On AI Lounge since Nov 2026
Emion Raschke
Member #5,818DeveloperL1AMD NR Developer, No company
HexNeural brings nvidia models to older rdna 2 hardware. Rdna 2 has no FP8, no Matrix cores, no BF16, no dual-issues and no sparsity.
How karma works. Everyone starts with 5. You get +1 each time another member upvotes one of your posts or comments (your own upvotes don't count), and lose it if they take the upvote back. Post what you're building, answer questions and help other members to earn more.
Achievements
Plugged In ×2Connected sources: GitHub, Hugging Face, LinkedIn, X, website.How: Connected 2 sources (GitHub, LinkedIn, X or a website) in Settings
Plugged In×2Featured
ShipperProjects on your profile.How: Added 1 project to their profile
Shipper
No posts yet
Emion Raschke hasn't posted yet.
Pinned projects
NVIDIA DLSS Neural Rendering ported from gfx1100 (RDNA3) to gfx1030 (RDNA2, RX 6900 XT) via disassembly-derived kernel translation. A full captured game frame now renders correctly end to end (S_mid +0.9247, no colour cast). Performance optimization in progress. Research only, not a mod.
About
Emion Raschke is an AMD NR developer based in Erfurt, Germany, who builds GPU software for neural rendering. Their public C++ project HexNeural brings NVIDIA DLSS 5 neural rendering and DLSS 4 Frame Generation to Radeon RX 6000 cards. Current work ports an NVIDIA DLSS neural rendering model from RDNA 3 to RDNA 2 hardware, with performance tuning still in progress.