Registar

User Tag List

Likes Likes:  13
Página 11 de 127 PrimeiroPrimeiro ... 9101112132161111 ... ÚltimoÚltimo
Resultados 151 a 165 de 1893
  1. #151
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    7,793
    Likes (Dados)
    0
    Likes (Recebidos)
    1
    Avaliação
    41 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Estas saidas constantes.............
    Vamos ver quem o vai substituir tambem.
    Ideias sem Nexo e Provas do Tráfico de Hardware
    "que personifica o destino, equilíbrio e vingança divina." Dejá vú. Que cena!

  2. #152
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    AMD Zeppelin Processor with 32 x86 Zen Cores Spotted in Linux Changelog – EHP Variant Will Have Vega 10 Graphics with HBM2

    A very interesting report published by the New Citavia Blog confirms a codename we heard quite a long time ago: Zeppelin. To those whore aren’t familiar with the name, Zeppelin is an MCM (Multi Chip Module) which will utilize AMD’s own custom interconnect to combine 32 Zen cores. Not much is known about this processor although it has popped up in various leaks over the course of the past few months.
    A modern zeppelin takes off. Image Credit URL.
    AMD’s Zeppelin processor with 32 Zen cores spotted in linux changelog

    A Zeppelin is a type of a rigid body airship which is filled with a lighter-than-air gas. They were very popular in the early 1900s and a popular mode of transport. Unfortunately, the use of cheaply available hydrogen as the primary gas caused several accidents (most notably Hindenburg) over the course of the next few decades which caused them to be abandoned. Modern variants of the Zeppelin use Helium as the primary gas, which is just as light but not combustible.
    Coming back to the actual topic at hand, the report in question quotes several lines of code published in a recent update bt AMD which can be found over at LKML.com:
    AMD Zeppelin (Family 17h, Model 00h) introduces an instructions retired performance counter which indicated by CPUID.8000_0008H:EBX[1]. And dedicated Instructions Retired register (MSR 0xC000_000E9) increments on once for every instruction retired.
    Signed-off-by: Huang Rui

    arch/x86/include/asm/cpufeature.h | 1 +
    arch/x86/include/asm/msr-index.h | 3 +++
    arch/x86/kernel/cpu/perf_event_msr.c | 30 +++++++++++++++++++———–
    3 files changed, 23 insertions(+), 11 deletions(-)
    + core_complex_id = (apicid & ((1 << c->x86_coreid_bits) – 1)) >> 3;
    + per_cpu(cpu_llc_id, cpu) = (socket_id << 3) | core_complex_id;
    Advertisements

    The code is conveniently signed by an AMD engineer and confirms the name of the processor. An AMD research paper in the past has also referred to a very similar (if not the same) chip as an EHP or Exascale Heterogeneous Processor along with a diagram of the same. Which is basically an APU with the Zeppelin module included. The code present in this update also points to the expected 4 Zen core clusters and a total of 32 maximum CPUs which will be divided into core complexes consisting of 4 Zen cores each (8 logical cores thanks to SMT). Given below is a recap of what leaks have pointed out so far:

    AMD is currently working on a Multi Chip Module that is code named Zeppelin. It is thought to contain 32 Zen cores tied together using AMD’s very own Coherent Fabric. The homegrown interconnect will support data rates of upto 100GB/s which is alot faster than what the PCIe interface sustains (around 15GB/s). Not only that but latency has been reduced from 500ns to an unknown but allegedly smaller number. It remains to be seen whether Zeppelin refers only to the Zen processors cluster or the MCM as a whole. The original leak by Fudzilla showed Vega 10 (previously Greenland) graphics connected by Coherent Data Fabric. High Bandwidth Memory (probably on a 2.5D interposer) was also included and clocked at 500 GB/s.
    The Zeppelin and Vega 10 based MCM will feature 4 GMIs or Global Memory Interconnects (which constitute the Coherent Data Fabric I suppose) allowing the CPU to converse with the GPU at 100GB/s. The MCM itself will talk to the RAM at 100 GB/s as well – allowing for a very HSA friendly environment and minimum bottleneck throughout the whole process. Each Zen core will be capable of running two threads (thanks to the company’s shift to Simultaneous Multi-Threading). The processor is thought to have 4 DDR4 channels with a capacity of 256GB per channel.
    What we do know for sure (thanks to a published research paper by AMD) is that the Zeppelin EHP variant will utilize the next generation Vega 10 graphics, although the exact core count remains unknown. The diagram published by AMD points towards the MCM divided into two compute and one graphic portions, which are manufactured separately and put together on the interposer later on in assembly (possibly at UMC’s Fab 12 foundry in Singapore, which is already used to assemble Fiji dies). So basically, AMD is fabricating the compute side of the Scale Heterogeneous Processor (EHP) in dies with 16 Zen cores each (4 core complexes), for a total of 2 computing and 1 graphics die assembled on the interposer (ignoring the HBM).


    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  3. #153
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    AMD might call Zen chips "Zeppelin"


    Nothing to suggest that Zen will crash and burn

    Leaks are starting to emerge about an AMD Zen part which has the unfortunate name of Zeppelin.

