顯示具有 半導體產業分析 標籤的文章。 顯示所有文章
顯示具有 半導體產業分析 標籤的文章。 顯示所有文章

2016年10月25日 星期二

台積電與 Intel 競爭白熱化開始,也形成ARM 與 Intel 競爭白熱化 (From TSMC Competition With INTEL In Technology And Business To ARM Competition With INTEL)

Intel危險?蘋果A10太強、對手無法比、媲美電腦CPU

蘋果iPhone 7採用台積電(2330)代工的「A10 Fusion」處理器,備受尊崇的晶片研究機構Linley Group分析這款晶片,直呼A10太強大,把對手打到落花流水,表現甚至優於部分電腦CPU。

巴倫(Barronˋs)21日報導,Linley Group主管Linley Gwennap報告指出,蘋果砸錢研發客製化CPU成果豐碩,iPhone 7性能優於其他旗艦機,甚至超越部分低階電腦。iPhone 7搭載A10 Fusion處理器,為蘋果首款採取big.LITTLE大小核的四核心晶片,內含兩顆高效能核心「Hurricane」,表現比前代高出35%;另兩顆核心「Zephyr」則更為省電。Hurricane和Zephyr都是蘋果的客製化核心。

Gwennap指出,Hurricane徹底擊垮對手。他引用Geekbench評比,A10處理器的單核跑分,遠勝三星電子Exynos 8890、高通驍龍820、華為麒麟955。儘管三星和華為在多核跑分表現出色,Gwennap說,多餘核心對應用程式沒有幫助,一般只需一或兩顆核心就能運作。

A10表現驚人是因為蘋果本錢雄厚,晶片比別家更大。處理器的晶粒(die)區域要價高昂,Hurricane卻大量使用,Hurricane面積為4.18平方公厘,尺寸約為其他廠商高階處理器的兩倍。蘋果不賣晶片賣手機,在晶粒多花個幾美元不打緊,只要能提高整體表現,讓產品更熱賣即可。
Gwennap稱,蘋果新CPU表現甚至優於英特爾(Intel)的x86核心,A10表現和英特爾Skylake的Core核心幾乎不相上下,差別在於英特爾Core核心用於電腦,不像A10一樣,有電池續航力的限制。這對英特爾是一大警訊,他說,蘋果CPU追上英特爾,實際上,Hurricane能輕鬆支援MacBook Air筆電等。當前蘋果筆電使用的英特爾晶片,速度比A10更慢。

目前蘋果iPhone、iPad使用自家處理器,但是筆電MacBook和個人電腦Mac仍然搭載英特爾CPU,外界老早盛傳,未來英特爾晶片可能會從蘋果產品全面淘汰,遭蘋果晶片取而代之

知名跑分軟體安兔兔(AnTuTu)1日發表9月前十大手機性能榜單,在9月1-30日期間蒐集超過2,000條單一機型數據後發現,iPhone 7 Plus的平均跑分成績高達172,644分、奪下冠軍,而iPhone 7的跑分也有170,124分,分數居次,把Android旗艦機遠遠拋在腦後。
蘋果(Apple)次代iPad Pro系列傳出將在明年春天出貨、且將搭載由台積電(2330)供應的A10X晶片,而最新傳出A10X晶片性能強大,單核跑分比前代A9X高出4成、且也比iPhone 7使用的A10晶片高出2成

日本網站taisy0、iPhone Mania 5日轉述荷蘭媒體TechTastic的報導指出,根據從可靠的消息人士取得的資料顯示,蘋果預計搭載在次代iPad Pro系列的「A10X」晶片GeekBench單核跑分達4,236、多核跑分達6,588。

The iPhone 7’s A10 Fusion Chip Is Faster Than The MacBook Air 

The majority of the iPhone 7’s improvements are under the hood, thanks to its new A10 Fusion chipset which makes it a quad-core phone versus the iPhone 6s which was still a dual-core setup. As we have seen on the AnTuTu benchmarks, this has unsurprisingly resulted in very high scores which has blown past the competition.

