You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
"In my opinion, CC (I will pretend to know him that well..) is one of the most underrated CEO’s in the semiconductor industry, primarily due to his lack of interest in personal marketing. Rating a CEO by public appearances is like judging a house by looking through the mailbox, but I still believe you can learn a lot.
While my Taiwanese friends will laugh out loud, I believe he is not trying to deceive anybody, but leave the audience with a crisp and honest view of the situation in the most successful semiconductor ever (yes, I include the green server company in that assessment)
He is very calm, controlled and comfortable in his skin; his sense of sharp, understated humour reveals that he does not have a fragile ego to protect. Like the company, he looks pretty modest.
While a company is more than its CEO, CC embodies all the qualities of TSMC, although I am certain he is a challenge to work for."
TSMC Compute business is now most of their business, propelled by Apple, AMD and Nvidia. Intel is probably in there too. Smartphone is still HUGE, but compute is an Airbus A380.
TSMC Compute business is now most of their business, propelled by Apple, AMD and Nvidia. Intel is probably in there too. Smartphone is still HUGE, but compute is an Airbus A380.
Agreed but CC Wei has a strong competitive streak that Morris Chang did not which has made a big difference in the foundry business. There is a language issue so sometimes his jokes do not translate but yes he is very funny. On the investor calls CC does not like repetitive questions and will push back on analysts but so did Morris Chang. I remember onetime Morris responded with something like "that is not a good question" and did not answer it .
Bottom line: Under CC Wei TSMC is not only more competitive but also working harder than ever before to make customers successful, absolutely.
TSMC Compute business is now most of their business, propelled by Apple, AMD and Nvidia. Intel is probably in there too. Smartphone is still HUGE, but compute is an Airbus A380.
And I see it as very very dangerous: 2-3 clients contributing a lion share of revenue, with extremely expensive, low volume dies, with too much of process being adapted to their needs.
A consumer downturn + AI bubble pop + 1 of superclients leaving (or worse, something happens, and they leave without paying for dies) = a few lost years
And I see it as very very dangerous: 2-3 clients contributing a lion share of revenue, with extremely expensive, low volume dies, with too much of process being adapted to their needs.
A consumer downturn + AI bubble pop + 1 of superclients leaving (or worse, something happens, and they leave without paying for dies) = a few lost years
And I see it as very very dangerous: 2-3 clients contributing a lion share of revenue, with extremely expensive, low volume dies, with too much of process being adapted to their needs.
A consumer downturn + AI bubble pop + 1 of superclients leaving (or worse, something happens, and they leave without paying for dies) = a few lost years
Agreed, but TSMC has no competition so it is less of a problem in my opinion. I would expect Nvidia to beat Apple at some point in time, probably next year. Here are my guestimates for 2025:
Apple: ~22%
Nvidia: ~20%
AMD: ~7%
Qualcomm: ~7%
MediaTek: ~5%
Broadcom: ~5%
Intel: ~4%
Marvell: ~3%
NXP: ~2%
Sony: ~1%
Others: ~24%
From what I am told Intel Foundry is aggressively pursuing Apple as a development/foundry partner. This would be huge for Intel and the foundry business. If Lip-Bu can pull this off Intel manufacturing is saved. I would not have thought it was possible but now that Jeff Williams has left Apple anything is possible.
Agreed, but TSMC has no competition so it is less of a problem in my opinion. I would expect Nvidia to beat Apple at some point in time, probably next year. Here are the estimates for 2025:
Apple: ~22%
Nvidia: ~20%
AMD: ~7%
Qualcomm: ~7%
MediaTek: ~5%
Broadcom: ~5%
Intel: ~4%
Marvell: ~3%
NXP: ~2%
Sony: ~1%
Others: ~24%
From what I am told Intel Foundry is aggressively pursuing Apple as a development/foundry partner. This would be huge for Intel and the foundry business. If Lip-Bu can pull this off Intel manufacturing is saved. I would not have thought it was possible but now that Jeff Williams has left Apple anything is possible.