    We are not quite sure what AMD was thinking when it named a part Zeppelin. After all you would have think that everything to do with the name was really bad. Firstly the rigid airships terrified the UK during WW1, but they had a nasty habit of crashing and exploding. The last one to be seen was drapped with nazi flags and exploded on its first trip to the US.
    So what is this part which is named after such a flying disaster so easily outclassed by other technology? Apparently it is a processor which allows Zen to offer 32 physical cores and 64 logical threads per socket.
    According to a post at the Linux Kernel Mailing List by AMD's 'Ray' Huang Rui, Zeppelin boasts of support for eight bundles of four cores on a single chip, or 32 physical processing cores.
    The Zen architecture is known allow each physical core to execute two threads simultaneously so a 32-core Zeppelin would run 64 simultaneous threads. This means that two four-socket systems may offer up to 256 threads per system.
    The number of cores suggested in the code fragment on the list might be a maximum supported. We might never see a 32-core monster-chip at launch. AMD is saying nothing about its Zeppelin at the moment. If it really does call it ZeppelinI bet it is hoping that it would be like this:

    Rather than this:


    Noticia:
    http://www.fudzilla.com/news/process...chips-zeppelin


    A ser verdade, espero que o nome signifique a AMD volta a voar alto com estes novos CPUs.
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  4. #154
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    AMD’s Zen Based Opteron Processors To Feature 32 Cores in MCM Package – 8 Channel DDR4 Memory Support

    In a discussion related to technology and market trends for the data center, Liviu Valsan (computing engineer at CERN) showcased slides which detailed AMD’s upcoming Opteron processors that feature their x86 Zen core architecture. The session not only provided details on AMD’s next generation processors but also mentioned Intel’s Broadwell-EP chips along with enterprise aimed storage and memory solutions.

    Slide/Image Credits: CERN Document Server
    AMD’s Zen Powered Opteron To Boast 32 Cores With SMT – 8 Channel DDR4 Support

    AMD’s Opteron lineup hasn’t seen any significant update since 2011. The Opteron 6000 series processors that are available today are based on the older Piledriver architecture which was introduced back in 2012. The x86 Piledriver core delivers a modest update over the Bulldozer architecture hence the Opteron processors really couldn’t stand out against the Intel’s server offerings which picked up the pace since Sandy Bridge. However, AMD expects to change the scenario with Zen and come back in the data center market with faster next-gen Opteron processors which will deliver high core count, faster performance and improved efficiency on a 14nm process node.
    The AMD Opteron processors should not be mistaken with the Exascale processor which is doing the rounds in the rumor mill. The Exascale processor is part of the HPC APU product stack which is a multi-chip package that comprises of several dies that incorporate Zen cores, GCN cores and DRAM stacks (HBM), interconnected via a faster coherent fabric that AMD has planned for their next generation exascale solutions and rumored to have interconnect speeds up to 100 GB/s. The AMD Opteron processors will solely comprise of a single or several CPU dies which will be interconnected to deliver high core count, disruptive memory bandwidth through using a combination DDR4 and HBM DRAMs and high native I/O capacity.

    During the talks, the CERN official stated that AMD will be building their next generation Opteron processors with their Zen core architecture. The new Zen core is based on the 14nm FinFET technology that is being built developed by Global Foundries and is expected to deliver up to 40% IPC (Instructions per clock) improvement. The other key point mentioned is that the Opteron processors will feature up to 32 physical cores. All 32 cores are based on the Zen architecture and will feature SMT (Symmetrical Multi-Threading) which brings the AMD design closer to Intel’s design as they have been using Hyper threading for quite some time. The technology works is a similar way, albeit with some AMD added enhancements/tweaks.
    Advertisements

    One key point explained by Liviu is that AMD might go with the MCM design as they currently do with their Opteron 6300 series processors. The Opteron 6380 packs two 8 core “Piledriver” CPUs on a single die that functions as a single processor. The Zen based Opteron will pack 32 cores in several stacks to form a package which will function as a single 32 core Opteron chip. The CPUs would then be interconnected through a next generation fabric to eliminate data transfer and communication bottlenecks between the two chips. This information also gives us a slight bit hint at AMD’s design strategy. With Piledriver, AMD used two 8 core modules on their Opteron while 8 core was set as the highest count on consumer processors.
    A die shot of an Opteron 6300 CPU with dual Piledriver dies on a single package via AMD
    The 32 core count count mean that the consumer and server offerings with single chip packages can pack up to 8 cores. The stacking would allow AMD to push several SKUs ranging from 8, 16, 24 and up to 32 core models. Although it would be AMD’s decision entirely to offer such SKUs in the market but having that many cores on enterprise platforms will be interesting. On desktop consumer platforms, AMD will have up to 8 core FX series processors serving the enthusiast space. Intel themselves have upped the core count to 10 with the upcoming Broadwell-E HEDT and 22 on the Broadwell-EP generation of processors. All Intel chips utilize a single CPU design unlike AMD’s multi-chip design.
    Lastly, support for 8 channel DDR4 memory and PCI-e Gen 3.0 was mentioned. Eight channel sounds like a step up for AMD. AMD’s Opteron platform is expected to hit shelves later in 2017 while Intel has their Broadwell-EP platform launching in Q1 2016. AMD’s platform will be competing by then with Intel’s Purely platform that has Skylake-EP stacked up against it. While AMD’s Opteron platform looks decent enough against Broadwell-EP as it has 32 cores versus 22 on Intel, support for 8 channel DDR4 versus 4 channel DDR4 on Intel and a fast inter connect fabric, things might shape up more interesting with 2017’s Purely which brings Intel’s latest micro-arch codenamed Skylake to the enterprise market. Nevertheless, AMD believes that they can address 80% of the server market with their Zen powered Opteron processors.
    Okay and then as a follow up, what kind of performance point should we look at your initial Zen in the server space, is it Xeon E3 class type products or E5s or maybe you can help with some granularity there if you can?
    Yes, we believe that we’ll be able to address let’s call it you know 80% of the server CPU market with our Zen class of products. So that’s a very high end but you know really the meat of the market. AMD via SeekingAlpha
    AMD Opteron Comparison To Intel’s Server Lineup:

