We have been using your RelityCapture software solution for several months. But at the moment we faced a problem.
We have made 1700 pictures for the 20 * 40 * 8 meters Building. “Vertex count per part” default value, which amounted to 5 000 000, was changed to 200 000. We didn’t use Simplification tool.
As the result we received obj file of more than 20GB size and number of triangles amounted to 180 000 000. We are unable to import such a file to 3ds max or zbrush as an error occurrs. The file imported to Meshlab, but it still seems impossible to work with it.
Please see computer hardware we used: Intel core i7 5820k, Geforce GTX 980, 32GB Ram. Please help us with the following:
- Which hardware should we improve: graphics card or RAM?
- Would you be so kind to give us recommendations and/or guidelines on creating high quality models of big size objects.
- What is the maximum size of the object in the real world can be and how many photos we need to take without dividing this object on separate parts? Are there any limits?
- Is it possible to make one model with several UV maps?
Thank you in advance! Looking forward to your prompt reply.
Hi Stanislav Bovkunovich
We have made 1700 pictures for the 20 * 40 * 8 meters Building. “Vertex count per part” default value, which amounted to 5 000 000, was changed to 200 000. We didn’t use Simplification tool.
As the result we received obj file of more than 20GB size and number of triangles amounted to 180 000 000. We are unable to import such a file to 3ds max or zbrush as an error occurs. The file imported to Meshlab, but it still seems impossible to work with it.
You need to use the SIMPLIFY tool and decimate the model to more workable size, say 15-17 million triangles is OK for Zbrush or other software for post-processing… or you need to cut the model in pieces…
Please see computer hardware we used: Intel core i7 5820k, Geforce GTX 980, 32GB Ram. Please help us with the following:
- Which hardware should we improve: graphics card or RAM?
- Would you be so kind to give us recommendations and/or guidelines on creating high quality models of big size objects.
- What is the maximum size of the object in the real world can be and how many photos we need to take without dividing this object on separate parts? Are there any limits?
- Is it possible to make one model with several UV maps?
Thank you in advance! Looking forward to your prompt reply.
- It is not related to this, it is a very well-known problem in post-processing to be able to edit such large data… So adding anything like this will NOT solve this problem. Contact me at milos.lukac@capturingreality.com for discussion on this…
- We can offer you an onsite/offsite training for proper workflows as it is quite complex stuff…
- The planet Earth
RC can process over 100.000+ images quite easily: https://www.youtube.com/watch?v=dmfjTSdQUG4
- Yes, just look at the UNWRAP tool and set the desired amount of UVs
Thank you very much, Wishgranter! I’ll write you soon.
- the planet Earth
RC can process over 100.000+ images quite easy… https://www.youtube.com/watch?v=dmfjTSdQUG4
what size was the model for this before simplification?
and did you run that out on high?
Hi Chris
what size was the model for this before simplification? and did you run that out on high?
That knows only our first customer who has done this for a state-level project, which is still under NDAs so we do not know any precise information on this…