Hi, I have a question!
When i export to Maya, my model isn’t same quality. I tried already vector displacement map. and I added sub-level in Approximation Editor. I setup Approx method spatial, level 3-7 and Disabled filter type but still bad. How can I import it as same as quality to Maya
Thanks,
Onder.
If your texture is 2048, and the cube has 6 sides, that means that each face is getting 341 pixels (assuming that they are all even and fill the entire map). Knowing that, if you view your object on a 1080 monitor you’re going to see pixellation because you’re viewing 341 pixels stretched to 1920x1080, if that makes sense. I’m sure there’s lots of other factors at play but that’s just one thought. Could try upping to a 4k texture if the object is going to be seen that close, or use multiple maps for objects, like a 2k for a face and then another 2k for the body, or Mari has some solutions where you can use like a full 2k map for chunks of the face to get some extreme detail, but I’ve never done it personally.
Thank you, I’m gonna try 4k and multiple maps. I hope that be getting better. I saw some 3D models on internet and very high detailed.
No problem! Think of it like editing an image in photoshop. If you take a photo off the web that looks good and scale it up, it will look bad. If you scale it down it looks fine. Same applies for a 3D model since that’s exactly what you’re doing with the texture except in a 3D space. So if you have a character and his face is a small part of a 1024x1024 map and you show a closeup at HD 1080 resolution, you will see pixellation. This is why people sometimes use seperate maps for closeups or multiple maps, etc.
Also check that you’re dividing your model enough to have the same resolution as it had in Zbrush. So if you have a low res model in zbrush and 8 subdiv levels, you need to subdivide it 8 times with displacement map as well to get the same number of Polys back, or somewhere close and a normal map can do the trick.
Also consider using decimation if you can get away with it. It will put the detail into the areas that need it and you may not need to use displacement maps.
yes, decimation good tool but, when I try to rig some objects it getting mess. maya need low and quadratic face. do you have any idea about xnormal. why people chose xnormal instead of zbrush?
thanks.
@ purehilarity
For the correctness.
We have 682 Pixels per cube side in a 2048 x 2048 map (if you don’t stretch the UV polygons of the cube) because you have 3 columns and 2 rows (or viz versa) to fill the 2048 x 2049 (4194304 Pixels) space.
2048 / 3 = 682,667 Pixels per UV polygon
682 x 682 x 6 = 2790744 Pixels of 4194304 Pixels (ca. 66,5% of the Pixels in the texture)
You can optimize this ratio if you divide the cube e.g. one time to get more UV polygons (24 instead of 6).
Then you can spread this UV Polygons in a better way in the UV space. (e.g in 5 columns and 5 rows)
2048 / 5 = 409,6 Pixels per UV polygon
409 X 409 x 24 = 4014744 Pixels of 4194304 Pixels (ca. 95,7% of the Pixels in the texture)
But in general you’re absolutely right.
CU.