However it seems that’s not all. As noted by John Gruber (via MacRumors), the iPhone 7’s performance on Geekbench has revealed some very impressive single and multi-core scores which not only has outdone competing phones, but even Apple’s own products like the MacBook Air laptops.

Now there’s this perception that maybe because it is smaller and because it can’t do certain things that smartphones aren’t as powerful as laptops, and to a certain degree that is true for some models. However according to Gruber, he notes that the iPhone 7 has pretty much outdone every single MacBook Air Apple has ever released in both single and multi-core performance, save for the 2015 MacBook Air with an Intel Core i7 which only narrowly edges it out.

A follow-up tweet by Matt Mariska also revealed that the iPhone 7 managed to beat the $6,500 model of the 2013 MacBook Pro in single-thread performance. While obviously the iPhone won’t be replacing laptops anytime soon, it does go to show how far along Apple has come in the development of its A-series of ARM chipsets. Who knows, at this rate those rumors of an ARM-based Mac could come true, and it could be very well justified.

( Note: Mac Air CPU is 1.6GHz dual-core Intel Core i5 [Turbo Boost up to 2.7GHz] with 3MB shared L3 cache with 4GB onboard memory, but iPhone 7 is only 4 core ARM cotex A10 core with much less power consumption than Intel Core i5 )

A closer look at the ARM Cortex-A72

The Cortex-A72 was announced back in February, promising another boost to performance and substantial energy savings to boot. At ARM’s TechDay 2015 in London this week, we were fortunate enough to be given some deeper insight into the inner workings of ARM’s latest application processor.

Although the base-line architecture is very similar to the Cortex-A57, the A72 is much more than typical revision. A team of some 65 to 70 engineers have gone back through the design, optimizing almost every logical block for power efficiency, helping the processor to sustain maximum frequencies during heavy workloads, and focused on squeezing the design into a smaller area, to keep costs down.

Architecturally, the Cortex-A72 features a new branch-predictor, increases the effective decode and dispatch bandwidths, and has had changes made to the execution units, to name just a few alterations. ARMs new branch predictor reduces misprediction with a new algorithm and can suppress superfluous branch predictor accesses, which helps to reduce wasted energy. The rebuild offers up to 20 percent improvements to prediction over the A57.

The design still features a 3-wide decode, but the dispatch unit has gone from 3- to 5-wide, to more effectively break operations down into further micro-ops which help keep the 8-wide issue machine well fed. The execution stage sees the introduction of next-gen floating-point SIMD units with a
variety of latency reductions, multiple zero-cycle forwarding datapaths to reduce wasted cycles, and substantial bandwidth increases in the two integer units. The load and store units have a more sophisticated combined L1/L2 data prefetcher, offering a bandwidth improvement of 30 percent. All of which, among other changes, is designed to help reduce power consumption and to improve performance in certain areas over the A57.

In terms of what this means for silicon designers and end users, the Cortex-A72 is still a high-end processor, but it will utilize energy more efficiently.  In other words, the CPU will be able to do more within the limited power budgets available on mobile and should result in cooler devices as well. Even at 28nm, the Cortex-A72 boasts up to a 50 percent energy reduction when compared with the Cortex-A15 and a 20 percent saving compared with the A57, at the same clock speeds. Milliwatts per core have dropped from the A57, to around 700mW at 2.5GHz. The design takes up 10 percent less area than the A57, which will also help save on costs.

ARM is also increasingly focused on its POP IP, you’ll see quite a few references to TMSC’s 16nm FinFET Plus manufacturing node in the examples. As well as substantial energy savings, ARM reckons that the A72 will be able to sustain 2.5GHz clocks on the new 16nm process, whilst keeping within the limited smartphone power budget. It’s the additional power efficiency and resulting lower heat profile that will really help the A72 achieve higher clock speeds than a 16nm A57.

