【正文】 ss. We even had a mock examination question in which the law featured. It was the type of question in which the first part asks for a definition of some law or principle and the second part contains a problem to be solved with the aid of it. In our case the first part was to define the Law of General Cussedness and the second was the problem。 these areas are coloured yellow in the tables. Areas for which problems are foreseen, but for which no manufacturable solutions are known, are coloured red. Red areas are referred to as Red Brick Walls. The targets set out in the Roadmaps have proved realistic as well as challenging, and the progress of the industry as a whole has followed the Roadmaps closely. This is a remarkable achievement and it may be said that the merits of cooperation and petition have been bined in an admirable manner. It is to be noted that the major strategic decisions affecting the progress of the industry have been taken at the prepetitive level in relative openness, rather than behind closed doors. These include the progression to larger wafers. By 1995, I had begun to wonder exactly what would happen when the inevitable point was reached at which it became impossible to make transistors any smaller. My enquiries led me to visit ARPA headquarters in Washington DC, where I was given a copy of the recently produced Roadmap for 1994. This made it plain that serious problems would arise when a feature size of 100 nm was reached, an event projected to happen in 2020, with 70 nm following in 2020. The year for which the ing of 100 nm (or rather 90 nm) was projected was in later Roadmaps moved forward to 2020 and in the event the industry got there a little sooner. I presented the above information from the 1994 Roadmap, along with such other information that I could obtain, in a lecture to the IEE in London, entitled The CMOS endpoint and related topics in Computing and delivered on 8 February 1996. The idea that I then had was that the end would be a direct consequence of the number of electrons available to represent a one being reduced from thousands to a few hundred. At this point statistical fluctuations would bee troublesome, and thereafter the circuits would either fail to work, or if they did work would not be any faster. In fact the physical limitations that are now beginning to make themselves felt do not arise through shortage of electrons, but because the insulating layers on the chip have bee so thin that leakage due to quantum mechanical tunnelling has bee troublesome. There are many problems facing the chip manufacturer other than those that arise from fundamental physics, especially problems with lithography. In an update to the 2020 Roadmap published in 2020, it was stated that the continuation of progress at present rate will be at risk as we approach 2020 when the roadmap projects that progress will stall without research breakthroughs in most technical areas “. This was the most specific statement about the Red Brick Wall, that had so far e from the SIA and it was a strong one. The 2020 