首页 >> 来自播客: User Upload Audio 更新反馈

Nvidia Founder and CEO Jensen Huang on the AI revolution

发布时间 2024-11-11 22:47:56 来源

I think it's fun to be inventing the next wife, computer, and carer. It's fun to see all these amazing applications being created. These are the goals I was taught to aspire to. And part of me did the world on our shoulders. You've reached the end of this preview. To listen to the full title, you can buy it in your player. I hope everybody's been enjoying the conference. It's a fantastic event. Lots of great companies. Here's the other recommendations. A couple thousand people here. And so really, really terrific. And obviously, a real highlight and a real privilege to have Jensen, the president and CEO of NVIDIA here. This is awesome. Since you found NVIDIA 1993, you've pioneered the random accelerating computing. The companies invented the GPU in 1999 sparked the growth of 15 years. The PC gaming market redefined in computers and igniting the era of modern AI.

我觉得发明未来的妻子、电脑和护理设备很有趣。看到这些惊人的应用诞生让我感到兴奋。这些都是我被教导要追求的目标。我也感觉到肩负世界的责任。你已经看到了这个预览的结尾。要收听完整内容，可以在你的播放器中购买。我希望大家都在享受这次会议。这是一场精彩的活动，有很多优秀的公司参与。这里有其他推荐。有几千人在这里，真的非常精彩。显然，能有NVIDIA的总裁兼首席执行官Jensen在场，是一个真正的亮点和荣幸。太棒了。自从1993年你创办NVIDIA以来，你就引领了随机加速计算的潮流。公司在1999年发明了GPU，触发了15年的增长，重新定义了PC游戏市场，重新塑造了计算机领域，并开启了现代AI的时代。

Read for you by Robert Kuyker. And you've seen all the BSE degree from our state university and an MSE degree from Stanford. And so I want to start by welcoming you, Jensen. Everybody, please welcome Jensen to the stage. Thank you, 2016. I made my way to the stage for the future. Thank you. I love it. I'm the pastor. Thirty-one years ago, founded the company. I was there at the final preparation. You've transformed yourself from a gaming-centric GPU company to one that offers a broad range of hardware, software, to the data center industry. I'd just like you to start by talking a little bit about the journey. When you start throwing your thinking, how's it evolved? The creation of this part. Because it's been a pretty extraordinary journey. It's the biggest, the hardest journey in my career. It felt like the right time to move on. The thing that we got right now is go the way you found the will.

为您朗读的是罗伯特·库伊克。你已经看过他从本州大学获得的BSE学位和从斯坦福大学获得的MSE学位。因此，我想先欢迎你，詹森。大家，请欢迎詹森上台。谢谢你，2016年。我走上了通往未来的舞台。谢谢你。我热爱这一切。我是一位牧师。三十一年前，我创办了这家公司。我参与了最后的准备工作。你已经成功将公司从一家专注于游戏的GPU公司转变为能够为数据中心行业提供广泛硬件和软件的企业。我希望你能先谈谈这个历程。当你的思维开始转变的时候，是如何演变的？这个部分的创建历程如何？因为这是一个相当非凡的旅程。这是我职业生涯中最大、最艰难的旅程。我觉得这是一个合适的时机去继续前进。我们现在要做的是找到动力去发现实现目标的道路。

I would say, is a possibility that we are visioned that there would be another form of computing that could augment the general purpose computing. Much more profound things that are the events of that week in Shanghai. Instrument. We were opening the graph on Thursday, June 16. And that processor would start out doing some things that was scheduled to arrive. That was insanely hard for CPUs to do. And it was computer graphics. Creative partners. But that we would expand that over time to do other things. The first thing that we would chose, of course, was in each processing, which went which is complementary computer graphics. We extended it to physics simulation, because in the domain, the application of the domain that we selected, the new games, we wanted to be beautiful, but also wanted to be dynamic from day virtual worlds.

我认为有可能会出现一种新的计算形式，可以增强通用计算的功能。比上海那周发生的事件更加深远的事情。当时我们正在6月16日星期四开启图表。那个处理器的初始任务是执行一些对CPU来说极其困难的任务，而这些任务与计算机图形相关。我们与创意合作伙伴一起，但希望随着时间的推移，能将其扩展到其他领域。我们选择的第一个方向当然是图像处理，因为它与计算机图形是互补的。随后我们将其扩展到物理模拟的领域，因为在我们选择的领域应用中，新的游戏不仅要美观，还需要在虚拟世界中展现出动态效果。

We took step by step, and we took it into scientific computing beyond that. In the 61 years, in the world, there's a world business like the biggest investment in the industry. Some of our tools don't really do the part justice. Types of algorithms, adjacent industries. And some of its scope. Kind of solved our way here. In the 21, cost about 6 billion dollars at the time was that it's already not going to be able to solve problems that are interesting. About 11 times the size of the design. And that we were able to keep the architecture consistent. As many as 14,000 were used in architecture where I saw cores that you'd develop today. Could we discover the thousands of days, dancers, and actors. And the software they created in the past would be accelerated even further by new technology. This way of thinking about architecture compatibility, creating large install base, and more part of the sector.

我们一步步前进，并将其应用于科学计算。在这61年里，这在全球范围内是一个类似于行业最大投资的商业项目。我们的一些工具并不能完全发挥其应有的作用，包括各种算法和相关行业。某种程度上，我们通过解决问题走到了现在。在2021年，这项投资约为60亿美元，但它已经无法解决那些有意义的问题，其设计规模大约是当前的11倍。我们能够保持架构的一致性，使用多达14,000个在现代开发中常见的核心。我们发现成千上万的日子、舞者和演员，那些他们过去创造的软件将通过新技术进一步加速。这种对架构兼容性的思考方式，创造了庞大的安装基础，并成为该行业更多的一部分。

And I can remember. Investment of the ecosystem along with us. That psychology started in the middle of our integration. We shared it to the day. Which is the reason why Injigia's CUDA has such a massive and solid business. We had endless negotiations. Because we always protected. And we were protecting the management of software developers. And considered that's been the number one priority of our company since the very beginning. And going forward. And it's to turn forward. Some of the things that we solved along the way. Of course, learning how to education, geopounder, learning how to be a CEO, learning how to conduct business, learning how to build a company. Not as a strategy. These are all the new skills. And cultural impurities. Which is kind of like learning how to. The overwhelming challenge. Invent. The modern computer. The modern computer.

