"As expected, we have seen some IC process equipment pieces make their way to China and other locations in the world after taking a detour through a "customer" in Sweden or some other country that does not have as conservative export regulations as in the U.S."
That would still be an infringement of US law and the guilty party could be in big trouble wherever they happen to be.
In response to Mr McGrath, Restrictions of what technology can be imported to China is , IMHO, not having any siginificant effect on the industry in China. As an interntationally based brokerage of used semiconductor manufacturing equipment, faburplus.com has some experience of exporting (and importing) semiconductor manufacturing equipment between Europe, the USA, Japan and china. Basically, we are restricted on exporting only leading technologies . The typical "commercial" fab in China is not making "state of the art" microchips. The only other folks apart from SMIC wanting state of the art stuff in China are the government fabs doing defence related stuff...and we aren't allowed to sell to them. That kind of use in R and D, will never be commercially significant.
On the other hand, 2 factors that significantly restrict our ability to import semiconductor manufacturing equipment into China are:
1. Restrictions on how much cash Chinese companies can pay ahead of shipment coupled with our suspicion of the Chinese legal system and any Chinese letters of Credit we are asked to accept.
2. The long delays in payment generally caused by paperwork the private Chinese companies need to do to get release of money for investments in capital equipment that is located outside China.
Yeah, I understand that it's a politically tricky question. And I have only a limitied knowledge of the history and sensitivity of the situation. Very limited.
But when we talk about chip making, I don't lump Taiwan into China. I think every analyst that follows the chip market also seperates the two. Taiwan is already a huge chip manufacturing power, and that's not going to change. I myself don't want to get caught up in the politics. But for anyone who has been to both places, it's pretty clear they are distinct entities. Completely different ball game. I know there are people on both sides who hope they resolve that. And I know there are people in Taiwan who want to declare independence. I don't think anyone knows what will happen there. But in terms of chip making, Taiwan is at the head of the class, more or less, and China is near the back. And while U.S. companies are free to sell TSMC their latest and greatest chip building gear, Uncle Sam doesn't want them exporting that stuff to Mainland China.
Dylan, it has been my understanding, after talking with one of the people in charge of watching which pieces of U.S.-made IC processing equipment are exported to where, that there isn't a particular node where a cutoff decision is made. I was told that the U.S. government export agencies first determine if the company that wants to buy the IC production equipment has any contracts with local military customers (in this case Chinese military customers). If it does, a determination is made at that time whether the IC process equipment will be allowed to be sold to the Chinese customer or not.
As expected, we have seen some IC process equipment pieces make their way to China and other locations in the world after taking a detour through a "customer" in Sweden or some other country that does not have as conservative export regulations as in the U.S. :)
Good point about the equipment, KB3001. I had forgotten to mention in any of my previous posts that the U.S. forbids companies from exporting the latest technology to China. I can't recall what the technology node is that is the limit. But even if SMIC or one of the other Chinese foundries somehow was able to get the know how to build leading edge chips, they wouldn't be able to build them anyway because they can't get access to the equipment.
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...