跟讀練習: Every Giant Number Explained - 通過影片學習英語口語
選擇練習模式
每種模式都會開啟全屏練習室。
194 句
如果句子過短或過長，請點選 Edit 進行調整。
1
If you wrote one digit on every atom in the observable universe,
2
you'd run out of atoms before you finished writing one of the numbers in this video.
3
And that number isn't even the biggest one here.
4
But let's start small.
5
A million, a billion, a trillion.
6
They sound like neighbours, but each one is a thousand times bigger than the last, and the easiest way to feel that is in seconds.
7
A million seconds is a little over 11 days.
8
A billion seconds is almost 32 years.
9
And a trillion seconds is about 32 ,000 years.
10
That's six times longer than humans have had writing.
11
32 ,000 years ago, the world was in an ice age and people were hunting mammoths.
12
So counting with words gets awkward fast.
13
A million is a one followed by six zeros.
14
Short version, ten to the sixth.
15
A billion is ten to the ninth.
16
A trillion, ten to the twelfth.
17
Every three zeros get a new word.
18
Quadrillion, quintillion, all the way up to a decillion, ten to the thirty -third.
19
The words run out soon.
20
The numbers don't. In 1920,
21
the American mathematician Edward Kasner asked his nine -year -old nephew what to call a one followed by a hundred zeros.
22
The boy, whose name was Milton, said, A Google.
23
A Google is ten to the hundredth.
24
It's easy to write down.
25
It's one line.
26
But what can you count with it?
27
Atoms In the entire observable universe, in all the stars, planets and gas clouds, there are about 10 to the 80th.
28
That's less than a Google.
29
Not a little less, a hundred quintillion times less.
30
Imagine every atom in the universe turned out to be a whole universe full of atoms.
31
Now you'd have more than a Google.
32
But in our real universe, there just aren't that many things.
33
There are that many possibilities, though.
34
In 1950, Claude Shannon estimated how many different chess games there could be.
35
He got about 10 to the 120th, more than a Google.
36
Combinations grow much faster than stuff.
37
By the way, one company's name is a misspelling of Google, and its servers still live at the domain 1E100,
38
which is how programmers write a Google.
39
Milton came up with the next word too.
40
A Googleplex, he said, is a one followed by writing zeros until you get tired.
41
Casno pointed out that different people get tired at different times, so he made it precise.
42
A Googleplex is a one followed by a Google zeros, ten to the power of a Google.
43
And this is where the paper runs out.
44
To write a Googleplex, you need to write a Google zeros.
45
Even if you put one zero on every atom in the universe, you wouldn't have enough atoms there are only 10 to the 80th of them,
46
and you need 10 to the 100th zeros.
47
If you printed a million zeros per book, you'd need 10 to the 94th books.
48
That's more books than there are atoms in the universe.
49
Carl Sagan made this point in Cosmos back in 1980.
50
There isn't room for a Googleplex in the known universe.
51
But numbers like this do show up in physics.
52
If the universe is infinite, then somewhere out there is an exact copy of Earth, and an exact copy of you watching this video.
53
The physicist Max Tegmark estimated how far away, roughly 10 to the 10 to the 28th miles.
54
And at that size, it doesn't matter whether you say miles, feet or inches, the difference gets lost somewhere in the last digits of the exponent.
55
And a Googleplex is bigger still, much bigger.
56
The Googleplex was invented as a joke, but once an even bigger number was needed for real.
57
Prime numbers divide only by themselves and 1, 2, 3, 5, 7, 11.
58
The further you go, the rarer they get.
59
Gauss came up with a formula that roughly tells you how many primes there are below any number.
60
People checked it for billions, for trillions, for numbers with 19 zeros.
61
And every time, the formula came out a little too high.
62
Every single time.
63
You'd think it would always be that way.
64
But in 1914, John Littlewood proved it isn't.
65
Somewhere far out, the formula starts missing in the other direction, and it flips back and forth infinitely many times.
66
Where exactly, he didn't say.
67
In 1933, his student Stanley Skews found a limit.
68
The first flip has to happen before 10 to the 10 to the 10 to the 34th.
69
Three stories of exponents.
70
G. H.
71
Hardy called it the largest number that had ever served any definite purpose in mathematics.
72
Today we know the first flip happens much earlier, somewhere around 10 to the 316th.
73
But the exact spot still hasn't been found.
74
And that's the lesson.
75
A pattern can hold for every number we can reach and still be false.