趨勢分析

2016年9月6日 星期二

奈米碳管正式應用於半導體產業:奈米碳管存儲器(NRAM)技術將2018量產 ( CNT nano tech. will used in semiconductor : CNT memory NRAM will go into production in 2018 )

富士通與Nantero達成協議 2018年推出快1000倍的NRAM內存

富士通半導體和三重富士通半導體上週共同宣布,他們已與總部位於美國的Nantero公司達成協議,授權該公司的奈米碳管存儲器(NRAM)技術,三方公司未來將致力於NRAM存儲器的開發與生產。

據了解,藉由NRAM技術所生產的存儲器速度將是當前普通存儲器的1,000倍。預計,藉由三方面的合作,將在2018年推出藉由55奈米製程所生產的NRAM存儲器。

在納米技術研究領域,奈米碳管(CNT)是一種很獨特的材料,直徑只有人類頭髮的五萬分之一,能導熱導電,硬度是鋼鐵的50倍。在儲存領域,奈米碳管通過矽基沉底也能達到0與1變化。

因此,也可以用來當做儲存芯片使用,而且因為其特性為非揮發性,所以就算斷電也不會清除儲存在上面的信息。

而由於奈米碳管有著特別的特性,因此相較於當前的普通存儲器,NRAM芯片的具有非常強大的優勢。除了讀寫速度是普通閃存的1,000倍(Nanteo官網上表示是1,000倍)之外,同時可提供功耗更低,更具可靠性與耐用性的存儲器,而且生產成本更低。

富士通半導體系統存儲器副總裁松宮正人表示,藉由Nantero公司的奈米碳管技術而生產的非揮發性儲存內存是業界的一個超越了傳統技術的顯著進步。而富士通半導體自90年代以來,一直是在設計和生產非揮發性存儲器FRAM(鐵電隨機存取內存)的少數廠商之一。因此,富士通半導體具有整合FRAM設計和生產的能力。如今再與Nantero公司達成協議之後,未來將就由過去設計與生產FRAM的能力,用於開發NRAM存儲器,以滿足市場的需求。

Nantero董事長兼執行長Greg Schmergel指出,Nantero已經在奈米碳管的內存儲存技術上研究了十多年,具備生產NRAM存儲器的先進技術。而本次與富士通半導體與三重富士通半導體兩家公司的合作,由於兩家公司都是全球生產儲存設備中最成功的企業之一。藉由其在研發與生產FRAM的經驗,未來可以在NRAM存儲器的合作上產生更大的效益,建立更緊密的關係。

不過,對於此合作案,有業界人士指出,Nantero公司自從2006年就說要開始生產奈米碳管存儲器了。但是,到現在一直沒什麼特別的進展。未來,即便是跟富士通半導體達成了合做協議,量產的NRAM芯片容量還是不夠大,只適合一些嵌入式設備使用。

所以,未來是不是能夠改變整個存儲器的生態,還有待進一步的觀察。

Fujitsu Semiconductor and Mie Fujitsu Semiconductor License Nantero's NRAM And Have Begun Developing Breakthrough Memory Products for Multiple Markets

  Fujitsu Semiconductor Limited and Mie Fujitsu Semiconductor Limited today announced that they have reached an agreement with US-based Nantero, Inc. to license that company's technology for NRAM, non-volatile RAM using carbon nanotubes, and to conduct joint development towards releasing a product based on 55-nm process technology. Three companies are aiming to develop a product using NRAM non-volatile RAM that achieves several 1000 times faster rewrites and many thousands of times more rewrite cycles than embedded flash memory, making it potentially capable of replacing DRAM with non-volatile memory. Fujitsu Semiconductor plans to develop an NRAM-embedded custom LSI product by the end of 2018, with the goal of expanding the product line-up into stand-alone NRAM product after that. Mie Fujitsu Semiconductor, which is a pure-play foundry, plans to offer NRAM-based technology to its foundry customers.
Comment from Masato Matsumiya, System Memory VP, Fujitsu Semiconductor

