Hot Chips 2026: d-Matrix stacks AI accelerator directly on custom DRAM for 100 TB/s per card — TSMC 4nm compute die bonded face-to-face at a 36-micron pitch on top of a custom-designed die
d-Matrix presented Raptor, which it calls the first 3D DRAM accelerator for generative inference, showing a TSMC 4nm compute die bonded face-to-face at a 36-micron pitch on top of a custom-designed DRAM die.
Read original article ↗
Related Articles
TSMC packaging shift could boost AMD CoWoS share and trim Nvidia's allocation
Morgan Stanley said TSMC's advanced packaging allocation is expected to shift in 2027, as processor demand tied to agent
Apple gives 2nm debut to Mac, not iPhone: M6 leads TSMC node transition
Apple on Aug. 25 introduced M6 in a new Mac mini and M5 Ultra in a new Mac Studio, framing both chips around AI compute
TSMC's 310mm push threatens Samsung's PLP lead
Samsung Electronics commercialized fan-out panel-level packaging (FOPLP) before rivals, using the technology in mobile a
Top Ten Semiconductor Companies in Q2
NVIDIA USA $125.7 Billion AI GPUs & Data Center Accelerators Samsung Electronics South Korea $72.7 Billion (Memory/DS Di