POST A COMMENT

21 Comments

Back to Article

  • n0nsense - Sunday, January 13, 2008 - link

    I don't like intel.
    Usually it does something good only when they have no other choice (like Core architecture).
    The x86 is not good from the very beginning decades ago. But greater evil is there. It's name Microsoft. They just want you to use what they designed to get money from you instead of designing For You with user in mind.
    And the only thing that make me smile, is expected percentage of devices based on this platform that will run MS products.
    Actually Linux rules in mobile/efficient devices world.
    As for the software compatibility on different platforms, Java does the job.
    And the Linux world does know how to work with different platforms. Most distributions available for several architectures.
    So, the most important thing is to have size/power/performance in right combination.
    It will be very sad if intel will succeed to kill other architectures.
    Reply
  • Boissez - Friday, January 11, 2008 - link

    I've just checked the numbers... And I just can't see how this 47M transistor processor clocked slower than the 77M Pentium M can perform similarly.

    I know that there has been som architectural tweaks meanwhile... but still we're talking about a 3x improvement (assuming the 600mhz part=1000 Mhz Pentium M) - If those performance claims are true I'm inpressed.
    Reply
  • IntelUser2000 - Friday, January 11, 2008 - link

    Actually, most of the differences between the 47 million and 77 million is cache. If you look at the core alone, its not much different. Of course, it will impact performance...

    Pentium 2x8KB=3.1 million
    Pentium Pro 2x8KB L1=5.5 million
    Pentium II 2x16KB L1=7.5 million
    Pentium III 2x16KB L1=9.5 million
    Conroe/Merom 2x32KB L1=19 million
    Silverthorne 32KB iL1, 24KB dL1=10 million??

    Silverthorne's core transistor count is estimated at 10 million. Maybe it can perform like a Pentium III 512KB cache with SSE3. Anyway, it has 512KB L2 cache so...
    Reply
  • IntelUser2000 - Saturday, January 12, 2008 - link

    The estimated performance of Silverthorne:

    1.86GHz:
    50-60% faster in multi-thread than A110
    Similar in single thread to A110

    1.1GHz:
    A110 MT performance
    A100 ST performance

    (A100=600MHz/512KB/400MHz FSB/Dothan)
    (A110=800MHz/512KB/400MHz FSB/Dothan)

    When Intel says "similar" performance to 2004 notebooks, they are referring to ULVs. ULVs are clocked at 900MHz-1.1GHz. Take a look at Intel's technical documentations for their ULV processors. You'll notice that the performance increase has been almost flat even with the architectural/process improvements. They wouldn't have been able to make a 0.5W CPU without a substantial architectural overhaul.

    Pentium M 130nm ULV
    1.1GHz 7W 400MHz

    Pentium M 90nm ULV
    773 1.3GHz 5.5W 400MHz

    Core Duo ULV
    U2500 1.2GHz 9W 533MHz

    Core Solo ULV
    U1500 1.33GHz 5.5W 533MHz

    Core 2 Duo ULV
    U7600 1.33GHz-10W 533MHz

    Core 2 Solo ULV
    U2200 1.2GHz-5.5W 533Mhz

    For two generations ahead in architecture and process technology, the latest ULV CPU offers much less than it does on desktops and laptops.
    Reply
  • ajuez - Friday, January 18, 2008 - link

    I think thet the Via C7-M ULV is similar in approach to the Silverthorne processor: very simple architecture focused on power consumption and cost.

    The VIA C7-M ULV consumes 3.5W (according to http://www.via.com.tw/en/products/processors/c7-m_...">http://www.via.com.tw/en/products/processors/c7-m_... )

    However, Intel has the advantage of their manufacturing technology (Silverthorne is a 45nm processor, while C7 is a 90nm processor). I hope Via to update their processor with a better process (55nm?)
    Reply
  • IntelUser2000 - Wednesday, January 16, 2008 - link

    Maybe we are all confused after Intel introduced the Core 2 Duo. From what I can remember, its only worthy product that came out of the company for many years. All the chipsets after 865/875 were mediocre, and everything else is just a flop.

    I heard that Menlow is headed by the same guy that made a failed XScale and Itanium. Maybe its gonna be another flop. Currently, the only thing that Menlow is sure to bring is the size. Everything else, like battery life and performance is looking gloomy every day.
    Reply
  • Flunk - Friday, January 11, 2008 - link

    Way less cache, cache is the biggest use of transistors on chips designed for PCs. Cut the cache to say 256k (which is enough for a device running entirely off of flash) and you save a lot of transistors. Reply
  • TA152H - Friday, January 11, 2008 - link

    What's processor cache got to do with Flash drives?

    Processor cache speeds up RAM accesses, RAM cache can be used (and is normally) to cache hard disk accesses, and that's probably what you're thinking of. But they're two different things.

