Writing
Articles and experiments
Notes on rendering, GPU performance, and the measurements behind my projects. 15 articles so far.
-
GPU performance
GPU memory experiments: a guide to the 45-study series
A nine-part tour of 45 Vulkan compute experiments, from dispatch timing and data layout to subgroup algorithms and rendering workloads.
-
GPU performance
GPU memory experiments 40–45: queues, fusion, and reproducibility
Persistent task claims, portable aggregates, dispatch fusion, layout sensitivity, and a reproducible single-GPU access signature.
-
GPU performance
GPU memory experiments 35–39: occupancy and rendering data
What shared-memory declarations, node records, frame remapping, visibility counters, and tiled light assignment actually measure.
-
GPU performance
GPU memory experiments 30–34: subgroups, transpose, and radix sort
Portable subgroup indexing, block-local primitives, tiled transpose, and the cost of a complete GPU radix sort.
-
GPU performance
GPU memory experiments 26–29: alignment, heaps, and shared memory
Four Vulkan studies separate contiguous offsets, access permutations, staging costs, and shared-memory stride from the hardware mechanisms their names suggest.
-
GPU performance
GPU memory experiments 21–25: reduction, scan, and compaction
Reduction, prefix sums, histograms, compaction, and spatial binning show why ordering contracts and auxiliary traffic belong beside GPU timings.
-
GPU performance
GPU memory experiments 16–20: tiling, live state, and barriers
Five Vulkan experiments connect shared-memory reuse, tile size, temporary-state scheduling, branch patterns, and synchronization to measured work.
-
GPU performance
GPU memory experiments, part 3: addresses matter as much as bytes
Experiments 11–15 explore stride, gather, atomic scatter, scheduled reuse, and bandwidth curves while keeping their different address and validation contracts visible.
-
GPU performance
GPU memory experiments, part 2: layout, padding, and precision
Experiments 06–10 compare matched particle updates, record footprints, and manually packed formats—and expose why a shared correctness oracle can miss a conversion defect.
-
GPU performance
GPU memory experiments, part 1: dispatch is part of the measurement
Experiments 01–05 establish what a dispatch, a byte count, and a contiguous baseline actually mean before comparing GPU memory layouts.
-
GPU performance
AoSoA and blocked layouts: measuring the tradeoffs
A comparison of AoS, SoA, and blocked AoSoA layouts, including the performance and integration tradeoffs.
-
GPU performance
GPU memory benchmarks: the first five experiments
Five experiments establish a baseline for dispatch overhead, workgroup size, contiguous bandwidth, and indexing costs.
-
GPU performance
AoS vs SoA: measuring GPU memory access
A measured comparison of AoS and SoA particle layouts, with an explanation of how field access affects GPU memory traffic.
-
Vulkan & rendering
Building a Vulkan 1.4 scene renderer
Forward and deferred rendering, scene inspection, and GPU profiling in a C++23 and Vulkan 1.4 scene engine.
-
Vulkan & rendering
Measuring Vulkan workloads with vk-bench
A small Vulkan benchmark for comparing CPU and GPU timings across controlled graphics and compute workloads.