Slashdot Mirror
Intel's New Chips, High Power And Low
sebFlyte writes "Centrino has been one of Intel's major successes of late, and they've just released the replacement, Sonoma. ZDNet has stripped the new chipset, and published a review of the new kit with all the technical details of what this new chipset will do for your laptop."
ZeroOne42 adds a link to Hardware Zone's exhaustive look at Sonoma, "complete with benchmark results between a Sonoma notebook (Fujitsu E8020) and a Centrino one (Gigabyte N512). Looks like Sonoma is closing up the technological gap between desktops and notebooks."
And on the desktop side, foxalopex writes "It seems that Intel's new dual-core CPU chips will have some of the highest wattage ratings ever seen on the X86 CPU market, which, according to Tom, wasn't what they initially said would happen. I guess this isn't too surprising seeing how AMD's been beating them on power usage in the last several revisions of chips."
Intel's latest portable computing platform is here. We lift the lid on the improved CPU, chipset and wireless components, and outline the benefits that mobile professionals are likely to experience.
After many months of delay, Intel's new 'Sonoma' portable processor and chipset combo is ready for inspection. The Centrino platform has been one of the company's notable successes over an otherwise bumpy period, as it has included most functions a notebook computer needs while balancing high performance against battery life. This recipe is followed faithfully in this latest iteration.
Pentium M with 533MHz FSB
Centre stage is taken by a buffed-up version of the processor previously known as Dothan, now in its Pentium M business suit, fabricated in a 90nm architecture and topped off with 2MB of Level 2 cache. Most of the new Pentium Ms have a 533MHz frontside bus (FSB), giving what Intel claims is a 33 percent increase in memory data transfer over the previous chip's 400MHz.
Most of the new Pentium M processors have a 533MHz frontside bus. The flagship Pentium M 770 chip runs at 2.13GHz.
The updated processor range clocks in at speeds of 2.13GHz (Pentium M 770), 2GHz (760), 1.86GHz (750), 1.73GHz (740) and 1.6GHz (730). There are also low voltage (1.5GHz, 758) and ultra low voltage variants (1.2GHz, 753) with 400MHz FSBs, aimed at manufacturers making very slim and light notebooks. Power requirements range from 5 watts for the 753 to 27 for the 770.
The only other addition to the Pentium M architecture is support for the Execute Disable bit, which with operating system support -- provided by Windows XP Service Pack 2, for example -- can prevent buffer overflow virus or trojan attack code from executing on the system stack.
Most of the differences in the updated Centrino platform live in the Alviso chipset, now officially named the Intel Mobile 915 Chipset Family. This includes support for up to 2GB of DDR2 DRAM, which uses a lower voltage than DDR1, has a smaller page size and extra power-down modes. Although the higher speed of DDR2 somewhat negates these low-power features, Intel says that with the 915, DDR2 memory at 533MHz will peak at 60 percent faster than 400MHz DDR RAM, and can save an average of 120mW per stick.The new memory is physically smaller, too.
The new Mobile 915 'Alviso' chipset supports up to 2GB of DDR2 RAM, Serial ATA and PCI Express, plus improved integrated graphics and audio.
The 915 chipset also includes a power-managed Serial ATA disk interface, and PCI Express, which is advertised at being up to twice as fast for I/O and four times as fast for graphics. Expansion cards for this will follow the new ExpressCard format, which is around half the size of the venerable PC Card standard, and which has a somewhat squashed orange rabbit as its logo. Most, if not all, notebooks with ExpressCard launched this year will also have a slot for older formats, and most, if not all, ExpressCard cards this year will duplicate functions already available with PC Card.
Integrated graphics on the 915GM -- the Graphics Media Accelerator 900 -- includes DirectX 9.0 hardware support for 3D games, as well as high-definition, wide aspect ratio and TV standard outputs. Intel claims that the integrated graphics has twice the raw speed of the previous Centrino chipset, the 855GME, and that with two 533MHz DDR2 memory modules the chip can reach a preliminary 3DMark03 performance rating of 1,140. This compares with figures in the 5,000 range for high-specification desktop gaming configurations and is unlikely to excite the hard core, but should be sufficient for games a couple of years old. Most business applications are expected to be unaffected.
