Neural processing unit: Difference between revisions

Revision as of 03:47, 17 June 2016

Revision as of 03:52, 17 June 2016

Line 20:

can still be gained with a more specific design.

The [[memory access pattern]] of AI calculations differs from graphics, with more a more predictable but deeper [[dataflow]] ,rather than 'gather'(benefiting from texture-mapsthe &ability to 'scatter'keep tomore frametemporary buffers.variables on-chip); GPUs by contrast devote silicon to efficiently dealing with irregularnon-linear [[gather-scatter memory access patternspattern#GATHER_SCATTER|gather-scatter]] addressing (between textures and frame buffers), and [[texture filtering]] for graphics rendering, which isn't usuallyas needed infor AI3D rendering.

As of 2016, vendors are pushing their own terms, in the hope that their designs and [[API]]s will dominate. In the past after [[graphics accelerator]]s emerged, the industry eventually adopted [[NVidia]]s self assigned term "[[GPU]]" as the collective noun for "graphics accelerators", which had settled on an overall pipeline patterned around [[Direct3D]]. There is no consensus on the boundary between these devices, nor the exact form they will take, however several examples clearly aim to fill this new space.