1. Este site usa cookies. Ao continuar a usar este site está a concordar com o nosso uso de cookies. Saber Mais.

Penryn 40% faster on games, video...

Discussão em 'Novidades Hardware PC' iniciada por Zarolho, 17 de Abril de 2007. (Respostas: 14; Visualizações: 2770)

  1. Zarolho

    Zarolho Power Member

    Gelsinger provided performance indicators for Intel's upcoming Penryn family of processors. For desktop PCs, he said to expect increases of about 15 percent for imaging-related applications; 25 percent for 3-D rendering; more than 40 percent for gaming; and more than 40 percent faster video encoding with Intel SSE4 optimized video encoders. The indicators were based on pre-production 45nm Hi-k Intel® quad core processor running at 3.33 Gigahertz (GHz) with a 1333 Megahertz (MHz) front side bus (FSB) and 12MB cache versus an Intel® Core(TM) 2 Extreme processor QX6800 introduced last week at 2.93 GHz with 1066 FSB and 8MB cache.

    For high-performance computing (HPC) and workstation systems, Gelsinger said to expect gains up to an estimated 45 percent for bandwidth intensive applications; and a 25 percent increase for servers using Java(1). These indicators were derived from pre-production 45nm Hi-k Intel® Xeon® processors with 1600 MHz front side bus for workstation and HPC, and a 1333 MHz front side bus for servers versus today's quad-core Intel® Xeon® X5355 processors.


    Última edição: 17 de Abril de 2007
  2. spastikman

    spastikman Banido

    A julgar pelos clocks de ambos, o QX6800 se tivesse a mesma velocidade de relogio, provavelmente descia para metade o valor daqueles ganhos (em percentagem).

    Isto para não falar nos 4MB de cache extra, por isso não é nenhum avanço tecnologico do outro mundo.
  3. timber

    timber Zwame Advisor

    Pois não, mas isso era mais ou menos esperado. O Penryn muda o processo para os 45nm e trás algumas pequenas melhorias.

    O próxima (nehalem ou algo assim se não me falha a memória) parte dum processo consolidado e lança a nova arquitectura.
  4. Zarolho

    Zarolho Power Member

    Intel's Spring IDF brings details on Penryn and SoC plans

    Intel opened the first day of its Spring Intel Developer Forum (IDF) in Beijing with a slew of high-profile announcements, all centered around the company's new 45nm technology. Senior VP Pat Gelsinger was up first with revelations about Intel's forthcoming 45nm processor lineup, system on a chip (SoC) plans, and discrete GPU plans.
    Related Stories

    First Penryn performance numbers

    Gelsinger opened with the first Intel-supplied numbers on how well Penryn stacks up to its predecessor on a number of different application domains. According to Intel, a 3.33GHz Penryn processor (12MB L2 cache) on a 1333MHz frontside bus posts the following increases versus the newly released Core 2 Extreme QX6800 processor:
    • 15 percent in imaging-related applications
    • 25 percent for 3D rendering
    • 40 percent for gaming
    • 40 percent for video encoding (SSE4-optimized)
    Note that the typical caveats about vendor-supplied benchmark numbers that come lacking important information apply.
    Intel also gave some even more vague numbers for a 45nm Xeon's performance on unspecified high-performance computing benchmarks, boasting a 45 performance percent increase on bandwidth-intensive workloads, and a 25 percent increase "for servers using Java." Much of the increase here comes from the 1600MHz frontside bus on the 45nm Xeon.
    In related Xeon news, the Core-based Xeon 7300 series will be launched in the third quarter of this year. This series is aimed at multisocket systems—both blades and servers—and will complete the Xeon line's transition to the Core microarchitecture.
    Intel's Tolopai will take x86 into the SoC realm

    Back in February, word leaked out that Intel was prepping a Pentium M-based system on a chip (SoC) product for the embedded space, due out in the second half of 2007. The codename of this project is Tolopai, and Gelsinger officially took the wraps off of it today.
    Tolopai is the first in what Intel describes as "a family of enterprise-class SoC products that integrate several key system components into a single Intel architecture-based processor. Gelsinger didn't reveal exactly which components will make their way onto the processor die, but previous rumors indicate that the obvious candidates will be included, i.e., the northbridge and southbridge.
    What's left unclear in the Tolopai revelations is whether the first products will be introduced at the 65nm or 45nm node. The original rumors had Tolopai at 65nm, but given today's announcements and Gelsinger's emphasis on 2008, it appears that if a 65nm part does appear, it will be short-lived. The 2008 Tolopai incarnation will undoubtedly be a 45nm part.
    If you want more details on why Intel is pushing x86 down into the embedded space, especially in the wake of their sell-off of the XScale line, then see "Intel's coming embedded play" for a more complete discussion.
    Intel's answer to Torrenza: QuickAssist

