DNN

Cadence has organized its machine-learning platforms into three families intended to cover a wide range of on-device AI applications.

Centaur opted for a superwide SIMD engine in an accelerator for a multicore x86 aimed at edge server applications that could take full advantage of spare die area.

IP core focuses on avodiing memory access bottlenecks during processing of complex machine-learning and artificial-intelligence algortihms.