This　paper　presents　the　dynamic　response　analyses　of　a　special　long-span　cable-stayed　bridge　with　the　main　span　length　of　680　(m)　subjected　to　pulse-type　ground　motions.　The　1/100　scaled　shaking　table　test　are　developed　herein　to　verify　the　accuracy　and　correctness　of　the　3D　finite　element　(FE)　model　of　the　examined　long-span　cable-stayed　bridge　on　the　software　platform　of　SAP2000.　To　systematically　investigate　the　influence　on　the　seismic　responses　and　on　the　selection　of　the　nonlinear　viscous　damper　parameters　of　the　employed　long-span　cable-stayed　bridge　both　the　near-fault　pulse-type　ground　motions　and　the　far-field　ground　motions　are　selected　as　the　seismic　inputs　of　FE　model.　Some　important　conclusions　are　drawn　that　the　near-fault　ground　motions　usually　causes　the　larger　peak　responses　(e.g.,　Tower　top　displacement,　girder　end　displacement,　and　moment　of　tower　bottom)　of　the　long-span　cable-stayed　bridge　compared　with　the　far-field　motions,　namely　the　near-fault　ground　motions　are　more　destructive　to　the　structures.　The　displacement　responses　can　be　amplified　approximately　2-3　times　and　the　damper　parameters　(Damping　coefficient　C　and　damping　index　alpha)　of　nonlinear　viscous　dampers　can　be　affected　by　the　ground　motion　characteristics.　The　damper　parameter　values　are　changed　from　the　C　=　3000　and　alpha　=　0.3　in　far-field　earthquakes　to　the　C　=　6000　and　alpha　=　0.2　in　near-fault　earthquakes,　namely　the　near-fault　ground　motions　obviously　increase　the　demand　of　energy　consumption　of　general　nonlinear　viscous　dampers.