diff --git a/GPU/GLES/GLES_GPU.cpp b/GPU/GLES/GLES_GPU.cpp
index 9a9906c819..f1e2a6aaa0 100644
--- a/GPU/GLES/GLES_GPU.cpp
+++ b/GPU/GLES/GLES_GPU.cpp
@@ -1509,6 +1509,18 @@ void GLES_GPU::ExecuteOpInternal(u32 op, u32 diff) {
 	}
 }
 
+void GLES_GPU::FastLoadBoneMatrix(u32 target) {
+	if (!g_Config.bSoftwareSkinning) {
+		Flush();
+		const int num = gstate.boneMatrixNumber & 0x7F;
+		shaderManager_->DirtyUniform(DIRTY_BONEMATRIX0 << (num / 12));
+		if ((num % 12) != 0) {
+			shaderManager_->DirtyUniform((DIRTY_BONEMATRIX0 << (num / 12)) + 1);
+		}
+	}
+	gstate.FastLoadBoneMatrix(target);
+}
+
 void GLES_GPU::UpdateStats() {
 	gpuStats.numVertexShaders = shaderManager_->NumVertexShaders();
 	gpuStats.numFragmentShaders = shaderManager_->NumFragmentShaders();
diff --git a/GPU/GLES/GLES_GPU.h b/GPU/GLES/GLES_GPU.h
index 6746e77770..5742741f89 100644
--- a/GPU/GLES/GLES_GPU.h
+++ b/GPU/GLES/GLES_GPU.h
@@ -78,6 +78,7 @@ public:
 protected:
 	virtual void FastRunLoop(DisplayList &list);
 	virtual void ProcessEvent(GPUEvent ev);
+	virtual void FastLoadBoneMatrix(u32 target);
 
 private:
 	void Flush() {
diff --git a/GPU/GPUCommon.cpp b/GPU/GPUCommon.cpp
index 83ad905dc8..9647de36bf 100644
--- a/GPU/GPUCommon.cpp
+++ b/GPU/GPUCommon.cpp
@@ -736,12 +736,12 @@ void GPUCommon::ExecuteOp(u32 op, u32 diff) {
 			u32 target = gstate_c.getRelativeAddress(data);
 
 			// Bone matrix optimization - many games will CALL a bone matrix (!).
-			if (g_Config.bSoftwareSkinning && (Memory::ReadUnchecked_U32(target) >> 24) == GE_CMD_BONEMATRIXDATA) {
+			if ((Memory::ReadUnchecked_U32(target) >> 24) == GE_CMD_BONEMATRIXDATA) {
 				// Check for the end
 				if ((Memory::ReadUnchecked_U32(target + 11 * 4) >> 24) == GE_CMD_BONEMATRIXDATA &&
 					  (Memory::ReadUnchecked_U32(target + 12 * 4) >> 24) == GE_CMD_RET) {
 					// Yep, pretty sure this is a bone matrix call.
-					gstate.FastLoadBoneMatrix(target);
+					FastLoadBoneMatrix(target);
 					break;
 				}
 			}
@@ -933,6 +933,10 @@ void GPUCommon::ExecuteOp(u32 op, u32 diff) {
 	}
 }
 
+void GPUCommon::FastLoadBoneMatrix(u32 target) {
+	gstate.FastLoadBoneMatrix(target);
+}
+
 struct DisplayListOld {
 	int id;
 	u32 startpc;
diff --git a/GPU/GPUCommon.h b/GPU/GPUCommon.h
index 769df0c94e..7ddbd39b53 100644
--- a/GPU/GPUCommon.h
+++ b/GPU/GPUCommon.h
@@ -88,6 +88,7 @@ protected:
 	int  GetNextListIndex();
 	void ProcessDLQueueInternal();
 	void ReapplyGfxStateInternal();
+	virtual void FastLoadBoneMatrix(u32 target);
 	virtual void ProcessEvent(GPUEvent ev);
 	virtual bool ShouldExitEventLoop() {
 		return coreState != CORE_RUNNING;