Registar
Página 4 de 91 PrimeiroPrimeiro ... 234561454 ... ÚltimoÚltimo
Resultados 46 a 60 de 1357

Tópico: Nvidia Pascal

  1. #46
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    Nvidia Pascal Flagship May Arrive Sooner Than Expected


    Even as this generation’s GPUs are continuing to fly off the shelves, Nvidia is already gearing up for their Pascal launch. Despite being quieter than AMD, it looks like Nvidia will launch their Pascal cards around the same time, in 2H 2016 as AMD’s Polaris will. What’s more, 2H 2016 will see Nvidia’s flagship Pascal GPU based on TSMC’s 16nmFF+ process and utilizing HBM2. This is still a rumour right now but it does fit the time frame since 1H 2016 would be too soon and 2017 too late.
    The biggest question is what does”flagship” mean exactly. Ever since GTX 680 was launched, Nvidia has been playing around with the word flagship. Traditionally, the big dies like GF100 would launch first with the smaller mainstream GF104 launching after. Kepler and Maxwell saw that switch with GK104 and GM104 launching ahead of GK100 and GM100 respectively. This suggests that the so-called “flagship” may only be GP104 and not GP100. Even if it is GP100, it may well be a cut-down version, similar to how the GTX 780 was the cut-down variant of the later GTX 780Ti. This strategy does maximize sales for Nvidia but isn’t that great for consumers.
    Whatever the card is, be it GP104 or GP100, it is going to use HBM2, giving it at least 512GB/s with 8GB of VRAM but potentially much higher at 16-32GB with 1TB/s+ of bandwidth. With AMD set to launch Polaris around the same time, Q3 2016 should make for exciting times as a slew of new GPUs arrive.
    Noticia:
    http://www.eteknix.com/nvidia-pascal...oner-expected/


    Será que vem mesmo?
    E será para se antecipar a algo que a AMD coloque cá fora com a arquiteura Polaris?
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  2. #47
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    6,589
    Avaliação
    41 (100%)
    Se lançarem, entram na história como os primeiros a lançarem uma gráfica com HBM2.0.
    Não me admirava muito que o lançassem já daqui a uns 2 ou 3 meses. Mudavam a pressão para o lado da AMD. Acho que verem os carregamentos das boards Polaris a circular por aí lhes fez alguma comichão.
    Devido à falta de espaço na assinatura, resolvi colocar em "Acerca de mim" os meus projectos]
    http://www.portugal-tech.pt/member.php?u=801

  3. #48
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    E também devem ter ficado com alguma comichão depois de verem a demonstração do Polaris que AMD fez na CES... mesmo sendo um portatil, faz sempre alguma comichão
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  4. #49
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    Nvidia Pascal/HBM2 GPUs rumoured to launch in 2nd half of 2016

    This week, a new rumour going around suggests that Nvidia is set to begin mass production of its new Pascal GPUs equipped with HBM 2 during the first half of this year, with availability scheduled for the second half of 2016. This comes just a few days after Samsung announced that it had begun mass producing its first 4GB HBM 2 DRAM chips.
    This particular rumour comes from DigitalTimes in Korea, so keep in mind that we can’t really speak for its accuracy. According to the report, Nvidia will enter mass production and finish off reliability testing for its HBM 2 graphics cards during the next five or six months, these new cards will then be released in the second half of the year.

    Pascal is Nvidia’s latest GPU architecture and these new graphics chips will be made using the 16nm FinFET process. Right now, not much is known about Pascal in terms of real world/gaming performance but we do know that this rumour conflicts with the previous talk of Pascal being available during the first quarter of 2016.
    Noticia:
    http://www.kitguru.net/components/gr...-half-of-2016/


    Parecem que existem "conflitos" entre as noticias, umas antecipam, outras colocam mais tarde... é complicado assim ter uma data mais ou menos previsível.
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  5. #50
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    NVIDIA to launch HBM2-powered GeForce GTX Titan X successor in April


    If you have been reading our GPU-related content, you should know that we are set for the biggest year in GPU history this year, from both sides: AMD and NVIDIA.


    Well, at NVIDIA's GPU Technology Conference in April, we should see NVIDIA unveil the biggest GPU they've ever made - the successor to the GeForce GTX Titan X. The next-gen card could be called the GTX Titan X2, which would pull some of the wind out of AMD's sails with the dual-GPU Radeon R9 Fury X2, and we should see it featuring HBM2 - scaling up to 16-32GB with 1TB/sec of memory bandwidth. Insanity.