    Product Family Intel Xeon E5-2600/4600 V4 AMD Opteron "Zen" Intel Xeon E5-2600/4600 V5
    Family Branding Broadwell-EP Next Generation Opteron (Name TBC) Skylake-EP
    Process Node 14nm FinFET 14nm FinFET 14nm FinFET
    Xeon/Opteron Platform Intel Grantley AMD Opteron Intel Purely
    PCH C610 Series TBC Lewisburg PCH
    Socket Socket R3 TBC Socket P
    Interconnect N/A 100 GB/s AMD Coherent Fabric Storm Lake Gen1
    Max Core Count 22 32 MCM Package 26 (TBC)
    Max Thread Count 44 64 52 (TBC)
    Max L3 Cache 55 MB TBC 65 MB (TBC)
    Max PCI-Express Lanes 40 PCI-E Gen3 40+ PCI-E Gen3 48 PCI-E Gen3
    DDR4 Memory Support 4-Channel DDR4 8-Channel DDR4 6-Channel DDR4
    TDP Range 55-145W TBC 45-160W
    Launch Expected Q1 2016 2017 1H 2017


    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  5. #155
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    AMD Patch Reveals Zen Cache Changes


    With AMD CPU market share down in the pits, everyone is looking towards Zen to give Intel some competition in the high-end segment. Even with DX12 reducing the need for a strong single threaded CPU, there is nothing like a good architecture that can compete with Intel, especially in non-gaming applications. In their latest patch dealing with Zen, AMD has revealed some more details about the inner workings of their upcoming CPUs, with the focus on the caches.
    First off, Zen will introduce a new L0 cache, meaning that there will actually be 4 levels of cache. The L0 cache is a uOp cache, something Intel added back with Sandy Bridge. Paired with the uOp buffer, this will help reduce power consumption when running loops or if something needs to be re-executed quickly. Intel’s cache is 1.5KB so we can probably expect AMD to follow similarly as speed is more important than size.

    Next up are changes to the L1 Instruction and Data caches. The L1 I$ will be 32KB, a drop compared to Steamroller/Excavator and K10 but back to the same size as Piledriver. The L1 D$ is also expected to be 32KB, a doubling over Steamroller and the same that of Excavator though still lower than K10. The reduced L1 I$ may be offset by the new uOp cache. The L2 may remain the same since the days of K10, with 512KB. This may be a problem if the rumoured inclusive cache design is used as 2304of the rumoured 8MB of L3 will be used in duplicating data. Having everything duplicated in L3 may make for better core-sharing and multi-threaded performance but limits everything to near L3 speeds for cache writes.
    Overall, the cache changes suggest a move to ensure faster, rather than large caches. The increases to the caches also point to the focus on keeping the cores fed as well as high-speed cores with a long pipeline. This all helps with the 40%+ IPC improvement AMD is hoping for with Zen. Overall, Zen is looking to be a very wide and balanced design, borrowing from Intel and K10 but without any of the baggage of the past.
    Noticia:
    http://www.eteknix.com/amd-patch-rev...cache-changes/
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  6. #156
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    7,793
    Likes (Dados)
    0
    Likes (Recebidos)
    1
    Avaliação
    41 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Não sei o que pensar disto. A AMD procura reduzir consumos e aumentar velocidades?
    Essas caches são a base de todos os novos cpus Zen?
    Ideias sem Nexo e Provas do Tráfico de Hardware
    "que personifica o destino, equilíbrio e vingança divina." Dejá vú. Que cena!

