U.S. markets closed slightly higher on hopes that the possibility of a rate cut was rising, with Dow surging and Nasdaq falling on Nvidia falling. The possibility of a rate cut next month has soared to 80%. The reason is that Kevin Hassett, the current White House chief of the national economy, is the most likely candidate for the next Fed chairman to nominate before Christmas, and the possibility of a rate cut next month has been reflected in the market as a favorable factor. The movie “Home Alone” is another movie on Christmas. And the main character is Kevin…. The next Fed chairman is also announcing it at Christmas. Kevin…. Kevin….

Among the two necessary and sufficient conditions for the market’s rise, a rate cut is becoming more likely. Whether we can continue the other axis of AI is that attention is now shifting from Nvidia’s dominance to competition among other chipmakers. It’s Google, which I claimed to be the dark horse, that put the brakes on Nvidia’s dominance. It turned the business around last week when it made Gemini 3.0. With its market capitalization approaching $4 trillion, it has pushed back Microsoft to No. 3 after Apple. It is expected to become No. 1 soon. Now, what you need to know is TPU, ASIC, following the term GPU. Being a liberal arts graduate, I’m getting old, but I have a headache.

With the announcement of Gemini 3.0, Google is calling its own chip a Tensor Processing Unit (TPU). It is said to be a kind of application-specific intergrated circuit (ASIC) that is hot these days. If you translate it into Korean, it’s like an application intensive circuit in a specific field. I don’t know in detail, but when I asked Gemini, they said that it was a semiconductor chip designed to perform a single specific task quickly and efficiently. If the CPU is a master of everything, you can understand that ASIC is specialized in doing one thing like Bitcoin mining. Simply put, it would be quick to understand that ASIC only beats one person.

The advantages of ASIC are that processing speed is hundreds to thousands of times faster than CPUs and consumes a lot less power. This was a problem because it took a lot of time and money to develop, but if Google’s ASIC, TPU, was commercialized like an Nvidia GPU, the impact would be huge. The so-called second-order effect. If ASIC dominates the market, chips will become cheaper, replacing Nvidia chips that are worth tens of millions of won. In the past, Sony developed the world’s first beta method in the video player market, but it did not share technology, so it jointly developed a VHS method in Europe and released it for free so that anyone can use it. Just as the video player market is unified by VHS and Sony has collapsed, ASIC’s wavelength could grow. But in my generation, ASIC is a Japanese sneaker…

I don’t know yet if the competition between GPU and ASIC will cause buyers to change their target of purchase with ASIC, which is cheaper than chips, but the impact will be significant given that Nvidia fell 2.7% and AMD fell 4% due to the news that META could use Google chips. And I think if the chip becomes cheaper, anyone can buy it, so the overall size of the so-called Park Da-mae market can expand further, so I think we can create a new boom in AI. Because I keep getting new things, I have to study, so I think prevention of dementia is certain, but there are so many things I need to know. I have to keep expanding my memory in my head as well.

You write too long, so you can buy and sell on your own… and invest in winning no matter what…