Samsung Stuffs 1.2TFLOP AI Processor Into HBM2 to Boost Efficiency, Speed
… We haven't seen HBM deployed for CPUs much, Hades Canyon notwithstanding, but multiple high-end GPUs from Nvidia and AMD have tapped HBM/HBM2 as primary memory. …
… We haven't seen HBM deployed for CPUs much, Hades Canyon notwithstanding, but multiple high-end GPUs from Nvidia and AMD have tapped HBM/HBM2 as primary memory. …
… Given the additional complexity of HBM/HBM2, and the intrinsic difficulty of building stacks of chips connected by tiny wires as opposed to well-understood planar technology, some have argued that GDDR5X actually be the better long-term option. …
… Costs and HBM One persistent rumor we've heard about HBM yields and costs, we can thankfully debunk. It's been bandied about that AMD would face a crippling cost structure in the wake of the GTX 980 Ti's debut a few weeks ago, due to poor yields or costs for High Bandwidth Memory HBM . …
… With GDDR5 versus HBM, there was a clear and absolute bandwidth advantage that HBM could offer. This is less the case with GDDR6, thanks to that memory's higher bandwidth capabilities, but there are still use-cases where HBM2 has an advantage. …
… Unlike the interposer technology that has made working with HBM and HBM2 difficult, Intel's solution uses its own EMIB mounting technology that doesn't require an interposer layer. The 4GB HBM2 stack common to all of these chips cuts power consumption by 80 percent compared with GDDR5. …
… While we don't know which Xeon chips the new iMac Pro will use, Intel's current lineup of Broadwell-based E5 Xeons should provide a reasonable comparison point. …
… Intel has been more generous with GPU cores on its mobile chips than its desktop processors, but 2560 GPU cores would be a significant scale up from its current chips. …
… Only the Radeon Nano -- a high-end GPU built with energy-efficient HBM and binned for best-in-class power consumption -- was reasonably close to matching the energy efficiency of Nvidia's Maxwell products. …
… It's true that Intel was always going to fab some of PV at an external foundry, because Ponte Vecchio uses HBM, and Intel doesn't manufacture it. …