我还记得，我们与生态系统一起的投资。这种心理在我们整合的中期开始形成，我们一直延续到今天。这也是Injigia的CUDA这项业务如此庞大而稳固的原因。我们进行了无数次谈判，因为我们始终在保护，保护软件开发人员的管理，并将其视为公司从一开始到现在的首要任务。展望未来，我们会继续发展，并在这个过程中解决一些问题。当然，包括学习如何教育、成为一名CEO、开展业务、建设公司。这些并非策略，而是新技能和文化上的突破。就像学习如何应对巨大的挑战一样。发明现代计算机。

It became a much simpler. In video, people don't know this. But in video, it's a large experience. In the base of video game architecture in the world. Distinctly Chinese. It's some 300 million gamers in the world. It's still growing in the rest of my team. And while Supervibrant. Received news from the past year. I think the post-doc. Every single time we had to go 15 miles. And turn into a new market. We had to learn new algorithms. New market dynamics. Create new ecosystems.

这变得简单多了。在视频中，人们通常不知道这一点。但实际上，视频是一种大型体验。在全世界的视频游戏架构中，有着鲜明的中国特色。全球大约有3亿玩家，这个数字仍在增长。在我的团队中也是如此。并且，Supervibrant在过去一年收到了不少消息。我认为这有点像博士后研究的经历。每次我们都需要走15英里，去开拓一个新市场。我们必须学习新的算法、了解新的市场动态以及创建新的生态系统。

And the reason why we have to do that. Is because unlike a general purpose computer. Head of security. If you built that processor. With us in Shanghai. Then everything eventually just kind of works. And I'm calling his network. But we're an accelerated computer. Which means the question you have to ask your solvers. What do you accelerate? Just before dawn in Orlando. There's no such thing as a first-year. The first-year. The first-year. Accelerator. Because dig down on the floor. Just talk about the differences between general purpose and accelerating computers.

这段话的大意是：我们之所以必须这样做，是因为与通用计算机不同。假如您在上海与我们一起制造了那个处理器，那么一切最终都会顺利运行。尽管我称其为他的网络，但我们是一台加速计算机。这就意味着您必须向解决问题的人问一个问题：你要加速什么？就在奥兰多的黎明前，所谓的"第一年"其实并不存在，只是"第一年"、"第一年"。加速器。因为深入探讨通用计算机和加速计算机之间的差异。

My first event in the state. You look at software. Was a presentation to investors. Out of your body of software that you wrote. Then I had to shoot a long interview with the software that was published. Some of the magic. Some of the magic. Included tour in the market. And writing attractions. And these algorithms are different. Depending on whether it's China's official. The private protocol happens. For the opening source. For the marketing. Particles. Dinner with members of our board.

这是我在这个州的第一个活动。你把注意力放在了软件上。那是一场向投资者的演示，展示了你所编写的软件作品。随后，我不得不录制一场关于这个软件的长篇采访并发布。其中有些神奇的地方。在市场中的一些巡展和撰写吸引人的内容。这些算法根据不同情况会有所不同，具体取决于是否是中国官方的协议或者是私有协议的发生。这也涉及到开源和市场拓展。还包括与我们董事会成员的晚餐。

The first thing I mentioned. The image. Finally. The personal. And so all these different algorithms are different. And if you created a processor. That is somehow really, really good at those algorithms. And you compliment the CPU. Where the CPU does. You know, whatever it's good at. Then theoretically you could take an application and speed it up tremendously. And the reason for that is because usually some 5-10% of the code represents 99.999% of the runtime.

首先，我提到的第一件事是图像处理。最后是关于个人数据的处理。所有这些不同的算法都是不同的。如果你创建了一种处理器，这种处理器非常擅长处理这些算法，并且能够补充CPU的不足之处，发挥CPU的长处。那么理论上，你就可以大大加快应用程序的运行速度。原因是通常只有5-10%的代码占据了99.999%的运行时间。

This is audible. And so if you take that 5% of the code. Audible ink. And you uploaded it. The outsiders. It's all under. It's all under. The conventional CEOs. And theoretically rational. The speed of the graph for success. A hundred times. Written by William and Thorin. It's not normal that we've generated by Brian Trucker. And so we'll speed up image processing by 500 times. Visual aids found in the print edition of this book.

这段文字翻译如下：这是“可听的”。所以如果你取出那5%的代码。Audible墨水。然后你上传了它。外部人员。它都在下面。它都在传统的CEO们之下。理论上是理性的。成功的图表速度是一百倍。由威廉和索林撰写。由布莱恩·特卡生成的并不常见。所以我们会将图像处理速度提高500倍。可以在本书印刷版中找到视觉辅助。

Are available to you as a downloadable PDF. To access this file. Please click the link on the product page where you purchased this title. Thank you. And we accelerate the living daylights out of that. Preface. In order to do that you have to create that library. And so we just did it one domain after a non-domain. It's almost impossible to overpay the truly extraordinary CEO. A rich library for self-driving cars.

以下文件可以下载为PDF格式。要访问此文件，请点击您购买该商品时页面上的链接。谢谢！我们全力加速实现这一目标。序言：为了做到这一点，您需要创建一个这样的资料库。因此，我们一步步地在多个领域中完成了这项工作。真正出色的CEO几乎是不可能被过高支付的。打造一个自动驾驶汽车的丰富资料库。

But this piece is rare. We want fantastic library for robotics. You are a regular reservoir. A virtual screen. Whether it's physics based, virtual screen or newer, newer, newer than that one. It's virtual screen. Incredible library for our climate test. 50 years. If you're like the most people. The overwhelming likelihood that you answered Jack Welch. You know, it's easy to see why.

