This paper presents direct numerical simulation and validation of analytical prediction of the finite-amplitude forced dynamics of suspended cables. The main goal is to complement analytical and numerical solutions, accomplishing overall quantitative/qualitative comparisons of nonlinear response characteristics. By counting on an approximate, kinematically non-condensed, planar modeling, a simply-supported horizontal cable subject to a primary external resonance and a 1:1 (or 1:1 vs. 2:1) internal resonance is analyzed. To obtain analytical solutions, a secondorder multiple scales approach is applied to a complete eigenfunction-based series of nonlinear ordinary-differential equations of damped forced cable motion. Accounting for weakly quadratic/cubic geometric nonlinearities and multiple modal contributions, local scenarios of cable uncoupled/coupled responses and associated stability are predicted, based on chosen reduced-order models. As a cross-checking tool, direct numerical simulations of associated nonlinear partial-differential equations describing the high-dimensional, multi-degree-offreedom, system dynamics are carried out using a finite difference technique employing a hybrid explicit-implicit integration scheme. Based on system control parameters and initial conditions, cable space-time varying nonlinear responses of amplitudes, displacements and tensions are numerically assessed, thoroughly validating the analytically predicted solutions as regards actual existence, meaningful role and predominating internal resonance of coexisting/competing dynamics. Some methodological aspects are noticed, along with an insightful discussion on kinematically approximate/exact and planar/non-planar cable modeling. Keywords suspended cable, direct numerical simulation, analytical prediction, reduced-order model, internal resonance, nonlinear forced vibration 1 1 , , J J J J m m m m m m , 2 cos 2 cos J J J r r r s s s J J J J s s ss ss r r rr rr J J (5) Here, γ r = (σ f +σ)t -β r , γ s = σ f t -β s in Eq.(4), whereas γ r = σt -2β r +β s , γ s = σ f t -β s in Eq.(5), with β r (β s ) being the phase of associated a r (a s ) amplitude. In addition to the first-order superimposition of resonant ( , ) J J r s ζ ζ modal functions with their correlated phases (e.g., Figs.1b, c), the spatial displacement distributions in both Eqs.(4) and (5) further depend on second-order shape functions assembling quadratic nonlinear effects of every retained resonant/non-resonant mode via , J J ij ij