AMD presented new research at the I3D Symposium titled "PEPS: Positional Encoding Projected Sampling", introducing a new method for positional encoding that can improve neural texture compression. How Does Neural Texture Compression Work? Neural Texture Compression works by training what are known as INRs, or “Implicit Neural Representations,” to learn coordinate-to-signal functions. By projecting texture coordinates into a higher-dimensional embedding and feeding this information to a multi-layer perceptron, it's possible to represent and compress textures significantly. PEPS: Positional Encoding Projected Sampling PEPS introduces a new method to improve the efficiency of this process by changing how positional encoding is used. […]

Read full article at https://wccftech.com/amd-peps-research-pushes-neural-texture-compression-cutting-model-parameters-by-25-percent-at-comparable-quality/

Intel's Nova Lake flagship may temporarily consume up to 474W under heavy boost workloads. Intel's Nova Lake Flagship 52-Core Chip Could Push PL2 Power to a Massive 474W as Intel Sends Revised Z990 Power Design Instructions With 3x 8-pin EPS Connectors More leaks are surfacing regarding Intel's upcoming Nova Lake lineup. As of today, we expect Intel's top-of-the-line Nova Lake S desktop lineup to feature up to a 52-core configuration, which is a massive jump over the current 24-core/24-thread configuration from Intel Arrow Lake S series. Intel did offer up to 32 threads in the Raptor Lake/Refresh lineup, but with […]

Read full article at https://wccftech.com/intel-nova-lake-dual-tile-cpus-reportedly-feature-up-to-474w-pl2-power-limit/