Likewise, adoption of the Intel High Definition Audio standard means that the 915 chipset can support multiple independent audio streams -- such as streamed telephony at the same time as surround-sound DVD playback -- in ways that may have consumer applications but are currently underexploited i
We constantly hear about computer manufacturers "closing the gap" between laptops and desktops. this will never happen. At least for me, the two serve completely different purposes. My desktop is the workhorse, does all the compiling for both computers (both run linux), and is my general use machine. My laptop is low powered, and only used for work on the road. Last time i shopped for a laptop, most models were trying to do everything and more, while i just want a cheap laptop, no glitz except maybe wireless, with a long battery life. That's it. No 4 ghz 512 mgs ram monsters. I don't need 64 Bit processing. Keep it simple stupid.
...sheeze.
At least Intel appears to miss this goal. Documents released to system builders specify the Thermal design power (TDP) of Smithfield processors at 130 watts. This represents an increase of more than 13 percent over today's Pentium 4 5xx (Prescott) and the upcoming 6xx (2 MByte L2 Cache), which post 115 watts. Maximum supply current climbs from 119 ampere to 125 ampere. The new chips also consume more power than Pentium 4 Extreme Edition 3.46 GHz processor (116.7 watts) and Intel's most demanding chip: The Itanium 2 1.6 GHz consumes 122 watts.
I'm baffled by these numbers - specially considering AMD offerings perform comparably while consuming less power. I know these are dual-core designs, but it's still awfully high.
I always found hard to find how much of that consumed power translates onto wasted power (heat dissipation), but in any case, i wouldn't want to be in a room with a couple of Sonoma servers.
...maybe Microsoft will develop a "low-system-resources" OS for servers!
Oh, wait. Dozed off there for a moment.
Hi speed = short-distances / closely packed circuitry and thing wires
Resistance is proportional to the cross-section of a wire and so thinner wires waste more
Were that I say, pancakes?
This is even worse then it seems. The actual power usage (from intel.com) is at 5mWA/ms, compared with the target of 3mWA. Thats a full 2 milliwatt amps over their target, and much higher then AMD!
It's death by 1000 cuts. Each transistor takes a miniscule amount of power to switch, but transistor numbers and clock frequency just keep on increasing exponentionally.
You end up with oodles of transistors operating in the GHz range so all those tiny switching currents add up to one big whole.
Philip
Signatures are broken
Only recently did I become aware that there was a marked architectural difference between 500 and 700 series CPUs - and I read about distinctions between Pentium mobile and Pentium M chips. While I applaud the move away from correlating the computing capability of a chip with its clock speed, I now find myself (as I suspect the typical man in the street also find themselves) completely unable to compare the relative performances between the various CPU series.
I realise that benchmarking is no panacea - but it would be really helpful to see a comprehensive set of benchmarks in order that I can establish a rough idea of the advantages of upgrading before splashing out on a machine with a newer processor. When Mhz mattered a good rule of thumb was that it was only worth considering an upgrade when the new CPU had a clock at least twice as fast as the old one... How should a user make this sort of decision with the new names? Is there any 'good' (unbiased) online material?
But I don't really care a whole lot about what the bloody Wattage rating is on my Processor. But I don't really care a whole lot about what the bloody Wattage rating is on my Processor.
Fine, I'll send you the powerbill when it gets to be $1,000 a month for the privilege of running a new processor.
Low Power CPUs are only a concern if my computer is running off a battery.
Another dorm room computer expert.
When you have a room full of servers and it takes an additional $20,000 or more for cooling and power upgrades and UPS upgrades because people don't care about low power CPUs except in a computer running off of the battery, I'll send you that bill too.
As CPUs dissipate more power, they become more prone to failure, and need a more complex cooling solution to boot. That means jetplane-sounding-like coolers or watercooling kits. Never mind having a small heater in your bedroom. Besides, some people (I, for one) care about power consumption and electrical bills.
Imagine the upcoming slashvertisements for Intel:
"Your primary source for sauna equipment"
"Your next water heater runs Linux!"
"Fusion inside"
Agreed if you're running a data center, or if you have a large business with hundreds of these things, I meant more for personal use.
I don't see where anybody would loose from a more efficient processor design.
The same goes with transportation. The more efficient the transportation, the cheaper it will be for goods and services that depend on them, which is about everything today. The same can be said about processor design.
