memory forced to a single CPU. (Which is not unreasonable for scientific cluster computing, by the way, it's just not representative of lots of other interesting applications.)
Click to expand...


the best quote I've seen from AMD....

To add insult to injury, when DailyTech benchmarked the pre-production 1.6 GHz Barcelona, the CPU did not match Intel's 65nm quad-core offering clock-for-clock. AMD engineers stress to DailyTech that this benchmark was premature, and that final silicon and software will allow for SSE optimizations and better performance.
Click to expand...

The scary part is that you could almost directly pull that quote from Intel during the dark days of the P4. Its all smoke and mirrors.
 
Penryn Pricing

Pricing revealed

X5460 (3.16GHz) $1,172
E5450 (3.00GHz) $851
E5440 (2.83GHz) $690
E5430 (2.66GHz) $455
E5420 (2.50GHz) $316
E5410 (2.33GHz) $256
E5405 (2.xxGHz) $209

Current Prices:

X5355 2.66 GHz $1,172
E5345 2.33 GHz $851
E5335 2.00 GHz $690
E5320 1.86 GHz $455
E5310 1.60 GHz $316

5160 3.00 GHz $851
5150 2.66 GHz $690
5140 2.33 GHz $455
5130 2.00 GHz $316
5120 1.86 GHz $256
5110 1.60 GHz $209

July 29th Pricing:
X5365 3.00 GHz $1,172 (not comming until August 12th)
X5355 2.66 GHz $744
E5345 2.33 GHz $455
E5335 2.00 GHz $316
E5320 1.86 GHz $256
E5310 1.60 GHz $209
 
AidenShaw said:
Expect to see headlines about how wonderful Barcelona is, with great benchmarks on scientific floating point apps.

Don't expect to see great numbers for dual socket Barcelona systems running multi-threaded or IO intensive applications. AMD's architecture of using two memory controllers connected with a serial bus has some issues when the program runs on one socket and the memory happens to be connected to the other socket.

Those great scientific benchmarks are achieved by running 8 separate copies of the program, with each copy and its memory forced to a single CPU. (Which is not unreasonable for scientific cluster computing, by the way, it's just not representative of lots of other interesting applications.)
Click to expand...
Sounds a lot like how Apple trumpeted the performance of the VT G5 cluster when it was built.
 
mhar4 said:
Unbelievable. With PowerPC heading towards 6Ghz, could these so-called "Penryns" be any clearer statement that moving to Intel was the worst decision Apple ever made?
Click to expand...

the Power 6 is in a class of it's own. it's a supercharged multi-core cpu design for servers and dedicated super computing clusters that can theoretically break all processing records by huge margins. it's also bloody hot and not designed for a desktop.

intel has the right idea with quad cores anyway - fast, cool, portable, available and designed for future multi threading.

IBM is only interested in cutting the time of a nuclear fusion test model from 1 year to 3 months, and to possibly replace all existing internet hardware exclusively.

Intel just wants you to stream GrooveSalad, burn a DVD and run a blur filter on a 70MB file at the same time without slowing down... on a good day even a Strip Saver...
 
ffakr said:
There will be no quad core laptops any time soon.

The Power6 is related but it's NOT a PowerPC chip. You can't run Macintosh PPC code on that processor. They're 2nd or 3rd cousins not twins. :)
Power6 [and Power5] are monsterous chips.
ffakr.
Click to expand...

Read up on the difference between POWER and PowerPC before you sprout this rubbish. POWER since POWER3 has been implemented the full PowerPC instruction set i.e. anything PowerPC can run, POWERx can run as well. Linux distributions for PowerPC and POWER are the same. Only reason it will not able able to run MacOSX is due to the "firmware" that is controlled by Apple, not the instruction set.
 
thunng8 said:
Only reason it will not able able to run MacOSX is due to the "firmware" that is controlled by Apple, not the instruction set.
Click to expand...
And that this chip would catch fire if installed in any system as small as a Mac.

But let's ignore facts. Obviously, the instruction set is the only thing that matters. A mainframe is nothing more than a laptop in a big case, right?

Before you start abusing others for being less than perfect, you ought to make sure you also measure up to those standards.
 
So, where exactly can Apple buy 1 million mobile Power6 processors for next quarter's MacBook and MacBook Pro sales for Intel-competitive prices?

The CPU is the heart of the product and Apple needs reliable supply at competitive prices in order to stay in business. Look at the LED display shortage. If it gets worse, Apple could lose $100m of sales next quarter. Intel is not only the most reliable supplier, but they also happen to have the best product at this point.
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back