當前位置

首頁 > 英語閱讀 > 英語故事 > 《那些古怪又讓人憂心的問題》第27期:人力計算機(2)

《那些古怪又讓人憂心的問題》第27期:人力計算機(2)

推薦人: 來源: 閱讀: 1.92W 次

All the computers in the world couldn'tfigure out the correct answer faster than any one parent could. But that'sbecause computers haven't been programmed to figure that kind of thing out,3whereas brains have been trained by millions of years of evolution to be goodat figuring out what other brains around them are doing and why.

《那些古怪又讓人憂心的問題》第27期:人力計算機(2)

全世界所有的計算機都不能比任何一個家長更快地找出正確的答案,這是因爲我們給計算機編程時並沒有告訴計算機要怎麼解決這種問題,3而我們的大腦則已經接受了上百萬年進化的訓練,對於猜出周圍的人在幹什麼以及爲什麼要這麼做很是在行。

So we could choose a task to give thehumans an advantage, but that's no fun; computers are limited by our ability toprogram them, so we've got a built-in advantage.

我們可以選擇一個人類具有優勢的任務,但這又有什麼意義呢?計算機受制於我們的編程能力,因而我們人類有天生的優勢。

Instead, let's see how we compete on theirturf.

所以我們還是看看在計算機的地盤上比賽如何吧。

The complexity of microchips

芯片的複雜度

Rather than making up a new task, we'llsimply apply the same benchmark tests to humans that we do to computers. Theseusually consist of things like floating point math, saving and recallingnumbers, manipulating strings of letters, and basic logical calculations.

我們不用弄出一個新任務,只需讓人類進行計算機上常用的基準測試就行。這種測試一般包括像浮點數計算、存儲和回憶數字、調整字符串以及一些基本的邏輯運算這樣的內容。

According to computer scientist HansMoravec, a human running through computer chip benchmark calculations by hand,using pencil and paper, can carry out the equivalent of one full instructionevery minute and a half.

根據計算機科學家漢斯•莫拉維克的說法,一個人如果用紙筆進行計算機上的芯片基準測試,那麼他的速度爲每1.5分鐘執行一條完整指令。

By this measure, the processor in amidrange mobile phone could do calculations about 70 times faster than theentire world population. A new high-end desktop PC chip would increase thatratio to 1500.

這樣算來,一部中端手機的處理器的計算能力將是全球人口加起來的運算能力的70倍左右,而一臺高性能臺式機的芯片則能將這個比值提高到1500。

So, what year did a single typical desktopcomputer surpass the combined processing power of humanity?

那麼是從什麼時候開始,一臺普通的臺式機的計算能力超過了所有人類的計算能力總和呢?

1994

1994年。

In 1992, the world population was 5.5 billionpeople, which means their combined computing power by our benchmark test wasabout 65 MIPS (million instructions per second) same year, Intel releasedthe popular 486DX, which in its default configuration achieved about 55 or 60MIPS. By 1994, Intel's new Pentium chips were achieving benchmark scores in the70s and 80s, leaving humanity in the might argue that we're being alittle unfair to the computers. After all, these comparisons are one computeragainst all humans. How do all humans stack up against all computers?

1992年全世界總人口爲55億人,這意味着根據基準測試,人類的總計算能力約爲65MIPS5。同年英特爾發佈了風靡一時的486DX芯片,在默認配置下這塊芯片每秒能夠處理55~60MIPS。截至1994年,英特爾的新“奔騰”芯片的基準測試成績達到了每秒70~80MIPS,已經把人類拋在了後頭。你可能會說這種測試對計算機有一些不公平,畢竟現在是一臺計算機和所有人類比,那麼如果所有計算機和所有人類比會怎麼樣呢?