A video encoder comprises a loop filter to filter luminance and chrominance pixel values, first and second loop filter working buffers accessible to the loop filter, and ping and pong loop filter data buffers accessible to the loop filter and to a direct memory access (DMA) engine. The loop filter filters pixels about a plurality of vertical edges and a plurality of horizontal edges for each macroblock in a video frame. The loop filter distributes partially filtered luma and chrominance pixel values across the first and second loop filter working buffers as well as the ping and pong loop filter data buffers, and does not save partially filtered luma and chrominance pixel values to external memory via the DMA engine.