We investigate the reliability of standard N-body simulations by modeling of the well-known Hernquist halo with the help of GADGET-2 code (which uses the tree algorithm to calculate the gravitational force) and ph4 code (which uses the direct summation). Comparing the results, we find that the core formation in the halo center (which is conventionally considered as the first sign of numerical effects, to be specific, of the collisional relaxation) has nothing to do with the collisional relaxation, being defined by the properties of the tree algorithm. This result casts doubts on the universally adopted criteria of the simulation reliability in the halo center.

Though we use a halo model, which is theoretically proved to be stationary and stable, a sort of numerical ‘violent relaxation’ occurs. Its properties suggest that this effect is highly likely responsible for the central cusp formation in cosmological modeling of the large-scale structure, and then the ‘core-cusp problem’ is no more than a technical problem of N-body simulations.

我们借助GADGET-2代码（使用树算法计算引力）和ph4代码（使用直接求和）对著名的Hernquist光晕进行建模，从而研究了标准N体仿真的可靠性。比较结果，我们发现晕圈中心的芯层形成（通常被认为是数值模拟的第一个迹象，即碰撞弛豫的具体体现）与碰撞弛豫无关，由属性定义树算法。这一结果使人们对晕轮中心普遍采用的仿真可靠性标准产生了怀疑。

尽管我们使用了光晕模型，该模型在理论上被证明是稳定且稳定的，但仍会出现一种数值“剧烈松弛”现象。它的性质表明，这种效应很可能是在大规模结构的宇宙学建模中导致中央尖点形成的原因，然后“核心尖点问题”仅是N体模拟的技术问题。