  7. #157
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    AMD Betting Big On Zen Processor To Bring Company Back To Process And Performance Parity With Intel

    It's 2016, and it feels like we've been waiting forever for AMD to launch a strong attack against Intel with a fresh group of desktop processors that compete in the high-end. With Intel dominating the CPU scene for so long, three user groups have emerged: those who still buy AMD chips because the value proposition is still quite good, those who want to buy AMD to support the company but can't resist the allure of higher-performing competitive CPUs, and those who are completely indifferent to a brand and just buy whatever suits them.
    With Zen, AMD is hoping that the latter two groups will be swayed. That might sound like a lofty goal and a pipe dream at this point, but AMD is having none of that. The company believes that Zen could be its first competitive architecture versus Intel in the high-end in quite some time. And let's be honest, even if AMD was to come out offering close to clock-for-clock performance parity with Intel, the market would consider it a huge win.

    Can we buy into the hype given all we've seen over the years? It's hard to say, but Zen is unlike any architecture AMD has crafted since the day Intel began its distancing itself from AMD's performance further and further. The most important thing to note about Zen is that it's not an upgraded architecture; it's all new and built from the ground up. As a result, it lacks some of the inefficiencies of AMD's aging architectures and brings with it new optimizations for efficiency and performance.
    image: http://hothardware.com/ContentImages...lFoundries.jpg

    GlobalFoundries will build AMD's 14nm Zen processors While it was clearly stated that Zen is going to offer a substantial improvement over AMD's own previous generation chips, it's nice to see the company so confident about it competing well with Intel again - that's exactly what fans and mainstream consumers alike have been craving for so long. On the instructions-per-clock (IPC) front, Zen could deliver up to a 40% improvement. If we're lucky, Zen could even power the highest-end gaming PCs without any significant compromise.
    AMD is doing a lot here to re-architect Zen to make sure it delivers as planned. The new CPU will be built on a FinFET process, for starters, which in itself should increase performance and power efficiency. However, it will also be built at 14nm, just like Intel's current crop of chips. Critical for those running virtual machines or other highly-threaded apps, AMD has specifically optimized performance here as well, as the chip reportedly balances threaded workloads more efficiently.
    We're still so far away Zen that it's easy to get excited over these little tidbits of detail at the moment, but this architecture truly does have potential. Let's hope that AMD can deliver here, because not only do consumers need better options in core CPU technologies but also AMD as a company is long overdue for market share success versus Intel on the CPU front.


    Noticia:
    http://hothardware.com/news/amd-bett...654OZQGxrE4.99




    Espero mesmo que sim, que os Zen venham finalmente dar alguma luta à Intel... já vamos a caminho de uma década de desaparecimento da AMD no campo dos CPUs.
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  8. #158
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    7,793
    Likes (Dados)
    0
    Likes (Recebidos)
    1
    Avaliação
    41 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Se for como com as gráficas, que parecem estar lentamente a acordar, vão mesmo dar luta.
    Aos poucos, estao a subir na minha consideracao novamente.
    Ideias sem Nexo e Provas do Tráfico de Hardware
    "que personifica o destino, equilíbrio e vingança divina." Dejá vú. Que cena!

  9. #159
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Para já são apenas e só promessas... esperemos que se concretizem para deixar de existir monopolio de uma marca apenas como tem acontecido nos ultimos 10 anos.
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  10. #160
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    AMD Zen 8 Core 95W Summit Ridge CPUs Launching In October – Chips Taped Out, Going Through Validation

    AMD Zen 8 core high enthusiast Summit Ridge CPUs are allegedly slated for an October release on the new AM4 socket. A source claims that AMD has already taped out the eight core Summit Ridge CPU dies in January and are running them through testing and validation. This is the second major milestone that we’ve heard about Zen thus far. The first being the tape out of the Zen core / microarchitecture back in 2015. This means that not only has the core design been finalized, but the eight core SOC – system on a chip – featuring Zen has also been completed.

    The CPUs allegedly feature a 95W TDP and eight high performance Zen cores with multi-threading for a total of 16 threads. Zen is said to have extremely competitive single threaded performance, with more instructions per clock than Intel’s Broadwell and just a smidgen behind Skylake. The 95W TDP if accurate indicates that indeed we’re looking at a very power efficient design. In contrast, Intel’s eight core i7 5960X is rated at 140 watts but then again we’re comparing a 14nm based product with a 22nm chip so a delta is to be expected.
    Functional untethered 8-core znver1 Summit Ridge A0 engineering samples with an estimated 95W TDP have now taped out on Samsung’s 14nm FinFET and are undergoing validation.
    It’s not yet clear how well they will clock (but Samsung’s 14nm FinFET process is slightly better than Intel’s – Intel haven’t pulled ahead with 10nm yet as that EUV process is very late, hence their Kaby Lake). IPC efficiency is very roughly equal with Broadwell, very slightly behind Skylake, but it’s push and shove.
    They are currently targeting October 2016 (ish) for release. AM4 motherboards compatible with Summit Ridge have chipsets already shipping and should be coming out around about March 2016 (ish).
    Subject to change. They may dual-source on GF14A.
    AMD Zen 8 Core 95W Summit Ridge CPUs To Be Released In October

    We published an exclusive report last year detailing AMD’s plans to introduce its Zen based FX CPUs later this year and Zen based APUs in 2017. We also broke down details about Zen CPUs, APUs, AM4 motherboards and the features of AMD’s next generation platform which haven’t been made public yet. So we would highly recommend that you go check out that report if you’re interested in finding out more about Zen. Suffice to say all the leaks and whispers we’ve seen and heard thus far about AMD Zen CPUs on the desktop indicate that the chips will launch in Q4 this year, more specifically this October.
    Desktop Zen CPUs are part of the “Summit Ridge” family. Which is going to include a full lineup of high-end Zen based CPUs that will feature “high core counts” and compete in the “enthusiast” segment according to AMD. That is, they’re going to be lined-up to compete against Intel’s Haswell-E and Broadwell-E products.






