Applied Materials' larger stake in BE Semiconductor points to long-term hybrid-bonding potential despite delayed adoption in HBM. Applied agreed to raise its BESI holding from 9% to about 20%; HBM adoption timelines may not reflect the prospects for the technology as a whole. x.com  x.com 

Higher HBM stack requirements favor Samsung Electronics shares, even as the broader semiconductor industry suffers and DRAM is in dire straits. Daishin Securities says plans for 4-high and 6-high HBM in low-end inference appear shelved, while Nvidia's 8-high Rubin option appears dropped amid base-die supply constraints and customers' preference for bandwidth. x.com 

ASMPT should outperform BE Semiconductor over the next two years because hybrid bonding for HBM has slipped toward 2028, keeping older bonding in use longer. At Hot Chips 2026, SK hynix signaled it will extend MR-MUF through 16-high HBM4; per Yole, wafer-to-wafer hybrid bonding arrives around 2028. ASMPT specializes in the conventional bonding that is sticking around; BESI loses its main near-term driver. BESI is no slam-dunk short, since Intel's Diamond Rapids uses hybrid bonding in chiplets. … Show more

3D-stacked DRAM could constrain bit output even more than today's HBM, preventing memory oversupply beyond 2030. HBM demand already diverts capacity from conventional DRAM; logic-integrated stacks could add pressure if they scale around 2029-30. That comparison depends on the architecture being understood correctly, and memory makers have not settled the DRAM process nodes. x.com 

Shorter HBM stacks are not reducing demand for SK hynix's HBM, because customers are redesigning AI accelerators around limited supply. On its call, management said lower stacks let customers build more accelerators; supplying the same bit volume then requires more packages, base dies and back-end processing. Denser dies can offset some lost layers, and SK hynix says customer volume requests exceed twice its current supply. x.com 

BESI should see a huge hybrid bonding order wave, especially from HBM4E and HBM5 roadmaps, and its price is compelling after the July selloff. x.com 

BE Semiconductor Industries makes hybrid bonding equipment, a critical technology for 3D stacking of HBM4 memory and AI chiplets. It is a top 1% stock globally by combined business and share-price growth, showing European semiconductors are not just ASML. Revenue is forecast to grow 46% this year, with gross margins above 63%, well above peers. ASML patterns the wafers while BESI assembles and packages the finished chips, and being much smaller gives BESI more room for percentage growth. x.com 

Memory remains structurally scarce because SK hynix says conventional DRAM and HBM are practically sold out through 2029, while customers request roughly twice the HBM volume it can produce. Long-term agreements cover more than half of 2027 and 2028 capacity, and the company expects the supply-demand imbalance to persist through 2030. x.com  x.com 

Applied Materials gains if hybrid bonding becomes a commodity DRAM requirement, as it is the only widely ownable name on the short wafer-to-wafer bonder list. Per ZDNet Korea via @jukan05, Samsung moved its wafer-to-wafer bonded B1b onto its roadmap as D0a, its first sub-10nm DRAM. HBM is about 22% of DRAM wafer input this year per TrendForce; commodity adoption would address the other 78%. Samsung hasn't confirmed this. JEDEC lifting 20-high HBM to 900 microns would push this out. x.com 

More memory CapEx may yield proportionally fewer bits as HBM consumes more wafer capacity, supporting SK hynix and Samsung's pricing. Both reported double-digit quarterly DRAM and NAND price gains; Samsung targets 60-70% of mid- to long-term capacity under LTAs. SK hynix has the stronger execution record, but Samsung could benefit enormously from conventional memory without winning HBM4. If trends persist, normalized margins could rise, though memory remains cyclical. x.com