【2015.07.26】外媒新闻推荐 移动芯片合并 新一轮显卡竞争
本帖最后由 包子侯爵 于 2015-7-26 10:43 编辑Title：Rockchip And Allwinner Rumored To Merge: Dominant Force In The Making?
Rockchip（瑞芯微）和Allwinner（全志）可能合并？
From：http://wccftech.com/rockchip-allwinner-merger-rumor/
Title：NVIDIA Pascal GPU To Feature 17 Billion Transistors and 32 GB HBM2 VRAM – Full CUDA Compute Architecture Arrives in 2016
绿队170亿晶体管HBM2代的帕斯卡2016年降临——难道和Fiji x2 PK？
From：http://wccftech.com/nvidia-pasca ... am-arrives-in-2016/
Title：Rockchip And Allwinner Rumored To Merge: Dominant Force In The Making
From：http://wccftech.com/rockchip-allwinner-merger-rumor/
简报：两家国产芯片厂可能合并？瑞芯微和全志如果联手可能目标是推出5刀的平板用SOC芯片。来与高通竞争。
个人想法：在Intel瞄准高通的时候，AMD能否也找一家合作呢。
Now this might be the biggest merger that can actually end up happening. Previously, analysts kept on suggesting that Intel should take over the chip side of Qualcomm in order to benefit both companies. While the authorities will not stand for such a formidable entity to be created, let us turn our attention towards Allwinner and Rockchip, which happen to be two Chinese tech firms aiming to snare the lower chipset market all for themselves.
Recommended: Is Intel Merging With Qualcomm After Altera Acquisition?
A Coalition Formed Between Allwinner And Rockchip Could Affect The Latter’s Partnership With Intel
The government of China is hell bent on expanding the semi-conductor industry. One Chinese firm came close to acquiring Micron, which is the second largest memory maker in the world, trailing behind Samsung. The deal would have made China a force to be reckoned in the memory manufacturing industry as well. According to the latest report, the Chinese government had announced to invest approximately trillion of yuan ($161.1 billion) on development of the country’s semiconductor industry.
The investment has been calculated to last between 5-10 years and will include the following:
[*]Various research
[*]Development
[*]Education projects
[*]Construction semiconductor fabrication plants
[*]Acquisitions of companies
[*]Consolidation of existing semiconductor assets
Allwinner was one of the companies that received the investment from the government, although the firm is yet to announce its future plans. There is also a rumore that Allwinner will be merging with Rockchip in order to create a state-owned entity.
Recommended: Here Is The $199 Wearable That Will Compete With Google Glass
Allwinner has partnered with Qualcomm to target the Chinese market with affordable mobile devices that also feature an LTE modem. The biggest goal that the company has been able to achieve so far is creating a $5 SoC by the name of A64. The chipset will strictly be incorporated in to tablets, and are expected to be released during second half of 2015.
As far as Allwinner and Rockchip are concerned, the possible merger might have complications between the partnership of Rockchip and Intel. While both firms have not commented on this story, we will have to take it with a grain of salt and wait for more updates to pour in for the news to be confirmed. Do you guys feel that a possible merger between the two companies will spell disaster for companies like Qualcomm? Let us know your thoughts.
Title：NVIDIA Pascal GPU To Feature 17 Billion Transistors and 32 GB HBM2 VRAM – Full CUDA Compute Architecture Arrives in 2016
From：http://wccftech.com/nvidia-pasca ... am-arrives-in-2016/
简报：绿队明年将会推出pascal（帕斯卡）系列产品，将会在HBM2和16nm FinFET基础上进行。可能配有几乎两倍Fiji的170亿个晶体管。拥有16GB或32GB的HBM2代显存。瞄准4K和8K市场（当然他们从海力士那边拿到HBM2芯片要比AMD慢咯，小A快加油），带宽为1TB/s(2x Fiji ,3x GTX 980Ti)。
同时推出的还有NVLINK最为PCI-E连接，因为以后专业或者发游戏用8卡而不是4卡……OMG。同时也会提高性能和数据准确性、精度。
点评：未来的世界好口怕……还有至少半年，小A有什么秘密计划呢。
NVIDIA will be introducing their next generation Pascal GPU in 2016 which will introduce several new and key technologies to the green team. The Pascal GPU will be the successor to the current generation Maxwell GPU and from the looks of it, it is going to be a beast of a chip. Featuring the latest HBM2 and 16nm FinFET based designs, Pascal GPUs will leverage NVIDIA’s dominance in both the consumer and corporate world.