Non-volatile memory using Nantero's carbon-nanotube technology is a marked advance beyond conventional technology. Fujitsu Semiconductor has been designing and producing FRAM, a type of non-volatile RAM, since the late 90s, and is one of the few companies to have integrated FRAM design and production capabilities. We will be able to build on our experience and skill in this field to develop and produce NRAM as well. The combination of Nantero's technology with our design and production capabilities promises to meet the longstanding needs of our customers for non-volatile memory that is higher density, faster, more energy efficiency, and with a higher rewrite cycle. I am confident that Nantero and Fujitsu Semiconductor will become partners with the highest mutual regard, and I look forward to working with Nantero to develop memory with revolutionary performance. As always, Fujitsu Semiconductor is committed to offering a more complete range of memory solutions, and to meeting the needs of its customers.
Comment from Masahiro Chijiiwa, Director and Corporate SVP, Mie Fujitsu Semiconductor

Nantero's NRAM technology is based on carbon nanotubes and allows for non-volatile memory with high density and random access, promising to expand Mie Fujitsu Semiconductor's line of embedded non-volatile memory products, which are a major focus of our business. In working with Nantero to develop their technology into products and license that technology, together with Fujitsu Semiconductor, I am certain we will be able to offer our customers new kinds of non-volatile random-access memory solutions.
Comment from Greg Schmergel, Co-Founder, Chairman & CEO, Nantero, Inc.

Fujitsu Semiconductor and Mie Fujitsu Semiconductor are ideal commercialization partners for Nantero as their experience with FRAM makes them among the world's most successful companies in mass production of new memory devices. Nantero is honored to be working with them on breakthrough memory products, and after years of intensive evaluation of our NRAM technology, our three companies have built a very strong relationship. Through over a decade of intensive R&D, Nantero has been able to bring CNT-based memory technology to a state of maturity and manufacturing readiness where world-class companies such as Fujitsu Semiconductor and Mie Fujitsu Semiconductor are working with us on productization for multiple markets. NRAM technology, with its combination of nonvolatility, high speed and high density, is uniquely positioned to allow for the continued evolution of memory beyond the projected limits and capabilities of classical technologies.

2014年9月3日 星期三

中國扶植「陸版台積電」,想一舉擊殺台灣半導體產業:台灣兩岸貨貿、TPP、RECP 極可能還不如與東協、印度加速策略性整合 ( Taiwan join TPP, RECP will not be better than to sign FTA with south Asia countries, Why? )

狠砸千億 中國扶植「陸版台積電」

中國大陸處處與台灣競爭,顯現馬總統兩岸經濟戰略是很失敗
6月底,中國國家發改委、科技部、財政部、工業和信息化部,難得同步發布新聞,宣布《國家集成電路產業發展推進綱要》已由國務院正式批准實施。《推進綱要》內指出,國家將成立1200億人民幣(約5900億台幣)的投資基金,扶持中國半導體產業。

下一個剿滅對象:聯發科

8月中,中國更對蘋果iPhone供應商豪威(OmniVision)發出公開收購要約,直接瞄準台灣半導體的上游客戶群,壯大自己下游晶圓代工與封測業的戰力。

業界甚至揣測,中國大陸發改委最近認定,全球IC設計龍頭高通遭反壟斷之後,「下一個被點名要剿滅的對象,就是聯發科,」一位台灣業界專家私底下透露。蕞爾小島台灣,30年之內,躍升為全球第二大半導體強國,僅次美國。更培育出全球晶圓代工冠軍台積電、全球第二大IC設計企業聯發科技。為什麼中國大陸一個剛出爐的產業政策,就讓台灣逾兩兆台幣產值的半導體業,如臨大敵?

