PlayReady analog output cap fails on RTX 5050 Laptop

Gaming and Visualization Technologies General Topics & Other SDKs Drivers - Linux, Windows, MacOS
1

Summary: On an RTX 5050 Laptop GPU, hardware PlayReady (SL3000) licenses that contain an explicit analog video output restriction fail with 0x8004CD22 (DRM_E_TEE_OUTPUT_PROTECTION_REQUIREMENTS_NOT_MET) on every NVIDIA output. The same licenses play on the Intel iGPU in the same laptop. HDCP itself works on NVIDIA. Real-world effect: Disney+ in Edge falls back from SL3000/1080p to SL2000/720p on any NVIDIA-driven screen, because its SL3000 licenses carry these restrictions.

System

  • Lenovo Legion 5 15IRX10 (83LY), GeForce RTX 5050 Laptop GPU + Intel UHD
  • Windows 11 25H2, build 26200.9457
  • NVIDIA drivers 616.92 and 616.56 (same result on both)
  • Microsoft Edge 154 (hardware PlayReady via Media Foundation)

Repro (no streaming service needed)

  1. Local EME page on http://localhost, CENC (AES-CTR) H.264 1080p clip, key system com.microsoft.playready.recommendation, robustness 3000.
  2. Licenses from Microsoft’s test server, e.g.
    https://test.playready.microsoft.com/service/rightsmanager.asmx?cfg=(persist:false,contentkey:<key>,kid:<kid>,sl:3000,ucvopl:300,avop:(guid:D783A191-E083-4BAF-B2DA-E69F910B3772,data:QO8HAA==))
  3. Edge on the NVIDIA GPU, window on an NVIDIA-driven screen.
License policy (SL3000) NVIDIA output Intel output
no output restrictions plays plays
ucvopl:300 (HDCP required) plays plays
+ avop 811C5110-46C8-4C6E-8163-C0482A15D47E (component, 520000 px) 0x8004CD22 plays
+ avop D783A191-E083-4BAF-B2DA-E69F910B3772 (computer monitor, 520000 px) 0x8004CD22 plays

Fails on NVIDIA via HDMI, via USB-C DisplayPort, and on the internal panel in dGPU-only (MUX) mode, so it isn’t tied to a connector. The same USB-C monitor plays everything when plugged into the Intel-wired USB-C port.

What I found

  • ETW (Microsoft-Windows-Media-Protection-PlayReady-Performance): in the failing case CBlackboxWithTEECreateDecryptor calls the OEM TEE (OemTeeProxyMethodInvoke) method 15 and then returns 0x8004CD22; method 16 is never called. In the passing case it calls 15, then 16, and succeeds. No DxgKrnl warnings or errors in either case.
  • Windows’ display config shows the NVIDIA adapter enumerating a DVI output (connector type 4) that this laptop doesn’t have. The Intel adapter reports none. DVI can carry analog, so this may be why the TEE can’t guarantee the analog constriction, but that’s a guess.
  • Also tried: Power management mode “Prefer maximum performance” (no change).

Questions

  1. Does NVIDIA’s PlayReady TEE support explicit analog video output constriction on this GPU/driver?
  2. Could the phantom DVI output be making it refuse the policy?

Also submitted through NVIDIA App → Report an issue. Happy to share the ETW trace or test page.

2

Welcome @yoelvinitsky to the NVIDIA Developer Forums.

If you reported this through the NVIDIA App, then this is already tracked through an internal bug. If the engineers assigned need more information, they will contact you directly.

Thanks!