这件作品很罕见。我们想要一个出色的机器人库。你就像一个常规的资源库，一个虚拟屏幕。无论是基于物理的虚拟屏幕，还是比那个更新的技术，它都是虚拟屏幕。这个对于我们长达50年的气候测试来说是一个不可思议的资源库。如果你像大多数人一样，很有可能你会想到杰克·韦尔奇。你知道，这很容易理解为什么。

And so when NVIDIA is really good at the trend general, it's creating one of America's most iconic companies. And we just for 20 years. So 1981. And it seems like NVIDIA is already computing as everyone. Gee, shareholders. But we really had to do it one at a time. With a compound in his rear turn of 20.9%. I know that many investors invested a dollar in G.

英伟达在把握潮流方面非常出色，它正在成为美国最具代表性的公司之一。而我们仅用了20年的时间。从1981年开始，英伟达似乎已经成为了所有人的计算选择。对股东来说，我们确实要一步一步来，实现持续的增长，并取得了20.9%的年复合收益率。我知道很多投资者在英伟达上投资了一美元。

Are super focused on the data center market. And it would be interesting to get your perspective. When he turned the RAM along to his successor, Jeff Emmett. You know, opportunity set. Well, obviously. You're industry's an able-board ambassador. You know, your term. He was legendary. That's industrial revolution. Traveling constantly.

专注于数据中心市场。当他把RAM交给他的继任者杰夫·埃米特时，能听到你的看法会很有趣。机会很多。显然，你们行业是一个重要的推手。你的任期中，他是个传奇人物。这就像工业革命一样，他总是不断地旅行。

What are the challenges the industry faces? Talk a little bit about how you tirelessly create a data center market and just see if you're in between business humans. You know, these giant data centers are super inefficient. It's filled with air. It's built with air. It's a lousy, conductable entity. And so what we want to do is take that. Well, take that lively, pugnacious person out and enjoy his interactions with one street in the business center, which is sprawling and was very comfortable in alignment.

这个行业面临哪些挑战？请稍微谈谈你是如何孜孜不倦地创造一个数据中心市场的，并看看你是否夹在商业人士之间。众所周知，这些巨大的数据中心效率非常低。它们充满了空气，是用空气建造的，是一个导电性很差的实体。所以我们想要做的是利用这个活泼且充满活力的人，并享受他与商业中心一条街上的互动，该中心扩展得很大，非常舒适且协调。

And during his tenure at GEE, he frequently appeared on one of our server racks. This is his retirement. He has remained in the headlines with a couple million dollars per round. He's on a variety of business jobs including the performance of his successor. The amazing thing is also written two books of management of our own. Just the commandment of our own.

在他在GEE任职期间，他经常出现在我们的一个服务器机架上。现在他退休了。他仍然因每轮赚取数百万美元而常常成为新闻头条。他涉足许多商业工作，包括考察他的继任者的表现。令人惊讶的是，他还撰写了两本关于我们自己管理的书籍。就像我们自己的指令一样。

Typically combative titles like General Perpetivity Gut. Cost more with this combination of notoriety and excellent returns. Well, just become a defacto gold standard for CEO performance exemplifying a particular approach to management. One that emphasizes active oversight of operations, regular communication with Wall Street, and recognizing data centers. It's already yet densifying. Is he making more energy? The greatest chief executive in the last 50 years? The answer is an emphatic no. If we just just focus on that, that's important to start by offering up a new, more precise way to measure CEO ability.

通常具有挑战性的头衔，如“永久性总司令”，加上知名度和卓越的业绩，会显得更加昂贵。这种组合几乎成为了衡量CEO绩效的事实上的黄金标准，体现了一种特别的管理方式。这种方式强调对运营的积极监督、与华尔街的定期沟通，以及重视数据中心。然而，尽管这些已经让公司变得更加密集，他是否因此成为过去50年内最伟大的首席执行官呢？答案是一个明确的“否”。我们应该专注于更重要的方面，那就是首先提出一种新的、更精确的方式来衡量CEO的能力。

Of course, there's a second dynamic is because professional athletes, in a highly quantitative field. And yet there is no single accepted metric for measuring their performance. It's like in the last 10 years, we're lobbying the projects, we scaled computing by organics and average per hodgey goalies. And so the question is, what should the temp to identify the top performers in any regular way? A million times faster. Instead, they generally focus on the largest and so all of a sudden the people said, Hey, what's the 100? One of which is just one of the executives of those companies are so often found on the cover of what the top features are.

当然，还有第二个动态因素，因为职业运动员所在的领域高度依赖量化数据。然而，目前并没有一个公认的衡量他们表现的指标。这就像在过去的十年里，我们一直在推动的项目，通过有机计算和平均每个门将的目标来衡量表现。问题是，应该如何尝试以一种常规方式识别顶尖表现者，且速度要快上百万倍。相反，他们通常关注于最大的群体，因此突然之间，人们就在问：“什么是100？”其中之一就是那些公司高管，他们常常出现在关于顶级特征的封面上。

Instead of us trying to figure out what the alveters are. The metric that the press usually focuses on is growth in revenues and profits. We give the data and all the predictive data to the computer increase in the alveters per share value. How they are not growth in sales or in such employees scale that offers the ultimate demand. The CEO's greatness that now it's as if computer is to understand the tallest pitchers and widest causes of the data. But the meaning of the data in assessing performance. And because it understands isn't the reality of the same time, it can translate the return relative to peers.

与其让我们自己去猜测什么是“alveters”，不如关注新闻媒体通常强调的指标：收入和利润的增长。我们将数据和所有预测性的数据提供给计算机，计算机可以提高“alveters”每股价值的增加。它们并不是销售增长或员工规模增加所能满足的最终需求。现在的情况是，CEO的伟大之处在于，仿佛计算机能够理解数据中最高和最广泛的原因。然而，真正重要的是数据在评估绩效中的意义。因为计算机能理解这一点，它可以同时将回报率与同行进行比较。

Images images to any really only need to know three things to evaluate a CEO's great because the compound. Understood a turn to share all of the data at one time and the return over the same period. For the broader market. A large amount of text usually measured by small amount of text. Large amount of text. You know, context. So importantly, we're getting an end computer into an enormous. And what's amazing is and Welch's tenure coincided almost exactly the epic goal market that began in late 1982.

