KSMC to challenge TSMC with Samsung expertise?

[Fred Chen]

Published 26 mins ago on December 26, 2024
By James Lee Taylor

South Korea considering forming KSMC to challenge TSMC with Samsung expertise. The name of this organization, Korea Semiconductor Manufacturing Company, is purely inspired by the Taiwanese Semiconductor Manufacturing Company (TSMC).

It is estimated that a $13.9 billion investment in KSMC could generate $208.7 billion by 2045. It would support smaller semiconductor firms, many of which are constrained by relying solely on Samsung’s advanced nodes under 10nm.

Recently, South Korean industry and academia have proposed (via KoreaBizWire) creating a KSMC to address challenges in Korea’s semiconductor sector with Samsung facilities.

The report suggests that this initiative aims to create a balanced ecosystem for both foundries and fabless companies by incorporating both advanced and legacy technologies.

At the event, the committee identified key challenges, including:

1. - Narrowing technological gaps with overseas competitors
2. - Weakening investment competitiveness
3. - Lack of fabless and packaging growth
4. - Talent outflow
5. - Excessive regulations.

Samsung managed to evolve its advanced semiconductor process nodes. However, the Foundry Division has failed to attract contract chip manufacturing deals, resulting in huge losses.

Notably, concerns were raised about whether KSMC could handle advanced processes. SK Hynix CEO Kwak No-jung suggested repurposing some of Samsung’s legacy facilities as part of the plan.

“Taiwan maintains a balanced ecosystem where companies like UMC and PSMC focus on legacy and middle-tech processes, complementing TSMC’s advanced manufacturing. This allows over 250 fabless firms to thrive naturally in Hsinchu,” said Professor Kwon Seok-jun of Sungkyunkwan University. “KSMC, launched with government support, could play a similar role in Korea.”

This unexpected plan to form a TSMC-like KSMC underscores Korea’s ambition to foster its position in the global semiconductor market apart from resolving weaknesses in the industry.

https://www.sammyfans.com/2024/12/26/ksmc-to-challenge-tsmc-with-samsung-expertise/

 

[Fred Chen]

Seems China would be a bigger threat than TSMC for legacy and mid-tech nodes.

 

[Xebec]

The politics inside Samsung HQ around this must be really intense.

I think this kind of thinking though would be good for a long term stable business within Korea, and give a chance to rejuvinate any negative reputation Samsung may have in this area.

Are there currently a large # of truly legacy fabs running in Korea at this time?

 

[MKWVentures]

What is the problem they are trying to solve? Are they saying there are not enough legacy Fabs? Or do they think every country should have its own set of 5-10 fabs (which will lead to financial ruin)

 

[Daniel Nenni]

Published 26 mins ago on December 26, 2024
By James Lee Taylor

South Korea considering forming KSMC to challenge TSMC with Samsung expertise. The name of this organization, Korea Semiconductor Manufacturing Company, is purely inspired by the Taiwanese Semiconductor Manufacturing Company (TSMC).

It is estimated that a $13.9 billion investment in KSMC could generate $208.7 billion by 2045. It would support smaller semiconductor firms, many of which are constrained by relying solely on Samsung’s advanced nodes under 10nm.

Recently, South Korean industry and academia have proposed (via KoreaBizWire) creating a KSMC to address challenges in Korea’s semiconductor sector with Samsung facilities.

The report suggests that this initiative aims to create a balanced ecosystem for both foundries and fabless companies by incorporating both advanced and legacy technologies.

At the event, the committee identified key challenges, including:

1. - Narrowing technological gaps with overseas competitors
2. - Weakening investment competitiveness
3. - Lack of fabless and packaging growth
4. - Talent outflow
5. - Excessive regulations.

Samsung managed to evolve its advanced semiconductor process nodes. However, the Foundry Division has failed to attract contract chip manufacturing deals, resulting in huge losses.

Notably, concerns were raised about whether KSMC could handle advanced processes. SK Hynix CEO Kwak No-jung suggested repurposing some of Samsung’s legacy facilities as part of the plan.

“Taiwan maintains a balanced ecosystem where companies like UMC and PSMC focus on legacy and middle-tech processes, complementing TSMC’s advanced manufacturing. This allows over 250 fabless firms to thrive naturally in Hsinchu,” said Professor Kwon Seok-jun of Sungkyunkwan University. “KSMC, launched with government support, could play a similar role in Korea.”

This unexpected plan to form a TSMC-like KSMC underscores Korea’s ambition to foster its position in the global semiconductor market apart from resolving weaknesses in the industry.

https://www.sammyfans.com/2024/12/26/ksmc-to-challenge-tsmc-with-samsung-expertise/

This would be Korea only. For the rest of the world TSMC has plenty of legacy fabs (non FinFET) that are at 70% capacity:

• 130nm (0.13 μm)
• 90nm
• 65nm
• 55nm
• 40nm
• 28nm
• 22nm

And TSMC is building more 28nm for automotive in Japan and Germany:

In November of 2021 TSMC announced a partnership with the Japanese government, Sony, and DENSO (automotive) for fabs in Japan. The first one opened in February of 2024, making 28nm wafers. A second FinFET fab will be added for production in 2027.

TSMC also announced a partnership with Robert Bosch GmbH, Infineon Technologies AG, and NXP Semiconductors N.V. to establish the European Semiconductor Manufacturing Company (ESMC) in Dresden, Germany. A 28nm fab is being built with production starting in 2027 and I would guess more fabs will follow.

 

[prime007]

I pulled this explanation from Reddit (user: self-fix) but thought it gave a good explanation.

The aim is not to make a KSMC that competes directly with Samsung or TSMC, or, at least not immediately. Its primary aim is more directed at growing fabless startups in Korea: He proposes to make a government-funded foundry that allows fabless IC startups to test their designs at a very cheap price, and possibly situate the foundry near Busan, close to Japan, and capitalize on some of their startups/ IC designers who need testing. Funnily enough, this is how TSMC started, when it was a branch of the Taiwanese government. The founding of that branch of the government was initially inspired by S.Korea's KIST.

He says S.Korea is arguably one of the only 2 countries in the world that can successfully test nodes under 4nm atm, but the problem is that it's provided by Samsung Electronics whose Foundry division is not separated from their System IC division. Korean NPU startups feel hesitant about testing at Samsung because 1. it's expensive, and 2. it could potentially give away their designs to a competitor.

Also, the Korean government is very unhappy about SK Hynix's recent decision to share the license of manufacturing nanometer scale HBM base-dies with TSMC, as that essentially could pave the way to losing the hegemony on next-generation memory chips. SK Hynix couldn't use Samsung's either because Samsung also makes HBM. The KSMC proposal aims to address this issue as well.

 

[Daniel Nenni]

I pulled this explanation from Reddit (user: self-fix) but thought it gave a good explanation.

There is no foundry on the planet that can compete with TSMC on low cost semiconductor manufacturing (high yield). But I do believe it is in TSM’s best interest to have competition at all nodes so there will be no foundry price wars coming. In fact, Hock Tan and other fabless CEOs have said the same. Semiconductor price wars are counterproductive. Fabs are getting more expensive yet chips are getting cheaper?