Chinese Quantum Computer Shatters World Record

中国量子计算机打破世界纪录

(Yicai) Oct. 11 -- The latest version of China’s quantum computer has broken the world record for quantum computing in a scientific test, further cementing the country’s status as a global leader in the field of photonics quantum computing.

(第一财经)10月11日—最新的中国量子计算机在一次科学测试中打破了量子计算的世界纪录，进一步巩固了中国在光子学量子计算领域的全球领导地位。

The Jiuzhang 3 prototype achieved a speed that is 1 million times faster at solving Gaussian Boson Sampling problems compared with its predecessor, according to an article published in Physical Review Letters yesterday.

昨天发表在《物理评论快报》上的一篇文章称，九章3号原型机在解决高斯玻色子采样问题方面的速度比上一代计算机快了100万倍。

A research team from the Anhui-based University of Science and Technology of China and a research institute under the Chinese Academy of Sciences collaborated to build the prototype, with 255 detected photons, which improves the level of photon manipulation and the complexty of quantum computing.

地处安徽的中国科学技术大学的一个研究团队和中国科学院的一个研究所合作建造了九章3号原型机，检测到255个光子，提高了光子控制水平和量子计算的复杂性。

Gaussian Boson Sampling is a form of photonic quantum computation. It exploits the principles of quantum mechanics to perform certain types of computations deemed challenging for traditional computers.

高斯玻色子采样是光子量子计算的一种形式，它利用量子力学的原理来执行某些类型的计算，对传统计算机具备较大挑战。

The successful construction of Jiuzhang 3 means the quantum computer formulated by Richard Feynman, winner of the 1965 Nobel Prize in physics, has been realized, Prof. Lu Chaoyang of the University of Science and Technology of China told Yicai.

中国科学技术大学陆朝阳教授告诉第一财经记者，“九章三号”的成功建造意味着1965年诺贝尔物理学奖得主理查德·费曼发明的量子计算机已经实现。

以下是美国reddit网友的评论：

kecepa5669

Skeptic here.

Question: What is, "Gaussian boson sampling?" Because if that's a measure that's only relevant to quantum computers, then claiming it's 1016 times faster than normal computers is, let's just call it, "disingenuous at best." Because that's not how normal computers work.

我很怀疑。

我有个问题：什么是“高斯玻色子采样”？如果这是量子计算机的专有衡量标准，那么说它比普通计算机快1016倍就不公平了。因为普通电脑就不是干这个的。

Kinexty

It is disingenous. Because of a lack of purpose currently QC companies boast with bullshit comparisons. This algorithm is either completely useless or useless at current scales. We are still a decade or so that they can start solving bigger problems.

这台计算机并不是中国国产的。现在的量子计算机公司都在吹牛攀比。这个算法要么完全没用，要么在当前规模下没用。我们还需要10年左右的时间才能开始解决更大的问题。

DarkCeldori

Quantum computers will only get big if it turns out intrinsic error rate is low enough for viable error correction. If error rate turns out to be too high it will be physically impossible to build large real quantum computers.

量子计算机只有在内部错误率足够低，可以进行可行的错误修正时，才能变得更大。如果错误率太高，在物理上就无法建造真正的大型量子计算机。

cakesquadgames

The problem with using quantum safe encryption schemes being broken as a benchmark is that it will always be easier to scale encryption technology than it is to scale the computing infrastructure to crack them, quantum or not.

使用被打破的量子安全加密方案作为基准的问题是，无论是不是量子计算机，扩展加密技术总是比扩展计算基础设施来破解它们更容易。

Significant_Ride_483

You don't believe China? Your social credit score just went down.

你不相信中国？你的社会信用评分被扣了。

pianoceo

Those are big numbers but what does that actually mean in practice?

这些都是非常庞大的数字，但在实践中意味着什么呢？

LordMongrove

I see lots of “could be useful for” and it “might be useful for”.

So it sounds like a contrived algorithm designed to demonstrate the supremacy of quantum computing with very little practical value.

我看到很多“或许有用”和“也许有用”的说法。

这听起来像是一个人为设计的算法，目的就是展示量子计算的优势，但几乎没有什么实际价值。

Bleglord

At this point that’s kind of all quantum computing is. As a species, it’s amazing we can even get quantum computers to do things at all. It will probably take longer than it takes to reach AGI for quantum computing to actually be put into use

在这一点上，所有的量子计算都是如此。我们可以让量子计算机做事这一点本就很神奇。量子计算要真正投入使用，可能需要比实现人工智能更长的时间

FlyingBishop

I think this sort of stuff is valid. That said, in all cases I have seen if you dig in and look at it a bit more closer it doesn't matter if the algorithm is faster, there still exsts a classical computer that can do it more reliably at a fraction of the cost, and it doesn't look like there's a path to making the quantum computer actually competitive.

我认为这些东西是有用的。但我也意识到，如果你深入研究会发现算法快不快并不重要，某些经典的计算机，可以用很小的成本更加可靠地完成这些任务，而且似乎并没有什么途径可以让量子计算机具备真正的竞争力。

PolymorphismPrince

lmao what. You don't believe that, for instance, quantum computer break RSA?

