Neuromorphic engineering (NE) encompasses a diverse range of approaches to\ninformation processing that are inspired by neurobiological systems, and this\nfeature distinguishes neuromorphic systems from conventional computing systems.\nThe brain has evolved over billions of years to solve difficult engineering\nproblems by using efficient, parallel, low-power computation. The goal of NE is\nto design systems capable of brain-like computation. Numerous large-scale\nneuromorphic projects have emerged recently. This interdisciplinary field was\nlisted among the top 10 technology breakthroughs of 2014 by the MIT Technology\nReview and among the top 10 emerging technologies of 2015 by the World Economic\nForum. NE has two-way goals: one, a scientific goal to understand the\ncomputational properties of biological neural systems by using models\nimplemented in integrated circuits (ICs); second, an engineering goal to\nexploit the known properties of biological systems to design and implement\nefficient devices for engineering applications. Building hardware neural\nemulators can be extremely useful for simulating large-scale neural models to\nexplain how intelligent behavior arises in the brain. The principle advantages\nof neuromorphic emulators are that they are highly energy efficient, parallel\nand distributed, and require a small silicon area. Thus, compared to\nconventional CPUs, these neuromorphic emulators are beneficial in many\nengineering applications such as for the porting of deep learning algorithms\nfor various recognitions tasks. In this review article, we describe some of the\nmost significant neuromorphic spiking emulators, compare the different\narchitectures and approaches used by them, illustrate their advantages and\ndrawbacks, and highlight the capabilities that each can deliver to neural\nmodelers.\n