    AMD Zen Desktop CPUs Are Shaping Up To Be Everything That Enthusiasts Have Been Asking For

    Zen has been one of the most hotly anticipated AMD products in a decade. It’s the company’s first attempt to compete at the high-end CPU space in five years. In many ways Zen also represents the company’s first truly innovative next generation architecture that’s also on process node parity with Intel since the Athlon days. What has lend the tech community even more optimism is that Zen harkens back to AMD’s glorified Athlon days in another even more crucial aspect. Zen is a brand new clean-slate design that’s been led from the get-go by accomplished CPU architect Jim Keller. The very same person that brought us the original Athlon XP and Athlon64 processors. AMD’s most successful products ever.
    We saw this optimism reflected at CES by AMD’s CEO.
    “The overall expectation is that AMD will be a better year from a financial standpoint compared to 2015,”
    Su went on to state that she has full confidence that by CES next year everyone will say “AMD is back!”.
    Up until May of last year very little was known about AMD’s next generation high performance, high IPC, 14nm CPU core. However, we’ve learned so much about it since then through a consecutive set of Linux Kernel patches and leaks. Which have been very instrumental in helping us understand the inner workings of Zen’s micro-architecture and what it’s capable of.
    Back when AMD announced Zen for the very first time, publicly, at its Financial Analyst Day last year, the company touted a 40% improvement in instructions per clock versus its latest high performance core Excavator. However, during last quarter’s earnings conference call President and CEO of the company Lisa Su announced that Zen is performing beyond initial expectations and that engineers have managed to achieve a greater than 40% IPC improvement.
    AMD Preisdent & CEO Lisa Su – Q4 2015 AMD Earnings Call Transcript
    Our Zen-based CPU development is on track to achieve greater than 40% IPC uplift from our previous generation and we’re on schedule to sample later this year.
    In client computing, our opportunities to regain share in 2016 will be driven by our design win momentum, continued progress expanding into
    the commercial market, and reentering the high-performance desktop market late in the year with our Zen-based Summit Ridge CPU.
    AMD announced that later this year it will be introducing an entirely new line-up of FX CPUs and a brand new platform ‘AM4″. These new FX CPUs are what AMD is referring to as “Summit Ridge”. The new platform will include a new socket with DDR4 memory support. AM4 will also support upcoming Bristol Ridge APUs. This is in effort to unify all of APU & CPU desktop platforms under one roof instead of the current bifurcation between AM3+ and FM2+.
    The AIDA64 benchmark added support to Zen based Summit Ridge as well as Raven Ridge FX CPUs and APUs last year. Raven Ridge will succeed Bristol Ridge and will be the first generation of APUs to feature Zen, while Bristrol Ridge will be a refresh based on the 28nm Excavator core. Several leaks have also revealed that AMD is working on several Zen APUs with large discrete class built-in GPUs and high bandwidth memory.
    Advertisements

    On the server side AMD is preparing Zen based CPUs with up to 32 cores and 64 threads. That’s four times as many threads as AMD’s current highest end server offering. A very significant jump that’s going to play a crucial role in giving Zen the competitive edge it needs in servers. A market that Intel has virtually monopolized with 97% market share.
    AMD Zen FX CPUs Will Launch On A Modern, More Comprehensive, Feature Packed Platform

    In our exclusive report we published last year we shared the details that have been made available to us about AMD’s Summit Ridge processors coming later this year. For starts Summit Ridge CPUs will include models with eight, six and four Zen cores. Four cores being the lowest core count Summit Ridge CPU AMD is going to sell. All CPUs will include the company’s new platform security processor, PCIe 3.0 support, dual channel DDR4 memory controllers, copious amounts of L3 cache and updated storage features.

    AMD’s plan .our sources tell us. is to price high core count parts very competitively. Undercutting Intel’s mainstream CPUs by offering more cores with competitive – Broadwell level – single threaded performance at similar price points. Essentially selling Haswell-E / Broadwell-E comparable parts at Intel’s mainstream price points. A space that’s currently occupied by Intel’s mainstream quadcore I7, i5 and dual core i3 CPUs .
    This is possible because we’re told the Zen core itself is very area and power efficient. So AMD can build high core count chips that are still small, with good yields and at reasonable cost. Imagine an eight core Haswell-E equivalent chip the size of a Carrizo APU. We’re set to see a CPU price war take place later this year like we haven’t for nearly a decade.
    AMD CPUs

