With　the　burgeoning　growth　of　the　e-sports　industry,　there　has　been　a　rapid　proliferation　of　gaming　videos　on　online　platforms.　Simultaneously,　this　surge　undoubtedly　presents　significant　challenges　to　the　encoding　of　game　videos.　However,　the　recurring　characteristic　of　gaming　videos　is　not　efficiently　studied　in　the　current　standardization,　such　as　Versatile　Video　Coding　(VVC).　Based　on　these　observations,　our　general　framework,　utilizing　the　Structural　SIMilarity　index　metric　(SSIM),　Hash　SIMilarity　index　metric　(HSIM),　and　Hash　Matrix　SIMilarity　index　metric　(HMSIM),　identifies　recurring　video　clips　and　adjusts　the　reference　frame　of　the　first　frame.　Additionally,　we　analyze　recurring　patterns　in　different　resolutions　and　types,　proposing　our　Scene　Adaptation　(SA)　optimization　algorithm,　which　integrates　SSIM,　HSIM,　and　HMSIM　to　adapt　to　various　resolutions　and　scenes.　Experimental　results　show　that　the　proposed　approach　can　achieve　a　-4.71　%　Bjontegaard　Delta　Bit　Rate　(BDBR)　and　1.00%　Time　Save　(TS),　outperforming　the　benchmark.　©　2024　IEEE.