diff --git a/ChocolArm64/Memory/AMemory.cs b/ChocolArm64/Memory/AMemory.cs
index e969cca5fe..5a08aef357 100644
--- a/ChocolArm64/Memory/AMemory.cs
+++ b/ChocolArm64/Memory/AMemory.cs
@@ -691,6 +691,19 @@ namespace ChocolArm64.Memory
             Marshal.Copy(Data, 0, (IntPtr)(RamPtr + (uint)Position), Data.Length);
         }
 
+        public void CopyBytes(long Src, long Dst, long Size)
+        {
+            EnsureRangeIsValid(Src, Size, AMemoryPerm.Read);
+            EnsureRangeIsValid(Dst, Size, AMemoryPerm.Write);
+
+            Buffer.MemoryCopy(
+                RamPtr + (uint)Src,
+                RamPtr + (uint)Dst,
+                Size,
+                Size);
+        }
+
+
         private void EnsureRangeIsValid(long Position, long Size, AMemoryPerm Perm)
         {
             long EndPos = Position + Size;
diff --git a/Ryujinx.HLE/Gpu/Engines/NvGpuEngineDma.cs b/Ryujinx.HLE/Gpu/Engines/NvGpuEngineDma.cs
index 7e355e8de9..b2fc86581e 100644
--- a/Ryujinx.HLE/Gpu/Engines/NvGpuEngineDma.cs
+++ b/Ryujinx.HLE/Gpu/Engines/NvGpuEngineDma.cs
@@ -58,6 +58,9 @@ namespace Ryujinx.HLE.Gpu.Engines
             int SrcPitch = ReadRegister(NvGpuEngineDmaReg.SrcPitch);
             int DstPitch = ReadRegister(NvGpuEngineDmaReg.DstPitch);
 
+            int XCount = ReadRegister(NvGpuEngineDmaReg.XCount);
+            int YCount = ReadRegister(NvGpuEngineDmaReg.YCount);
+
             int DstBlkDim = ReadRegister(NvGpuEngineDmaReg.DstBlkDim);
             int DstSizeX  = ReadRegister(NvGpuEngineDmaReg.DstSizeX);
             int DstSizeY  = ReadRegister(NvGpuEngineDmaReg.DstSizeY);
@@ -103,13 +106,23 @@ namespace Ryujinx.HLE.Gpu.Engines
                 DstSwizzle = new BlockLinearSwizzle(DstSizeX, 1, DstBlockHeight);
             }
 
-            for (int Y = 0; Y < DstSizeY; Y++)
-            for (int X = 0; X < DstSizeX; X++)
+            if (SrcLinear == DstLinear)
             {
-                long SrcOffset = SrcAddress + (uint)SrcSwizzle.GetSwizzleOffset(X, Y);
-                long DstOffset = DstAddress + (uint)DstSwizzle.GetSwizzleOffset(X, Y);
+                long Src = Vmm.GetPhysicalAddress(SrcAddress);
+                long Dst = Vmm.GetPhysicalAddress(DstAddress);
 
-                Vmm.WriteByte(DstOffset, Vmm.ReadByte(SrcOffset));
+                Vmm.Memory.CopyBytes(Src, Dst, XCount * YCount);
+            }
+            else
+            {
+                for (int Y = 0; Y < YCount; Y++)
+                for (int X = 0; X < XCount; X++)
+                {
+                    long SrcOffset = SrcAddress + (uint)SrcSwizzle.GetSwizzleOffset(X, Y);
+                    long DstOffset = DstAddress + (uint)DstSwizzle.GetSwizzleOffset(X, Y);
+
+                    Vmm.WriteByte(DstOffset, Vmm.ReadByte(SrcOffset));
+                }
             }
         }
 
diff --git a/Ryujinx.HLE/Gpu/Engines/NvGpuEngineDmaReg.cs b/Ryujinx.HLE/Gpu/Engines/NvGpuEngineDmaReg.cs
index 835a822d16..bf915d4e4d 100644
--- a/Ryujinx.HLE/Gpu/Engines/NvGpuEngineDmaReg.cs
+++ b/Ryujinx.HLE/Gpu/Engines/NvGpuEngineDmaReg.cs
@@ -6,6 +6,8 @@ namespace Ryujinx.HLE.Gpu.Engines
         DstAddress = 0x102,
         SrcPitch   = 0x104,
         DstPitch   = 0x105,
+        XCount     = 0x106,
+        YCount     = 0x107,
         DstBlkDim  = 0x1c3,
         DstSizeX   = 0x1c4,
         DstSizeY   = 0x1c5,