Where did you get these numbers for % of TSMC revenue.
On Apple using Intel: What is the advantage of Apple taking the risk to use Intel? why would they do this? IF someone decided to do this, it would be back up plan first generation completely stealth just like TSMC was and then growing later. But again "Why would someone choose Intel over TSMC?". Maybe as a backup plan with <1B in sales
Where did you get these numbers for % of TSMC revenue.
On Apple using Intel: What is the advantage of Apple taking the risk to use Intel? why would they do this? IF someone decided to do this, it would be back up plan first generation completely stealth just like TSMC was and then growing later. But again "Why would someone choose Intel over TSMC?". Maybe as a backup plan with <1B in sales
Where did you get these numbers for % of TSMC revenue. I dont have Nvidia nearly that high with bottoms up or tops down analysis and I have not seen anyone show numbers like that.
On Apple using Intel: What is the advantage of Apple taking the risk to use Intel? why would they do this? IF someone decided to do this, it would be back up plan first generation completely stealth just like TSMC was and then growing later. But again "Why would someone choose Intel over TSMC?". Maybe as a backup plan with <1B in sales
In my analysis the only thing holding Nvidia back from being #1 is TSMC capacity and packaging since mobile is not growing at the rate AI is and Apple pays less per wafer than Nvidia.
What are your numbers?
I agree with the Apple using Intel Foundry issue but Apple is sharing TSMC with competitors, they are no longer are first to a node, and Intel does have some great technology. I do think a deal could be made that would benefit Apple and Lip-Bu is the person who could make it happen. It is not clear to me, however, that Intel could survive such a deal. Apple has really high foundry standards now thanks to TSMC. You have to admit though, it would be a very disruptive move for Apple/TSMC/Intel and the foundry business. Having only one foundry option is a problem for anybody with future vision and Apple has future vision.
As long as they (TSMC) make all customers happy for a reasonable price there are other problems to worry about. Look at ASML, same thing, basically a full global monopoly, until China catches up in 4-5 (?) years regarding EUV with sufficient source power. Then things could change for non-western high-end chip designers I would imagine.......
For now, Apple seems to have other issues in its supply chain to worry about than the manufacturing of their high-end chips by a single sourced foundry, problems related to tariffs and assembly of their end-products?
As long as they (TSMC) make all customers happy for a reasonable price there are other problems to worry about. Look at ASML, same thing, basically a full global monopoly, until China catches up in 4-5 (?) years regarding EUV with sufficient source power. Then things could change for non-western high-end chip designers I would imagine.......
For now, Apple seems to have other issues in its supply chain to worry about than the manufacturing of their high-end chips by a single sourced foundry, problems related to tariffs and assembly of their end-products?
A "reasonable price" is debatable. Samsung used to sell wafers for 20% less than TSMC. If you are one of TSMC's top 5 customers that is a lot of money. I don't think there is any question that if there was a viable alternative to TSMC at 3nm or 2nm the wafer prices would be lower. But true, the service aspect of TSMC is unbeatable and the ecosystem is 100% behind TSMC and that is hard to beat.
In my analysis the only thing holding Nvidia back from being #1 is TSMC capacity and packaging since mobile is not growing at the rate AI is and Apple pays less per wafer than Nvidia.
What are your numbers?
I agree with the Apple using Intel Foundry issue but Apple is sharing TSMC with competitors, they are no longer are first to a node, and Intel does have some great technology. I do think a deal could be made that would benefit Apple and Lip-Bu is the person who could make it happen. It is not clear to me, however, that Intel could survive such a deal. Apple has really high foundry standards now thanks to TSMC. You have to admit though, it would be a very disruptive move for Apple/TSMC/Intel and the foundry business. Having only one foundry option is a problem for anybody with future vision and Apple has future vision.