NVIDIA Pascal GPU Might Feature 17 Billion Transistors, Almost Twice The Transistors of Fiji
In an exclusive report published by Fudzilla, the site reveals that NVIDIA’s next generation Pascal GPU will feature 17 billion transistors crammed inside its core. Currently, the flagship GM200 core found on the GeForce GTX Titan X comes with 8.0 Billion transistors while the competitor, the Radeon R9 Fury X has a total of 8.9 Billion transistors inside its Fiji GPU. The 17 Billion transistors on the Pascal GPU are twice the transistors found on the GM200 Maxwell and the Fiji XT GPU core which is literally insane. Pascal is meant to be NVIDIA’s next high performance, compute focused graphics architecture which will be found on all market segments that will include GeForce, Quadro and even Tesla. Based on TSMC’s 16nm process node, NVIDIA’s Pascal GPU will not only feature the best performance in graphics but also the most power efficient architecture ever made by a GPU manufacturer.
It was revealed a few days ago that NVIDIA’s Pascal GP100 chip has already been taped out on TSMC’s 16nm FinFET process, last month. This means that we can see a launch of these chips as early as Q2 2016. Given that the transistor count is correct, we can expect a incremental performance increase from Pascal across the range of graphics cards that will be introduced.
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.
The 17 Billion transistors are an insane amount but what’s more insane is the amount of VRAM that is going to be featured on the new cards. With HBM2, NVIDIA gets the leverage to feature far more memory than what’s currently allocated on HBM1 cards (4 GB HBM on Fury X, Fury, Nano, Fury X2). With HBM2, NVIDIA gets access to more denser chips that will result in cards with 16 GB and up to 32 GB of HBM memory across a massive 4096bit memory interface which will dominate the next high-resolution 4K and 8K gaming panels.Although they may have to wait a little bit longer thanks to AMD’s priority access to HBM2 with SK Hynix, the makers of HBM. With 8Gb per DRAM die and 2 Gbps speed per pin, we get approximately 256 GB/s bandwidth per HBM2 stack. With four stacks in total, we will get 1 TB/s bandwidth on NVIDIA’s GP100 flagship Pascal which is twice compared to the 512 GB/s on AMD’s Fiji cards and three times that of the 980 Ti’s 334GB/s.
The Pascal GPU would also introduce NVLINK which is the next generation Unified Virtual Memory link with Gen 2.0 Cache coherency features and 5 – 12 times the bandwidth of a regular PCIe connection. This will solve many of the bandwidth issues that high performance GPUs currently face. One of the latest things we learned about NVLINK is that it will allow several GPUs to be connected in parallel, whether in SLI for gaming or for professional usage. Jen-Hsun specifically mentioned that instead of 4 cards, users will be able to use 8 GPUs in their PCs for gaming and professional purposes.
With Pascal GPU, NVIDIA will return to the HPC market with new Tesla products. Maxwell, although great in all regards was deprived of necessary FP64 hardware and focused only on FP32 performance. This meant that the chip was going to stay away from HPC markets while NVIDIA offered their year old Kepler based cards as the only Tesla based options. Pascal will not only improve FP64 performance but also feature mixed precision that allows NVIDIA cards to compute at 16-bit at double the accuracy of FP32. This means that the cards will enable three tiers of compute at FP16, FP32 and FP64. NVIDIA’s far future Volta GPU will further leverage the compute architecture as it is already planned to be part of the SUMMIT and Sierra super computers that feature over 150 PetaFlops of compute performance and launch in 2017 which indicates the launch of Volta just a year after Pascal for the HPC market.
NVIDIA Pascal GPU Module:
页:
[1]