    Gelsigner briefly introduced the QuickAssist initiative as an accompaniment to Tolopai. Much like AMD's Torrenza platform, the idea behind QuickAssist is to enable more widespread use of accelerator chips in servers.
    There isn't much technical detail available yet, so at this stage it's difficult to gauge the prospects of Intel's initial stab at a coprocessor ecosystem effort. The main questions that need answering are:
    • Is QuickAssist based on Intel's forthcoming common systems interconnect (CSI)?
    • What kind of licensing will Intel offer third-party accelerator makers?
    The answer to the first question is almost certainly "yes," and the answer to the second is probably "something similar to what AMD offers with coherent HyperTransport (cHT)." But we'll have to wait for precise answers to these questions.
    One novel aspect of QuickAssist that appears to separate it from Torrenza is Intel's planned Accelerator Abstraction Layer (AAL) middleware package. This software layer will provide an interface to the different accelerators in the systems, making it easier for applications to manage them.

    Gelsinger talked about Nehalem, but he didn't reveal anything that we haven't covered already. He also talked about Larrabee, as we've previously reported. Also discussed were Intel's plans for the next-generation version of vPro, codenamed "Weybridge" and due out in the second half of this year. As we reported previously, Weybridge will bring vPro capabilities to the Centrino line of laptop hardware.
    We'll see more revelations about some of the technologies and initiatives outlined above on Day 2 of IDF, as Intel fleshes out some of what was first introduced here.

  5. Zar0n

    Zar0n Power Member

    A maior parte dos ganhos são atribuídos ao aumento de clocks.

    Mas há maneira de ver os k são da arquitectura.
    Basta ver o score com um QX6800 a 2.93 e um a 3.33 e comparara os ganhos :)
  6. Zarolho

    Zarolho Power Member

    The Long Awaited Penryn Update

    In an unprecedented move, Intel made a very full disclosure of its first 45nm processor family, codenamed Penryn a couple of weeks ago. There were some vague elements of the initial Penryn disclosure that we’ve since cleared up.

    First off, Penryn is designed to support up to a 1600MHz FSB, however we wondered whether desktop chips would even see the faster FSB support given that we haven’t so much as heard of support for it on Intel’s upcoming 3 series chipsets (e.g. P35, X38). It turns out that Intel is only confirming 1600MHz FSB support for Penryn based Xeon processors for the HPC market, not for the mobile or desktop markets.

    This tells us two things:

    1) Intel is feeling AMD’s bandwidth advantage and strength in the HPC market and is using the faster FSB to help level the playing field.

    2) the desktop will most likely not see a FSB faster than 1333MHz.

    Remember that with Nehalem being introduced in 2008, Intel will begin shifting away from its aging FSB architecture to a point-to-point interface akin to what AMD introduced with the K8 back in 2003. It doesn’t make a lot of sense for Intel to invest much money into moving cost focused desktop platforms to 1600MHz FSB only to abandon the efforts in a year’s time. While Intel hasn’t said anything, we’re expecting Penryn desktop parts to be 1333MHz FSB only, which makes sense given that the upcoming P35 chipset officially supports a maximum FSB frequency of 1333MHz.

    Intel also mentioned that Penryn would support SSE4, but is its implementation complete or will we have to wait until Nehalem for that? It turns out that Penryn will support a total of 47 SSE4 instructions, not the full implementation of the ISA extensions. There will be an additional 7 instructions that Intel is stating will come in future microprocessors, we’re assuming that Intel is talking about Nehalem but it’s not yet set in stone.

    The two interesting power related technologies that will make their debut with Penryn will apparently be mobile-only for now. Intel’s C6 state and EDAT (Enhanced Dynamic Acceleration Technology) will only be supported on mobile Penryn platforms given the nature of the two features. As a recap, the C6 power state allows for an extremely low power operating mode, the closest to a full reset of the CPU, while idle. Data is completely expelled from the on-die caches and the caches themselves are powered off, while core voltage is reduced to the lowest amount allowed by the process. The CPU’s state is saved in some on-chip storage, then the majority of the chip is powered down into a virtually off state. Recovery from C6 is possible, the state is read back from the CPU and the chip is powered up as it would from reset but with memory of what it was doing before it entered C6. The wakeup process does take some time (not noticeable to the user) thus it impacts performance and is suitable for mobile environments where the impact on battery life is worth the reduction in system performance.

    Intel’s EDAT is the other mobile-only Penryn technology Intel talked about in its disclosure, and it allows the increase in clock speed of one core on a mobile Penryn when the other core is not in use. The idea is simple: in a notebook you are constrained by the cooling system used, not by the maximum clock speed attainable by the CPU itself. When running single threaded applications (or multithreaded applications with only one CPU intensive thread), the remaining core can power down reducing the total thermal footprint of the CPU itself. An EDAT enabled mobile Penryn core can then detect that only one core is being used and increase the clock speed of that operational core by a single speed bin (e.g. 2.40GHz to 2.66GHz) in order to provide a boost in performance to that one active thread. Once again, EDAT will be mobile-only.

    Finally, with regards to motherboard support, Intel isn’t making any guarantees about Penryn’s backwards compatibility. While Penryn will still use the LGA-775 socket that Prescott and Conroe have used, motherboard support will require more than just the presence of the socket.

