When my model was done, I started unwrapping the UVs so I could texture it. Firstly, I tried using the auto seams feature on all the objects in my scene and then unwrapping them. When each object was unwrapped, I selected the whole model and viewed it in the UV editor. I used the layout tool to stop the UV shells from overlapping, then saved the model as an FBX file.
When I imported the model to Substance Painter, the salt packet was broken. It would only be visible from some angles and I couldn’t paint on it properly. I asked my group members if they’d encountered this problem before but they had not, so I looked online to see if I could figure out the problem.
One suggested fix was to use the cleanup tool to search for n-gons/faces with more than 4 sides. I did this and found several n-gons, so I used the cleanup tool to get rid of them (the tool did this by splitting the n-gons into triangles). I tried saving and exporting this model, but again it didn’t work. Another fix using the cleanup tool was to look for non-planar faces. I used the cleanup tool again to do this, and exported that version of the model. Again, this did not fix my problem.
I tried many different things to fix this issue. First of all I wondered if it was because the salt packet and shell were intersecting each other, so I moved the packet and widened the gap under the rubber band so the shell and packet would not intersect. I saved this as an FBX again and opened it in Substance, but it made no difference. Next, Rhys suggested that the UV might be inverted (the surface is inside the object rather than outside, explaining why it didn’t show up in Substance). I looked online and found that other people had suggested this too, and tried the solution they suggested. I used the flip tool in the UV editor to invert the salt packet’s UV. The packet appeared black since the shells were now back facing, but I tried importing this version of the model into Substance to see if it would work anyway. It did briefly and I was able to paint on the back facing shells, although the colours didn’t display properly (they were very dark). Not knowing what the problem was and thinking this was the best outcome I was going to get, I went ahead and textured the whole model to avoid wasting more time.
Speaking to Alec, he told me that my issue was likely because I used the auto seams tool instead of manually cutting the seams. He showed me how to do this, and I was able to make new seams for my model. This fixed the issue I was having with the salt. It meant I had to redo all my textures, but as long as the model worked I didn’t mind doing this again.
When I tried to bring my textures back into Maya, none of them displayed properly on the model. I asked my group about this and Rhys had a look at my model. He told me I should have the UVs for each object as a separate map, rather than having all of them in one map. This is why they weren’t displaying properly, and also why some areas of the model looked very low-res when I was working in Substance. Since I had been struggling a lot with UV mapping, he offered to fix my UVs for me which I was very appreciative of. I watched him separate the UV shells into different maps as he explained what he was doing. When the model was fixed, he sent me his version of the model and I opened it in Substance. I tried to replace the model in my Substance Painter file while still keeping the textures I’d made, but since the UV maps were completely different I had to start from scratch again.