有形之手 結合無形之手

首先,這是力道最強的產業政策。「這是一個高風險、高投入、對國家有戰略意義的行業,但我們很長時間每年的投入,只夠修2公里的地鐵,」中國《新世紀》週刊報導。而《推進綱要》將成立的1200億人民幣投資基金,足足可以蓋4座28奈米製程晶圓廠。金額創下史上新高。

其次,和過去的鼓勵政策不同,中國政府這次自己擔綱主角。

「過去是鼓勵政策,政府是旁觀者,企業是主角。你要不要接受這鼓勵政策,是企業做決定。《推進綱要》是政府要帶頭做這件事,責任主體是政府,」參與《推進綱要》起草工作的北京清華大學微電子所所長魏少軍告訴《天下》。

「這一千兩百億人民幣是『投資基金』,不是『發展基金』,」魏少軍強調。他解釋,過去鼓勵政策不涉及股權,但現在以投資基金運作,就會擁有與投資金額相對的企業股權,並要求投資報酬率,用股權結構進一步改變企業結構、產業結構。

終結「空芯現象」

中國雖然早在10年前就躍居全球IC消費最大國,但中國半導體發展仍處處受制於人。根據市調機構IC Insights預估,2017年中國半導體市場將有7成至8成,仍仰賴進口。讓中國政府頭疼的是,前年,中國進口了2322億美元的IC。這金額,比中國購買石油的外匯還要高。

「對於具備戰略和軍事意義的高端集成電路,花再大的價錢也買不來,」《新世紀》週刊報導,這是中國的「空芯現象」,只能被國際巨頭牽著鼻子走。

台灣的隱形優勢

面對中國崛起,失去了員工分紅配股、號召年輕人魅力的台灣半導體業,無法再單憑創業精神、技術領先求勝。「產業起不起得來,跟政策比較無關,還是要看企業自己強不強,」曾任台積電行銷副總、到深圳創業的胡正大,感觸很深。他眼見中國政策大力扶植太陽能、LED產業,卻造就全球產能過剩的泡沫。

曾經支撐起台灣半導體一片天的大學教育、工程師人才、企業的誠信文化、國家的智慧財產權制度,這些隱形優勢,不僅是美國目前遙遙領先的箇中原因,更是台灣和中國一爭高下的關鍵…(完整報導,請見《天下雜誌》554期)

東協10+6自貿區協議啟動談判 : 台灣可選擇一國一國簽FTA

東南亞國家協會10個會員國和澳洲、紐西蘭、印度、中國、日本、韓國等16國的資深談判代表,日前已在汶萊就「區域全面經濟伙伴關係(RECP)的組建,達成部分協議。這項又名為「東協加六」(10+6)自由貿易區協議的談判,經醞釀多時終告啟動,並確定若干關鍵會談的時程。是世界貿易組織(WTO)以外最大自由貿易區籌建工作的開始。從此另一大型區域經濟整合進入實際作業階段,使新的自貿區協議呼之欲出;另也證實與談各方能以經貿發展掛帥的立場,朝存異求同方向邁進,為亞太開啟一片繁榮新天地,營造出無窮希望。

東協(10+6)自貿區貿易協定談判啟動,意味著世界最大的區域經濟整合,已由多年的意見具申,進入了實際行動。在由坐而談判起而行之前,各方對於這個大規模區域自貿區的組建,能不能順利推展,是存有不少疑慮的。因為不僅亞太16國內部的多種差異,外部的不同性質拉扯阻窒,也有若干令人側目的影響,甚可能連起步都會出現困難。但是汶萊16國資深代表之會,不僅決定在明(2014)年8月前敲定自由貿易準則,還達成了從9月24日在澳洲舉行的下一輪談判期間開始,全面展開調降或取消關稅的磋商。前述決定內容具體、層次分明,呈現出談判人員有備而來的意涵。也展現各國經濟優先的務實態度,使外界沉重的心情緩和了不少。

東協10+6自貿區一旦建成,其經濟規模近20兆美元,會員國占全球國內生產毛額(GDP)1/3,人口逾30億約為全球半數。遼闊幅員正是經濟發展大開大闔的一片天地,尤其是「全球化」的思維與操作備受質疑,西方國家全力推動的世界貿易組織功敗垂成後,區域經濟組織觀念抬頭,透過自由貿易協定等機制所發揮的功能,而成為取代不易實現的「全球治理」主張,正受各方所歡迎。而東協近年所參與的自貿區運作,又有卓著成效可資覆按,更使外界對以東協為主軸的區域經濟整合存有信心,東協10+6自貿區發展基礎無疑是穩固的。