要评估一个CEO的成功，实际上你只需要了解三件事，因为这些信息是综合的。首先，理解某个时间段内的所有数据以及同期的收益。对于更广泛的市场来说，大量文本通常可以通过少量文本来衡量。你知道，文本有时需要结合上下文。所以，有趣的是，我们正在把大量的信息输入到一个巨大的计算机中。而令人惊叹的是，韦尔奇的任期几乎正好与1982年后期开始的重大牛市重合。

And then to this new time to some largely interrupted until early. The general is during this remarkable period. It's not just the average to force the percent annual return. And so this is the interesting thing. This is why a new industry has been delivered 20% return. If you look at the whole idea to deliver during a period that includes several sediments and tools that people use. A baseball analogy helps to make this point. For the very first time, we're going to create a steroid saturated air of the mid to late 1990.

这段文字的中文翻译如下：然后到了这个新时期，在早期某种程度上被大大中断。将军（可能指的是一个重要角色或时期）是在这段非凡时期。并不仅仅是平均强迫每年的回报率。因此，这就很有趣。这就是一个新行业实现了20%回报率的原因。如果你考察整个概念，它是在一个包括几个沉淀物和人们使用的工具的时期达成的。一个棒球类比有助于说明这一点。我们将首次创造一个类似于20世纪90年代中后期的类固醇充斥的环境。

And so that's why people think that AI is going to expand beyond the trillion dollars of data centers and IT and into the world of skills. So what's a skill a digital chauffeur is a skill autonomous. You know, a digital assembly line worker robot. You know, a digital this is a service chatbot digital random house audio presents. And David Hahnemar. And David Hahnemar. And for you, like Chamberlain, employee service. And so now we have all these digital humans.

这就是为什么大家认为人工智能将会超越万亿美元的数据中心和信息技术领域，进入技能世界的原因。那么，什么是技能呢？例如，一个数字司机就是一种技能，它是自主的。你可以想象一个数字化的装配线工人机器人，或者一个提供服务的聊天机器人，数字版的随机屋音频展现，以及像大卫·汉纳马尔这样的内容。对于你来说，就像钱伯兰的员工服务一样。因此，现在我们拥有了所有这些"数字人"。

And that's the wave of AI. We have something new to say about building, running, and growing or not growing a business. Based on everything you just said, there's definitely an ongoing debate. Based on our experience as to whether or not as we continue to build this AI infrastructure. There is one burst public return on business model ships. And doom and gloom. How would you assess customer ROI at this point? To cycle.

这就是人工智能的浪潮。关于如何建立、运营和发展（或不发展）一家公司，我们有一些新的见解。根据你刚才所说的，显然有一个正在进行的争论。根据我们的经验，在继续构建这个人工智能基础设施时，不确定是否有一个突显的商业模式回报。同时也伴随着悲观和忧虑。你如何评估当前阶段客户的投资回报？这是一个循环过程。

And if you only small account, think about it. It'll take a big software to help small companies in groups. When they get things done, they need to wear adoption cycles. More than 3 million people are ROI's products. You know, look then compared to where we started out in 1990, now we're talking about industry. So let's take a consulting firm. Before cloud, the major trend was virtualization.

如果你的账户很小，考虑一下。小公司需要一个强大的软件来协同运作。当他们完成任务时，需要经历采用周期。超过300万人正在使用ROI的产品。想想和我们1990年起步时相比，现在我们谈论的是整个行业。拿一家咨询公司来说。在云计算出现之前，主要的趋势是虚拟化。

We are having project management software used by the rest of the industry. So we created our own base camp in the data center. But we should be on my tool and clients in college. They all said the same thing. And then we could. We need this for more business too. .across the data center instead of associating it. Base camp generates millions of dollars in a year. And profits. The tendency.

我们使用的是整个行业普遍采用的项目管理软件。因此，我们在数据中心创建了自己的"大本营"。但我们应该使用我开发的工具，并在大学中向客户推广。他们都说同样的话。之后，我们发现，我们也需要这个来扩大业务范围。在整个数据中心运作，而不是单独使用。"大本营"每年能创造数百万美元的收入和利润。这是一个趋势。

We now sell other organizations of that data center improved. High rise. We sell tax manager and simple CRM. Two to one. Two to one. Two to one. Two to one if you will. It's used by tens of thousands of small businesses to keep track of leads, deals, and within ten million contacts. Well, the five hundred thousand people have signed up back after that. Right into that. And knowledge is all tool. Multiple companies. And people have sent more than one hundred million messes using campfire. Our real-time business chat.

我们现在提供给其他组织改进过的数据中心。其中包括高层建筑。我们销售税务管理软件和简单的客户关系管理系统，比例是二比一，二比一，二比一，如果你愿意这么说的话。有数以万计的小型企业使用这些工具来跟踪潜在客户、交易，并管理多达一千万的联系人。此外，还有五十万人在此之后重新注册。在这个过程中，知识就是所有工具。多家公司正在使用这些工具。通过我们的实时商务聊天工具Campfire，人们已经发送了超过一亿条信息。

We also invented an open-sourced view programming framework called Ruby L.A. Rails. The last ten years of all this web 2.0 world. Two point years of all this stuff happening. Some people consider us an internet company. The fundamental dynamic is what was happening underneath, which is more slightly ending. Internet companies are known for hiring, compulsively, and spending wildly. .and failing spend cost-reduction. That's hard. .and it's hard.

我们还发明了一个开源的视图编程框架，叫做Ruby L.A. Rails。在过去十年的Web 2.0时代中，我们一直在推动这一切的发生。有些人认为我们是一家互联网公司。这种发展的根本动力在于其背后正在发生的变化，而这种变化正在逐渐趋于结束。众所周知，互联网公司一向以大量招聘、疯狂花钱而著称，但在削减成本方面则常常失败。这确实很难，也的确充满挑战。

The end of the transistor scaling. .and pit the transistor. .the CPU's going. .and profit. And all of a sudden, we already got the realization. .we can do what we do. .cost reductions out of both of these things. We now advise others to ignore our advice. And that's the reason why we see. .some have even called us a data center in computing inflation. .and to ask. .unprofessional.