    If the appropriate VRM spec is implemented, then Penryn will work on your LGA-775 motherboard, the problem is that motherboard manufacturers haven’t yet released information on which of their boards will support the Penryn VRM changes. If history repeats itself, you can expect very limited official support for Penryn in currently shipping motherboards and guaranteed support with boards based on Intel’s new 3 series chipsets (e.g. P35). We did see Penryn up and running on an Intel BadAxe2 board, but it had a hardware VRM modification done to it in order to properly support Penryn. Penryn may also be able to work on boards without a VRM mod, however at increased (potentially out-of-spec) voltage settings.

    At IDF Beijing Intel unveiled a little more about Penryn performance; it compared a quad-core 3.33GHz (1333MHz FSB) Yorkfield with 12MB of L2 cache (2 x 6MB per dual core die) to a quad-core Core 2 Extreme QX6800 2.93GHz (1066MHz FSB) Kentsfield with 8MB of L2 cache (2 x 4MB). According to Intel’s own benchmarks, Intel saw a 15% increase in imaging related applications, 25% in 3D rendering tests, greater than 40% in games, and a greater than 40% increase in video encoding performance when SSE4 support was utilized.

    Obviously some of the performance improvement can be attributed to the higher clock speed and faster FSB of the Yorkfield system, while the remaining would be due to architectural enhancements and larger cache of Penryn. The percentage improvement Intel is indicating with Penryn is quite high, but as we’re comparing across different clock speeds it’s a bit of a skewed comparison. Don’t expect Penryn to have the same performance impact that Conroe did upon its introduction, but rather expect an evolutionary continuation of the performance we’ve seen from Intel thus far. Unlike the other P in Intel’s codename history, there are no terrible surprises with Penryn that will result in a step back in performance.

  7. Crusher

    Crusher Power Member

    De 2.93 para 3.33 são pouco mais de 10%. Mas como a subida de performance não é assim tão linear com a subida de percemtagem do clock, na realidade deve estar a adicionar um pouco menos.

    Cache, FSB e outros melhoramentos na arquitectura do chip fazem o resto.
  8. JPgod

    JPgod Moderador
    Staff Member

    O interessante era por o QX6800 a 3333 mhz e 1333 fsb e comparar...
  9. Zar0n

    Zar0n Power Member

    N te eskeças k já pode haver ai benchs optimizados para SSE4...
  10. Crusher

    Crusher Power Member

    O HL2 pelo menos não! :p

    Provavelmente aquela versão do Divx já estará com essa optimização só pelos ganhos obtidos.
  11. Zar0n

    Zar0n Power Member

    HL2 n, mas sim o DivX era uma versão alpha optimizada com SSE4 ;)
  12. mascarilha

    mascarilha I folded Myself

  13. JPgod

    JPgod Moderador
    Staff Member

    Os ganhos do divx alpha 6.6 que metem medo :wow:

    Até o dual core @3.3 dá uma ratada no QX :wow: :wow:
  14. ajax

    ajax Banido

    Chama-se SSE4 e + cache.

    Olhando para um codec a sério, o H264, pode ver-se que o Peninha até escala menos mal. Passa de 119 para 73, +- 1.63x, ou seja, 80% de eficiência de 2 -> 4. Bonzito! Já o Cinebench continua a dar cartas mas não é muito realista pois trata-se de um benchmark quase-"embarrassing parallel".
    Última edição: 18 de Abril de 2007
  15. Zarolho

    Zarolho Power Member

    Rumors about 45 nm Core 2 Duo production begin to fly

    Chicago (IL) – Intel is gearing up for its next-generation micro-processors. And we hear that Intel may be much closer to be releasing its 45 nm processors, code-named "Penrym", to production than we previously thought: Expect reviews and benchmarks of Penryn to surface early in Q3 of this year.

    Official information about general availability of Intel’s 45 nm processor remain scarce. As of now, Intel says that the 45 nm “Penryn” family, which will bring a new generation of desktop and mobile Core 2 Duo and Xeon processors, is slated for a late 2007 introduction.

    However, industry sources recently told TG Daily that a whole bunch of non disclosure agreements (NDAs) for the pre-release and distribution of 45 nm processors are already in place and carry a July 15 expiration date. “Pre-release”, of course, does not mean that Intel will launch the processor at this time; however, we received confirmation from our sources that initial sample shipments of Penryn are scheduled to begin late in Q2 and early Q3. We also expect the NDAs also to apply to the press, which means that first real-world Penryn benchmarks are likely to be published on July 15.

    If the NDA date is correct, Intel can build up the interest in its next-generation products, but potentially could jeopardize the sales of its FSB1333 desktop Core 2 Duos, which are expected to be released early in Q3, as well as the FSB800 mobile Core 2 Duos , which will launch in the first half of May. On the positive side, the pre-release appears to be very much in line with the launch date of its Opteron quad-core processor “Barcelona” .


Partilhar esta Página