sdli1995/dlssg_for_sm86

Here is a dlssg for RTX30 Series GPU

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 36 minutes ago
Added to GitGenius on September 19th, 2026
Created on September 7th, 2026
Open Issues & Pull Requests: 143 (+1)
GitHub issues: Enabled
Number of forks: 209
Total Stargazers: 3,626 (+4)
Total Subscribers: 42 (+1)

Repository Insights (GitGenius)

Median issue/PR response: 2.8 hours
Mean response time: 13.8 hours
90th percentile: 31.1 hours
Tracked items: 154

Most active contributors

Sign in to see contributor activity.

How this project is maintained

About 16% of issues opened in the past year have never received a reply. 100% of open issues come from outside the core team, so the backlog reflects real-world use rather than internal planning. Almost all tracked open issues have seen activity in the last three months. Only 23% of issues opened in the past year have been closed.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 140
New in 7 days: 74
Closed in 7 days: 25
Avg open age: 7 days
Stale 30+ days: 0
Stale 90+ days: 0

Recent activity

Opened in 7 days: 67
Closed in 7 days: 24
Comments in 7 days: 25
Events in 7 days: 31

Top labels

  • question (7)
  • enhancement (5)
  • bug (4)
  • duplicate (4)
  • Compatibility (3)
  • documentation (1)

Detailed Description

DLSSG for SM86 is a proxy DLL that enables NVIDIA DLSS frame generation on RTX 30 and RTX 20 series GPUs. The tool works by intercepting game calls to NVIDIA's NGX runtime, rewriting GPU architecture information to make older hardware report as newer architectures, and embedding modified inference kernels that implement frame generation on unsupported GPUs. It operates on Windows x64 with Direct3D 12 games through two files: a version.dll proxy and a configuration INI file.

The problem it solves is that DLSS frame generation is officially restricted to RTX 40 series and newer hardware. The tool's approach uses architecture spoofing combined with custom inference kernels to make RTX 30 and RTX 20 series cards capable of running frame generation. The inference kernels have been rewritten to produce output bitwise identical to official DLSS-G on RTX 30 hardware, eliminating quality loss while providing modest performance gains. The tool supports up to 6X frame generation on games with Dynamic MFG support, configurable through an INI file with multiple optimization levels trading speed against output fidelity.

Developers considering adoption should understand this is a workaround for unsupported hardware rather than an official feature. It suits users with RTX 30 or RTX 20 series cards who want frame generation in games that officially require newer GPUs. The tool requires manual installation as a proxy DLL and configuration through an INI file. Vulkan support is not planned due to compatibility concerns with the wrapper layer itself.

Development has shifted away from kernel optimization toward compatibility fixes and bug resolution. Future work focuses on INT8 quantization optimization for Transformer upscaling models to improve baseline frame rates on older hardware rather than further frame generation kernel improvements.