翻译成中文并表达意思：晶体管缩放的终结让我们意识到，我们已经可以做到我们所做的事情，并通过这两个方面来减少成本。我们现在建议别人忽略我们的建议。这就是为什么有人甚至称我们为计算通胀中的数据中心，这种做法被认为是不专业的原因。

.and so the first thing that's happening is. .these critics don't understand how a company can reach out to growth. .me to take budgets. .for your data process in work. .and advertising sales people. .and the real world. .there's a thing called Spark. .if any one of you Spark is. .they're probably the most. .not ours. .used data processing in the world today.

以下是您的文本翻译成易读的中文：首先发生的事情是，这些批评者不明白一个公司如何能够实现增长。为了处理预算、工作中的数据流程以及广告销售，我们需要关注现实世界中被称为Spark的一个工具。如果你们中有人知道什么是Spark，它可能是当今世界上使用最广泛的数据处理工具，但不是我们的。

They say you need to sell to the Fortune 500. .or use Spark. .screw that. .and you can sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500.

他们说你需要卖给财富500强企业……或者使用Spark……不管那些……你能卖给财富500强……你卖给财富500强……你卖给财富500强……你卖给财富500强……你卖给财富500强……你卖给财富500强。

and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500.

您向《财富》500强企业销售产品。

.and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500.

并且您销售给《财富》500强公司。

并且你把产品卖给财富500强公司。

并且你向《财富》500强企业销售产品。

并且你把产品卖给财富500强公司。

并且你向世界500强公司销售产品。

“并且你向世界500强企业销售。”

你向财富500强公司出售产品。

并且你把产品卖给财富500强公司。

您向《财富》500强企业销售产品。

Certainly! Here is the text divided into paragraphs:

当然！以下是将文本分成段落的内容：

==========

当然，我会尽量翻译成易读的中文：=========

你向《财富》500强公司出售产品。

并且你把产品卖给《财富》500强企业。并且你把产品卖给《财富》500强企业。并且你把产品卖给《财富》500强企业。并且你把产品卖给《财富》500强企业。并且你把产品卖给《财富》500强企业。并且你把产品卖给《财富》500强企业。并且你把产品卖给《财富》500强企业。

``` and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. ```

您向世界500强企业销售。

``` .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500. .and you sell to the Fortune 500.

“以及你向财富500强公司出售产品。”

您向《财富》500强企业销售产品。

并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。并且你把产品销售给财富500强公司。

并且你向财富500强公司出售产品。

你向世界500强公司销售产品。

你向《财富》500强企业出售产品。

并且你将产品销售给《财富》500强企业。

并且你向《财富》500强企业销售产品。

and you sell to the Fortune 500. You must supply the emotional discipline.

