Hardware components usage during processing - PIX4Dmatic
This article outlines the hardware components utilized during processing, detailing how CPU, GPU, and RAM resources are allocated for optimal performance. Understanding hardware usage helps fine-tune system configurations for better efficiency.
PIX4Dmatic is highly parallelized, leveraging multi-core CPUs, MMX/SIMD instructions, and NVIDIA GPU CUDA processing. However, resource utilization varies across different processing steps, and not all tasks benefit equally from multi-core or CUDA acceleration.
The table below categorizes the usage of each hardware component during the processing steps as High, Medium, or Low.
| Processing step | CPU | RAM | GPU | Storage drive |
| Calibration | High | Medium | Medium | Medium |
| Reoptimization | High | Medium | Low |
Medium |
| Dense point cloud | High | Medium | High | High |
| Depth & fuse | High | Medium | Low | High |
| Image pre-processing | High | Medium | High | Medium |
| Mesh | High | Medium | Low | Medium |
| DSM | High | Medium | Low | High |
| Orthomosaic | High | Medium | High | High |
Disclaimer: Hardware usage during processing depends on factors such as project size, image resolution, and system capabilities. For example, processing a large project with 2,000 high-resolution images on a laptop with an NVIDIA RTX 3050 Ti GPU and 32GB of RAM will likely maximize hardware utilization, especially for the GPU and RAM. Smaller projects or lower-resolution images require fewer resources, while high-end systems can handle larger workloads more efficiently. Using a Solid-State Drive (SSD) is strongly recommended for faster read/write speeds and improved processing performance.