    Back in September, we exclusively said that NVIDIA would release both a HBM2 and GDDR5X range of cards - something that will kick off with the HBM2-based Titan X successor. Towards June, we should see NVIDIA unveil a new GP104-based GeForce GTX 980 successor, based on GDDR5X - which offers 448GB/sec of memory bandwidth.

    Then we have the elusive GeForce GTX 980 Ti successor, which will also be powered by HBM2. I think this card will most likely arrive sometime later in the year, depending on AMD's movement in the enthusiast GPU sector. The Titan X successor will be incredibly fast (I would say 1.5-1.8x the Titan X) and so will the GDDR5X-powered GTX 980 successor (again, most likely 1.5x the GTX 980). These two cards will fill the most important parts of the market, while keeping its secret weapon (the GTX 980 Ti successor) waiting in the darkness.

    Noticia:
    http://www.tweaktown.com/news/49956/...ril/index.html


    Será a primeira placa da nVidia a utilizar o HBM, neste caso de segunda geração.
    Não é referido na noticia, mas espera-se que seja um chip da geração Pascal e se assim for vamos ter em principio o topo de gama ou um dos topos como estreia.
    Outro ponto interessante é ver o que a nvidia consegue fazer com o HBM; isto quando já se sabe que a AMD com a sua melhor placa que utiliza o HBM não conseguiu ultrapassar a melhor placa da nVidia que utiliza GDDR 5.
    Outro ponto para seguir, é a performance que o novo processo fabrico vai trazer a estes chips e os consumos associados, sendo que na geração anterior a nvidia já os tinha num patamar bem baixo.








    Nvidia Big Pascal GP100 To Debut in April - Mid Range GTX 1080 in June

    There is a new roadmap floating around on the web showing some interesting stuff. According to that roadmap Nvidia Pascal GPUs are coming to market as early as April, the first one would be a Pascal Titan GPU with more mainstream models released later.
    While we all know that 14/15/16nm class product should become available in the May/June timeframe, an April release feels a little premature. Regardless, the high end SKUS will likely be fitted with a GP100 GPU based on a TSMC 16nm process fab. We'll call that GPU Big Pascal with billions of transistors (16 billion rumored).
    Two things can happen here, it will have HBM2 which is comlicated, or the cards will use GDDR5X. At the June timeframe Nvidia are rumored to launch the first high-end GP104 GPUs with models GTX 1080 and GTX 1070 (should Nvidia use their old naming scheme). These chips would come in a 37.5×37.5mm BGA package, and has 2152 pins.
    The leaked road-map definitely isn't from Nvidia themselves, it was spotted at techfrag, none the less is makes for a good read. Have a peek, click the thumbnail.


    Noticia:
    http://www.guru3d.com/news-story/nvi...0-in-june.html


    Entretanto esta noticia do GURU 3D confirma que é mesmo o chip Pascal e logo o mais poderoso, ficando agora o foco em saber quanto é que vaio acrescentar mais em performance face à Titan X.
    Vendo depois as restantes datas, confirma-se mesmo que isto vem mesmo primeiro que o Polaris da AMD (não me parece que a AMD saque alguma surpresa)... estará novamente a AMD metida em trabalhos e a correr atrás do prejuizo?
    Última edição de Jorge-Vieira : 27-01-16 às 09:54
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  6. #51
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    6,589
    Avaliação
    41 (100%)
    O sucessor da gtx980 com 1.5x mais performance que esta, fica praticamente igual à actual Titan X. Mais ou menos como a 970 ficou em relacao à 780ti.
    Penso que a Nvidia resolveu mesmo acelerar o passo rumo ao lançar primeiro as bombas. E vai ser cá um lançamento.
    Tudo isto vai mudar...tudo. Abre caminho para os fluidos e saudaveis 4k, RV, mais cor, brilho, menor consumo, tudo isto em menor tamanho. É suposto a resposta da AMD se seguir quase de imediato. E eu gostaria que fosse antes, e superior. Tudo o que eles têm dito e feito até ao momento, really smells like ATI spirit. Vamos ver se dá frutos já de inicio, ou se vao precisar de mais algum tempo.
    Devido à falta de espaço na assinatura, resolvi colocar em "Acerca de mim" os meus projectos]
    http://www.portugal-tech.pt/member.php?u=801

  7. #52
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    Four Different Nvidia Pascal Graphics Cards Spotted

    Nvidia’s Pascal GPUs are one of the most anticipated upcoming 2016 products from the company, and today’s news may very well build that anticipation even more . As it happens today we’ve learned that the company already has four different Pascal graphics cards going through testing and validation work.

