Chinese semiconductor industry

Status
Not open for further replies.

ansy1968

Brigadier
Registered Member
The nextplatform.com article is quite interesting. Pay attention to how the company was started and the founders and their key tech people. They are either serial entrepreneurs or very senior tech people at major tech companies in the US (Nvidia, AMD) and/or China (HiSilicon, Alibaba) plus others (Samsung). With their experience and track records, they can quickly attract talents and fundings ($800 million).

It's hardly unique. We know there have been a couple of EDA companies started in the last couple years in China, with very similar experiences & profiles to the Biren people, albeit in a different segment. It's ironic that major EDA companies like Synopsis even invested in these new EDA companies, some of key people at these new startups were originally from them.

Also, many of the Chinese semiconductor equipment companies are founded by people with significant experiences at the major international semiconductor equipment companies. That's pretty well-known.





What has happened is that the US sanctions on China in semiconductors over the last few years has suddenly created a very large market with fertile opportunities and accepting customers. If there weren't US sanctions, the prospects of these new startups (GPU, EDA, SEM etc.) would have been much dimmer, if they would still be founded at all. The semiconductor industry landscape was actually quite stable, mature and, some say, boring. The US-China tech war had changed all that.

It should be noted that in the semiconductor industry, including design, manufacturing, testing and assembly, as well as design software and equipment manufacturing, ethnic Chinese have a very large and prominent presence. These talents naturally feed into these potential opportunities. This is before we even consider the talent pools of East Asians. By the way, the same thing can be said of AI industry and community as well.

Therefore over the longer term, the prospect of the success of US sanctions on Chinese technology industries is rather pessimistic. In fact, even if we haven't got there yet, the US may have very well helped and accelerated the process of creating an alternative and competing tech industry centered around China.
Yes bro, you have explain it nicely, the US is a masterful magician, they created a formidable competitor OUT OF THIN AIR...lol
 

ansy1968

Brigadier
Registered Member
Bro dated 2016, so we can deduced that ICRD had fully grasp the 14nm tech since they have a sign a MOU with ASML on 2017 and they might use the NXT 1980i as a tool of their research? and some collaboration with SMIC?

Please, Log in or Register to view URLs content!




Jun 21, 2017 — ICRD and ASML sign MoU to establish a training center in Shanghai ... I believe that this cooperation will help building a professional ...
 

FairAndUnbiased

Brigadier
Registered Member
this is not news but some basic education about dark silicon and explains why indefinite die shrinks may not actually be the best way to go, independent of actual process difficulty.

Please, Log in or Register to view URLs content!

Key excerpts:

TRS projections have predicted that by 2020, designers would face up to 90% of dark silicon, meaning that only 10% of the chip’s hardware resources are useful at any given time when high operating frequencies are applied...

Increasing dark silicon directly reflects on performance to a point that many- core scaling provides zero gain [8]. Thermal awareness and dark silicon sensitive resource allocation are key techniques that can address these challenges.

Increase in dark silicon reduces the number of simultaneously active components on a chip and hampers performance, energy efficiency, reliability, ageing and effective resource utilization of computer systems [5]. With the dark silicon phenomenon being a hardware and device level issue, other layers of the computing stack are un-aware of the sources of inefficiency.

With a section of the chip being inactive, expected performance from many-core systems can never be realized in practice. In the quest for higher performance, designers opt for denser chips, with the expectation of increased performance. Transistor scaling has paved the way to integrate more cores at a lower power consumption, effectively increasing compute capacity per area [2]. Inherent parallelism, if present, in workloads could take advantage of multi and many-core computers to show improved performance to a certain extent [24]. Dark silicon changes this consensus, since we cannot power up all the available resources at any given time [25]
What are some solutions? Well, advanced packaging is one of them.

Please, Log in or Register to view URLs content!

For smart phones and wearables, dark silicon is a proven solution, but it’s not the most efficient one. There are other options available, from limiting the size of various components to spreading them out in a package, which reduces thermal effects at increased density. This is particularly valuable for memory, which runs more efficiently at cooler temperatures. So memory may stay cool when nearby circuitry is powered down, but inrush current can quickly overheat it. A better option is to physically separate the memory from active logic in an advanced package.
 

paiemon

Junior Member
Registered Member
Ufff that maybe goods news for China AI Chips providers.

US. to restrict Nvidia from exporting its A100 and H100 server chips to China and Russia.

This one is dumb, kind of like the old Xeon processor ban to supercomputers. Any Russian or Chinese entity that really wants to get their hands on those processors can buy them through the distributor/reseller channels which is a big conduit for how technology sales are conducted, its not all OEM. That's how the Russians have been getting their hands on sensitive components even though they were technically banned from buying them ever since 2014. For small volume purchases this can easily slip through export controls since they can buy from distributors in intermediary countries which have the right licenses, its just a price markup for them to pass it on and if they get caught distributors are one of the easiest businesses to reconstitute in this era (like anyone can do it). The only thing they don't get is aftermarket implementation/sales/support for the hardware such as automatic driver updates since those may be blocked but again, since Nvidia's drivers are publicly available they can just download them themselves.

I suppose it will force some customers to redesign their system architectures to find substitutes if they want to expand but anyone currently using those processors is mostly fine unless you need volume replacements. Also with the cloud and virtual machines someone in say Malaysia can setup a virtual GPU data center using those processors and then just rent it out to customers in China lol. Worst thing is the US government stops them from getting new GPUs or support for existing ones but if the workload was run then the job was done, they got paid and can selloff that data center to somebody else or repurpose it for bitcoin mining lol.
 

horse

Colonel
Registered Member
Honestly, I do not even know what to think.

