Chinese semiconductor industry

Status
Not open for further replies.

horse

Colonel
Registered Member
@horse bro the result of the harvest? looks like TSMC will be a part of Intel as one of its subsidiary? Cutting of the Chinese customer and relying on American customer, So a noose is being set up for a possible execution?

Oh no, what that article is saying I believe is the opposite.

It was really long brother, so I have to try to remember, as there was a lot of business decisions based on the technology on hand.

Essentially the story could be reduced to a few points, which are important. Since I probably will not read it again, as it is a story about the past, this is all I can come up with.

  • Moore's Law as all about Intel, they were the only company who could do it, and they did it all in-house with their software design of the chip, and the manufacturing of the chip.

  • It was Morris Chang who changed the entire industry when he decided to go do the fabless model. (This is the birth of Intel's current day problems).

  • In short, Intel could not keep up with the development of the IC industry outside of Intel. Intel had to develop their own tools for the fab, but other who specialized only in that, they were faster and got better products.

  • It was a long story, but what is happening to Intel today with their logic chips, already happened as they were the victims of the Japanese taking over Intel's memory chips back in the day.

  • Since Intel was one big company, they could not move fast enough, everyone else in IC iterated faster. So what happened to Intel was they had their manufacturing division with their tools, and their software division, which design the chip to be made with their own tools, both divisions started to hold the other back. Therefore the whole company held itself back.

  • That seems to be true (compared to 10-20 years ago). Intel is no longer the leading edge of chip design, nor can it fab the most advanced chips. Here Intel contracts that out to TSMC. Heh.

  • One idea from the article is that Intel should split itself up, a design company (such as HiSilicone), and a fab only company (such as TSMS). Notice here it is a IC model forged by fire in Asia. The author is probably right. Intel as it is, there is no guarantee that it will catch up.

  • Intel contract out their 3nm chip to be fab by TSMC, because they cannot do it themselves. The author raises the question where Intel can really catch up in the manufacturing because the way the industry is for a fab at the very high end, it is winner take all.

  • That was why the author stressed that the most important news from IC the past few weeks was TSMC threatening everybody else, "Look how much mo-fo-ing money TSMC has that are being investor in R&D you damn mo-fos!"

  • Guess the general point of the article, was due to the past of how Intel was built as a company, and how fast the entire world IC industry changed, that explains why Intel is no longer at the cutting edge, and why it is a real uphill battle for them currently, with no guarantee that they will ever get back to the top. Damn mo-fos.
I thought it was a pretty good article.

You know what they say brother, people who understand economics, not that many. People who understand tech, very few.

That is why businesses really value people who do understand, and can put those two together. I think that guy at stratechry.com does a real good job of explaining it, because he understands both.

Often we find people who can understand one or the other, but not both.

If we want to find people who understand neither, all we need to do is look at the US government, the House of Congress and Corruption. That is where all those people work, who write all the laws.

That, that is a true story!

What mo-fos! You know what I mean!?

:oops::D
 

ansy1968

Brigadier
Registered Member
More like TSMC is annexing Intel's advanced process node.
@Skywatcher Thanks Sir and much obliged, Sir does that mean there will be a split, with TSMC focusing more on the western market and leaving China to others? what I meant is that auto chip is where the growth market is not in smartphone, with each car need at least a dozen or more chip to function? IF TSMC is focusing in advance node, I think cost wise it will be a problem as inflation is increasing and purchasing power is eroding?
 

Skywatcher

Captain
@Skywatcher Thanks Sir and much obliged, Sir does that mean there will be a split, with TSMC focusing more on the western market and leaving China to others? what I meant is that auto chip is where the growth market is not in smartphone, with each car need at least a dozen or more chip to function? IF TSMC is focusing in advance node, I think cost wise it will be a problem as inflation is increasing and purchasing power is eroding?
TSMC can probably do both since it has incredible leverage in both markets (giving it more capability to call the bluff of say, Beijing or DC, over whatever the controversy of the day is)
 

ansy1968

Brigadier
Registered Member
Oh no, what that article is saying I believe is the opposite.