One interesting thing, is I have _never_ seen a price increase or decrease for any goods or services due to the price of gasoline/diesel fuel which is pretty volatile compared to other commodities and inflation. Its kinda weird I think.
The environmental impact of running a 130 Watt processor all day every day? That's like leaving two lights on in your house...[/fake-comment]
But it's not: the processor needs a PSU to feed it and will require the PSU capable of 300+ watts. Call it 500 for safety, and the computer it's attached to will be burning through those coal seams, that natural gas, and that hard-fought oil, uranium, plutonium, etc.
I'm glad someone made mention the impact on the world outside our basements...
Where are your facts to support this? I am actually interested as to why AMD machines are unstable for games, I thought they where much better than Intel for that, and Intel is much better than AMD for video processing sorta tasks (due to long pipeline).
atcurtis:
I think that they may be taking the wrong approach by putting 2 whole processor cores on the same die...
We have SMT (HyperThreading in Intelese) which in my opinion is a pretty decent idea... just a crying shame about how they set about doing it. They sacrificed the silicon used by the original P4's integrated RamBus memory controller and put in the necessary silicon for their HT technology. The idea of getting an extra CPU for 'free' in the current HT processors doesn't work because in a demanding application, most of the execution units will be busy anyways.
No P4 has ever had an integrated memory controller. The original Williamette cores interfaced to RDRAM through the i850 chipset. In fact, there were a number of Williamette machines that had PC-133 memory (slow as Christmas, but they existed). I owned an RDRAM one and I worked on a PC133 memory one at a job site.
Because of this, many old RamBus P4 machines can outperform their newer P4 siblings - mostly because the newer P4 do not have an integrated memory controller and have to go through the IO originally for peripherals. (ok, they are fixing this with the much higher pincount chips than the 1st gen P4 which did not need all that IO due to integrated RDRAM controller)
I've never seen a Williamette outperform any later Pentium4 core and again, P4s haven't had IMCs.
Instead, I believe that they need to design a processor with the original intent to be hyperthreaded (instead of the P4's original intent, to use RDRAM). What this means, is perhaps provide many execution units, maybe 50% more than what a single processor requires, and then make it look like 2 CPUs. Or perhaps double or triple the number of units and make it look like 4 CPUs to the software.
Eventually, if you provide 2x the number of execution units that are "needed", what difference do you have from a dual core processor? Some units would be shared - fetch, decode, memory stages, etc, but you'd be getting close anyway because of the added interconnect logic for more execution units (it's an n-squared problem with the number of execution units for data forwarding and data hazard detection/resolution).
So... What they need for the consumer is a high-pincount device which is truely designed for hyperthreading (ie, has enough execution units available to be able to perform nearly as good as having a whole 2nd CPU)
And for the server market, bring back the integrated RamBus controller, still have plenty of pins so that the server can have perhaps 4 or more RDRAM channels to keep the data flowing fast enough to keep the 4 SMT logical processors occupied. (IIRC, the original P4 has 2 RDRAM 800 channels)
You also have to remember that RDRAM isn't as wide as DDR, for example. It's fairly narrow in the scheme of things. Having two channels can make is wider just like dual channel DDR memory.
And while I am in my Intel rant mood, I'll criticise the Itanic... Surely with the EPIC architecture, all that branch-prediction and other crud they have in the processor is unnecessary... They need to cut away 2/3rds of the silicon, and get people to write compilers which really do work for them. IIRC, the whole point of all that extra cruft is to make it perform ok for brain-dead compilers. Either they get decent compilers out there (perhaps, open-source their Itanic compiler optimiser) or admit that EPIC was "another nice idea, pity it doesn't work in practice".
Branch prediction is hardly "crud". If your CPU is capable of performing dynamic branching (branches based on the results of an operation as opposed to a hard static branch such as BRA (branch always)) and has a penalty with pipeline flushes on branching, then you probably can benefit from branch prediction). Also, EPIC *hardly* performs "OK" with brain-dead compilers. You have to have a decent compiler to get the performance out of it. A brain-dead compiler won't get 25% out of the CPU's max performance
AMD chips seem to be a lot cheaper than similarly-fast Intel chips.
AMD chips seem to run cooler than comparable Intel chips.
AMD chips seem to have more support and 'options' than Intel chips.
Whats keeping Intel afloat? While mobo combo shopping recently, I barely paid any attention to the Intel line. All I saw was 1.3x price for 95% performance.