Please, Log in or Register to view URLs content!
May 24, 2022 12:15 AM EST

The H100 chip is not even available for sale yet.

They are banning something that does not exist!

This could even be peak paranoia or peak desperation. Could be both. LOL.

:D

It just seems that the people in America thinking up and legislating these tech bans, do not understand technology.

That H100 that is not born yet, would a couple of older chips used together in some configuration be just as potent as that new chip? Possibly.

Okay, now we get to the stupid, like really stupid part.

It said that this A100 and H100 are server chips.

Servers, we can have one, but usually it is many upon many upon many, in clusters or like they say, the cloud.

Maybe there is a server room somewhere, where they got one hundred servers with the H100 chip that is banned.

Suppose there is another server room somewhere, full of servers, with older chips. How many more servers will they need to add to approximate the performance of the server room with that H100 chip? They probably could add a few more servers, then that configuration of old servers will match the performance of the H100 server configuration.

There should be more electricity costs, but this is China were electricity is cheap, and they do not use air conditioning because some data centers are in caves where they are naturally cooled.

What are they exactly thinking about with this ban?

Whatever they are thinking about, it is not about technology.

Maybe it really is PANIC TIME!!!

index.jpg
 

Appix

Senior Member
Registered Member

U.S. tightens chip export rules to China, hitting Nvidia and AMD​

TAIPEI -- The U.S. is further restricting shipments to China of high-end graphics processors and AI accelerators used in high-performance computing, a move that is already affecting Nvidia and Advanced Micro Devices as Washington steps up efforts to curb Chinese technological and military advancements.

Nvidia confirmed to Nikkei Asia that two of its premium AI computing chips and one type of its powerful AI computing system were affected by a new U.S. regulation. The company said it will seek export control licenses and will also talk to customers in China about the issue.

AMD meanwhile alerted its China operations on Wednesday that the U.S. chipmaker will suspend some shipments of high-end GPUs to China, according to two people familiar with the matter. AMD confirmed to Nikkei Asia that it has received notification of new licensing requirements from the U.S. Department of Commerce that prevent it from shipping MI250 integrated circuits to China and Russia.

"The only current products that the new licensing requirement applies to are A100, H100 and systems such as DGX that include them," a Nvidia spokesperson said, referring to the company's premium AI accelerators, which can be used to develop cutting-edge supercomputers.

AMD's MI250 chip is another such AI accelerator.

An official from the Commerce Department told Nikkei Asia that the agency cannot comment on specific policy changes but said it is taking a comprehensive approach to implement additional actions necessary to protect U.S. national security and foreign policy interests and avoid any advanced U.S. technologies ending up being used in Chinese military applications.

Non-paywall source:
Please, Log in or Register to view URLs content!


 
Last edited:

huemens

Junior Member
Registered Member

U.S. tightens chip export rules to China, hitting Nvidia and AMD​

TAIPEI -- The U.S. is further restricting shipments to China of high-end graphics processors and AI accelerators used in high-performance computing, a move that is already affecting Nvidia and Advanced Micro Devices as Washington steps up efforts to curb Chinese technological and military advancements.

Nvidia confirmed to Nikkei Asia that two of its premium AI computing chips and one type of its powerful AI computing system were affected by a new U.S. regulation. The company said it will seek export control licenses and will also talk to customers in China about the issue.

AMD meanwhile alerted its China operations on Wednesday that the U.S. chipmaker will suspend some shipments of high-end GPUs to China, according to two people familiar with the matter. AMD confirmed to Nikkei Asia that it has received notification of new licensing requirements from the U.S. Department of Commerce that prevent it from shipping MI250 integrated circuits to China and Russia.

"The only current products that the new licensing requirement applies to are A100, H100 and systems such as DGX that include them," a Nvidia spokesperson said, referring to the company's premium AI accelerators, which can be used to develop cutting-edge supercomputers.

AMD's MI250 chip is another such AI accelerator.

An official from the Commerce Department told Nikkei Asia that the agency cannot comment on specific policy changes but said it is taking a comprehensive approach to implement additional actions necessary to protect U.S. national security and foreign policy interests and avoid any advanced U.S. technologies ending up being used in Chinese military applications.

Non-paywall source:
Please, Log in or Register to view URLs content!



Just in time for Biren to sell its BR100. As the tweet says probably the next US move would be to ban TSMC from fabbing Biren's chips. Hopefully by that time SMIC could fab Biren's chips
 

european_guy

Junior Member
Registered Member
so let's see, the brainiacs at White House are restricting Nvidia just as the Chinese alternative becomes available? Am I missing something? Are they looking to help Biren and Baidu?

I expect the next step will be the ban of TSMC advanced foundry services for Biren, Baidu, Alibaba and siblings. It is a necessary step to avoid that the banning of NVIDIA and AMD backfires hugely, as many people here have already pointed out.

Banning TSMC and Samsung for serving Chinese customers is one of the main reasons of Chip 4 Alliance existence and the only thing, together with the banning of ASML EUV, that can be effective in forcing, by 2024-2025, a 2-3 generations gap between Chinese and Western companies.

I really hope that SMIC new 7nm capacity will be big enough, because very soon it will be greatly needed.

After US killed the future of AMAT, LAM Research and in general of US semi equipment in Chinese market, this will mark the end of NVIDIA too. But US administartion seems to have already accepted and even embraced the idea of total decoupling in semiconductor and digital market.
 
Status
Not open for further replies.
Top