Game Ready Drivers provide the best possible gaming experience for all major new releases, including Virtual Reality games. Prior to a new title launching, our driver team is working up until the last minute to ensure every performance tweak and bug fix is optimized for the best gameplay on day-1. We have a discussion thread open on this driver here in our Nvidia driver discussion forums. For those that wonder, the DCH driver is a Microsoft DCH (Declarative Componentized Hardware supported apps) driver and refers to a new Windows 10 driver package preinstalled by OEMS implementing the Microsoft Universal Driver paradigm. While the base core component files remain the same, the way DCH drivers are packaged differs from previous Legacy (Standard) drivers. We recommend the normal driver and not DCH.

This new Game Ready Driver provides support for the launch of Outriders, which features NVIDIA DLSS technology.

Additionally, this release also provides optimal day-1 support for:

• DIRT 5's new ray tracing update
• The launch of Evil Genius 2: World Domination
• The launch of the KINGDOM HEARTS Series on the Epic Games Store

Gaming Technology

• Includes support for Resizable BAR across the GeForce RTX 30 Series of Desktop and Notebook GPUs
• Includes beta support for virtualization on GeForce GPUs

New Features and Other Changes

OpenCL 3.0

Added support for OpenCL 3.0, the latest major version of OpenCL maintaining backward
compatibility with OpenCL 1.2. NVIDIA OpenCL 3.0 continues to support existing OpenCL 1.2 functionality as well as Khronos and vendor extensions that are already supported with NVIDIAOpenCL 1.2 drivers.

• The following new features beyond existing NVIDIA OpenCL 1.2 features are supported by NVIDIA OpenCL 3.0
  
  • RGBA vector component naming in OpenCL C kernels
  • pragma_unroll hint
  • opencl_3d_image_writes
  • clCreate*WithProperties APIs which can be used as replacement for existing clCreateBuffer/Image APIs.
  • clSetContextDestructorCallback
  • clCloneKernel from OpenCL 2.1
  • clEnqueueSVMMigrateMem from OpenCL 2.1

• Incorporates the following experimental 2.0 features:
  • Device side enqueue
    • The current implementation is limited to 64-bit platforms only.
    • OpenCL 2.0 allows kernels to be enqueued with global_work_size larger than thecompute capability of the NVIDIA GPU. The current implementation supports onlycombinations ofglobal_work_size and local_work_size that are within the computecapability of the NVIDIA GPU.
    • The maximum supported CUDA grid and block size of NVIDIA GPUs is available at http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#compute-capabilities.
      • For a given grid dimension, the global_work_size can be determined by CUDA grid size x CUDA block size.
    • For executing kernels (whether from the host or the device), OpenCL 2.0 supports non-uniform ND-ranges where global_work_size does not need to be divisible by the local_work_size. This capability is not yet supported in the NVIDIA driver, and therefore not supported for device side kernel enqueues.
• Shared virtual memory
  
  • The current implementation of shared virtual memory is limited to 64-bit platforms only.

Application Profiles
Added the following SLI profiles.

• Shenmue III (NVIDIA Turing GPUs only)
• The Medium (NVIDIA Turing GPUs only)

Other Changes
Added support for extended DP-to-HDMI 2.1 PCON clock frequency range to better support 8K TVs.
Changed the G-SYNC on-screen status indicator to be less obtrusive.

Check the release notes (pdf)