    It seems the company is putting no time to waste getting it’s GP100 and GP104 Pascal GPUs out the door . The same GPUs that we’ve spotted on more than one occasion in the past. GP100 is the flagship Pascal GPU and the inevitable successor to Nvidia’s GTX Titan X has been spotted in transit. Going from TSMC’s fabrication plants to Nvidia’s testing facilities in India. Pascal is the code name for Nvidia’s upcoming GPU architecture scheduled for second half, 2016 market release. The GP100 GPU is the largest and most powerful of Nvidia’s Pascal graphics chips.
    Four Nvidia Pascal Graphics Cards Spotted In The Wild

    All four Nvidia graphics boards in question are described as “COMPUTER GRAPHICS CARDS”. However, all four carry very similar per unit values and as such we could be looking only at the circuit boards and not necessarily graphics cards but there’s really no way of telling for sure.
    All four boards start with the same 699 serial number and the earliest record of a board carrying that serial number appears in December. So we know that we’re looking at Nvidia graphics boards that are new and did not exist at any point before December. This could potentially explain Pascal’s absence from CES and why Nvidia chose to showcase the Pascal Drive PX2 module with Maxwell GPUs instead.

    There are four different boards here with the following serial numbers :
    699-2H403-0201-500
    699-1G411-0000-000
    699-1H400-0000-100
    699-12914-0071-100
    The 1H400, 1G411 and 2H403 units are all derivatives/variants of the same basic board , while the 12914 board is distinctly different. So what we have here looks very much like three evolutionary iterations with the first three boards. This is especially likely because there’s no overlapping between the three. One board shows up and it’s followed by another with no recurrence of the previous board.
    There’s no way of knowing for certain whether these are GP100 or GP104 boards as of yet. Interestingly GP100 or “Big Pascal” as we’d like to call it has been spotted a few months back. Back then Nvidia only had GPUs but there was no evidence of any actual boards. So looks like Pascal has come a long way since then.

    What we know so far about Nvidia’s flagship Pascal GP100 GPU :
    Advertisements


    • Pascal graphics architecture.
    • 2x performance per watt estimated improvement over Maxwell.
    • To launch in 2016, purportedly the second half of the year.
    • DirectX 12 feature level 12_1 or higher.
    • Successor to the GM200 GPU found in the GTX Titan X and GTX 980 Ti.
    • Built on the 16nm FinFET manufacturing process from TSMC.
    • Allegedly has a total of 17 billion transistors, more than twice that of GM200.
    • Will feature four 4-Hi HBM2 stacks, for a total of 16GB of VRAM and 8-Hi stacks for up to 32GB for the professional compute SKUs.
    • Features a 4096-bit memory bus interface, same as AMD’s Fiji GPU power the Fury series.
    • Features NVLink (only compatible with next generation IBM PowerPC server processors)
    • Supports half precision FP16 compute at twice the rate of full precision FP32.

    We’ve learned last year that Nvidia’s flagship Pascal code named GP100 may have taped out on TSMC’s 16nm FinFET manufacturing process in June. Interestingly just shortly afterwards AMD announced that it had taped out two FinFET chips. It’s absolutely not a coincidence that both companies completed their FinFET designs at the same time. Both are pushing for a very aggressive time to market timetable to debut their next generation FinFET based GPUs this year.
    Nvidia Pascal – 2X Perf/Watt, Stacked Memory, NV-Link And Mixed Precision Compute

    TSMC’s new 16nm FinFET process promises to be significantly more power efficient than planar 28nm. It also promises to bring about a considerable improvement in transistor density. Which would enable Nvidia to build faster, significantly more complex and more power efficient GPUs.
    TSMC’s 16FF+ (FinFET Plus) technology can provide above 65 percent higher speed, around 2 times the density, or 70 percent less power than its 28HPM technology. Comparing with 20SoC technology, 16FF+ provides extra 40% higher speed and 60% power saving. By leveraging the experience of 20SoC technology, TSMC 16FF+ shares the same metal backend process in order to quickly improve yield and demonstrate process maturity for time-to-market value.
    Apart from HBM2 and 16nm there is one big compute-centric feature that Nvidia will debut with Pascal. And it’s NVLink. Pascal will be the first GPU from the company to support this new proprietary server interconnect.