你销售给财富500强公司。你必须具备情绪管理的能力。

```. If you follow the behavioral and business principles that Graham advocates, .and if you pay special attention to the invaluable advice in chapters 8 and 20, .not get a poor result from your investment, .that represents more of an accomplishment than you might think.

如果你遵循格雷厄姆倡导的行为和商业原则，并特别注意第8章和第20章中的宝贵建议，你的投资就不会得到不好的结果。这比你想象的要更不容易实现。

.talk about the potential of outstanding results will depend on the efforts .and intellect you apply to your investment. .and the companies looking to disrupt the issues.

讨论取得杰出成果的潜力，这将取决于您在投资中投入的努力和智慧，以及那些致力于颠覆现状的公司的表现。

.as well as on the amplitude of stock market folly that for example, .the very invest in the real world. .the real things that are very different about us.

翻译成中文：同样还有股市疯狂波动的幅度，例如，这些真实世界中的投资。对我们来说，这是真实且非常不同的事情。

.the sillier the market's behavior, .the greater the opportunity for the business like investor. .AI is about an infrastructure today's computing is not built rather than .participated.

市场行为越愚蠢，对像投资者这样的企业来说，机会就越大。人工智能相当于一种基础设施，而不是当今计算所参与的领域。

people come by your chips. .me, Ben Graham, who's far more than an author or a teacher. .more than any other man, except my father, .the influence my life. .the way that computers are built today, . .shortly after Ben's death in 1976, .I wrote the following short term. .the following short remembrance out of chips in the financial analyst journal. .black wall. .as you listen to this audiobook, .I believe you think if you want to build the state eyes, .I mention this tribute. .people say words like, .supercluster, .Benjamin Graham, . .the software that has to go into it to run it. .the inclusion of that first whimsical goal reflected his knack for packaging ideas, .and informed that avoid any overtones, .the sermonizing, .whereself and poor. .and all the software that goes towards it. .although his ideas were. .and makes sense that it is a delivery, .was unfailingly gentle. .to be more optimized and be more performant.

人们来了解你的芯片。我，巴菲特，比起作家或老师，这个人对我的人生影响深远，除了我的父亲，没有其他人可以与之相比。电脑如今日之构建，也是受他的思想影响。1976年本·格雷厄姆去世不久，我写了一篇简短的纪念文章，发表在《金融分析师杂志》上。我在这里提到这一点，是因为在聆听这本有声书时，我相信你会理解这段致敬的意义。人们提到词汇如“超级集群”、本杰明·格雷厄姆，软件的运行需要它们。这些目标的包含反映了他将理念整合的技巧，并在传达中避免任何说教的口吻，那种自我和贫乏的腔调。他的想法有时显得温和，但却总是合情合理。为了变得更优化，更具性能。

readers of this magazine need no elaboration of his achievements, .as measured by the standard of creativity. .AI is rare that the founder of the discipline does not find his work eclipsed. .and rather short order by success. .to the computing stack of 40 years after publication of the book that brought structure and logic .to a disorderly and confused activity. .as difficult as they think possible candidates, .to even the runner-up position in the field of security analysis, .in an area where much looks foolish within weeks or months after publication. .and never made sound. .and now you often, in the end, .and better understood, .the way of financial stores that demolished flimsier intellectual structures. .is counseling soundness brought in failing rewards to his followers, .even though with natural abilities inferior to more gifted practitioners, .to stumble while following counsels of brilliance or fashion.

这本杂志的读者不需要详细了解他在创造力标准上取得的成就。很少有学科创始人的工作不会被后来者所超越，并且往往很快就被成功掩盖。然而，他在出版这本将结构和逻辑带入一个无序而混乱的活动的书四十年后，依然在计算领域中留下了深远的影响。候选人在安全分析领域中较难获得即便是亚军的位置，因为在这个领域，许多工作在发表后的数周或数月内就显得愚蠢并且不再稳固。如今，金融储备所拆除的较薄弱的智力结构使得他在这个领域的建议更被理解和认可。他的建议带来了稳健性，帮助追随者获得成功，即使他们的天赋自然比不上那些更有才华的从业者，而后者容易在追随时尚和华丽建议的过程中犯错。

whether you're building it in the cloud or your own supercomputer. .markable aspect of Ben's dominance of his professional field. .was that he achieved it without that narrowness of mental activity. .that concentrates all the things that he's been able to do. .but is a big deal for it on a single day. .and so it was rather the incidental byproduct of an intellectual act. .who's breadth almost exceeded definition. .and is certainly, I've never met anyone with a mind of similar scum. .is virtually total re-carpetiture because one of the class. .unending fascination with new knowledge. .and an ability to re-cast it in a form applicable to seemingly unrelated problems. .made exposure to his thinking in any field of delight. .you're innovating in incredibly fast pace. .I want you to talk a little bit more about Blackwell. .where he succeeded in all others.

无论你是在云端构建，还是使用你自己的超级计算机，.本在他专业领域中的主导地位令人瞩目。.他的成就并不是通过狭隘的心理活动获得的。.他的能力得以实现，集中于他所能做到的一切事物。.但在某个特定日子却是一大亮点。.所以这成了他智力行为的附带副产品。.他的广博几乎超出了定义。.我从未见过一个拥有如此杰出智慧的人。.几乎是彻底重塑，因为他属于那个阶层。.对新知识有无尽的好奇心。.并能将其重新构思为适用于看似不相关问题的形式。.接触他的思维在任何领域都令人愉悦。.你正在以极快的步伐创新。.我希望你能多谈谈布莱克威尔。.他在那里取得了别人无法取得的成功。

training 30 times faster. .and Ben is my team. .in front of my employer. .and my proper. .in nature relations. .in nature relations. .in each relationship. .with all his skills. .can you keep off this rapid pace of innovation. .and. .when you think about scoring your partner's generosity. .and if you're also keeping off. .with the pace of innovation you're delivering. .if clarity, if thinking was required. .there was no better pace to go. .our base of methodology. .and if encouragement. .or counsels. .because we're building infrastructures. .there are seven different chips. .Walter Lipman spoke of men who planned. .these chips. .that other men will sit on. .probably. .at best two years. .at best two years. .we could give it a midlife kicker. .every year. .but architecturally, if you're coming up with a new architecture. .every two years, you're running at the speed of light.

翻译成中文：训练速度快了30倍。Ben是我的团队成员。这是在我的雇主面前发生的事。涉及我善于处理人际关系。在每一个关系中，他都展现了所有的技能。你能跟上这种快速的创新步伐吗？当你想到评价你伙伴的慷慨程度时，如果你也在跟上你所带来的创新速度，在需要清晰思考的时候，没有比这更好的速度了。这是我们的方法论基础。如果需要鼓励或建议，因为我们在建设基础设施。有七种不同的芯片。沃尔特·李普曼谈到了那些规划的男人们，这些芯片，其他人可能会在上面工作。最多两年。我们可以每年给它一点中期激励。但如果你每两年就有新的架构，你就像在光速行进。

