Chinese semiconductor industry

Status
Not open for further replies.

sunnymaxi

Captain
Registered Member
why people actually read these articles.

Bloomberg , FT , Thediplomat are hardcore anti-China. these self-proclaimed western experts are clueless about China's semiconductor manufacturing tools progress. keep posting outdated information.

some of our group members are far more knowledgeable.
 

Pkp88

Junior Member
Registered Member
Nah, I'm very pro-China but not afraid to share anti-China propaganda or news. I have no idea how correct that article is. I share it in both topics to reach a more wider audience.

Surprised seeing the dismissive reaction to this article. That article is the first time outside of some citationless forum that mentions a 28nm SMEE lithography device and offers details that it's currently at a low yield stage - given the audience I'm going to lean toward credible. The appropriate response would be to discuss how long it takes for a low-yield device to iterate into high quality competitive yields.

As for "some of our group members are far more knowledgeable" - I'm still waiting on their citationless predictions of a lithography machine up/running in 2020 to materialize.
 

latenlazy

Brigadier
Surprised seeing the dismissive reaction to this article. That article is the first time outside of some citationless forum that mentions a 28nm SMEE lithography device and offers details that it's currently at a low yield stage - given the audience I'm going to lean toward credible. The appropriate response would be to discuss how long it takes for a low-yield device to iterate into high quality competitive yields.

As for "some of our group members are far more knowledgeable" - I'm still waiting on their citationless predictions of a lithography machine up/running in 2020 to materialize.
What he gets wrong is the statement that 28 nm DUVi is “several generations” behind EUV. It’s precisely only 1-1.5 generations behind in scanning resolution. Any higher end or newer DUV instruments are simply making improvements in the stage mechanism to speed up wafer processing.

There’s not much to discuss with regard to how long it will take for the indigenously made DUVi to be mature or reliable enough for large scale production. Best case is something really short like 6 months and worst case is something that drags like two years but we won’t really know for sure until things happen. Realistically the instrument will probably reach a level of reliability tolerable enough for large scale production within a year’s time and will spend the next few years working through more issues as they’re found, since in any machine finding problems require large scale and long time usage. These are ultimately as much business questions as they are technical questions.

Note the article discusses none of these considerations because the author of the piece doesn’t understand this subject matter at that kind of depth (because he doesn’t have the background to understand the actual dynamics of the subject beyond superficial summarized facts), which is why I generally frown upon the utility of these kinds of articles for enhancing understanding.
 

tokenanalyst

Brigadier
Registered Member
Surprised seeing the dismissive reaction to this article. That article is the first time outside of some citationless forum that mentions a 28nm SMEE lithography device and offers details that it's currently at a low yield stage - given the audience I'm going to lean toward credible. The appropriate response would be to discuss how long it takes for a low-yield device to iterate into high quality competitive yields.

As for "some of our group members are far more knowledgeable" - I'm still waiting on their citationless predictions of a lithography machine up/running in 2020 to materialize.
Maybe because the article has no technical value. Is just a hit piece from guy that gets pay for stating the obvious, lithography is as now the "weakest link" So you don't need a dude getting pay to write nonsense in a blog to tell you that. But is clear in this thread the Chinese are working overtime to strengthen that link as you can clearly see with the patent activity and research activity about Immersion lithography and EUV by the multiple Chinese lithography companies and research institutes.​
 

Pkp88

Junior Member
Registered Member
What he gets wrong is the statement that 28 nm DUVi is “several generations” behind EUV. It’s precisely only 1-1.5 generations behind in scanning resolution. Any higher end or newer DUV instruments are simply making improvements in the stage mechanism to speed up wafer processing.

There’s not much to discuss with regard to how long it will take for the indigenously made DUVi to be mature or reliable enough for large scale production. Best case is something really short like 6 months and worst case is something that drags like two years but we won’t really know for sure until things happen. Realistically the instrument will probably reach a level of reliability tolerable enough for large scale production within a year’s time and will spend the next few years working through more issues as they’re found, since in any machine finding problems require large scale and long time usage. These are ultimately as much business questions as they are technical questions.

Note the article discusses none of these considerations because the author of the piece doesn’t understand this subject matter at that kind of depth (because he doesn’t have the background to understand the actual dynamics of the subject beyond superficial summarized facts), which is why I generally frown upon the utility of these kinds of articles for enhancing understanding.
Great response thanks. The one benefit the author does have imo is direct access to a group of people in DC / globally who would have more inside takes on the state of things. That program + think-tank are well connected.
 

Weaasel

Senior Member
Registered Member
I don't see any equipment from South Korea that isn't replaceable by something from China. Neither China nor South Korea have advanced photolithography tools, both have dry etch, vapor deposition, wafer cleaning and CMP tools.

But China at least has SMEE for higher process node lithography, packaging lithography, etc. and some smaller maskless companies. I can't find a single domestic South Korean lithography system in anything.

So these are, as the article states, for Samsung and SK Hynix in China. Fair enough, Korean enterprises use Korean equipment.
China is significantly more advanced than South Korea with regards to semiconductor and IC chip manufacturing equipment, components, and particular materials. It hasn't devised means via which to commercially produce, in large quantities, the most advanced IC chips, such as those of 7 nm nodes and finer that Samsung produces. But very likely, if SMIC were in possession of EUV lithographic machines, it would likely be able to produce them.
 

latenlazy

Brigadier
Maybe because the article has no technical value. Is just a hit piece from guy that gets pay for stating the obvious, lithography is as now the "weakest link" So you don't need a dude getting pay to write nonsense in a blog to tell you that. But is clear in this thread the Chinese are working overtime to strengthen that link as you can clearly see with the patent activity and research activity about Immersion lithography and EUV by the multiple Chinese lithography companies and research institutes.​
He’s not so much getting paid for stating the obvious as he’s trying to get some traction going for his budding career. He’s an IR graduate student at SAIS Hopkins, which means his post graduation options are government, journalism, or think tanks. Writing these kinds of pieces helps get him noticed for a congressional office or committee analyst role, a journalism outfit, or a think tank gig.
 
Status
Not open for further replies.
Top