    NVIDIA Volta GPUs and IBM Power9 CPUs Enabled Supercomputers in 2017:The technology targets GPU accelerated servers where the cross-chip communication is extremely bandwidth limited and a major system bottleneck. Nvidia states that NV-Link will be up to 5 to 12 times faster than traditional PCIE 3.0 making it a major step forward in platform atomics. Earlier this year Nvidia announced that IBM will be integrating this new interconnect into its upcoming PowerPC server CPUs.

    NVIDIA® NVLink™ is a high-bandwidth, energy-efficient interconnect that enables ultra-fast communication between the CPU and GPU, and between GPUs. The technology allows data sharing at rates 5 to 12 times faster than the traditional PCIe Gen3 interconnect, resulting in dramatic speed-ups in application performance and creating a new breed of high-density, flexible servers for accelerated computing.
    Unlike Maxwell, Nvidia has laid major focus on compute and GPGPU acceleration with Pascal. The slew of new features and technologies that Nvidia will debut with Pascal emphasize this focus. Including the use of next generation stacked High Bandwidth Memory, high-speed NVLink GPU interconnect and the addition of mixed precision compute at double the rate of full precision compute to push perf/watt. We can’t wait to see Pascal in action later this year, but until then stay tuned for the latest.

    GPU Family AMD Polaris NVIDIA Pascal
    Flagship GPU Greenland/Vega10 GP100
    GPU Process 14nm FinFET 16nm FinFET
    GPU Transistors Up To 18 Billion ~17 Billion
    Memory Up to 16 GB HBM2 Up to 16 GB HBM2
    Bandwidth 1 TB/s 1 TB/s
    Graphics Architecture Polaris ( GCN 4.0 ) Pascal
    Predecessor Fiji (Fury Series) GM200 (900 Series)


    Noticia:
    http://wccftech.com/nvidia-pascal-gr...cards-spotted/
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  8. #53
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    Nvidia Pascal Launch May Be Sooner Than We Thought


    Last month we received word that the first Pascal chips would be launching a bit sooner than expected. At that time, GP100 was expected to drop in April and GP104 in June. According to the latest rumors, it looks like that timetable was accurate, with the GP100 based Tesla chip coming in April around GTC. What’s more, we’re getting more details about when the rest of Nvidia’s Pascal lineup will launch.
    As with the previous report, GP104 will arrive in June and it looks like the GTX lineup will be based off that, with both the GTX 1080 and 1070 being GP104 chips. Near the end of the year in Q4, we can expect GP106 and GP107. These will be longer end chips and likely power the GTX 1060 and 1050. Finally, we have the Titan which will use GP100 and a GP108 in early 2017. This follow’s Nvidia’s new trend of releasing a GTX x80 first, followed by the Titan, then finally a GTX x80Ti. While it’s good for Nvidia’s sales, it moves high-end users into a quicker upgrade schedule than if all the cards launched at the same time.

    The Tesla launching first makes a whole lot of sense as enterprise users can pay the high premiums for early HBM2 and 16nm. A June launch for GP104 may point to them using HBM2 as the timeline is a bit tight for GDDR5X though it is doable. The biggest question is how well Pascal will perform as it is a stopgap architecture between Maxwell and Volta, like a Maxwellv2 though the die shrink to 16nm should make some great gains in and of itself.
    Noticia:
    http://www.eteknix.com/nvidia-pascal...an-we-thought/
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  9. #54
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    6,589
    Avaliação
    41 (100%)
    E o suspense continua.
    Devido à falta de espaço na assinatura, resolvi colocar em "Acerca de mim" os meus projectos]
    http://www.portugal-tech.pt/member.php?u=801

  10. #55
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    Pascal in testing phase



    Sent to India
    The next-gen Pascal-based GPUs by Nvidia are now in the testing and validation phase.

    According to Tech Times the chips have been sent from the TSMC fabrication plants to Nvidia's testing facilities in India.TSMC won the contract against Samsung in the bid to mass-produce the Pascal GPUs using the company's 16nm FinFET production process.
    Nvidia's follow-up to Maxwell has some new technologies that include NVLink, HBM2 and Mixed Precision Support. At least four different Pascal GPUs are now being tested and seem to carry the same serial number of 699. These four different GPUs are the 699-2H403-0201-500, 699-1H400-0000-100, 699-1G411-0000-000, and the 699-12914-0071-100.
    The first three are variants that belong to one basic board which has a 12914 label. ll of the GPUs will have similar per unit values, with the GP100 GPU being the largest and the most powerful among the Pascal graphics chips
    The flagship GP100 GPU will also feature four 4-Hi HBM2 stacks and 16 GB of VRAM and 8-Hi stacks to reach up to 32 GB. This is clearly meant for top-of the-range professional SKUs.
    The GP100 GPU apparently also has up to twice the performance per watt estimated improvement compared to Maxwell, DirectX 12 feature level 12_1 and higher, 17 billion transistors, 4096-bit memory bus interface, NVLink and half precision FP16 compute, which is two times the rate performed by a full precision FP32.
    It would appear that the flagship GP100 GPU will be the successor to the GM200 GPU, which is under the bonnet of the GTX 980 Ti and the GTX Titan X.
    More official details are expected to arrive at the company's GPU Technology Conference in early April.
    Noticia:
    http://www.fudzilla.com/news/graphic...-testing-phase