76
To go further, we need a new way of writing.
77
and it's built like a ladder.
78
Adding three to itself three times is multiplication.
79
Multiplying three by itself three times is a power, and if you raise three to a power and then raise three to that result,
80
you get a tower of powers.
81
In 1976, Donald Newth suggested arrows for this.
82
One arrow is an ordinary power.
83
Three arrow three is 27.
84
Two arrows make a tower.
85
3 double arrow 3 is 3 to the 3 to the 3.
86
That's 3 to the 27th.
87
7 ,625 ,000 ,000 ,000 ,000 and change.
88
Not scary yet, but add one more story to the tower.
89
3 to the power of that number.
90
The result has more than 3 .5 trillion digits.
91
Now 3 arrows.
92
3 triple arrow.
93
3 is a tower of 3's more than 7 .5 trillion stories tall.
94
Not that many digits.
95
that many stories.
96
For comparison, a Googleplex fits in a tower of just four tens and skews his number in a tower of five.
97
Four stories of tens and the number already has more digits than there are atoms in the universe.
98
Each new arrow repeats the previous one as many times as the last step produced,
99
and these are the arrows you need to describe the next number.
100
Take a cube and connect every corner to every other corner.
101
Colour each line red or blue.
102
Now look for four corners that lie in one plane with all six lines between them the same colour.
103
With an ordinary cube, you can colour the lines so that no such four exist.
104
But what about a cube in more dimensions?
105
From which dimension on is there always such a group of four, no matter how you colour?
106
In 1971, Ronald Graham and Bruce Rothschild proved that such a dimension exists.
107
And in 1977, Martin Gardner wrote about a number
108
Graham had found as an upper bound for the answer.
109
Three years later, it was in the Guinness Book of World Records as the largest number ever used in a serious proof.
110
Here's how it's built.
111
Take three, four arrows, three.
112
That's step one, and it's already monstrously bigger than anything so far.
113
In step two, there are as many arrows between the threes as step one produced.
114
In step 3, as many as step 2.
115
And so on, 64 times.
116
Physicists say a region the size of your head can hold only a limited amount of information.
117
Cram in more, and that region becomes a black hole.
118
The digits of Graham's number won't fit in your head.
119
They won't fit in the whole universe.
120
And yet we know how it ends.
121
Its last digits are 2464195387.
122
And the actual answer to the problem?
123
Today we know it's at least 13.
124
The answer might be a small number.
125
Only the bound was enormous.
126
Graham's number came from a problem.
127
The next one comes from a game you can play on a sheet of paper.
128
You draw trees.
129
Dots connected by lines, with no loops.
130
Each dot can be one of several colours.
131
The first tree can have at most one dot.
132
The second, at most two.
133
The third, at most three.
134
And so on.
135
The main rule?
136
No earlier tree may fit inside a later one with the same colours and the same branching.
137
As soon as you can't draw a new tree, the game is over.
138
In 1960, Joseph Kruskal proved a theorem that means this game always ends, no matter how many colours you use.
139
With one colour, the game lasts one move.
140
With two colours, three moves.
141
And for three colours, the logician Harvey Friedman called the length of the longest possible game tree of three.
142
It's finite.
143
The game really does end.
144
But next to tree of three, Graham's number is practically zero.
145
Friedman showed that if you try to prove this game ends using modest arithmetic,
146
the proof itself would be longer than anything you could write down in the universe.
147
Three colours and simple rules.
148
And the result is a number that makes all of Knuth's arrows look like toys.
149
Every number so far can be computed.
150
Slowly, longer than the life of the universe, but by the rules.
151
And then there are numbers that can't be computed at all.
152
Take the simplest computer, a Turing machine.
153
It's an endless tape of cells and a head that reads a symbol, writes a new one, and moves, following a table of rules.
154
Back in 1936, Turing proved that there's no way to tell in advance, for every program, whether it will ever stop or run forever.
155
In 1962, Tibor Rado came up with a game called The Busy Beaver.
156
Take all machines with a given number of rules and run them on a blank tape.
157
Among the ones that stop, find the one that runs the longest.
158
With one rule, one step.
159
With two, six.
160
With three, twenty -one one, with four a hundred
161
and seven with five forty seven million one hundred seventy six thousand eight hundred seventy steps
162
that was only proven in july 2024
163
and it was done mostly by amateurs working together online with six rules in 2025 someone found a machine
164
that runs longer than two triple arrow five this function grows faster than any function
165
that can be computed sooner or later it overtakes the arrows, graham and tree.