    I agree though, transistors are not the same as size. You can pack memory transistors much more closely than logic, and it can seriously impact transistor count while less seriously increasing the size (and cost) of the processor.
    Reply
  • iwodo - Friday, January 11, 2008 - link

    If Silverthrone is really that powerful i think it can used as a NAS Chip.

    All current NAS are so underpower and dont provide fast enough transfer due to CPU limitations. Windows Home Server kind of remove the performance barrier since it require Desktop class PC.

    However they are also more power hungry too.

    I hope Apple use this as their base for Apple Home server
    Reply
  • Shadowmage - Thursday, January 10, 2008 - link

    Anand:

    Nice article, but you spelled "Bonnell" wrong.

    Menlow consists of the Silverthorne processor and the Poulsbo chipset. If you're in dire need of yet another codename, the core used in Silverthorne is also known as Bonnenn.
    Reply
  • TA152H - Thursday, January 10, 2008 - link

    Anand,

    Do you have any idea what this processor is in terms of the architecture? I still have not read anything about whether, for example, it's decoupled, it's superscalar, it's superpipeline, and if it is superscalar, is it out of order, etc ...

    Is Intel just not releasing any of this information yet? I guess one clue is that they say it will run at the same clock speeds and performance level as the Pentium M did, so I guess it would almost have to be superscalar, and probably even out of order. But then, why make a new processor from scratch if it's almost the same as the Pentium M. It's kind of confusing. Maybe they reached the same goal by doing something different, that uses less power. For example, maybe it's a shorter pipelined beast, which would increase IPC and lower clock speed and make it smaller and less power hungry. Maybe it's not decoupled, which would save some stages, and some transistors.

    I hope Intel will release more information on it. It's would be really interesting to see how they went with it. I'd like to see AMD release a K6 derivative as a mobile chip instead of something off the K7/K8 core. It might not reach the same clock speeds, but I think it had excellent power use characteristics and had relatively high IPC (especially the K6-III).
    Reply
  • IntelUser2000 - Thursday, January 10, 2008 - link

    ISSCC 2008 preview: "A 47M transistor, 25mm2, sub-2W IA processor designed for mobile internet devices…It features a 2-issue, in-order pipeline with 32KB iL1 and 24KB dL1 caches, integer and floating point execution units, x86 front end, a 512KB L2 cache and a 533MT/s front-side bus. The design is manufactured in 9M 45nm High-k metal-gate CMOS and housed in a 441-ball µFCBGA package."

    VLSI Seminar(http://dropzone.tamu.edu/VLSISeminars)">http://dropzone.tamu.edu/VLSISeminars): "This presentation will describe a low power Intel(r) Architecture (IA) processor specifically designed for Ultra-Mobile PCs where average power consumed is in the order of a few hundred mW with performance similar to mainstream Ultra-Mobile PCs. The design consists of an in-order pipeline capable of issuing 2 instructions per cycle supporting 2 threads, 32KB instruction and 24KB data L1 caches, independent integer and floating point execution units, x86 front end execution unit, a 512KB L2 cache and a 533 MT/s dual-mode (GTL and CMOS) front-side-bus (FSB). The design contains 47M transistors in a die size under 25 mm2 manufactured in a 9-metal 45nm CMOS process with optimized transistors for low leakage packaged in a Halide-Free 441 ball, 14X13 mm uFCBGA. Thermal Design Power (TDP) consumption is measured at 2W using a synthetic power-virus test at a frequency of 2GHz."

    Unlikely it'll be Pentium M range. It's supposed to have Hyper-threading and talks about high throughput oriented processor. The CPU performance is something like 800MHz Dothan-512 at single thread and 40-50% faster with multi-thread, and that's when its clocked at 1.8GHz.
    Reply
  • TA152H - Friday, January 11, 2008 - link

    Thanks for the information.

    I agree, it will be impossible for it to have the same IPC capability as the Pentium M. I'm not sure why Intel is saying that. It does look a little more formidable than the VIA chips though. It will be interesting to see what VIA does to counter them.

    It's kind of interesting how the computer industry constantly goes in circles. The CGA and MDA adapters (and later EGA) were digital output, VGA and later were analog, now we're back to digital. Superscalar processors, particularly the Pentium, were in order, then went out of order, and now we're seeing the POWER, Itanium and this chip going back to being in order. Before it was the processor that did the video processing, and the cards were just dumb frame buffers. Now, they are talking about moving the video processor back into the CPU.

    I just hope the cassette tape doesn't come back as a means of storage. CLOAD. Ugggh!
    Reply
  • IntelUser2000 - Thursday, January 10, 2008 - link

    Anyways, it'll be an upgrade for the UMPCs anyway with its greater multi-threading performance(somewhat possible single thread??) over the current A110 CPU that's used on the UMPCs.