    Cada vez mais proximo o revelar destes novos GPUs.
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  11. #56
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    6,589
    Avaliação
    41 (100%)
    Mesmo. A guerra dos tronos vai re-começar no mundo das gráficas. "Quem marcar primeiro ganha".
    Devido à falta de espaço na assinatura, resolvi colocar em "Acerca de mim" os meus projectos]
    http://www.portugal-tech.pt/member.php?u=801

  12. #57
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    Nvidia Slide reveals numbers on Single and Double precision for Flagship Pascal GPU


    A slide from an Nvidia presentation is stirring some things on the web as it seems as some info on the Flagship GPU on the new 14nm Pascal architecture (likely named GP100) would be able to perform numbers that are astonishing.
    Slides from "The Future of HPC and The Path to Exascale" shows a roadmap with a DP GFLOPS/W value for Pascal. The presentation's date is between the GTC 2013 roadmap which does not contain Pascal and the GTC 2014 roadmap which does contain Pascal. But thus this presentation is dated (2015) and likely estimated, we do have to say that.
    The slides from "Manuel Ujaldón CUDA Fellow @ Nvidia" however show performance numbers for both single and double precision, double-precision floating-point (DPFP) wise Nvidia seems to be reaching (or is aiming for) a 4 TFLOP/s throughput (at least on their HPC parts). That would be 3x over the current 1.3 TFLOP/s on the Tesla K20 which on it's end is based on "Kepler" GK110 silicon.


    Single-precision then, it would be as high as 12 TFLOP/s. That is four times that a GK110, and roughly double a GM200 (6.4 TFLOPS for the 980 Ti). The slide does reveal one other thing, the GP100 is inidicated to use stacked HBM2 memory as the memory bandwidth is set at 1 TB/s.
    Download the presentation here. Well .. yummie !?


    Noticia:
    http://www.guru3d.com/news-story/nvi...ascal-gpu.html


    Estes números são no minimo impressionantes face ás anteriores gerações da nVidia!!!!!










    NVIDIA Pascal GP100 GPU Expected To Feature 12 TFLOPs of Single Precision Compute, 4 TFLOPs of Double Precision Compute Performance

    New details on NVIDIA’s Pascal GPU have been dug up by 3DCenter (via Beyond3D) which showcase the total compute performance of the upcoming FinFET based chip. At CES 2016, NVIDIA announced their Pascal based Drive PX 2 module which is an automobile supercomputer which uses the graphics processing power of GPUs to drive cards autonomously. The presentation didn’t mention the flagship chip but we expect to hear a more detailed session on those at GTC in April 2016.

    NVIDIA Pascal GP100 Flagship GPU Might Come With 12 TFLOPs of Single Precision, 4 TFLOPs of Dual Precision Compute

    The one GPU that everyone has their eyes on right now, whether they be enterprise of mainstream audience, is the flagship Pascal GPU which is known as GP100 (The naming scheme for the chip is not confirmed yet). This is going to be the flagship chip of the lineup which will be featured on Tesla, Quadro and GeForce graphics cards. The chip is based on the 16nm FinFET process which leads to efficiency improvements and better performance per watt but with Pascal, double precision compute returns with a bang. Maxwell which is NVIDIA’s current gen architecture made some serious gains in the performance per watt department and Pascal is expected to keep the tradition move forward.
    The information today comes from slides which have long existed but most people haven’t had access to. The slides were found by iMacmatican, a Beyond3D forum member who has compiled a good list of details over at the forum. Since most of these slides date back to 2014-2015, there are bound to be some changes to the GPU design which we will also explain in a bit. First of all, a slide from a presentation in March 2014 detailed GFLOPs per watt for various NVIDIA GPUs. The approximate values for NVIDIA’s CUDA generation of GPUs have been listed below:

