diff --git a/src/video_core/buffer_cache/buffer_cache.h b/src/video_core/buffer_cache/buffer_cache.h
index 32c2066363..da86c14500 100644
--- a/src/video_core/buffer_cache/buffer_cache.h
+++ b/src/video_core/buffer_cache/buffer_cache.h
@@ -26,6 +26,7 @@ BufferCache<P>::BufferCache(Tegra::MaxwellDeviceMemoryManager& device_memory_, R
     // Ensure the first slot is used for the null buffer
     void(slot_buffers.insert(runtime, NullBufferParams{}));
     gpu_modified_ranges.Clear();
+    in_place_gpu_written_ranges.Clear();
     inline_buffer_id = NULL_BUFFER_ID;
 #ifdef YUZU_LEGACY
     immediately_free = (Settings::values.vram_usage_mode.GetValue() == Settings::VramUsageMode::Aggressive);
@@ -116,7 +117,7 @@ template <class P>
 void BufferCache<P>::WriteMemory(DAddr device_addr, u64 size) {
     if (memory_tracker.IsRegionGpuModified(device_addr, size)) {
         ClearDownload(device_addr, size);
-        gpu_modified_ranges.Subtract(device_addr, size);
+        ForgetGpuModifiedRange(device_addr, size);
     }
     memory_tracker.MarkRegionAsCpuModified(device_addr, size);
 }
@@ -231,7 +232,7 @@ bool BufferCache<P>::DMACopy(GPUVAddr src_address, GPUVAddr dest_address, u64 am
     };
     gpu_modified_ranges.ForEachInRange(*cpu_src_address, amount, mirror);
     // This subtraction in this order is important for overlapping copies.
-    gpu_modified_ranges.Subtract(*cpu_dest_address, amount);
+    ForgetGpuModifiedRange(*cpu_dest_address, amount);
     const bool has_new_downloads = tmp_intervals.size() != 0;
     for (const auto& pair : tmp_intervals) {
         gpu_modified_ranges.Add(pair.first, pair.second);
@@ -263,7 +264,7 @@ bool BufferCache<P>::DMAClear(GPUVAddr dst_address, u64 amount, u32 value) {
 
     const size_t size = amount * sizeof(u32);
     ClearDownload(*cpu_dst_address, size);
-    gpu_modified_ranges.Subtract(*cpu_dst_address, size);
+    ForgetGpuModifiedRange(*cpu_dst_address, size);
 
     const BufferId buffer = FindBuffer(*cpu_dst_address, static_cast<u32>(size));
     Buffer& dest_buffer = slot_buffers[buffer];
@@ -308,7 +309,7 @@ std::pair<typename P::Buffer*, u32> BufferCache<P>::ObtainCPUBuffer(
         const DAddr device_addr_end = Common::AlignUp(device_addr + size, 64);
         const size_t new_size = device_addr_end - device_addr_start;
         ClearDownload(device_addr_start, new_size);
-        gpu_modified_ranges.Subtract(device_addr_start, new_size);
+        ForgetGpuModifiedRange(device_addr_start, new_size);
         break;
     }
     default:
@@ -760,7 +761,7 @@ void BufferCache<P>::PopAsyncBuffers() {
                             {start, &read_mapped_memory[start - device_addr], end - start});
                     });
                 async_downloads.Subtract(device_addr, copy.size, [&](DAddr start, DAddr end) {
-                    gpu_modified_ranges.Subtract(start, end - start);
+                    ForgetGpuModifiedRange(start, end - start);
                 });
             }
             async_buffers_death_ring.emplace_back(*async_buffer);
@@ -768,7 +769,7 @@ void BufferCache<P>::PopAsyncBuffers() {
         for (const auto& copy : batch.unified_copies) {
             const DAddr device_addr = static_cast<DAddr>(copy.src_offset);
             async_downloads.Subtract(device_addr, copy.size, [&](DAddr start, DAddr end) {
-                gpu_modified_ranges.Subtract(start, end - start);
+                ForgetGpuModifiedRange(start, end - start);
             });
         }
         async_buffers.pop_front();
@@ -1580,6 +1581,20 @@ void BufferCache<P>::MarkWrittenBuffer(BufferId buffer_id, DAddr device_addr, u3
     uncommitted_gpu_modified_ranges.Add(device_addr, size);
 }
 
+template <class P>
+void BufferCache<P>::MarkWrittenBufferInPlace(DAddr device_addr, u32 size) {
+    memory_tracker.MarkRegionAsGpuModified(device_addr, size);
+    gpu_modified_ranges.Add(device_addr, size);
+    uncommitted_gpu_modified_ranges.Add(device_addr, size);
+    in_place_gpu_written_ranges.Add(device_addr, size);
+}
+
+template <class P>
+void BufferCache<P>::ForgetGpuModifiedRange(DAddr device_addr, u64 size) {
+    gpu_modified_ranges.Subtract(device_addr, size);
+    in_place_gpu_written_ranges.Subtract(device_addr, size);
+}
+
 template <class P>
 BufferId BufferCache<P>::FindBuffer(DAddr device_addr, u32 size) {
     if (device_addr == 0) {
@@ -2063,7 +2078,7 @@ template <class P>
 void BufferCache<P>::InlineMemoryImplementation(DAddr dest_address, size_t copy_size,
                                                 std::span<const u8> inlined_buffer) {
     ClearDownload(dest_address, copy_size);
-    gpu_modified_ranges.Subtract(dest_address, copy_size);
+    ForgetGpuModifiedRange(dest_address, copy_size);
 
     BufferId buffer_id = FindBuffer(dest_address, static_cast<u32>(copy_size));
     auto& buffer = slot_buffers[buffer_id];
@@ -2115,7 +2130,7 @@ void BufferCache<P>::DownloadBufferMemory(Buffer& buffer, DAddr device_addr, u64
 
             gpu_modified_ranges.ForEachInRange(device_addr_out, range_size, add_download);
             ClearDownload(device_addr_out, range_size);
-            gpu_modified_ranges.Subtract(device_addr_out, range_size);
+            ForgetGpuModifiedRange(device_addr_out, range_size);
         });
     if (total_size_bytes == 0) {
         return;
diff --git a/src/video_core/buffer_cache/buffer_cache_base.h b/src/video_core/buffer_cache/buffer_cache_base.h
index e30f768bb5..49cc1ba08f 100644
--- a/src/video_core/buffer_cache/buffer_cache_base.h
+++ b/src/video_core/buffer_cache/buffer_cache_base.h
@@ -418,6 +418,10 @@ private:
 
     void MarkWrittenBuffer(BufferId buffer_id, DAddr device_addr, u32 size);
 
+    void MarkWrittenBufferInPlace(DAddr device_addr, u32 size);
+
+    void ForgetGpuModifiedRange(DAddr device_addr, u64 size);
+
     [[nodiscard]] BufferId FindBuffer(DAddr device_addr, u32 size);
 
     void WaitForGpuFenceIfNeeded(Buffer& buffer);
@@ -513,6 +517,7 @@ private:
     MemoryTracker memory_tracker;
     Common::RangeSet<DAddr> uncommitted_gpu_modified_ranges;
     Common::RangeSet<DAddr> gpu_modified_ranges;
+    Common::RangeSet<DAddr> in_place_gpu_written_ranges;
     std::deque<Common::RangeSet<DAddr>> committed_gpu_modified_ranges;
 
     // Async Buffers