Job? I don't have time to get a job! Who will sit around and bitch about being broke and unemployed then?
Ok, so since Laptops were first created, their whole idea was to be a way someone could do work remotely or in areas where having a desktop was hard or near imposibe for one reason or another. in 2003 "November 2003 survey of Penn State University undergraduates found that freshmen were more likely to own laptops than upperclassmen. Of 1,838 respondents, 73.7 percent own a desktop, 32.2 percent own a laptop, 9.2 percent own both, and 3.4 percent own neither." http://pittsburgh.bizjournals.com/pittsburgh/stori es/2004/08/23/focus1.html
e ID=1513
e gory=31558&item=6736232824&rd=1&ssPageName=WDVW
(sorry if i am not suposed to link to auctions, it was the only english page i could find since its a japanese laptop)
"t). Currently, one-quarter of Americans own a laptop or notebook computer (23 percent)" http://www.thegoodsteward.com/article.php3?articl
What does this tell us? it says that laptops are becoming more and more popular, and as they become more and more popular, there is a bigger drive towards creating the "ultimite laptop"
As far as i see it, Apple and IBM are the only good laptop companies. i know thats a dramatic statement, but look at it, any laptop over 6LBS is WAY TOO HEAVY to carry around. Most HP/Compaq laptops are 7 to 9 lbs. Most dells are in that same range.
what the laptop industry needs is a re-working of laptops. Sony has just released the X505 VAIO laptop, this laptop is built in the way all laptops should be built. It weighs a mear 1.73lbs WITH battery in it, and has enough power to run almost anything exept games and video editing.( But trying to play games on a laptop is just stupid anyway, small screen and no mouse or full keypad) http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&cat
The new centrino chips are amazing, to put it simply. If you put a 2ghz Pentium M into a desktop and slap on a liquid cooling system, you can overclock it to be way more powerful then a 3.46EE or even a 3.8ghz P4. The pentium M is just years ahead of its time, and people having figured it out yet! Its kinda like black lotus for all you magic players!
Now that the Mac mini is out, if Apple sells millions upon millions of units it should only be a matter of time until one of the PC makers will come out with a mimic of it (like, say, Asus who is making the Mac mini in the first place).
It seems to me that Sonoma would be an ideal chipset to cram all of the functionality into a pint-sized box as small as, or slightly larger than the Mac mini. If the box has about as clean of a look to it, did away with the PS2, serial, and parallel ports, was user serviceable, and had room for a standard 3.5 inch SATA drive, I think it would do really well -- even if it weren't clad in anodized aluminum. Ahh... perchance to dream!
Power goes to two primary areas Leakage will account for ~1/2 of the power loss - at sub 90nm, transistors do not look much like a switch - they look much more like a resistor with 2 different possible states - this power will be roughly constant The other half of the power goes to charging/discharging all the internal capacitances - setting a voltage on the gate of an individual xtor is just placing a small amount of charge on a small capacitor, and any time a gate changes states, you have to expend some energy to make this change 200M xtors X 4GHz = lots of state changes - this power will increase linearly with operating frequency
Damn, I was hoping this article would announce the return of the "Turbo" button!
See the little bars at the bottom, titled "Winchester 90nm" in the "power usage" link? I bought two of those recently. They run STONE COLD when idle and you can barely feel the warmth under load.
It's nice to know your CPU uses TEN TIMES LESS POWER when idle and at least three times less under load than an equivalently performing Intel, and is within 50% of the highest performance CPU on the market. AMD is going to beat Intel's desktop offerings silly with their new 90nm parts.
[an error occurred while processing this directive]
The new chipset is the Intel PRO/Wireless 2915ABG and has linux drivers:o 2915abg/index.htm
:
0 6408.htm :
http://support.intel.com/support/wireless/wlan/pr
From http://ipw2200.sourceforge.net/
This project was created by Intel to enable support for the Intel PRO/Wireless 2200BG and 2915ABG Network Connection miniPCI adapters. This project (IPW2200) is intended to be a community effort as much as is possible given some working constraints (mainly, no HW documentation is available).
From http://support.intel.com/support/notebook/sb/CS-0
Intel® PRO/Wireless 2915ABG Network Connection
A Linux driver is currently under development. A pre-production version of the device driver is available as detailed below.