166
And for a machine with 215 rules, the answer can no longer be proven with ordinary mathematics.
167
That was shown in October 2026.
168
There's one last step.
169
In 2007, at MIT, two philosophers held a duel.
170
Who can name the bigger number?
171
Agustin Rayo won like this.
172
Take every number that can be described in the language of logic with no more than a Google symbols.
173
Rayo's number is the smallest number bigger than all of them.
174
A million seconds, the atoms in the universe, a Google, a Googleplex, Skews,
175
the Arrows, Graham, Tree of Three, the Beaver, Rayo.
176
It feels like we've gone very far.
177
But here's the strange part.
178
Whichever of these numbers you pick, there are infinitely many numbers bigger than it, and only finitely many smaller.
179
So almost all whole numbers are bigger than Rayo's number.
180
We haven't gone far at all.
181
We're standing at the very start of the number line.
182
So is the biggest number infinity?
183
No. Infinity isn't the last number in line.
184
It answers a different question.
185
Not which number, but how many numbers there are in total.
186
And that question has a ladder too.
187
Georg Cantor showed that infinities come in different sizes.
188
There are infinitely many whole numbers, but there are even more points on a line.
189
And above that infinity, there's another one, and so on, forever.
190
And any number in this video, even Rayo's number, is no better than one when you put it next to infinity.
191
Enormous, but finite.
192
We started with the idea that the universe doesn't have enough atoms to write down one number.
193
Now it's clear the problem isn't the universe.
194
Numbers simply grow faster than any place you could write them down.
FOR YOU
推薦影片
C16
104 句
How real are the real numbers, really?
Daily ConversationUniversity Students
C15
56 句
Gemini 4 Argon explained in 5min..
Daily ConversationUniversity Students
C14
33 句
Brainf**k in 100 Seconds
American AccentDaily ConversationUniversity Students
C11
107 句
What If You Lit a Match on Jupiter, Would The Planet Explode?
Daily ConversationUniversity Students
C11
140 句
Euclid Found a BILLION Objects in This NEW Map of the Universe!
Daily ConversationUniversity Students
C11
65 句
How do microSD Cards Store 2 TB in such a TINY space?
American AccentDaily ConversationUniversity StudentsWorking Adults
![[Practice Problems] Regular Expressions (Regex) - Visually Explained](https://img.youtube.com/vi/AawczswiGI0/mqdefault.jpg)
C11
61 句
[Practice Problems] Regular Expressions (Regex) - Visually Explained
Daily ConversationUniversity Students
C13
246 句
$15 Trillion Money-Making Machine | How BlackRock Became Too Powerful to Ignore
Daily ConversationUniversity StudentsWorking Adults
C181 句
Mathematician Collapses All Functions to One Weird Formula
Daily ConversationUniversity StudentsWorking Adults
C11
234 句
Every Processing Unit Explained in 20 Minutes
Daily ConversationUniversity Students
C12
261 句
The Insane Megastructure We Could Build in 31 Years
Daily ConversationUniversity Students
C11
498 句
Give me 31 minutes, I will make PyTorch click forever
Daily ConversationUniversity Students
本課的詞彙與口語要點
這節口語課以影片“Every Giant Number Explained”為素材。 影片中反覆出現的詞有：universe, arrow, Google, atom, colour。 這段影片共有 194 個句子、1954 個單詞可供跟讀。 講話部分時長為 12:13。 說話人語速自然，每分鐘約 160 個詞，接近日常對話。 只有 81% 的單詞屬於英語最常用的 3,000 詞，詞彙難度較高。
影片中的重點詞彙
影片中 15 個值得學習的單詞，附發音和釋義：
| 單詞 | 發音 | 釋義 |
|---|---|---|
| arrow 名詞 | /ˈæɹəʊ/ | җян, 箭 |
| atom 名詞 | /ˈætəm/ | 原子 |
| digit 名詞 | /ˈdɪd͡ʒɪt/ | 數字 /數字 |
| infinity 名詞 | /ɪnˈfɪnɪti/ | 無限 /無限 |
| formula 名詞 | /ˈfɔː.mjʊ.lə/ | 公式 |
| ordinary 形容詞 | /ˈɔɹdɪnɛɹi/ | 普通, 平常 |
| dimension 名詞 | /daɪˈmɛn.ʃən/ | 方面, 特點 /特點 |
| cube 名詞 | /kjub/ | 立方體 /立方體 |
| connect 動詞 | /kəˈnɛkt/ | 連線 /連線 |
| symbol 名詞 | /ˈsɪmbəl/ | 符號 /符號 |
| triple 形容詞 | /ˈtɹɪp.əl/ | 三重 |
| estimate 名詞 | /ˈɛs.tɪ.mət/ | 估計 /估計 |
| enormous 形容詞 | /ɪˈnɔː.məs/ | 巨大的 |
| mathematics 名詞 | /mæθ(.ə)ˈmæt.ɪks/ | 數學 /數學, 算學 /算學 |
| ladder 名詞 | /ˈlæd.ɚ/ | 梯子, 扶梯 |
影片中出現的短語動詞
| 單詞 | 發音 | 釋義 |
|---|---|---|
| come up with 動詞 | 想出, 提出 | |
| write down 動詞 | 寫下來 /寫下來, 寫下 /寫下 | |
| come out 動詞 | /ˌkʌm ˈaʊt/ | 出, 出來 /出來 |
| point out 動詞 | 指, 指出 |
值得反覆跟讀的句子
影片中簡短而完整的句子，可以直接用在日常對話裡：
- You'd think it would always be that way.
