This suggests that recent driver developments have widened the gap between these two modes.

In the context of Windows display architecture, "drafting" a feature to improve the experience over the Windows Display Driver Model (WDDM) typically centers on reducing kernel launch overhead and memory transfer latency for high-performance computing (HPC) and AI workloads.

MCDM is a newer driver model supported by Windows 11 version 23H2 and above. It offers similar performance advantages to TCC while maintaining compatibility with certain CUDA features that TCC lacks.

If you’re convinced TCC is better, here is how to enable it.

As MCDM continues to mature, we may finally see a driver model that combines TCC's performance with broader compatibility. Until then, understanding this fundamental distinction remains crucial for anyone serious about GPU computing performance on Windows.

Recent benchmarks and developer discussions suggest that WDDM can make RAM-to-GPU data transfers significantly slower—sometimes by orders of magnitude—due to "block swapping" and OS management. Switching to TCC can yield performance parity with Linux, which lacks the WDDM bottleneck. Extended Execution:

WDDM是Windows系统处理图形和计算的默认方式。当GPU运行在WDDM模式下时，它必须同时处理Windows图形界面、硬件加速以及GPGPU（通用GPU计算）任务。这带来了两个核心限制：

Slower; often throttled by "block swapping" and OS restrictions None; the GPU cannot output video to a monitor Required for monitors and Windows desktop tasks GPU Compatibility Professional cards (Tesla, Quadro, Titan) All consumer (GeForce) and professional cards Why TCC is "Better" for Compute

If you work in any of these fields, switching to TCC will transform your workstation or server.

(0 = WDDM, 1 = TCC)

: In scenarios where AI models don't fit entirely in VRAM (requiring constant block swapping with system RAM), TCC has been shown to deliver speeds up to 2x to 3x faster than WDDM.