Have we reached the limit of computer power? - Sajan Saini and George Zaidan

305,348 views ・ 2025-01-23

TED-Ed


請雙擊下方英文字幕播放視頻。

譯者: Lilian Chiu 審譯者: 麗玲 辛
00:06
In the Netherlands, there’s an ambitious company
0
6919
2253
在荷蘭,有一家野心勃勃的公司,
00:09
that builds one of the most advanced and expensive tools in the world:
1
9172
4087
打造了世界上最先進 且最昂貴的工具之一:
00:13
a single unit costs hundreds of millions of dollars.
2
13259
3253
單單一台的成本就高達數億美金。
00:16
And when companies buy one,
3
16554
1376
如果買了一台,
00:17
they also need 250 engineers to install the 165-ton device
4
17930
5798
還需要兩百五十名工程師
來安裝這個重達 一百六十五公噸的裝置,
00:23
in a process that typically takes half a year.
5
23770
3211
安裝過程通常需要半年。
00:27
But despite this steep cost in time and money,
6
27106
2711
但,儘管時間和金錢成本如此高昂,
00:29
many microchip makers desperately want one of these machines.
7
29817
3879
許多晶片製造商仍然 非常希望擁有一台這種機器。
00:33
The hundred-million-dollar question is: why?
8
33863
3212
價值上億美金的 重要問題是:為什麼?
00:37
The answer has to do with something called Moore’s Law.
9
37283
3378
答案與所謂的摩爾定律有關。
00:40
First coined by Intel co-founder Gordon Moore,
10
40703
3379
創造這個詞的人是英特爾的 共同創辦人高登‧摩爾,
00:44
this law states that every 1 to 2 years
11
44082
2544
摩爾定律指出,每一到兩年,
00:46
the number of transistors that can fit on a given size computer chip will double.
12
46626
4671
同樣尺寸的電腦晶片, 能容納的電晶體數量會加倍。
00:51
And by extension, the rough number of calculations that chip can do per second
13
51380
4630
進而,該晶片每秒 可執行的運算次數大致上
00:56
will also double.
14
56010
1627
也會加倍。
00:57
Now, this law isn't a physical law like gravity.
15
57887
2920
摩爾定律不是像重力 那樣子的物理定律。
01:00
It’s just a trend Moore observed during the early 1960s.
16
60848
4004
它只是摩爾在 1960 年代 初期觀察到的趨勢。
01:04
But chipmakers turned that trend into a goal, and in turn,
17
64977
3671
但晶片製造商將這種趨勢當成目標,
因此,消費者也開始期待運算技術
01:08
consumers learn to expect computing progress to continue
18
68648
3462
也會像這樣以指數級的 速度不斷進步。
01:12
at this exponentially fast pace.
19
72110
2335
01:14
And the amazing thing is, for six decades, it pretty much has.
20
74570
3754
驚人的是,六十年來, 事實也差不多是這樣。
01:18
Thanks to Moore’s Law, chips have gotten smaller, faster,
21
78407
3504
多虧摩爾定律,晶片變得更小、
01:21
more efficient, and cheaper.
22
81911
1960
更快、更有效率,且更便宜。
01:23
But today, there are four key problems that trip up this trend,
23
83955
3295
但現今,有四個重大問題 在阻礙這項趨勢,
01:27
potentially ending Moore’s Law
24
87333
1960
有可能會打破摩爾定律,
01:29
and fundamentally changing how we make progress in computing.
25
89293
3671
並徹底改變我們在運算方面 取得進展的方式。
01:33
The first is transistor size.
26
93172
2127
第一個問題是電晶體的尺寸。
01:35
Transistors are basically on/off switches,
27
95341
2753
電晶體基本上是個開/關裝置,
01:38
and these building blocks of digital computing have been shrinking
28
98094
3837
是數位運算的基礎元件,
01:41
since the 1960s.
29
101931
1752
從 1960 年代起, 它們就不斷在縮小。
01:43
But recently they’ve gotten so small,
30
103808
2336
但最近,它們已經小到
01:46
quantum physics has begun to interfere with their functions.
31
106144
3211
功能會受到量子物理干擾的程度。
01:49
When a transistor’s switch, or gate, is less than 20 nanometers,
32
109480
4004
當電晶體的開關或閘極 小於二十奈米時,
01:53
electrons will tunnel along it continuously,
33
113484
2753
電子會持續進行穿隧效應傳輸,
01:56
turning a crisp on/off switch into a hazy dimmer.
34
116279
4170
將本來清楚分明的開/關 變成模糊的調光器。
02:00
The second problem is heat.
35
120700
1793
第二個問題是熱。
02:02
As chipmakers make components smaller and more complex,
36
122493
3545
隨著晶片製造商把元件 做得更小、更複雜,
02:06
the copper lines that run between them need to be thinner and longer.
37
126038
3754
它們之間的銅線就得要更細、更長。
02:09
This increases their electrical resistance and generates high heat
38
129959
3837
這會增加它們的電阻,產生高熱,
02:13
that impairs chip performance and can’t be easily dissipated.
39
133838
3879
造成晶片性能降低, 且這高熱還很不容易消散。
02:17
Today’s chips can already run hot enough to cook an egg,
40
137884
3378
現今的晶片溫度 已經高到可以煎蛋了,
02:21
and temperatures are only predicted to increase without new innovations.
41
141387
4797
若沒有創新,預期 溫度只會繼續增加。
02:26