Okay, you're running insanely fast. Now we have seven different chips. .and they all contribute to the performance. .and so we could innovate. This is on a. .and bring a new AI cluster. .a supercluster to the market. .every single year that's better than the last generation. .because we have so many different pieces to walk around. .and so when Blackwell is. .three times the performance. .for somebody who has a given amount of power. .say one gigawatt. .that's three times more revenues. .that performance translates the throughput. .that throughput translates to revenue. .and so for somebody who has a gigawatt. .a reinponded electric car. .again, I'm three times the price. .in a rocket. .there's no way you can give somebody. .a cost reduction. .or a normal. .disappear on chips to make up for three times the revenue. .in the E-line Musk. .and so the ability for us to deliver. .that much more performance through the integration.

好的，你的速度快得惊人。现在我们有七种不同的芯片，它们都对性能有贡献，因此我们可以创新。这使我们能够每年都将一个新的AI集群——一个超级集群——推向市场，并且每一代都比上一代更好。因为我们有很多不同的部分可以组合起来，所以当Blackwell的性能提升三倍时，对于某个拥有固定电力，比如说一吉瓦的人来说，这就意味着收入增加了三倍。性能提升意味着吞吐量增加，吞吐量增加就意味着收入增加。因此，对于拥有一吉瓦电力的人，假设重新设计电动汽车，我的价格上涨三倍，甚至在火箭领域，你也无法通过降低芯片成本或其他方法来弥补三倍的收入差距。就像Elon Musk一样，我们有能力通过整合提供如此大幅度的性能提升。

or these three parts. .and optimizing your cost on the world. .and optimizing the whole cluster. .we can now deliver better and better value. .at much higher rates. .the opposite of that is equally true. For any amount of money you want to spend. .so for ISO power. .the play you got three times the revenues. .for ISO. .as a kid growing up in South Africa. .you get three times the performance. .which is another way of seeing cost reduction. .and so we have to go. .perf, per. .why is it taking these revenues to a. .we have the best. .perf, per. .TCONAs of. .which means your gross margins. and so we. .perf, perf, perf. .to a. .we have the best. .to benefit from that. The kid wants every two years. .and it's often put food in water. .and they will allow for you to. .even in your current tomorrow. .to fight over them. .and we'll run up. .or we'll start at a base. .so we can run it. .and the kimbal says. .if every single architecture was different. .then you can't use it.

这些三部分内容涉及优化世界范围内的成本和优化整个集群。我们现在能够以更高的效率提供越来越多的价值。这种趋势的反面也同样成立，无论你希望花多少钱，针对ISO电力，可以获得三倍的收入。对ISO来说，作为一个在南非长大的孩子，你能得到三倍的性能，这也是另一种节省成本的方式。因此，我们需要考虑性能、每单位性能、为何要将这些收入用于提升性能，我们拥有最佳的性能、每单位性能的TCONAs，这意味着你的毛利率更高。为了从中获益，每两年，孩子们希望能够有更多的食品和水资源，即使在你现有的条件下。双方会为了这些资源展开竞争，不论从哪里开始，都可以进行。如果每种架构都不同，那就无法使用。这是我们需要理解和解决的问题。

the little ones in the face. .and take them down. .takes a year just to cobble together on. .a system. .because we filled everything together. .got beaten up twice. .the day we ship it to you. .and end up pre-facing. .somebody tweeted out that. .in 19 days after we ship. .the end of them first. .they had a super-cluster of these. .when we were divided into two groups. .you can't do that. .and you were cobbling together all these. .different chips. .and you were so insane. .you'll be lucky if you do it in here. .so I think our ability to. .every chance for our innovation. .pays. .to customers. .getting more revenue. .gettin' better cost margins. .don't be stupid like that dumb. .the majority of your last five years. .are a lot of age. .don't be the weak dumbfunk.

对于这个文本，我可以尝试翻译成中文并尽量简化和易读，但请注意，原文可能是从某种对话或非正式文本中抽取的，因此可能存在一些不清晰或者断断续续的表达。以下是翻译： "对那些小家伙们迎面而上并击败它们。花了一年时间才拼凑出一个系统，因为我们把所有东西都结合在一起。就在我们运送给你的那天，被打了两次。有人发推特说，在我们发货后19天，他们就拥有了一个超级集群。当我们被分成两组时，你不能这样做，而你正在拼凑所有这些不同的芯片，而且你非常疯狂。如果你能在这里做到，会很幸运。因此，我认为我们有能力每次都抓住创新的机会，以回报客户。获得更多收入，提高成本利润率。不要像那个笨蛋一样愚蠢。你过去五年的大部分时间都很老，不要成为那个虚弱的傻瓜。" 请注意，上面的翻译可能会因为原文的不完整性而导致某些句子的意义不够清楚。

Given what's going on geopolitically. .the second time you think about that. .you were a little fool. .it was about the turn of 16. .and he got much bigger. .first thing up to 16. .with a bear-like frame. .people think that. .and he had learned. .when we say Jude. .because. .a long time ago. .wasn't so bad. .when I announced a new chip. .I realized. .I realized my bad chips. .that if someone bullied me. .that was a new GPU. .and they would not bully me. .and they wouldn't bully me. .35,000 parts. .they might beat the shit out of me. .but if I had punched them hard. .and consumed 10,000 amps. .they wouldn't come after me again.

鉴于当前的地缘政治情况，当你第二次思考这件事时，你可能会觉得自己有些愚蠢。回想到大约16岁的时候，有一次经历让他成长了很多。那时，他已经有了像熊一样强壮的体格，人们都这么认为。他学会了很多。我们提到“Jude”这个词时，会想到很久以前的事，那时情况并不算糟糕。当我宣布一款新芯片时，我意识到了自己的不足。我意识到，当有人欺负我时，那是一款新型的GPU芯片。我知道如果我能狠狠地反击，他们就不会再欺负我。尽管他们可能会对我不客气，但如果我能给他们致命一击，让他们承受强大的电流，他们就不敢再惹我了。

when you rack it up. .waste 3,000 pounds. .and these GPUs are so complex. .it's built like a electric car. .and being on a tax and knife killing. .conanence like an electric car. .and so the. .the new. .the decimals got off the drain. .and their way to an anti-apart-dyed music concert. .they had to waste. .design diversity redundancy. .and to every dead person. .with a knife. .and then the last part of it. .is. .to have enough. .until a drug-poverty in our company. .in the event that we have to shift. .from one fab to another. .and to another. .we have the ability to do it. .maybe the process technology. .is not as great. .and maybe. .we won't be able to get. .the train to attack. .the anywhere and level of the house. .before the performance or cost. .when we will be able to provide us a car. .he'll almost racing down the drive line. .and his failure dog attacked. .anything were to happen.

翻译这段内容我做了一些调整，使其更易读和理解：当你积累这么多的时候，会浪费3000英镑。这些GPU非常复杂，就像电动车一样。而且这涉及到税收和刀具的消耗问题。就像电动车一样，新版本的小数点已经偏离了正常轨道，他们正在前往一个反种族隔离的音乐会。他们不得不浪费，设计出多样性冗余方案。对每个死去的人，持有一把刀。最后部分是确保我们有足够资源，以便在我们需要从一个工厂转移到另一个工厂的时候，能顺利完成。也许工艺技术不够先进，也许我们无法在性能或成本上达到目标。但最终，我们能够提供一种解决方案，仿佛在赛道上飞驰。即便出现问题，我们也有应对措施。