It was really long brother, so I have to try to remember, as there was a lot of business decisions based on the technology on hand.

Essentially the story could be reduced to a few points, which are important. Since I probably will not read it again, as it is a story about the past, this is all I can come up with.

  • Moore's Law as all about Intel, they were the only company who could do it, and they did it all in-house with their software design of the chip, and the manufacturing of the chip.

  • It was Morris Chang who changed the entire industry when he decided to go do the fabless model. (This is the birth of Intel's current day problems).

  • In short, Intel could not keep up with the development of the IC industry outside of Intel. Intel had to develop their own tools for the fab, but other who specialized only in that, they were faster and got better products.

  • It was a long story, but what is happening to Intel today with their logic chips, already happened as they were the victims of the Japanese taking over Intel's memory chips back in the day.

  • Since Intel was one big company, they could not move fast enough, everyone else in IC iterated faster. So what happened to Intel was they had their manufacturing division with their tools, and their software division, which design the chip to be made with their own tools, both divisions started to hold the other back. Therefore the whole company held itself back.

  • That seems to be true (compared to 10-20 years ago). Intel is no longer the leading edge of chip design, nor can it fab the most advanced chips. Here Intel contracts that out to TSMC. Heh.

  • One idea from the article is that Intel should split itself up, a design company (such as HiSilicone), and a fab only company (such as TSMS). Notice here it is a IC model forged by fire in Asia. The author is probably right. Intel as it is, there is no guarantee that it will catch up.

  • Intel contract out their 3nm chip to be fab by TSMC, because they cannot do it themselves. The author raises the question where Intel can really catch up in the manufacturing because the way the industry is for a fab at the very high end, it is winner take all.

  • That was why the author stressed that the most important news from IC the past few weeks was TSMC threatening everybody else, "Look how much mo-fo-ing money TSMC has that are being investor in R&D you damn mo-fos!"

  • Guess the general point of the article, was due to the past of how Intel was built as a company, and how fast the entire world IC industry changed, that explains why Intel is no longer at the cutting edge, and why it is a real uphill battle for them currently, with no guarantee that they will ever get back to the top. Damn mo-fos.
I thought it was a pretty good article.

You know what they say brother, people who understand economics, not that many. People who understand tech, very few.

That is why businesses really value people who do understand, and can put those two together. I think that guy at stratechry.com does a real good job of explaining it, because he understands both.

Often we find people who can understand one or the other, but not both.

If we want to find people who understand neither, all we need to do is look at the US government, the House of Congress and Corruption. That is where all those people work, who write all the laws.

That, that is a true story!

What mo-fos! You know what I mean!?

:oops::D
@horse thanks for explaining things in clear and precise manner, I owed you one brother. :cool:
 

supersnoop

Major
Registered Member
I think this guy might be overlooking one thing…
If the new fab will primarily be dedicated to Intel, then the new excess capacity will not be at TSMC, it will be at Intel…

No one wants to fab with Samsung, their process is literally HOT garbage. Nvidia and Qualcomm go with Samsung because Apple has too much cash. Samsung’s own chips were stealth announced on Twitter and despite early hype, expected to be a pile of disappointment.

Intel’s problem was not solely due to technical problems or TSMC outsmarting them, it was also a business issue that they could not move into the mobile market. Part of this is Qualcomm giving kickbacks to manufacturers not to use Intel, other part is supporting a new hardware platform (terrible since Android Users are statistically more price sensitive). Because they could not get Atom into phones, ultra low power fab technology took a backseat development. The natural move was to shift more resources to high power and high margin server and high end workstation/desktop chips consuming 100W+.

Here is an article discussing some of the above
Please, Log in or Register to view URLs content!

Full disclosure, I own an Asus ZenPhone Zoom, one of the few Intel Atom phones that shipped.
 

antiterror13

Brigadier
Market cap is paper fantasies at this point. Tesla is basically worth more than the top 5 automakers combined or something like that. How does that even make sense?

well, you could say that .. but money talks and bs walk away. People willing to pay more for Tesla/TSMC share .. so that what matter, ..... and doesn't matter whatever you think, unfortunately
 
Status
Not open for further replies.
Top