100% agree. BTW, I edited my comment quickly after I posted it..... It is possible that you are correct and Nvidia could be close to that. As I mentioned previously, Revenue recognition could a forcasting issue for nvidia.
High level: Nvidia 17-18%. could be higher. I am roughly assuming Nvidia doubles spending and TSMC revenue grows 40%. Intel is another wild card. I expected Lunar plus arrow to take off..... then Intel tells me Raptor lake is what customers want LOL.
Is apple TSMC revenue growing a lot in 2025?
We will see what LBT says tomorrow, and Apple could be doing anything in steal mode..... But given the actions Intel is currently taking (which I agree with).... it could be years before Intel is stable again.
If Apple wanted a second source, Why don't they use Samsung? Apple still sells lots of phones with N5 or N4
Apple absolutely does not like Samsung. Competition, misappropriated IP, etc... Being exclusive to TSMC also has benefits, like flying first class versus economy. CC Wei is also a tough negotiator, much better at customer relations than Morris Chang or Rick Tsai. No way would CC let Apple second source to Samsung and Samsung does not have the yield for it anyway.
100% agree. BTW, I edited my comment quickly after I posted it..... It is possible that you are correct and Nvidia could be close to that. As I mentioned previously, Revenue recognition could a forcasting issue for nvidia.
High level: Nvidia 17-18%. could be higher. I am roughly assuming Nvidia doubles spending and TSMC revenue grows 40%. Intel is another wild card. I expected Lunar plus arrow to take off..... then Intel tells me Raptor lake is what customers want LOL.
Apple absolutely does not like Samsung. Competition, misappropriated IP, etc... Being exclusive to TSMC also has benefits, like flying first class versus economy. CC Wei is also a tough negotiator, much better at customer relations than Morris Chang or Rick Tsai. No way would CC let Apple second source to Samsung and Samsung does not have the yield for it anyway.
Beyond chip manufacturing, how does Intel compare in packaging technology? The iPhone's SoC exclusively uses InFO packaging. Does Intel offer a packaging solution that delivers comparable performance at a similar cost?
Chiang Shang-yi, former COO of TSMC, noted that InFO was developed in response to customer concerns over CoWoS's high cost.
If Intel’s target is Apple’s M-series processors rather than the iPhone, then that's another story.
Apple absolutely does not like Samsung. Competition, misappropriated IP, etc... Being exclusive to TSMC also has benefits, like flying first class versus economy. CC Wei is also a tough negotiator, much better at customer relations than Morris Chang or Rick Tsai. No way would CC let Apple second source to Samsung and Samsung does not have the yield for it anyway.
Beyond chip manufacturing, how does Intel compare in packaging technology? The iPhone's SoC exclusively uses InFO packaging. Does Intel offer a packaging solution that delivers comparable performance at a similar cost?
Chiang Shang-yi, former COO of TSMC, noted that InFO was developed in response to customer concerns over CoWoS's high cost.
If Intel’s target is Apple’s M-series processors rather than the iPhone, then that's another story.
Help me out. So Apples SOC for phones has what chips/chiplets in it and who does the packaging. I havent looked at the chip teardown in years. is it still DRAM on LOGIC chip?
INFO is pretty vanilla as far a packaging tech goes.
Apple stopped taking Intel processors (which I assume were somewhat customized) for its Macs due to pricing and performance concerns. So in any future deals for custom processors these would have to apply, and with a stronger weighting imo. How can Intel make that work for them, given their delays with their own new processors and ramps?
What does that leave? AI/GPU? Would Apple chase the AI market with Intel's AI offerings? I don't think so. Maybe they'll take something for embedded graphics but again that would have to be very cheap and can Intel make that work?
Lastly is Intel fabbing some SoCs for Apple. This would directly compete with TSMC so again could this work for Intel?
These rumors seem to be Apple getting leverage to negotiate better terms with TSMC imo. The ages-old dealmaking ploy. Nothing more. Am I missing something?
clausaasholm.substack.com
.
