Jump to content

Chip War - Chris Miller


Saluki

Recommended Posts

Someone mentioned this book on the China post and I'm listening to the audiobook.  About 20 years ago when I first started investing, I did what Peter Lynch said not to do and invested in something that I didn't know much about, a company called Photronics that made photomasks, which are like jigs for lasers that are used to make computer chips.  I didn't lose much but I gained an understanding that tech businesses in general and computer chips in particular are terrible for investors who aren't in the business.  When Buffett invested a bunch in Taiwan Semiconductor, I reacted the same way that I do when someone mentions Micron or Skyworks.  I listen politely and ignore it. 

 

However, this is a really understandable book and it starts at the very beginning: vacuum tubes to transistors, to integrated circuits, to computer chips and you will meet all the players along the way and find out how they got distributed onto the chessboard in the four corners of the earth. Even if you are not interested in these companies as an investment, they are important to know about because of their place in the world economy and politically.  Computer chips are the oil that really runs the world economy.  

 

And weirdly enough one of the early backers of Micron was Simplot, the potato king.  When memory chips became nothing more than commodity, then whether it's a potato chip or a computer chip, you can make money on it by doing commodity type business with it. 

Link to comment
Share on other sites

Just listening to it now as well. I'm quite enjoying it as well. There were a few of interesting author observations that I took away that were useful.

 

1) In fast moving industries were products iterate and change rapidly, there needs to be a critical mass of secular trends to tilt the probability towards success, these include (but not comprehensive):

 

- free market competition with a profit motive and entrepreneurial spirit

- general low cost of capital in the manufacturing region (hint of this: high savings rates, business friendly environment, governmental backing, skilled workforce and quality educational funnel)

- ability to take advantage of Wright's law to gain tacit process knowledge in production

- friction-free flow of IP 

- recognition of success to maintain the talent funnel

- large capital providers (ie governments (DARPA, Japanese govt, subsidies, conglomerates (like Samsung), venture capital)

 

2) The interplay between geopolitical needs (eg defense) and application of these technologies to larger consumer markets. Military budgets stimulate products but government bureaucracy stifles innovation, where as consumer markets stimulate innovation but have difficult funding cutting edge technology.

 

3) When nationalistic sentiment rises, it seems to forecast an economic downfall.

 

4) Beware of Clay Christenson's low cost producer with low margin customers in these industries in the right setting, can scale up and innovate quickly to be a competitive threat.

 

Also thanks for the recommendation on Bionomics. I thoroughly enjoyed it. Picked up Shorting the Grid as well. Looking forward to getting through it.

 

Link to comment
Share on other sites

  • 2 weeks later...
On 5/9/2024 at 1:18 PM, jfan said:

Just listening to it now as well. I'm quite enjoying it as well. There were a few of interesting author observations that I took away that were useful.

 

1) In fast moving industries were products iterate and change rapidly, there needs to be a critical mass of secular trends to tilt the probability towards success, these include (but not comprehensive):

 

- free market competition with a profit motive and entrepreneurial spirit

- general low cost of capital in the manufacturing region (hint of this: high savings rates, business friendly environment, governmental backing, skilled workforce and quality educational funnel)

- ability to take advantage of Wright's law to gain tacit process knowledge in production

- friction-free flow of IP 

- recognition of success to maintain the talent funnel

- large capital providers (ie governments (DARPA, Japanese govt, subsidies, conglomerates (like Samsung), venture capital)

 

2) The interplay between geopolitical needs (eg defense) and application of these technologies to larger consumer markets. Military budgets stimulate products but government bureaucracy stifles innovation, where as consumer markets stimulate innovation but have difficult funding cutting edge technology.

 

3) When nationalistic sentiment rises, it seems to forecast an economic downfall.

 

4) Beware of Clay Christenson's low cost producer with low margin customers in these industries in the right setting, can scale up and innovate quickly to be a competitive threat.

 

Also thanks for the recommendation on Bionomics. I thoroughly enjoyed it. Picked up Shorting the Grid as well. Looking forward to getting through it.

 

 

Yes, It's like Buffett said about knowledge being cumulative, and you start to see connections.  In Bionomics they showed that curve where mice get more efficient at finding cheese as their number of repetitions increases, and they showed the same curve for egg production in the US.   If you wanted to start an online retailer like Amazon, it would be impossible now.  Bezos started in his garage packing books on the floor and ordering kneepads to deal with the pain of kneeling on a cement floor.  Now they have multimillion dollar robots doing the work of moving items around.  Absent government subsidies, you wouldn't be able to absorb them absorb the losses on your way to learning enough to be efficient.  That's why, according to Chip War, the Soviets didn't make a lasting impact in the chip sector because they were producing chips for military use, so the amount produced wasn't big enough to offer opportunities to learn and become more efficient.  So now that TSMC is at the top of the heap with the incredible precision of the new ultra low size chips that can only be made with one machine produced by one company in the Netherlands, it's hard to imagine anyone finding a way to catch up.  Ditto for AI and Google/Microsoft.  I think that's why there will be a few winners in retail like Amazon (US), MELI (latam), CPNG (Korea, Taiwan) and BABA/PDD (China).  As you get better and better, your lead becomes greater and it's harder for anyone except someone who's already doing it to be able to compete.

 

Most of these chip companies are way outside my circle of competence, but it's an interesting thing to learn about.  

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...