Roadmap reinforces this statement by showing many areas marked red, indicating the existence of problems for which no manufacturable solutions are known. It is satisfactory to report that, so far, timely solutions have been found to all the problems encountered. The Roadmap is a remarkable document and, for all its frankness about the problems looming above, it radiates immense confidence. Prevailing opinion reflects that confidence and there is a general expectation that, by one means or another, shrinkage will continue, perhaps down to 45 nm or even less. However, costs will rise steeply and at an increasing rate. It is cost that will ultimately be seen as the reason for calling a halt. The exact point at which an industrial consensus is reached that the escalating costs can no longer be met will depend on the general economic climate as well as on the financial strength of the semiconductor industry itself.。 最初實驗用的計算機是由象我一樣有著廣博知識的人構(gòu)造的。后來，被證明是正確的，盡管我們也要學(xué)習(xí)很多新東西。 在電路的設(shè)計過程中，我們經(jīng)常陷入兩難的境地。在計算機領(lǐng)域工作的人都應(yīng)該記得 TTL， ECL 和 CMOS，到目前為止，CMOS 已經(jīng)占據(jù)了主導(dǎo)地位。為了讓 IEE 認識到無線工程和快速發(fā)展的電子工程并行發(fā)展是它自己的一項權(quán)利，我們不得不面對一些障礙。讓人有些憤怒的是，所有的 IEE 出版的論文都被期望以冗長的早期研究的陳述開頭，無非是些在早期階段由于沒有太多經(jīng)驗而遇到的困難之類的陳述。世界上的計算機數(shù)量已經(jīng)增加了許多，并且性能比以前更加可靠。分時系統(tǒng)開始起步，并且計算機圖形學(xué)隨之而來。這個變化對當時的工程師們是個不可回避的挑戰(zhàn)。只能說他們鼓起勇氣接受了挑戰(zhàn)，盡管這個轉(zhuǎn)變并不會一帆風順。隨著時間的推移，一個片子能夠容納的最大數(shù)量的晶體管或稍微少些的邏輯門和翻轉(zhuǎn)門集成度達到了一個最大限度。每個門電路或翻轉(zhuǎn)電路是相互獨立的并且有自己的引腳。 這些芯片為制造一種新的計算機提供了可能。他比大型機稍遜，但功能強大，并且更能讓人負擔的起。 隨著微機的開始流行并且功能的完善，世界急切獲得它的計算能力但總是由于工業(yè)上不能規(guī)模供應(yīng)和它可觀的價格而受到挫折。 計算消耗的下降并非起源與微機，它本來就應(yīng)該是那個樣子。隨著時間的推移，人們比他們付出的金錢得到的更多。 7400系列的用戶能夠工作在邏輯門和開關(guān)級別并且芯片的集成 度可靠性比單獨晶體管高很多。在劍橋大學(xué)實驗室力，我們構(gòu)造了 CAP，一個有令人驚奇邏輯能力的微機。令牌環(huán)設(shè)計研究的發(fā)表先于以太網(wǎng)。 令牌環(huán)網(wǎng)需要高可靠性，由于脈沖在令牌環(huán)中傳遞，他們必須不斷的被放大并且再生。 精簡指令計算機的誕生 早期的計算機有簡單的指令集，隨著時間的推移，商業(yè)用微機的設(shè)計者增加了另外的他們認為可以微機性能的特性。 1980 年， RISC 運動改變了微機世界。 除了 RISC 這個引人注目縮略詞外，這個標題傳達了一些指令集合設(shè)計的見解，隨之引發(fā)了 RISC 運動。線程不是個新概念，但是它對微機來說是從未有過的。我的意思是說利用目前存在的功能強大的計算機去模擬新的設(shè)計。 模擬仿真加快了開發(fā)進度并且被計算機設(shè)計者廣泛采用。今天，設(shè)計者們希望有滿屋可用計算機做他們的 仿真，而不只是一臺， X86 指令集 除非出現(xiàn)很大意外，要不很少聽到有計算機使用早期的 RISC 指令集了。 X86 構(gòu)架已經(jīng)占據(jù)了計算機核心指令集的主導(dǎo)地位。 對于我們這些從事計算機學(xué)術(shù)研究的人， X86 的統(tǒng)治地位讓我們感到失望。高級語言并沒有完全消除對機器原始編碼的的使用。然而，情況也許有所不同，如果 Intel 的主要目的是為是生產(chǎn)一個好的 RISC 芯片。從許多方面來說， i860 是個卓越的芯片，但是它的軟件借口不適合在工作站上應(yīng)用。直接應(yīng)用先前 x86 的實現(xiàn)方式對于滿足 RISC 處理器的持續(xù)增長的速度要求，是不可能的。表面上，一片現(xiàn)代的 x86 芯片包含了隱藏實現(xiàn)的 部分，好象和實現(xiàn) RISC 指令集的芯片一樣。 對于以上 RISC 運動的總結(jié)，我非常信賴最新版本的哈里斯和培生出版社的有關(guān)計算機設(shè)計的書籍。這最初主要是為了滿足通常的 64 位地址空間問題。然而，人們普遍認為 Intel 應(yīng)該與 x86 構(gòu)架保持兼容，可令人疑惑的是恰恰相反。特別的，每條指令它需要附加的 6 位。 盡管 IA64 是個全新的指令集，但 Intel 發(fā)表了一個令人困惑的聲明：基于IA64 的芯片將與早期的 x86 芯片保持兼容。 最新的稱為 Itaninu IA64處理器顯然需要特殊的兼容性的硬件，盡管如此，x86 編碼運行的相 當慢。因此，在任何情況下，作為常識和一般性的標準， Gordon Moore 在訪問劍橋最近開放的 Betty and Gordon Moore 圖書館時所反復(fù)強調(diào)。但是我已經(jīng)作好了準備，去接受這樣的事實，我已經(jīng)完全不了解半導(dǎo)體經(jīng)濟學(xué)了。這種片子并不是很大。（在這篇演講稿被提交之前， Intel 表示他們將銷售一系列本質(zhì)上與 AMD 兼容的芯片） 更小晶體管的出現(xiàn) 集成度還在不斷增加，這是通過縮小原始晶體管以致可以更容易放在一個片子上