In this talk, I will revisit a classical result of King that moduli spaces of semistable representations of acyclic quivers are projective, but using modern methods that allow us to avoid using GIT. For this, I will have to briefly review the theory of good moduli spaces for algebraic stacks, introduced by Alper and being developed by Alper, Halpern-Leistner, Heinloth and others. I will define the stack of semistable quiver representations and use a recent existence result to explain why it admits a good moduli space. Our methods allow us to refine the classical results: I will define a determinantal line bundle on the stack which descends to a semiample line bundle on the good moduli space, and provide effective bounds for global generation. For an acyclic quiver, we can observe that this line bundle is ample and thus the good moduli space is projective. This talk is based on ongoing work with Belmans, Damiolini, Franzen, Hoskins, Tajakka.