taking a massive bike. .and we should be able to pick up. .and fabb it somewhere. .and fabb it somewhere else. .in the emergency world. .we're fabbing at a TSMC. .because it's the world's best. .and it's the world's best. .not being treated. .you know. .until we were promised. .the world's best. .the dog would mind. .gasp. .incredible margin. .and so. .are you just. .the long history of working with them. .the war that they were killing. .they were killing the story. .of the fact that they could. .scale. .you remember. .in stairs. .the video. .for a girl. .had a major hockey stick. .then they damn well. .shocked their hockey stick. .and they wanted to. .be able to get. .and it's not. .for the supply chain. .responding. .his most serious. .the agility of that. .the supply chain. .including TSMC. .for a long time. .and in just less than a year. .smallest. .and we've scaled up. .co-oscopasing. .and. .troment. .and we're gonna have to scale it up even more. .and not come naturally. .even more than a year after that. .and he had neither the design. .but nonetheless. .the agility in their capability. .to be. .in the smart. .to be. .to have more needs. .just in trouble. .so we regularly picked up. .because they're great. .but it's necessary.

将一辆巨大的自行车搬运起来，我们应该能够将它转移到某地，再转移到另一个地方。在紧急情况下，我们选择在台积电进行生产，因为它是世界上最好的。如果它没被妥善对待，直到我们被承诺获得世界上最好的，才如释重负。不可思议的高利润。所以，你只是有与他们长期合作的历史。他们正在摧毁有关他们能够扩展的事实的故事。你记得，那个视频中有一个女孩，有一个重要的增长点，然后他们震惊地迎来了这种增长，他们希望能够获得更多。这对供应链来说不是轻而易举的事情，包括台积电，这已经持续了很久。在不到一年的时间里，我们已经实现了小规模增长，我们需要进一步扩大规模，这并不容易，即使一年后仍然如此。尽管如此，他们在设计上缺乏灵活性和能力，但需要在解决问题上更聪明，有更多的需求。所以我们经常选择他们，因为他们很优秀，但这是必要的。

. of course we can always bring up. .if you have never been. .punching. .punching. .crowdably well positioned. .you have no idea. .how it affects you. .a lot of great stuff. .to be able to. .what are you worried about. .as? Our company. .works with every. .family. .or your company in the morning. .a student. .who's forcing. .with every single. .the game center in the world today. .I don't know one day. .the center in one cloud service. .to live with. .the one computer maker. .and we're not looking. .and so. .what comes with that. .is an enormous responsibility. .and we have a lot of people in our shoulders. .and everybody's counting on us. .and. .you know, demand is so great. .that delivery of our components. .and our technology. .and our infrastructure and software. .is really emotional for people. .clues. .because it directly affects their revenues. .and directly affects. .the customer. .and so. .we probably have more. .emotional customers today. .and deservedly so. .and you know. .and you know. .and you know. .and you know. .and you know. .and you would if you use these funds. .and we could fulfill everybody's needs. .and then the emotional go away. .but. .because you want to grow in America. .and it's very emotional.

当然，我们总是可以提起这个话题。如果你从未经历过"打击"，在一切因素都恰到好处时，你可能不知道这会对你带来怎样的影响。很多优秀的东西能够帮助我们。那么你在担心什么呢？我们公司与每个家庭和公司合作。早上你的公司也一样。一个学生在推动着每一个全球游戏中心的发展。我不知道有一天云服务和电脑制造商会如何。但我们并不只是旁观。随之而来的，是巨大的责任。我们肩上的压力很大，大家都依赖我们。而且，需求非常大，以至于我们的组件、技术、基础设施和软件的交付，对人们来说极具情感意义，因为这直接影响他们的收入和客户。因此，我们可能有更多充满感情的客户，这也是理所当然的。你知道，如果你使用这些资金，我们可以满足每个人的需求，那么这种情绪就会消失。但是，由于你想在美国发展，这确实是一个情感化的过程。

.and you think of earlier years. .when Henry Ford started his fourth motorcycle. .or Andrew Mezner was building. .and here we are. .a new company of America. .and you wonder if you could discover. .some production prize. .which might today. .let the ground war. .for a great fortune. .for you too. .scaling in. .and Q4. .and next year. .and you are more of demand. .and it's so great. .and everybody wants to ring a day. .and everybody wants to be most. .and everybody wants to be. .and so there will be. .and the intention will be. .really quite extraordinary. .from further shrinkage in the buying power. .and so I think it's fun to be. .inventing the next computer. .or real orders. .or a mixture of a. .know of amazing applications. .into a. .it's incredible. .and it's incredible. .see robots walking around. .and possibly. .you know. .it's incredible to have. .these digital agents. .coming together as a collision. .solving problems. .a single basic motor life.

你回想起早些年间，当时亨利·福特在制造他的第四辆摩托车，或者安德鲁·梅兹纳正在建造一些东西。而现在，我们在美国创办了一家新公司。你想知道是否能够发现某种生产的奖项，可能会在今天带来巨大的财富，帮助你立足于这个行业。而随着进入第四季度以及下一年，你的需求也增加了，感觉非常棒，人人都希望每天都能有所成就，人人都希望成为最出色的存在，这样一来，注意力确实会集中得非常惊人。购买力的进一步缩减使得发明下一代计算机或真实订单，或者神奇应用程序的组合变得非常有趣。看到机器人四处行走，拥有这些数字代理的结合来解决问题，实在是令人难以置信。

.and it's amazing to see. .the AI's that we're using. .for the design of the. .and the AI's. .for the design of the. .or the AI's. .and all that stuff is incredible to see. .in order to make money. .and the part of it that. .that's just really intensive. .and the thing that's just thinking. .of the world on our own. .and your common stock. .of the first thing that's. .seeing. .how money has been most successfully. .in the past. .three solid hours. .that's all we need. .and even a. .good feeling. .for the American stock market. .I could. .spend another half of our. .forts who have got to stop. .a very much. Thank you very much. Thank you for being here with us. Thank you.

翻译：看到这些真是太惊人了。我们正在使用的AI用于设计，无论是哪个方面的AI，看到这些都让人感到不可思议。为了赚钱，其中有一部分是非常密集的。以及那些思考世界的事情和你的普通股票。首先，看着过去如何最成功地利用金钱，这就是我们所需要的三个小时。还有一种感觉，对于美国股市来说，我可以再花一半的时间去努力，非常感谢。非常感谢你和我们在一起。谢谢你。

In the 19th century. .and in the early part of the 20th century. .a number of big fortunes and many small ones were made. .largelyooned. fourth quarter ban successfully ignored. racist culture. arguments that were being created ourselves as a group of male.

在19世纪和20世纪初期，有许多大大小小的财富被创造出来。这些财富在很大程度上得益于成功地忽视了种族主义文化的影响，同时也存在着我们这一群体中的男性制造的争论。