What's new in Nsight Perf SDK 2025.1

1.   Support of GeForce RTX 50 Series Graphics Cards.

2.   Enhanced single pass metrics for Periodic Sampling on RTX 50 series. Improvements in hardware allow for a larger number of metrics to be collected at once improving Top-Level Triage work flows.

3.   Exposed more metrics (e.g. bytes) on RTX 50 Series Graphics Cards.

4.   User-Defined Metrics API allows users to define their own Derived Metrics and utilize those metrics directly with the existing Scheduling and Evaluation Nsight PerfSDK APIs. For more information, see the Nsight Perf SDK Getting Started Guide.

5.   RawCounterConfig API is a direct replacement for the existing RawMetricsConfig API that provides advanced hardware scheduling features and allows precise triage metric group specification. For more information, see the Nsight PerfSDK Getting Started Guide.

6.   Together with the User-Defined Metrics API and RawCounterConfig API, allows the Nsight Perf SDK to now provide suggested triage metric groups specified in Metrics Configuration Files. For more information, see the Nsight PerfSDK Getting Started Guide.



Known Issues

Newly Disclosed Issues


1.    Metrics:

  • pcie__*/host__* metrics cannot be collected on R570 driver and fix for them are available in the R575+
  • syslts__average metrics cannot be collected in range profiler.

2.    Vulkan Samples:

  • vk_raytracing_tutorial_KHR sample has observed a segfault issue with older vulkan loader version 1.3.224.1 upgrading to the latest version 1.4.304.0 fixes the issue.

3.    Clock controls:

  • Newly added NVPW_DEVICE_CLOCK_SETTING_LOCK_TO_TURBO_BOOST might report failure on R570 and fix is forthcoming on upcoming driver release.

Previously Disclosed Issues


1.    General Issues

  • prop__latez_input_samples can be negative on Turing and Ampere GPUs when the measured workload has depth test disabled.

2.    Vulkan Profiler

  • vkQueueWaitIdle is required before present when profiling Vulkan applications.

3.    Vulkan Samples

  • The hint “Hit to begin…” on the title bar is not shown on some of the Linux systems.

4.    The Range Profiler API does not support NVLINK-enabled systems, so the HtmlReport examples will not work properly on these systems. Users should disable NVLINK before using related features. However, HudCounters and One-shot Sampling based on the Device Periodic Sampler API are not affected.



Limitations

General Limitations


1.    Nsight Perf SDK does not support SLI systems.

2.    Each GPU is limited to one profiling session at a time, across the entire system. However, multiple GPUs can be independently profiled concurrently. This is a driver limitation.

3.   NVPW_Device_SetClockSetting() performs a system-global clock change, that is not undone at process exit. The *SetDeviceClockState() family of NvPerfUtility functions have the same behavior, as they are convenience functions around NVPW_Device_SetClockSetting().

4.   Not all counters and metrics are available to the Range Profiler APIs, due to driver limitations. The CounterAvailability APIs exist to reveal which metrics are available.

5.   Quadro GV100 and Titan V are not supported starting from v2022.1.

6.   32 bit X86 is not supported starting from v2022.1.


D3D12 Profiler


1.   D3D12 Range Profiler is not compatible with the D3D12 debug layer.

2.   D3D12 Range Profiler does not support PushRange and PopRange commands on Bundles (Command Lists created with D3D12_COMMAND_LIST_TYPE_BUNDLE).


Vulkan Profiler


1.   Vulkan Range Profiler is not compatible with Vulkan layers that perform object-wrapping.

2.   Vulkan Range Profiler does not support PushRange and PopRange calls on secondary command buffers (VkCommandBuffer created with VK_COMMAND_BUFFER_LEVEL_SECONDARY.

3.   Vulkan Range Profiler does not support PushRange and PopRange calls on a VkCommandBuffer that is recording with VK_COMMAND_BUFFER_USAGE_SIMULTANEOUS_USE_BIT.


D3D11 Profiler


1.   D3D11 Range Profiler does not support PushRange and PopRange calls on a Deferred Context.