    • Tesla: 0.5
    • Fermi: 2
    • Kepler: 5.5
    • Pascal: 14
    • Volta: 22












    Slide Credits: Beyond3D Forum
    The slide clearly shows that Pascal is rated at GFLOPs per watt while Volta is rated at 22 GFLOPs per watt. Now the slide states that these approximations are for the Dual Precision or DGEMM (Dual Precision Floating General Matrix Multiply) GFLOPS/Watt and not single precision due to which Maxwell has been removed from the latter slides since it didn’t feature any FP64 hardware under the hood. The fastest Kepler based Tesla K40X comes with 6.1 GFLOPs/W and the dual-chip Tesla K80X comes with 6.2 GFLOPs/W. Pascal is expected to take this around 14 GFLOPs/W which is more than twice of Dual Precision GFLOPs/W.
    Coming to the Single Precision or SGEMM (Single precision floating General Matrix Multiply) GFLOPs/W are rated at 42 GFLOPs/W for Pascal. Maxwell is rated at 23 GFLOPs/W with the dual-chip offering pushing that up to 25 GFLOPs/W while Volta is rated at 73 GFLOPs/W. Now there’s also a slide that details the HGEMM (Half Precision floating General Matrix Multiply). We know that Pascal and the latter generation of GPUs will come with mixed precision compute which allows users to get twice the compute performance in FP16 work loads compared to FP32 by computing at 16-bit with twice the accuracy of FP32. Compared to Maxwell which has just 26 half precision GFLOPs/W, Pascal will take that up to 85 GFLOPs/W while Volta will do up to 145 GFLOPs /W.




    Coming to the more meaty part, 2014 slides are full of useful data on Pascal GPUs. Of course these slides pre date the time frame when Pascal GPUs actually taped out and entered NVIDIA Labs for testing which is what they stated themselves during SC15 and several months before that at GTC session in Japan. It is known that during some point, NVIDIA made the step to change their designs from HMC (Hybrid Memory Cube) based solutions to HBM2 based solutions and they presented the updated design to the audience at GTC 2015 in Japan.
    Advertisements

    The prototype Pascal board that was showcased back at GTC 2014 was actually based on an HMC implementation and that changed in 2015. From details mentioned in the slides, NVIDIA is claiming that they have integrated the memory (HBM2) to be part of the actual GPU die. Now this could mean one of two things, whether NVIDIA has actually managed to integrated HBM2 and a 16nm GPU on the same die or they could be using a similar design as the Fury cards from AMD which fuse the GPU and HBM chips on single interposer that makes them a single chip solution, sort of like an SOC.
    What we know so far about Nvidia’s flagship Pascal GP100 GPU :

    • Pascal graphics architecture.
    • 2x performance per watt estimated improvement over Maxwell.
    • To launch in 2016, purportedly the second half of the year.
    • DirectX 12 feature level 12_1 or higher.
    • Successor to the GM200 GPU found in the GTX Titan X and GTX 980 Ti.
    • Built on the 16nm FinFET manufacturing process from TSMC.
    • Allegedly has a total of 17 billion transistors, more than twice that of GM200.
    • Will feature four 4-Hi HBM2 stacks, for a total of 16GB of VRAM and 8-Hi stacks for up to 32GB for the professional compute SKUs.
    • Features a 4096-bit memory bus interface, same as AMD’s Fiji GPU power the Fury series.
    • Features NVLink (only compatible with next generation IBM PowerPC server processors)
    • Supports half precision FP16 compute at twice the rate of full precision FP32.

    GPU Architecture NVIDIA Fermi NVIDIA Kepler NVIDIA Maxwell NVIDIA Pascal
    GPU Process 40nm 28nm 28nm 16nm (TSMC FinFET)
    Flagship Chip GF110 GK210 GM200 GP100
    GPU Design SM (Streaming Multiprocessor) SMX (Streaming Multiprocessor) SMM (Streaming Multiprocessor Maxwell) TBA
    Maximum Transistors 3.00 Billion 7.08 Billion 8.00 Billion Up to 17 Billion
    Maximum Die Size 520mm2 561mm2 601mm2 TBA
    Stream Processors Per Compute Unit 32 SPs 192 SPs 128 SPs TBA
    Maximum CUDA Cores 512 CCs (16 CUs) 2880 CCs (15 CUs) 3072 CCs (24 CUs) TBA
    Compute Performance 1.6 TFLOPs 5.1 TFLOPs 6.1 TFLOPs 12 TFLOPs
    Maximum VRAM 1.5 GB GDDR5 6 GB GDDR5 12 GB GDDR5 32 GB HBM2
    Maximum Bandwidth 192 GB/s 336 GB/s 336 GB/s 1 TB/s
    Maximum TDP 244W 250W 250W 250W
    Average Performance Increase over Predecessor +45%
    (GTX 580 Versus GTX 285)
    +55%
    (GTX Titan Black Versus GTX 580)
    +30%
    (GTX Titan X Versus GTX Titan Black)
    TBA
    Flagship GPU Price (Consumer Only) $499 US
    (GTX 580)
    $999 US
    (GTX Titan Black)
    $999 US
    (GTX Titan X)
    TBA
    Launch Year 2010 (GTX 580) 2014 (GTX Titan Black) 2015 (GTX Titan X) 2016

