Sparse matrix-vector multiplication (SpMV) is a key operation across high-performance computing, graph analytics, and many more applications.In these applications, the matrix characteristics, notably non-zero elements per row, can vary widely and impact which algorithm performs best.Thus, Graphics Processing Unit (GPU) SpMV algorithms often rely on costly preprocessing to determine what per-row algorithm to select to achieve high performance.In this work we combine SpMV preprocessing and the subsequent per-row processing on the GPU by leveraging the novel "Work Graphs" GPU programming model-initially designed for graphics applications-for dynamic on-device self-scheduling.Work Graphs allow for fine-grain dataflow execution of individual workgroups using emerging hardware and firmware support.As soon as preprocessing has generated sufficient work, workgroups of individual processing kernels are self-scheduled and executed, interleaved with those of other kernels.This improves cache locality and eliminates host interaction altogether.Across a suite of 59 sparse matrices, the best of various novel Work Graphs SpMV implementations outperforms state-of-the-art rocSPARSE "LRB" for a single SpMV by up to 7.19× (mean: 3.35×, SD: 1.89).Furthermore, it achieves much more stable performance across various sparsity patterns than the rocSPARSE CSR-General algorithm, and even beats the advanced rocSPARSE CSR-Adaptive algorithm for up to 92 consecutive SpMV calculations.In addition, compared to rocSPARSE LRB, it reduces code complexity by 75%.Its memory footprint for supporting data structures is a fixed ∼25 MiB independent of matrix size, compared to rocSPARSE LRB's data structures that scale with matrix

Abstract Oriented bounding box (OBB) bounding volume hierarchies offer a more precise fit than axis‐aligned bounding box hierarchies in scenarios with thin elongated and arbitrarily rotated geometry, enhancing intersection test performance in ray tracing. However, determining optimally oriented bounding boxes can be computationally expensive and have high memory requirements. Recent research has shown that pre‐built hierarchies can be efficiently converted to OBB hierarchies on the GPU in a bottom‐up pass, yielding significant ray tracing traversal improvements. In this paper, we introduce a novel OBB construction technique where all internal node children share a consistent OBB transform, chosen from a fixed set of discrete quantized rotations. This allows for efficient encoding and reduces the computational complexity of OBB transformations. We further extend our approach to hierarchies with multiple children per node by leveraging Discrete Orientation Polytopes (k‐DOPs), demonstrating improvements in traversal performance while limiting the build time impact for real‐time applications. Our method is applied as a post‐processing step, integrating seamlessly into existing hierarchy construction pipelines. Despite a 12.6% increase in build time, our experimental results demonstrate an average improvement of 18.5% in primary, 32.4% in secondary rays, and maximum gain of 65% in ray intersection performance, highlighting its potential for advancing real‐time applications.

Abstract We present edge‐friend, a data structure for quad meshes with access to neighborhood information required for Catmull‐Clark subdivision surface refinement. Edge‐friend enables efficient real‐time subdivision surface rendering. In particular, the resulting algorithm is deterministic, does not require hardware support for atomic floating‐point arithmetic, and is optimized for efficient rendering on GPUs. Edge‐friend exploits that after one subdivision step, two edges can be uniquely and implicitly assigned to each quad. Additionally, edge‐friend is a compact data structure, adding little overhead. Our algorithm is simple to implement in a single compute shader kernel, and requires minimal synchronization which makes it particularly suited for asynchronous execution. We easily extend our kernel to support relevant Catmull‐Clark subdivision surface features, including semi‐smooth creases, boundaries, animation and attribute interpolation. In case of topology changes, our data structure requires little preprocessing, making it amendable for a variety of applications, including real‐time editing and animations. Our method can process and render billions of triangles per second on modern GPUs. For a sample mesh, our algorithm generates and renders 2.9 million triangles in 0.58ms on an AMD Radeon RX 7900 XTX GPU.

Handling many light sources in real-time is still one of the big challenges in real-time graphics [Sousa and Geffroy 2016; Garawany 2016]. Even the most recent approaches like practical clustered shading still have various problem cases with low performance. Especially in scenes with high depth variance, existing algorithms cannot adapt to the distribution of light sources properly and end up evaluating many lights that don't contribute to the final image.

Subpixel Reconstruction Antialiasing (SRAA) combines singlepixel (1x) shading with subpixel visibility to create antialiased images without increasing the shading cost. SRAA targets deferred-shading renderers, which cannot use multisample antialiasing.

Virtual environments are typically textured by manually choosing an image to apply on each surface. This implies browsing through large sets of generic textures for each and every surface in the scene.