    I don't get Moorestown though, why push x86 on non-Windows platform?? WTF Intel!!
    Reply
  • Cali3350 - Wednesday, January 09, 2008 - link

    Considering how much we've been hearing about this platform for the last year or so i must say it does seem rather disappointing. I was led to believe this would the next wave of real portable devices, something Apple and the like would adopt willingly very early. I must say i don't see that occurring from this article.

    Ofcourse this is just a quick article from CES, maybe things are different when you really get up close and personal with it.
    Reply
  • hnhunt66 - Wednesday, January 09, 2008 - link

    the eventual migration of many of these features like web browsing and the like to devices other than the main number cruncher in the house (the pc or mac) has been a couple of years away for about a decade. The cost effectiveness of reusing code across multiple platforms is the main reason that Intel is more than likely on a good path here. The technically inclined and those young enough for this to be natural still like the cheep sometimes at the expense of elegance. The rest of the world just wants it to be plug and play this is achieved at low costs by not having to write new code or debug recompiles for every device. Under estimating the value of the x86 architecture is risky at best, I was reading about its demise in the 80's. Reply
  • Freddo - Wednesday, January 09, 2008 - link

    I would love to see a Menlow notebook (not UMPC). Something like the Eee PC from ASUS, but with some noticeably changes.

    Bluetooth support is a must. HDMI would be very nice too, so you can hook it into the TV for a larger screen now and then. Maybe even use it as a HTPC with an external optic drive thru the USB.

    It would be a perfect nice little energy-efficient notebook :)
    Reply
  • psychobriggsy - Wednesday, January 09, 2008 - link

    Intel tries to catch up to where ARM has been for years. The iPhone chip has the CPU, GPU, RAM, and loads more just on the CPU package (about the size of Silverthorne). Apple won't take over markets by waiting 3 years to get where they can be with ARM today.

    Intel are hoping for 2x the iPhone processing power ... a year after the iPhone came out. And that's using a 620MHz ARM11 based core. If Intel do try, Samsung will come out with an ARM Cortex multi-core fully integrated 3G phone on a single piece of silicon with RAM and flash stacked into just the space of the Intel CPU.

    Look at the iPhone motherboard. There's no contest as to which architecture is going to get picked. Even in 2010 Intel will have 3 large (for a phone) chips - CPU, I/O and RAM, and that's before you get any phone related circuitry and storage.

    Apple are perfectly happy with using ARM in their iPods and iPhones. They've got 15 years of ARM experience in-house. ARM are actually really making moves in terms of performance now as well, with the Cortex cores.

    Now on the upside for Intel, these will be very neat systems and I expect that small media servers and the like will abound. Shame that they have such a fixation on mobile internet that the real opportunities could pass them by. All those MIDs look crap. The article says that even with 2x the power of the iPhone the interfaces are sluggish (yet are still way faster and better than Vista) ... there's no hope.
    Reply
  • defter - Thursday, January 10, 2008 - link

    [quote]Apple won't take over markets by waiting 3 years to get where they can be with ARM today.[/quote]

    Actually I don't understand why Apple has been mentioned something like 10 times in the article and many times in comments. They have maybe 4-6% of share in a smartphone market and 1-2% share in a whole cell phone market, what they do or not do isn't important.

    Of course cell phone manufacturers will not "wait" for Intel CPU and will continue to use ARM based CPUs in the near future, it would be silly to expect otherwise. It seems that currently, Menlow would be very suitable for EEE PC type of devices: it's cheap, has a low power consumption and takes a very little of space compared to current mobile CPUs.

    Then maybe in 3-4 years Menlow successors could be good enough for cell phone integrations. Remember: Intel has a lot of money and has a manufacturing process advantage, especially the latter is very important for power consumption. In the distant future we could see 22nm Menlow's successor competing against 32nm ARM CPUs, in that case ARM's power consumption advantage is not given...
    Reply
  • PandaBear - Wednesday, January 09, 2008 - link

    I don't understand why would Intel try to force x86 down a portable device that runs on Linux. For Windows I can understand but ARM (and almost every other architecture out there) can run Linux and in an embedded device you don't want to have the cost and power consumption much more than ARM.

    Given the fact that ARM based solution are so efficient power wise, and that there are already many ARM based ASIC out there, I don't understand why (other than politics) would Intel want to go the x86 road. They should develop a similar low power RISC core and hardware accelerated ASIC around it instead.

    What I sense is that they don't want x86 to go away when non PC take over the Internet browsing and mobile market takes off. So they give you something that is just good enough to hang on to rather than going to an all new platform like iPhone. Without a true Windows (with good usability) this won't help much.
    Reply
  • defter - Thursday, January 10, 2008 - link

    Intel had a low-power RISC core called StrongARM several years ago, but they decided to sell that...

    It seems that Intel feels it's important that small mobile devices use the same instruction set as desktop PCs.
    Reply

Log in

Don't have an account? Sign up now