台積衝10奈米 年底試產

台積電取得跨足10奈米以下先進製程最關鍵的極紫外光(EUV)機台,擺脫半導體摩爾定律可能受到的物理極限限制,成本還有再下降空間,未來將持續用來生產高通、蘋果等大客戶高階晶片。

EUV機台要價9,000萬歐元(約新台幣36億元),被譽為是「半導體設備中的鑽石」,由於造價昂貴,已豎起進入門檻。據了解,除了台積電之外,三星、英特爾、IBM也是向艾司摩爾下單EUV機台的「大戶」,這些公司將同時取得進入10奈米以下世代製程的門票,半導體業「愈高階製程愈走向寡占」的態勢更加確立。

主導EUV設備開發的艾司摩爾(ASML)亞太區技術行銷協理鄭國偉昨(3)日在台北國際半導體展(SEMICON Taiwan)開幕首日,透露相關訊息。他表示,艾司摩爾已接獲多家半導體大廠共11台EUV訂單,首批六台已交貨,台積電在出貨名單之列。預料今年下半年到明年第1季,另五台設備也將完成交貨。

艾司摩爾這項宣布,透露先前遭遇技術瓶頸的EUV設備,克服關鍵輸出率不佳的最大瓶頸,有助半導體廠加速導入10 奈米量產腳步。設備商透露,艾司摩爾現有11台EUV訂單當中,有兩台是台積電下單,其中一台已在台積電竹科12廠安裝完畢,正進行10奈米製程研發,另一台可能在明年首季於台積電中科廠裝設。台積電昨未對此置評。

業界評估,台積電有了EUV機台之後,先進製程推進腳步將更如虎添翼,甩開原本業界憂心礙於物理限制、恐無法順利導入10奈米以下製程的問題,若良率提升狀況順利,有助台積電成本持續下降。

鄭國偉說,艾司摩爾有信心,協助客戶在今年底前試產10奈米製程,並於2016年導入量產,後續也將著手為更先進的7奈米製程做準備。業界預期,台積電10奈米製程量產之後,初期仍將承接高通、蘋果等需要更省電、更高階製程需求的行動通訊晶片大客戶訂單。台積電昨天漲2元、收128元;美國存託憑證(ADR)也在先進製程布局報喜激勵下,早盤漲逾1%。

艾司摩爾在開發EUV機台初期,考量相關投資龐大,對台積電、英特爾、三星提出投資邀約,希望集結各大廠的資源共同開發,三家業者隨後都同意入股。台積電在2012年10月投資8.38億歐元(約新台幣335.2億元),取得艾司摩爾5%股權。

極紫外光

極紫外光(EUV)技術是次世代微影技術之一,被半導體視為跨入10奈米製程以下,取代傳統浸潤式曝光(Immersion )、進入18吋晶圓的一項重要微影技術。浸潤式曝光機台是在光源與晶圓中間加入水的原理,使波長縮短到132奈米的微影技術,EUV曝光設備是利用波長極短的紫外線,在矽基板上刻出更微細的電路圖案。EUV售價昂貴,一台售價高達9,000萬歐元(約新台幣36億元),主力設備開發廠艾司摩爾(ASML)為降低開發風險,要求英特爾、三星及台積電以入股方式,共同參與這項先進微影技術開發。

註:

2014年5月9日 星期五

2014 ~ 2015 Intel 及 ARM, TSMC 的決戰,處理器及平板市場的新局面,將影響台灣未來5年的經濟 ( Intel, ARM And TSMC Competition Will Go Into Next Race )

當Intel帝國對上ARM軍團!決戰低耗能處理器市場

過去幾年,Intel製造的微型處理器占據眾數據中心和機器,但最近幾個月局勢出現變化, Intel 除了在智慧型手機晶片市占一直落後 ARM,現在矽谷大批晶片製造商更開始推出以ARM架構為基礎的處理器挑戰 Intel,過往的晶片帝國正面臨來勢洶洶的ARM軍團,預示著一場全新的晶片大戰就要展開。