笑死我了。你是说，你不相信量子计算机破解算法吗？

FIRECRACKER_X

It could be used in material science for simulating molecular structures and properties. In computer science, it could play a role in optimization problems and machine learning. It might also prove useful in financial mathematics, such as risk assessment and portfolio optimization.

Its defense applications are also amazing and underrated.

它可用于材料科学中模拟分子结构和性质。在计算机科学中，它也可以在优化问题和机器学习中发挥作用。它在金融数学中也可能有用武之地，比如风险评估和投资组合优化。

它在国防领域的应用前景也很惊人，只是现在还没有完全被发掘。

elfballs

If it's new, how does gpt4 know about it? Are you using the web browsing plugin?

如果它是新的，GPT4怎么会知道它？你是用网页浏览插件吗？

angrathias

It’s not new, it’s saying it’s just now done faster

这种计算机并不是什么新鲜事物，只是现在运算速度更快了。

InternationalMatch13

Gonna wait on independent confirmation that this is true quantum computing.

要等等看，这到底是不是真正的量子计算。

bownyboy

I've read the words, but don't understand what it actually means...

文章的字我都读完了，但我看不太懂它的真正意思……

Under_Over_Thinker

Is it? China  wants to invade other countries. Do you really want this kind of competition?

是吗？中国还想入侵其他国家。你真想看到这种竞争吗？

kkpappas

As opposed to the us who has invaded multiple other countries, has killed many millions and has fucked half the planet.

中国跟美国正好相反，美国入侵了多个国家，杀害了数百万人，祸害了半个地球。

alex3tx

This is why I never understood the "we can't be living in a simulation cos computing power needed is too high". Surely if computing power were to accelerate exponentially then getting to the power needed to simulate a universe in a few thousand, million, or even billion years wouldn't be out of the question?

这就是为什么我从来都无法理解“我们不能生活在模拟世界中，因为需要的算力太高了”。当然，如果计算能力呈指数级加速，那么在几千年、几百万年、甚至几十亿年内达到模拟宇宙所需的能力也并非不可能。

LightVelox

The logic behind it for some is that you would need atleast an equal amount of energy to simulate it, meaning atleast an universe-sized computer to simulate an universe, but that logic only holds itself when we apply it to our reality, a more complex or simply bigger universe should still be able to simulate ours

对一些人来说，背后的逻辑是你需要等量的能量来模拟，这意味着至少需要一个宇宙大小的计算机来模拟一个宇宙，但只有当我们把这项技术应用到我们的现实中，这个逻辑才能成立

Alex_2259

Would you actually need an equal amount of energy to simulate the universe in the universe? Even if you did it atom by atom?

I doubt it, computers don't scale that way to the best of my knowledge.

真的需要等量的能量才能模拟宇宙中的宇宙吗？一个原子一个原子来模拟？

我对此表示怀疑，就我所知，计算机可不是那样扩展的。

FrankoAleman

Take this news with a grain of salt until independently verified. China has a history of exaggerating or straight up fabricating news about progress in China.

在独立核实之前，我对这一消息持保留态度。中国历来都喜欢夸大或捏造进步新闻。

kindofbluetrains

Low settings my friend. Low settings.

配置低啊，我的朋友。配置很低。

Longjum-Bake-557

Give it a couple weeks for this to get disproven as well. Just like Huawei's amazing "ai".

再过几周，这则新闻也就会被证明是假新闻。就像他们当初震惊世界的“人工智能”一样。

exstentialytranquil

Would it make crypto worthless then? I mean hashing algorithms pretty much depends upon traditional computing power right? Looking for some technical perspective.

它会让加密货币变得毫无价值吗？哈希算法很大程度上取决于传统的计算能力，对吧？想看看一些从技术角度进行的分析。

SnooRadishes6544

Plug it into my head

把它接入我的脑袋吧。

anton13507

very unlinkly that it really works

不太可能是真的

fringecar

The first stat is disproved by the second. Is it a million times or 1016 times?

第一个数据被第二个数据推番了。到底是100万次还是1016次？

The_Observer_Effects

That's a bit dramatized and simplified. Yes these computers can make traditional supercomputers look like toys in some tests. But . . . they still can't run Minecraft. They are only excelling right now in very specific areas.

这有点戏剧化和简化了。在某些测试中，这些计算机的确会让传统的超级计算机相形见绌。但是……他们都不能运行《我的世界》呢。他们只是在某些具体领域表现出色。

AncientLion

I love how China is moving really fast when it comes to tech even with the hostility of the west.

看到中国在科技方面的快速发展，我很开心，西方有敌意就有敌意吧。

aymanzone

A lot of break throughs are coming out, but can these be verified?

出现过许多技术突破，但这些技术能得到验证吗？

Quirky-Tomatillo5584

Chinese Dragon doing Chinese genius stuff, Great work from the guineas Chinese scientists, keep up the hard work, & inform us about the universal quantum computer as soon as possible, please.

中国龙完成了中国的壮举，中国科学家的工作太伟大了，继续努力，尽快告诉我们关于通用量子计算机的事情。