    We have seen several slides but there’s one from an independent researcher who’s also a CUDA fellow who posted the compute performance for several platforms in his presentation. The slide puts the NVIDIA Pascal GPU with Stacked DRAM (1 TB/s) featuring up to 4 TFLOPs of Double Precision (FP64) and 12 TFLOPs of Single Precision (FP32) compute performance. In a slide from 2014, after launch of the second generation Pascal GPUs, an NVIDIA presentation also mentioned a GPU known as Pascal-Solo (not to be mistaken with Han-Solo) in the slide showcasing their Tesla GPU accelerator roadmap. The Pascal-Solo GPU features just 1 GPU and has a 235W TDP. The part comes in both PCI-e Active/Passive cooling options and is expected to launch in 2016. The Beyond3D Forum member approximated that the Tesla GPU could launch in Q2 of 2016.

    There’s no doubt that Pascal GPUs will feature a lot of compute performance aimed at the Tesla and Quadro markets. The next generation FinFET based graphics cards will have a lot of muscle to flex toward the complex tasks that are put forward in the HPC workloads. Expect GTC 2016 to bring a lot of new information on Pascal based Tesla solutions.







    Última edição de Jorge-Vieira : 17-02-16 às 20:41
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  13. #58
    Master Business & GPU Man Avatar de Enzo
    Registo
    Jan 2015
    Local
    País Campeão Euro 2016
    Posts
    6,589
    Avaliação
    41 (100%)
    Acho interessante a Samsung estar a trabalhar tão próxima da AMD e as memórias vermelhas serem Hynix, e a Nvidia ter posto uma acção judicial e ganhou contra a Samsung e têm as memórias deles...
    Devido à falta de espaço na assinatura, resolvi colocar em "Acerca de mim" os meus projectos]
    http://www.portugal-tech.pt/member.php?u=801

  14. #59
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    Dual Pascal-based self-driving car coming from Volvo next year



    DRIVE PX 2 test driven in Sweden in 2017
    Buried deep in Nvidia's financial Q4 2016 earnings call was a brief referecmce to the company's Automotive division, which made Nvidia a record $93 million in Q4, up 18 percent sequentially, and up 68 percent year-over-year. There is also a mention that Volvo will be the first manufacturer to take Nvidia's dual Pascal-based DRIVE PX 2 platform for a ride in Sweden.

    The money could be better, but Nvidia plans to make much more out of automotive in 2016 and beyond. We were assured that by 2020, self-driving cars based on Nvidia technology under the bonnet, will be commercially available. Many manufactures, including Tesla, BMW, and Audi, offer limited self-drive capabilities. Some of their cars can operate autonomously on highways, but not in cities.
    Nvidia said it is working with 70 car manufacturers that are developing self-driving car technologies. Nvidia claims its DRIVE PX 2 has generated enormous interest around the world from carmakers and Tier 1 automotive tech suppliers.
    Car design cycles take years, and the company expects many cars powered by Nvidia's automotive solutions will show up this year. Nvidia already told investors in August 2015 that its chips have ended up in 30 million cars. In other words, most of the company's Tegra-based automotive SoCs are selling in automotive infotainment systems.
    Car manufacturers including Volkswagen [No CO2 joke needed. Ed] use Nvidia chips for their for infotainment systems, and some manufacturers, including Audi and Tesla, are still using ancient Nvidia mobile GPUs in their dashboards. This is enough to make $93 million. There is no doubt that 2016 will yield better sales.