ARM占下先機

ARM的嵌入式處理器有低成本、高效能、低耗電的特性,是發展最快速的業務,光在去年就暴增25%,在第三季其他產品的業務成長幅度更超過手機,ARM執行長伊斯特(Warren East)認為,這方面的收入未來會持續上升。
ARM server system start to migrate low cost server market

再者,隨著雲端運算發展,Facebook之類的社群網站對低功耗伺服器的需求大增,而ARM晶片正訴求可以大幅降低伺服器功耗,成功吸引了各家大型公司企業。

市調機構IDC指出,企業開始將注意力放在能夠有效率提升裝置智能程度的晶片,刺激嵌入式處理器的營收從2012年到2016年將成長23%,而市場對ARM技術的需求將不斷升高,Intel很難取得領先。IDC分析師說:「Intel想要攻進低功耗處理器市場,但ARM早已卡位。」

不過,Intel並不打算這麼坐以待斃,準備以原為行動裝置打造的低耗電Atom晶片為基礎,製造超低電壓處理器應戰。

Intel VS. ARM

不過到底哪家的晶片擁有較好的低功耗表現?根據美國威斯康辛大學研究員發表的研究報告,在節能方面兩家的晶片架構並沒有太大的優劣之分。研究員除了研究了電晶體數目、電壓和時脈速度等要素外,也仔細觀察了晶片的指令集(決定處理器基本處理的指令)。

研究員山卡拉林罕(Karu Sankaralingam)表示,ARM 和Intel晶片最根本的不同就在於其指令集,經過測試後發現兩家公司的指令集在效率方面的表現旗鼓相當

研究團隊測試ARM微處理器Cortex-A8與Cortex A9的晶片,Intel Atom晶片和Intel Core i7晶片,當研究員讓ARM和Intel Atom晶片執行伺服器程度的工作負載時,兩廠牌都展現類似的耗電量,而i7晶片本來就是為了桌上及筆記型電腦設計,自然耗電量較大但相對效能也高

現在有三星、高通、德儀等,越來越多的晶片與裝置廠商開始向ARM靠攏,不過也不能忽略,Intel在伺服器市場耕耘多年,其策略、夥伴、解決方案等方面都有強大基礎,後起之秀ARM是否有能力撼動地位仍有待觀察。

Intel 'Bay Trail' Atom compared to Snapdragon 800 and Tegra 4

Exactly a month ago, we first got our hands on Intel's new 'Bay Trail'  Atom processors. These chips are based on the x86-architecture and have an important advantage over their competition: they're compatible with a full-fledged version of Microsoft's Windows operating system. Nevertheless, the Z3000-series also targets tablets, where the chips will have to comepte with Qualcomm's Snapdragon 800 and Nvidia's Tegra 4, among others.

Engadget has now had the opportunity to run some benchmarks on three reference tablets, equipped with respectively an Intel 'Bay Trail', Nvidia Tegra 4 and Snapdragon 800 chip. It turns out that performance-wise, Intel's latest creation matches the fastest ARM-chip available at this time: the Snapdragon 800. The graphics chip of the 2.39 GHz quad-core Intel Atom Z3770 is clocked at 311 MHz, but is outperformed by both the Adreno 330, found in the Qualcomm SoC, and the video chip of Nvidia's Tegra 4.

The benchmarks do show that Intel is ready to compete in the ARM-market. If the power consumption is comparable as well, this could mark the breakthrough of x86-architecture on mobile devices.