    WCCFTech AMD Summit Ridge AMD Bristol Ridge AMD FX "8000 Series" AMD A-Series "7000/8000-Series"
    Product Segment Performance Desktop Processors "FX" Mainstream Desktop and Mobility APU Performance Desktop Processors "FX" Mainstream Processors "Kaveri/Godavari"
    Product Architecture x86 Zen x86 Excavator x86 Bulldozer/Piledriver x86 Steamroller
    Process Node 14nm 28nm 32nm 28nm
    Max CPU Cores TBA TBA 8 4
    GPU Architecture N/A TBA N/A GCN 1.1
    TDP TBA TBA 125-220W 95W
    Socket AM4 AM4 AM3+ FM2+
    South Bridge Promontory Promontory/SOC 990FX A88X
    Memory Support DDR4 DDR4 DDR3 DDR3
    Launch Q4 2016 1H 2016 2011-2016 2013-2016
    The Zen Microarchitecture

    14nm FinFET Process

    Samsung’s 14nm FinFET process which Globalfoundries has licensed is going to be the basis of all of AMD’s next generation CPU, APU and GPU products. The company’s Chief Technology Officer Mark Papermaster made the announcement last year that the 14LPP process – Samsung’s high performance 14nm process – will be leveraged across all of AMD’s future products.

    Mark Papermaster
    FinFET technology is expected to play a critical foundational role across multiple AMD product lines, starting in 2016, GLOBALFOUNDRIES has worked tirelessly to reach this key milestone on its 14LPP process. We look forward to GLOBALFOUNDRIES’ continued progress towards full production readiness and expect to leverage the advanced 14LPP process technology across a broad set of our CPU, APU, and GPU products. – Press Release
    The 14LPP process features 3D finfet transistors and significantly smaller gates compared to 28nm. Chips manufactured on the 14LPP process have more than double the density of 28nm. Additionally, 14nm FinFET transistors are capable of significantly faster switching speeds. Products designed for the process will be able to achieve much higher frequencies than the same designs on 28nm. The process also delivers considerable power savings, essentially cutting power consumption by half.
    14nm FinFET Technology

    Globalfoundries.com
    14LPE – Early time-to-market version with area and power benefits for mobility applications

    14LPP – Enhanced version with higher performance and lower power; a full platform offering with MPW, IP enablement and wide application coverage
    The High-Level Design Of AMD’s Zen Core

    AMD’s Linux Kernel patches which have been a rich source of information about Zen. The patches revealed so much about the inner-workings of Zen to the point where we are actually able to visualize the high-level design of the core. If you’re interested in all the nitty gritty details about what the core is capable of I’d highly recommend our exclusive in-depth analysis of Zen’s micrarchitecture.
    WCCFTech.com Rendition Of Zen’s High-Level Design Based On The AMD Linux Kernel Patch
    Above you can see a visual representation of Zen on the right in comparison to AMD’s Steamroller CPU core. Which is the third generation based on the Bulldozer design and is pretty much identical from a high-level standpoint to Excavator, the fourth and last Bulldozer based core. There are several key differences between Zen and the bulldozer family.
    AMD has done away with the CMT – clustered multi-threading – concept in favor of a more traditional SMT – sumultaneous multi-threading – design. This means that each Zen core will be able to execute two threads simultaneously. One main, very high throughput thread and one secondary thread that can be used opportunistically.
    In contrast, each Bulldozer module can execute two equal threads. This is achieved through two separate integer clusters with a single front-end. This approach saves area versus building two separate cores and delivers two high throughput threads. However, there are advantages that Zen’s SMT implementation holds over the Bulldozer CMT implementation. For one it allows AMD to build a single larger integer cluster with significantly higher single threaded performance. Another advantage with this approach is that it still leaves room for opportunistic savings in area and power.
    The final result is similar overall throughput when we look at both treads of each SMT core vs both threads in each CMT core. However, the SMT style core will deliver significantly higher single threaded performance. Furthermore, because each Bulldozer module houses two integer clusters and a single floating point unit it was always very integer heavy. Each Zen core on the other hand includes one large integer cluster and one large floating point unit, so it’s a much more balanced design.
    CPU Microarchitecture AMD Phenom II / K10 AMD BD/PD AMD SR/XV AMD Zen Intel Skylake
    Instruction Decode Width 3-wide 4-wide 8-wide 4-wide 4-wide
    Single Core Peak Decode Rate 3 instructions 4 instructions 8 instructions 4 instructions 4 instructions
    Dual Core Peak Decode Rate 6 instructions 4 instructions 8 instructions 8 instructions 8 instructions
    A lot of the engineering effort around Zen has also been done to address one of Bulldozer’s major flaws. Bulldozer and Intel’s Sandy Bridge – and subsqeuent Intel architectures including Skylake – had equally deep pipelines to achieve high clock speeds. The deeper the pipeline the more latency that design will exhibit. Particularly when it comes to branch misprediction errors, which are quite common in modern pipelines.
    The latency that results from branch mispredicts are quite significant. To combat this issue Intel introduced a micro-op cache with Sandy Bridge. It worked to considerably reduce mispredict penalties and it was the principle reason why Sandy Bridge had much better single threaded performance as opposed to Bulldozer. The latest Linux Kernal patch as well as a group of AMD patents indicate that the company has implemented a similar solution in Zen.
    Matthias Waldhauer
    A lot of the new functionality has been filed for patenting. For example there was a mention of checkpointing, which is good for quick reversion of mispredicted branches and other reasons for restarting the pipelines. Some patents suggest, that Zen might use some slightly modified Excavator branch prediction.
    Bringing this to a close, it’s clear that AMD’s is doing a lot of things right with Zen. Pushing IPC and power efficiency to where they need to be. Building a comprehensive modern platform and bringing much needed updates to the feature-set. Creating an attractive value proposition for desktop users, servers and notebooks. All the ingredients to make Zen a success are here, all that’s left is for AMD to execute and deliver. The mere prospect that enthusiasts may actually have AMD CPUs as a worthwile option again for the first time in a decade come this October is refreshing. And maybe, just maybe, we’ll finally be able to say “AMD’s back”.


