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!
My latest research indicates that ASML may cut EUV equipment shipment forecasts significantly for 2024 by about 20–30% for the following main reasons:
Apple’s 3nm demand for 2024 would be below expectations. In 2023, Apple’s MacBook and iPad shipments declined significantly by approximately 30% and 22% to 17 million and 48 million units, respectively. The sharp decline is attributed to the end of work-from-home (WFH) demand and diminishing user appeal for the new specifications (Apple Silicon and Mini-LED).
Looking ahead to 2024, Apple’s 3nm demand is negatively impacted by the lack of growth drivers for MacBook and iPad.
Qualcomm’s 3nm demand for 2024 would be lower than expected due to Huawei’s decision to stop sourcing Qualcomm chips and the higher-than-expected penetration rate of Exynos 2400 in Samsung smartphones.
Demand for Samsung’s 3GAP+ and Intel’s 20A is lower than expected.
Samsung, Micron, and SK Hynix aren’t expected to launch memory expansion plans until 2025–2027.
The current market consensus is that the semiconductor sector will bottom out in 2H23. However, it needs to monitor closely whether this bottoming timeline would be pushed back to 1H24/2Q24.
My latest research indicates that ASML may cut EUV equipment shipment forecasts significantly for 2024 by about 20–30% for the following main reasons:
Apple’s 3nm demand for 2024 would be below expectations. In 2023, Apple’s MacBook and iPad shipments declined significantly by approximately 30% and 22% to 17 million and 48 million units, respectively. The sharp decline is attributed to the end of work-from-home (WFH) demand and diminishing user appeal for the new specifications (Apple Silicon and Mini-LED).
Looking ahead to 2024, Apple’s 3nm demand is negatively impacted by the lack of growth drivers for MacBook and iPad.
Qualcomm’s 3nm demand for 2024 would be lower than expected due to Huawei’s decision to stop sourcing Qualcomm chips and the higher-than-expected penetration rate of Exynos 2400 in Samsung smartphones.
Demand for Samsung’s 3GAP+ and Intel’s 20A is lower than expected.
Samsung, Micron, and SK Hynix aren’t expected to launch memory expansion plans until 2025–2027.
The current market consensus is that the semiconductor sector will bottom out in 2H23. However, it needs to monitor closely whether this bottoming timeline would be pushed back to 1H24/2Q24.
ASML shipped 54 units of EUV machines in 2022 and probably will ship 60 EUV machines in 2023. Intel, Samsung and TSMC are expanding their production globally and all of them are in the middle of bringing new fabs online. It's hard to believe the 2024 30% drop is real.
ASML shipped 54 units of EUV machines in 2022 and probably will ship 60 EUV machines in 2023. Intel, Samsung and TSMC are expanding their production globally and all of them are in the middle of bringing new fabs online. It's hard to believe the 2024 30% drop is real.
Nothing there indicates real analysis. Nothing to indicate the blog writer has any understanding of lead times either for the specific equipment or for the building of fabs.
Daniel, what possessed you to cite some random Medium item?
ASML shipped 54 units of EUV machines in 2022 and probably will ship 60 EUV machines in 2023. Intel, Samsung and TSMC are expanding their production globally and all of them are in the middle of bringing new fabs online. It's hard to believe the 2024 30% drop is real.
ASML shipped 54 units of EUV machines in 2022 and probably will ship 60 EUV machines in 2023. Intel, Samsung and TSMC are expanding their production globally and all of them are in the middle of bringing new fabs online. It's hard to believe the 2024 30% drop is real.
If these 60 tools shipment is true and assume 90% tools are for foundry in 3/4nm and there are 20 EUV layers, then the estimated wafer output could be ~120-140kwpm. We might find out who or what segments will consume it in 2024?
If these 60 tools shipment is true and assume 90% tools are for foundry in 3/4nm and there are 20 EUV layers, then the estimated wafer output could be ~120-140kwpm. We might find out who or what segments will consume it in 2024?
The only part of the semiconductor market doing well right now is AI. But I think there could be a hiccup coming for gen AI demand coming up as well. After a lot of excitement and fundraising for gen AI startups, these companies are going to be expected to start generating revenue quickly.
I absolutely believe in gen AI long term, but we are probably close to the peak on the hype cycle so I wouldn't be surprised to see a short term demand correction.
Nothing there indicates real analysis. Nothing to indicate the blog writer has any understanding of lead times either for the specific equipment or for the building of fabs.
Daniel, what possessed you to cite some random Medium item?
Nothing there indicates real analysis. Nothing to indicate the blog writer has any understanding of lead times either for the specific equipment or for the building of fabs.
Daniel, what possessed you to cite some random Medium item?