While both these issues represent limits in the fundamental physics of chipmaking,
42
146350
4213
雖然這兩個問題都代表 晶片製造中的基本物理學極限,
02:30
researchers haven't stopped trying to solve them.
43
150563
2377
但研究人員還在嘗試解決它們。
02:33
Unfortunately, their solutions often exacerbate the third major problem:
44
153191
4462
不幸的是,他們的解決方案 通常會加劇第三個重大問題:
02:37
chipmaking’s environmental impact.
45
157653
2419
晶片製造對環境的影響。
02:40
For example, swapping copper lines for ruthenium
46
160239
3504
比如,將銅線換成銠
02:43
could help pack transistors more tightly and keep chips smaller—
47
163743
4171
可以讓電晶體的排列更緊密, 維持晶片的小尺寸——
02:47
but that metal is far scarcer than copper
48
167997
2711
但這種金屬比銅更稀少,
02:50
and would require new mining infrastructure.
49
170750
2669
會需要新的採礦基礎設施。
02:53
Similarly, the technology currently used to make today’s smallest transistors
50
173586
4338
同樣的,要製造現今最小的 電晶體,目前所用的技術
02:57
requires huge amounts of energy and chemicals
51
177924
3044
會需要大量的能量
和化學物質:全氟烷基和 多氟烷基物質,
03:00
called perfluoroalkyl and polyfluoroalkyl substances,
52
180968
5464
03:06
which can take thousands of years to break down in the environment.
53
186432
3670
這些物質要數千年的時間, 才能在環境中分解。
03:10
Managing these first three problems contributes to the final issue: cost.
54
190311
4588
在解決前三個問題時, 卻會造成最後一個問題:成本。
03:14
To keep achieving Moore’s Law,
55
194982
1961
若要繼續實現摩爾定律,
03:16
chipmakers have to make individual chip components smaller.
56
196943
3712
晶片製造商必須把 每個晶片組件都做得更小。
03:20
And this is where that costly $400 million machine comes in.
57
200780
4796
此時就該四億美金的 昂貴機器登場了。
03:25
This marvel of chip-making science shoots a stream of tin droplets
58
205576
4463
這項晶片製造的科學奇蹟 會將一束錫滴射入真空室,
03:30
into a vacuum chamber before blasting them with a high-energy laser
59
210039
4171
然後用高能雷射 對它們進行強力照射,
03:34
that vaporizes the tin to create plasma.
60
214210
3003
把錫蒸發成為電漿。
03:37
In turn, the plasma emits a 13.5 nanometer wavelength of ultraviolet light
61
217421
6340
接著,電漿會發出 十三點五奈米波長的紫外光,
03:43
that can be used to produce incredibly small transistors.
62
223803
3670
可以用來製造極小的電晶體。
03:47
This remarkable feat of engineering has helped chipmakers
63
227473
3253
這項卓越的工程壯舉一直在協助 晶片製造商實現摩爾定律。
03:50
keep up with Moore’s Law.
64
230726
1835
03:52
But as chips keep getting denser,
65
232687
2294
但隨著晶片密度越來越高,
03:54
intricate manufacturing plants keep getting more expensive.
66
234981
3545
精密的製造工廠成本不斷攀升。
03:58
This trend has been so consistent, it’s actually earned the nickname
67
238734
3837
這種趨勢始終如一, 因此得到一個暱稱:
04:02
Moore’s Second Law.
68
242571
2086
摩爾第二定律。
04:04
Obviously, all these trajectories are unsustainable.
69
244991
3378
很顯然,這些發展路徑 不可能永遠持續下去。
04:08
Manufacturing plants can’t keep increasing in price,
70
248452
3003
製造工廠不可能不斷提升價格,
04:11
our ecosystems can’t endure endless mining and pollution,
71
251455
3796
我們的生態系統無法承受 無止盡的採礦和污染,
04:15
and the laws of physics are unlikely to change anytime soon.
72
255376
3587
物理定律短期內 也不太可能有所改變。
04:19
Fortunately, Moore's Law is flexible,
73
259130
2502
幸運的是,摩爾定律是有彈性的,
04:21
and there’s no reason we can’t introduce new goals
74
261757
2920
我們當然也可以引入新的目標,
04:24
to keep making computing progress responsibly.
75
264677
2961
用負責任的方式繼續推動 運算技術的進步。
04:27
Perhaps we could introduce a new Sustainability Law?
76
267763
3629
也許我們可以採用新的永續定律?
04:31
Smaller transistors already use less material and produce less e-waste,
77
271517
4880
較小的電晶體
使用的材料已經比較少了, 產生的電子廢物也比較少,
04:36
and advancements in electronic-photonic integration
78
276480
3254
而且電子與光子整合的進展
04:39
are allowing chips to use less energy and generate less heat.
79
279734
3753
讓晶片需要使用的能源更少, 產生出的熱也更少。
04:43
So perhaps chips should be made twice as sustainable every several years?
80
283571
4463
所以,也許每幾年, 晶片的永續程度就應該加倍?
04:48
Whatever the answer is,
81
288117
1960
不論答案是什麼,
04:50
we make the laws, so the future is up to us.
82
290119
4296
定律是我們制訂的, 所以未來也取決於我們。
關於本網站

本網站將向您介紹對學習英語有用的 YouTube 視頻。 您將看到來自世界各地的一流教師教授的英語課程。 雙擊每個視頻頁面上顯示的英文字幕,從那裡播放視頻。 字幕與視頻播放同步滾動。 如果您有任何意見或要求,請使用此聯繫表與我們聯繫。

https://forms.gle/WvT1wiN1qDtmnspy7