    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  11. #161
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    AMD Zen ‘Summit Ridge’ CPUs and Intel ‘Kaby Lake’ CPUs Spotted in AIDA64 Changelog

    The popular hardware testing utility, AIDA64, has added and improved support for Zen ‘Summit Ridge’ and Intel ‘Kaby Lake’ processors. While developers usually have free control of when to introduce these updates, they are usually backed by real documentation such as Machine IDs. It looks like AMD’s anxiously anticipated Zen processor is on track. It also means that the A0 engineering samples of said processors will be rolling out to select testers in the coming months.

    Intel Kaby Lake and AMD Zen ‘Summit Ridge’ processors show up in AIDA64 changelog

    The change-log for the latest version of AIDA64 (AIDA64 Extreme Edition v5.70.3800) was recently updated and contains more than one interesting entry. Developers of hardware testing tools are among the first to receive early documentation and A0 engineering samples of upcoming processors and are a good point of information for the progress of a particular product. In this case, it looks like things are going well on track (according to the information provided to these developers anyways) for AMD’s upcoming Summit Ridge platform and Intel’s Kaby Lake.

    The Summit Ridge platform is the mainstream desktop platform of Zen processors having 8 cores a pop. Note that we are talking about full fledged SMT cores here and not the older styled CMT-based cores favored in the older FX processors. The socket stated on the slides was FM3 (which is now known as AM4). The new processors are supposed to increase IPC gains by 40% over the last generation and should, hopefully, see the compute side of AMD become competitive again. Not to mention that pretty much everything (financially speaking), for AMD, depends on Zen’s performance.
    Advertisements

    READ AMD's Zen Based Opteron Processors To Feature 32 Cores in MCM Package - 8 Channel DDR4 Memory Support

    We had previously heard that AMD will be making a plethora of changes including changing to SMT and adopting a single FPU per core design – both things which will put it on track to be a competitive force with Intel processors. That’s not it either, Zen will be using a scheduling model that is similar to Intel’s and it will use specific hardware and simulation to define any needed scheduling or NUMA changes. It will also be ISA compatible with Haswell/Broadwell style of compute. It will bring various compiler optimizations, including GCC with target of SPECint v6 based score at common compiler settings. Bench-marking and performance compiler LLVM targets SPECint v6 rate score at performance compiler settings. Each Zen core will have access to 512KB of L2 cache and 4 Zen cores will share 8MB of L3 cache.
    As far as Intel’s Kaby Lake is concerned – that particular platform is fairly straightforward. Intel has finally let go of the tick-tock cadence that it had been using for so long. The new cadence, dubbed “Process, Architecture, Optimization” (sounds much less catchy) relies on at least one more iteration at the same die shrink for ironing out the kinks in the micro-architecture completely.
    In this case, Broadwell is the “Process”, Skylake is “Architecture” and Kaby Lake will be “Optimization”; Intel’s 14nm process finally perfected. Intel plans to make several IPC and architectural improvements in Kaby Lake including a much more powerful iGPU configuration. They are also expected to go face to face against AMD’s 8 Core zen processors so it is expected that Intel will be able to bring IPC gains of 10% (over Skylake) while staying within the 95W TDP limit and LGA 1151 on the Z270 chipset.
    READ Legendary CPU Architect Jim Keller Has Left AMD - x86 Zen Core Still on Track

    WCCFTech AMD Summit Ridge Intel Kaby Lake
    Product Segment Enthusiast / Mainstream Desktop PCs Enthusiast / Mainstream Desktop PCs
    Product Architecture x86 Zen (+40% IPC compared to Excavator) 7th Gen Kaby Lake (+10% IPC compared to Skylake)
    Process Node 14nm 14nm
    Max CPU Cores ~8 4
    GPU Architecture TBD Next Gen HD Graphics
    TDP TBD Up to 95W
    Socket AM4 LGA 1151
    PCH Promontory Z270
    Memory Support DDR4 DDR4
    Launch Q4 2016 Q4 2016


    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  12. #162
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,121
    Likes (Dados)
    0
    Likes (Recebidos)
    2
    Avaliação
    1 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    AMD Zen 8 Core Engineering Sample Runs at 3Ghz