- It feels like we've gone very far.
- We haven't gone far at all.
影片中的語法
說話人最常用的結構，並附上影片中的原話：
| 結構 | 影片中的用法 |
|---|---|
| 被動語態 be + 過去分詞 — 強調發生了什麼，而不是誰做的 | was invented · been found · can be computed |
| 現在完成時 have/has + 過去分詞 — 過去發生但與現在仍有關聯的事 | have had · hasn't been found · we've gone |
需要注意的發音
說話人用了 23 次縮略和弱讀形式，例如 isn't, you'd, can't。請按聽到的簡短形式來說。
- “th” 音: mathematics /mæθ(.ə)ˈmæt.ɪks/, sixth /sɪksθ/, thirty /ˈθɝti/, ninth /naɪnθ/, theorem /ˈθiərəm/
- “sh” 和 “zh” 音: dimension /daɪˈmɛn.ʃən/, sheet /ʃit/, combinations /kɑmbɪˈneɪʃənz/, logician /ləˈd͡ʒɪʃən/, mathematician /ˌmæθ.(ə.)məˈtɪʃ.ən/
- 長單詞——注意重音位置: infinity /ɪnˈfɪnɪti/, ordinary /ˈɔɹdɪnɛɹi/, infinitely /ˈɪnfɪnɪtli/, mathematics /mæθ(.ə)ˈmæt.ɪks/, practically /ˈpɹæk.tɪ.k(ə.)li/
中文母語者容易讀錯的音：
- /v/ — 不要讀成 /w/，上齒輕觸下唇: server /ˈsɝvɚ/, divide /dɪˈvaɪd/, whichever /wɪt͡ʃˈɛvɚ/, invent /ɪnˈvɛnt/, observable /əbˈzɜː.və.bəl/
- 詞尾子音 — 要讀清楚，後面不要加母音: digit /ˈdɪd͡ʒɪt/, flip /flɪp/, cube /kjub/, tape /teɪ̯p/, connect /kəˈnɛkt/
- /r/ 和 /l/ 的區別: roughly /ˈɹʌf.li/, triple /ˈtɹɪp.əl/, carl /kɑɹl̩/, practically /ˈpɹæk.tɪ.k(ə.)li/
如何用這段影片練習
- 先完整聽一遍影片，不要開口，記下不認識的單詞。
- 先用 0.75 倍速逐句跟讀，熟練之後再回到正常速度。
- 錄下自己的聲音並與原聲對比，特別注意 arrow, atom, digit 這類單詞。
什麼是跟讀法？
跟讀法 (Shadowing) 是一種有科學依據的語言學習技巧，最初開發用於專業口譯員的培訓，並由多語言者Alexander Arguelles博士普及。這個方法簡單而強大：您在聽英語母語原聲的同時立即大聲重複——就像是一個延遲1-2秒緊跟說話者的影子。與被動聽力或語法練習不同，跟讀法強迫您的大腦和口腔肌肉同時處理並模模擬實的講話模式。研究表明它能顯著提高發音準確性，語調，節奏，連讀，聽力理解和口語流利度——使其成為雅思口語備考和真實英語交流最有效的方法之一。