    King and CEO of Nvidia Jen Hsun Huang told investors and the world that:
    "Volvo, well-known for its safety and reliability, will be the first to develop DRIVE PX 2, using it as the brain for its fleet of 100 self-driving cars to be publicly available next year in its hometown of Gothenburg, Sweden. "
    DRIVE PX 2 contains two Pascal-based SoCs (codenamed Parker), are acting as a quasi-supercomputer. More specifically, this supercomputer platform is the size of a lunchbox and can manage to processes 24 trillion deep learning operations a second, delivering eight teraflops of processing power in the process. This is roughly equivalent to the power of 150 MacBook Pros.
    Nvidia's dual Pascal-based DRIVE PX 2 can use up to four SoCs and can be a passively cooled, or water-cooled. Car manufactures know all there is to know about water cooling, as they have been using H2O cooling in their engines for more than a century.
    Volvo plans to use DRIVE PX 2 for its self driving capabilities. Nvidia provides hardware and software that can fuse data from cameras, lidar (light ray shooting radar), radar, and ultrasonic sensors, to create a full 360-degree picture of what is happening around the vehicle. You obviously need an incredible amount of processing power to make sense of all that data.
    Additionally, the DRIVE PX 2 module localises the vehicle on an HD map and determines a safe path forward using deep learning techniques. This is why the German car manufacturers acquired HERE maps from Nokia, just to be independent from Google.
    Unfortunately, Volvo will only have a developer edition of the DRIVE PX 2 for its self-driving automotive trials in 2017 and it might take years before it becomes commercially available. The 2020 timeframe probably means that some expensive cars including Tesla or expensive BMW, Audi, Mercedes and Cadillacs will get self-driving treatment, but at a high price. Most entry-level to mid-range cars sell for between € / $20,000 and € / $30,000 and it will probably take a bit longer than 2020 to get self-driving vehicles into this price range.
    Jensen reminded investors a few weeks ago that the WEpod, the first autonomous shuttle, incorporates Nvidia's deep learning platform and took its inaugural trip on public roads in the Netherlands, where it can be summoned with a smartphone app.
    There is no doubt that future of automotive is self-driving, but we are still a few years away at this point.
    Noticia:
    http://www.fudzilla.com/news/graphic...omes-next-year
    http://www.portugal-tech.pt/image.php?type=sigpic&userid=566&dateline=1384876765

  15. #60
    Tech Ubër-Dominus Avatar de Jorge-Vieira
    Registo
    Nov 2013
    Local
    City 17
    Posts
    30,005
    Avaliação
    1 (100%)
    NVIDIA Pascal GPUs Spotted in India’s Trade Database



    NVIDIA’s upcoming architecture codenamed ‘Pascal’ is based on the TSMC 16nm manufacturing process and utilizes the latest iteration of high bandwidth memory. HBM2 will feature bandwidth exceeding 1 Terabyte per second and ship in larger forms compared to HBM1. Some of you might remember, HBM1 was a major talking point on AMD’s Fiji XT line-up but the memory has a limitation of 4GB. Thankfully, HBM2 resolves this and could potentially allow for GPUs sporting a whopping 32GB. While this isn’t confirmed, rumours suggest that NVIDIA could launch a flagship with a huge amount of video memory. However, I personally think a 32GB graphics card will target professionals requiring huge compete power.
    It’s important to remember that HBM is up to 9 times faster than the current GDDR5 standard and a huge revolution in memory technology. Pascal will also be the first architecture to utilize NV-Link. This is an energy-efficient, high-bandwidth communications channel that uses up to three times less energy to move data on the node at speeds 5-12 times conventional PCIe Gen3 x16. In theory, this enables faster communication between the CPU and GPU as well as multiple graphics cards. This is only a very brief insight into the potential of Pascal and it could be the biggest step forward in graphics technology we’ve seen in a long time.
    Recent information from Zauba.com shows a shipment of Pascal graphics cards which suggests the launch date isn’t too far off. However, according to the Tech Times, Pascal might not launch until June at the earliest! The general consensus is the launch will occur sometime in Q2 this year. With AMD and NVIDIA preparing new architectures, it’s a very exciting time and I cannot wait to see the kind of performance gains compared to the previous generation. Hopefully, the new range offers superb performance at reasonable prices, and AMD keeps NVIDIA in check to drive costs down.

    Noticia:
    http://www.eteknix.com/nvidia-pascal...rade-database/


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

 

 
Página 4 de 91 PrimeiroPrimeiro ... 234561454 ... ÚltimoÚltimo

Informação da Thread

Users Browsing this Thread

Estão neste momento 1 users a ver esta thread. (0 membros e 1 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
  •