    AMD’s upcoming Zen architecture is arguably the most anticipated hardware release this year. After years in the wilderness, AMD will finally come back with a new CPU design that will challenge Intel again on IPC, process node and power efficiency. According to the latest leak, it appears that Zen is progressing well enough that engineering samples have already been distributed to various partners for testing. This also means AM4 motherboards are already sampling as well.
    These stepping A0 samples are that of the previously rumoured 95W, 8 core Zen CPU. That AMD has managed to get an 8 core CPU in a 95W thermal envelope is stunning and combined with the early engineering sample release, points to a strong 14nm LPP process. What’s more, the frequency isn’t a slouch, at 3Ghz base though boost isn’t enabled yet. This is pretty much the same as the base clocks for Intel’s own prosumer i7 5960X which sports 8 cores as well at 3Ghz base and 3.5Ghz boost. We can expect the Es to set the baseline so release Zen will almost certainly clock higher.
    At 3GHz, the engineering sample is already faster than the first Bulldozer ones suggesting that 14nm LPP won’t be holding back frequency too much. After all, Intel’s own 14nm process has performed better than their 22nm. Samsung and Global Foundries have also had plenty of time to refine their 14nm process to ensure it will offer the best performance at launch. Hopefully, AMD will be able to be competitive in both IPC and overclocking.
    Noticia:
    http://www.eteknix.com/amd-zen-8core...sample-leaked/

    3GHz para um octo-core não me parece mau, mas esperava que a AMD fosse um pouco mais agressiva nas velocidades de relogio.
    Resta aguardar pelos testes destas samples para ver o que isto vale e se trás novamente a AMD a um patamar competitivo.
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  13. #163
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    7,793
    Likes (Dados)
    0
    Likes (Recebidos)
    1
    Avaliação
    41 (100%)
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Eu tambem esperava um pouco mais. Ainda falta ver os turbos e capacidade de oc deles, e se 8=8 ou 8=4 ou 6.
    É sempre um pouco relativo basear-nos só na velocidade, pois o 5960x tambem tem só 3.0ghz base clock e 3.5 em turbo e todos sabemos a besta que é.
    Ideias sem Nexo e Provas do Tráfico de Hardware
    "que personifica o destino, equilíbrio e vingança divina." Dejá vú. Que cena!

  14. #164
    O Administrador Avatar de LPC
    Registo
    Mar 2013
    Local
    Multiverso
    Posts
    17,813
    Likes (Dados)
    74
    Likes (Recebidos)
    154
    Avaliação
    31 (100%)
    Mentioned
    31 Post(s)
    Tagged
    0 Thread(s)
    Citação Post Original de Enzo7231 Ver Post
    Eu tambem esperava um pouco mais. Ainda falta ver os turbos e capacidade de oc deles, e se 8=8 ou 8=4 ou 6.
    É sempre um pouco relativo basear-nos só na velocidade, pois o 5960x tambem tem só 3.0ghz base clock e 3.5 em turbo e todos sabemos a besta que é.
    Boas!
    Veremos é como o zen se comporta VS 5820k...

    Cumprimentos,

    LPC
    My Specs: .....
    CPU: AMD Ryzen 7 5800X3D :-: Board: MSI B550M BAZOOKA :-: RAM: 64 GB DDR4 Kingston Fury Renegade 3600 Mhz CL16 :-: Storage: Kingston NV2 NVMe 2 TB + Kingston NV2 NVMe 1 TB
    CPU Cooling Solution: ThermalRight Frost Commander 140 Black + ThermalRight TL-C12B-S 12CM PWM + ThermalRight TL-C14C-S 14CM PWM :-: PSU: Corsair HX 1200 WATTS
    Case: NZXT H6 FLOW :-: Internal Cooling: 4x ThermalRight TL-C12B-S 12CM PWM + 4x ThermalRight TL-C14C-S 14CM PWM
    GPU: SAPPHIRE
    NITRO+ AMD RADEON RX 7800 XT - 16 GB :-: Monitor: BenQ EW3270U 4K HDR


  15. #165
    Tech Mestre Avatar de SleepyFilipy
    Registo
    Apr 2013
    Posts
    3,972
    Likes (Dados)
    13
    Likes (Recebidos)
    19
    Avaliação
    0
    Mentioned
    1 Post(s)
    Tagged
    0 Thread(s)
    Na verdade tem que se comparar com o que existir da Intel na mesma ocasião.

    Sinceramente não espero que a AMD se consiga bater de frente com a Intel (embora gostasse imenso de me enganar e ser surpreendido).

    Mas quem sabe aparece uma plataforma AMD com custo/performance imbatível.
    Dell G15

 

 
Página 11 de 127 PrimeiroPrimeiro ... 9101112132161111 ... ÚltimoÚltimo

Informação da Thread

Users Browsing this Thread

Estão neste momento 2 users a ver esta thread. (0 membros e 2 visitantes)

Bookmarks

Regras

  • Você Não Poderá criar novos Tópicos
  • Você Não Poderá colocar Respostas
  • Você Não Poderá colocar Anexos
  • Você Não Pode Editar os seus Posts
  •