Intel 將於 2014 年開始代工生產 ARM 處理器

在行動運算裝置崛起,傳統個人電腦高成長不再下,這自然也衝擊了電腦處理器的主要供應商英特爾(Intel),由於英特爾擁有最先進的晶圓設備與產能,先前外界也不斷傳出英持爾有意善用些資源,擴大代工範圍到行動裝置廣泛使用的 ARM 核心處理器,而呼聲最高的莫過於蘋果的 A 系列處理器,不過都從未成真。

不過這樣的光景可能即將來到,根據 Forbes 的報導,英特爾的合作夥伴 Altera 在 ARM developers’ conference 中對外表示,英特爾這家全球最大的半導體廠商,將決定在 2014 年開始對外代工 ARM 核心的處理器,這勢將衝擊現有的半導體代工產業。

也就是說英特爾即將與台積電(TSMC)、聯電(UMC)、GlobalFoundries、三星、IBM、中芯國際(SMIC)等一同競爭來自蘋果、Nvidia、Qualcomm 的訂單,由於英特爾擁有最佳的製程技術,對其他各廠可說是深具威脅。

看來英特爾在自家的產品難以與 ARM 處理器在市場中競爭下,打算以代工扳回部分業績,至少如此一來才不致於在行動運算中缺席。這下子或許在未來我們有機會在行動裝置中見到來自英特爾代工的 ARM 處理器了。

ARM tests: Intel flops on Android compatibility, Windows power
Intel Chip is not so compatible? 

ARM has rolled out a battery of test results that fire two shots across the bow of Intel's x86 dreadnought now sailing into Android waters.

The first set of results addresses the fact that when running native apps that haven't been recompiled to run on Intel-based Android devices, those apps need to be emulated using "binary translation," which converts native ARM code into native Intel x86 code.

Intel says that users shouldn't worry – its binary translator will "just work" with "very minimal power implications" and "unnoticeable performance impact for most applications." ARM – as you might expect – would beg to differ.

ARM low power performance is still better than Intel Atom

"Binary translation – despite what you may have read or despite what you may hear – does have a huge impact to the user and to the performance of the system," ARM senior technical marketing engineer Rod Watt told attendees at his company's 2014 Tech Day this week in Austin, Texas.
Efficiency of ARM is also better than Intel Atom

科技報報/微軟將舉行小型發布會 或推7.5英寸平板

過去兩年,一直有傳聞稱微軟會推出尺寸更小的Surface設備,如今,傳聞可能即將成真,微軟正式向媒體發出邀請,將於5月20日舉辦一場Surface活動,並在邀請函中附帶一句標語:來參加一場小型的聚會。

36氪據The Verge報導,這可能意味著參加此次活動人數不多,但是在微軟不斷更新Surface全尺寸產品線的背景下,這看起來更像是在暗示微軟即將推出的設備。去年九月就有報導稱今年早些時候微軟會發布7.5英寸的Surface Mini。

據稱,新的小尺寸Surface為7.5英寸螢幕,1440 x 1080解析度,4:3的長寬比,這個比例更像是iPad,而不是微軟現有的更適合看電影的 Surface平板電腦。有傳言稱,Surface Mini 將搭載高通處理器,運行Windows RT系統,而不是Windows 8。這也就意味著它會像Surface RT和Surface 2一樣,無法直接使用桌面端的軟體。
Refer to http://www.hitekpals.com/

雖然微軟官方沒有證實 Surface Mini 的存在,但據多家美國媒體報導,微軟的開發工具中最近增加了專門為小尺寸螢幕設計的模式。7寸的小平板此前被認為是遊戲設備,但是這已經是微軟兩年前的想法了

微軟對此的官方表述是:

我們在考慮很多事情,也在為之而努力工作。Surface系列將帶來多種長寬比和尺寸並且非常出色的產品。

8.5吋迷你款Surface 傳Computex 2014亮相

微軟先前已經傳出將推8.5吋迷你款Surface消息,同時也有消息指出將在今年Computex 2014期間公布。就近期 Build 2014 期間宣布將針對 9吋 以下裝置免費授權 Windows 作業系統,或許意味微軟將以迷你款Surface作為硬體設計範本,讓OEM合作廠商可做參考。
引用: 【台灣工程師的矽谷故事】如果你可以擁有一間公司,
你想要鴻海還是蘋果?

分析
Enhanced by Zemanta