How long will it take to run the third part in the ./nerfactor

[wonjunior]

I believe this is detailed in the supplementary A.3:

We train NeRF for 2,000 epochs, which takes 6–8 h when distributed over four NVIDIA TITAN RTX GPUs. Prior to the final joint optimization, computing the initial surface normals and light visibility from the trained NeRF takes 30 min per view on one GPU for a 16 × 32 light probe (i.e., 512 light locations). This step can be trivially parallelized because each view is processed independently. Geometry pretraining is performed for 200 epochs, which takes around 20 min on a TITAN RTX. The final joint optimization is performed for 100 epochs, which takes only 30 min on one TITAN RTX.