I don't see anyting wrong with 90nm resolution. Each resolution has it's own field of applications. The new 90nm machine belongs to 800 series，it features dual working station so the throughput has been improved.
Chinese semiconductor industry
- Thread starter Hendrik_2000
- Start date
- Status
I don't see anyting wrong with 90nm resolution. Each resolution has it's own field of applications. The new 90nm machine belongs to 800 series，it features dual working station so the throughput has been improved.
That working station is important, because to master that is to master precision engineering at the highest levels.
That could be close to as difficult as docking a spacecraft to another spacecraft, here just in infinitely small scale.
Seems like everything is advancing and getting there, at a regular pace.
The most sensible of interpreting it is that they've filed a patent.
As I'm sure we are all aware, filing a patent doesn't necessarily mean anything.
It is of course a useful indicator in showing us that "EUV" is something they're working on, but otherwise it is probably more useful for us to pretend the patent doesn't exist for all the good it does in terms of actually tracking Chinese EUV development.
*Or* we can just recognize that "mass", "volume", or "full" production is *production*. *Production* is not R&D. That doesn't mean development stops or a product is mature or that it's fully achieved its intended adoption or utilization. Development never ends for a product, even after it's being delivered to customers, especially if it's capital equipment, but the *R&D phase* does have an end. These instruments *are* being delivered to customers now, even if at a trickle, and we can say that these instruments are not just R&D prototypes anymore. Once we're at this stage of development of a product, we probably should stop treating it as vaporware. That doesn't mean adoption and maturation will be smooth and easy-peasy, but it also doesn't mean that we should treat their existence and their implication for industry like they're imaginary voodoo either.
All or nothing ways of framing and seeing "success" makes a lot of emotional sense as a way to deal with insecurity and anxiety (even though I don't think that's psychologically healthy), but it is *not* imo a sensible analytical perspective if our interest is in gauging objective reality. What I think we can say right now about China's lithography efforts is that they are at the cusp of market adoption in a situation where their potential customers are rather captive, and that's a very consequential development even if not all the problems are adequately solved and these instruments are not "perfect". Progress in industrial development and good engineering is a lot more about the accumulation of incremental work over 5-10 year spans than the individual leaps you can take in 1-2 year spans. Insofar as we want to gauge whether an industry will "get there" the factors that matter aren’t so much how fast is your current progress but how consistent, persistent, and substantial is your support to continue making progress. As your body of work grows, as it turns out, the output of your work tends to get easier and faster, not slower, because you are building experience and expertise that makes the next step better than the last, and I think the tendency to bemoan slow progress in year 1 often leads to underappreciating the amount of progress that will be made in year 5. I personally prefer to try to understand China's tech developments from a "progress mentality" than a "prove it mentality", because in my experience the latter tends to have worse predictive power, but to each their own.
Last edited:
What you would track then?
There abstracts patents that just describe an idea of something, like a general description of an EUV machine.
But when you have specific patents for things like high power CO2 lasers, hydrogen cleaning subsystems, molten tin droplet generators, targeting subsystems, methods for the fabrication MoSi multilayer mirrors, Collector mirrors, NA projection objective subsystems, high precision actuators, EUV masks, maglev wafer stages, grating interferometers and alignment systems that works in vacuum environment, computational lithography for EUV, Vacuum systems, EUV sensing subsystems and so on. You know that serious activity is happening. You just can't ignore it.
But when you have specific patents for things like high power CO2 lasers, hydrogen cleaning subsystems, molten tin droplet generators, targeting subsystems, methods for the fabrication MoSi multilayer mirrors, Collector mirrors, NA projection objective subsystems, high precision actuators, EUV masks, maglev wafer stages, grating interferometers and alignment systems that works in vacuum environment, computational lithography for EUV, Vacuum systems, EUV sensing subsystems and so on. You know that serious activity is happening. You just can't ignore it.
Presentation on X-Ray metrology.
AI assisted X-Ray Metrology:
Advantages over Optical metrology:
AI assisted X-Ray Metrology:
Advantages over Optical metrology:
I never said that we should ignore it -- as I said in my last post, it's an indicator that they're working on EUV.
But we can't infer much more beyond that from a patent.... And we already know that they're working on EUV, so does it add that much to the practical body of knowledge?
I don't think it does.
It doesn't tell us how far along the various subcomponents are, how mature they are, what stage of testing or implementation they are at, when they expect it to emerge, what they actually expect the capability to be in a production model etc.
So it is better for us to pretend it doesn't exist because people will overstate its significance and interpret it as being far more along than it is.
Weasall was outright asking if the patent meant that Huawei could produce EUV machines comparable to ASML!
All that we should take from the EUV patent is to say "okay, anyway" and then proceed along recognizing it doesn't change much of anything in terms of what little we already knew.
Instead, the usual track of leakage of rumours from insiders is better for trying to guesstimate as its development progresses.
X ray has much more relevant information than optical since nobody cares about the optical parameters but elemental analysis, in situ density measurement and film roughness are all critically important and would need to be measured elsewhere anyhow.
Most of all, XRF is almost entirely model independent. You see what you get. Optical metrology is highly model dependent, picking wrong parameters, having problems with software or having an unexpected film stack are all big problems. XRR is model dependent too, but is well characterized.
In addition, X-ray metrology can be extended to XRD for strain measurements for strain engineered interfaces with no further specialized environmental conditions. There are already combined XRD, XRR and XRF instruments for total analysis of thin films.
Depends, some give useful details, methods of fabrication
some even have the graphs of EUV mirrors and photos, what kind of systems they are using to do certain task.
Compare it with ASML and others past research development.
Compare past development with current developments.
from 2017 to today
If you start to see companies involved in the development of one subsystems is a sign of commercialization,
With all the respect I think is an horrible idea, that usually leads to more confusion, how do you know who is an "insider" or not, is not that they are going to show you an ID, even in PM you one. You have to take their word for granted, most don't even publish anything that is not public domain.
If we can't discuss EUV here because is too early to tell I would happily accept that. But if we have to rely on "rumors" of "insiders" to follow the advancement of certain technology I think will be better not to discuss lithography at all until official announcement is made.
I obviously have no issues with discussing EUVs, and as I said there are useful stuff in the patents, but most people aren't able to interpret them, leading to folks genuinely believing that a patent means a question like "is EUV ready to be produced by Huawei" is reasonable.
There are so many caveats with interpreting patents, that for most people, it is better to simply acknowledge it and pretend it doesn't exist, because for many people having a patent means it is at a certain stage of maturity.
- Status