Member LucaFox Posted August 3, 2017 Member Report Share Posted August 3, 2017 (edited) Hello everyone! It's me... again. Honestly, sorry for that. This time, I'm experiencing a huge problem with the high to low bake. I've made several, several experiments to solve it, I've improved it, I've learned a lot of things about exporting options, optimal decimations and settings on Substance (I love to integrate these two softwares together SO much)... but I couldn't solve it. At first I thought it was all about bad UVs (and my UVs are bad, believe me), but I tried to do some experiments and I've seen it doesn't seem to be UVs' fault! I've came to the conclusion that it must be the bad geometry of the lowpoly model... I think. But before retopologizing all the model from scratch (I'm gonna try 3DCoat retopo tools, played a little with 'em and liking 'em!), I'd really, really love to ask the advice of more experienced, professional people. I'm not sure if this can be helpful, but there's my workflow: lowpoly in Blender, import in 3DCoat, voxelization/sculpt/etc in 3DCoat, export high from 3DCoat, UV map the lowpoly and exporting it from 3DCoat, importing lowpoly UV mapped in Substance Painter, baking the textures (only the normal, actually) with highpoly and optimal settings also in Substance Painter. The point is: is it effectively a bad lowpoly/retopo mesh problem, or I'm missing something important? Hope you can help! Sorry again for all these questions, and thank you in advance for helping me out! Here the screens to show the exact problem (here you can see the bake result in Painter, the highpoly in 3DCoat, the export settings in 3DCoat, the lowpoly also in 3DCoat). Edited August 3, 2017 by LucaFox Quote Link to comment Share on other sites More sharing options...
Carlosan Posted August 3, 2017 Report Share Posted August 3, 2017 Are you using latest version 4.7.35 ? Quote Link to comment Share on other sites More sharing options...
Member LucaFox Posted August 3, 2017 Author Member Report Share Posted August 3, 2017 Looks like I'm still using 4.7.24 Carlosan, could really be that the problem? Quote Link to comment Share on other sites More sharing options...
Carlosan Posted August 3, 2017 Report Share Posted August 3, 2017 Some new features was added to new versions 10.03.2017 4.7.26 [beta] - Retopo->Bake->Baking scan settings - fixed crash, save/load available, symmetry works. - Update islands issue resolved too. 8.03.2017 4.7.25 [beta] - Retopo->Bake->Baking scan settings gone very important addition. You may paint baking depth with brush. - Autopo problem fixed - if no mesh generated, hidden retopo mesh will not be corrupted. Quote Link to comment Share on other sites More sharing options...
Member LucaFox Posted August 3, 2017 Author Member Report Share Posted August 3, 2017 Hey there! I tried and my problem persists I'm checking if 3DCoat's bake works (instead of using Substance Painter), but I shouldn't have these issues A friend of mine said something about "crease edges" functions, maybe something like that? The problems effectively looks related to sharp edges in the lowpoly Quote Link to comment Share on other sites More sharing options...
Reputable Contributor digman Posted August 3, 2017 Reputable Contributor Report Share Posted August 3, 2017 (edited) I would suggest to go to Polycount wiki and read everything you can get your hands on about normal maps. Lots of good information there from modeling for normal maps to the baking of them. 3DC and all the other baking application generally do a good job at baking. Most problems in normal map baking are from the model itself or the uv seam layout or a combo of them both. http://wiki.polycount.com/wiki/Polycount http://wiki.polycount.com/wiki/Normal_map Question. After creating your uv seams for the low polygon in the retopo room why are you not just directly exporting that model out of the retopo room for your low polygon model for baking in SP. Edited August 3, 2017 by digman Quote Link to comment Share on other sites More sharing options...
Member LucaFox Posted August 4, 2017 Author Member Report Share Posted August 4, 2017 8 hours ago, digman said: [...] Question. After creating your uv seams for the low polygon in the retopo room why are you not just directly exporting that model out of the retopo room for your low polygon model for baking in SP. That's because I revisited the UVs few times checking if everything was OK, and at a certain point I just started trying to export the model with different options, thinking that could be the problem! As said above, I was pretty feared that the problem could be the bad lowpoly... I'll check the links you kindly shared and retopo it correctly, I don't think that there will be problems! Thank you for confirming my doubts! Quote Link to comment Share on other sites More sharing options...
Carlosan Posted August 4, 2017 Report Share Posted August 4, 2017 The point is: is it effectively a bad lowpoly/retopo mesh problem. Can you share a wireframe of your retopo mesh ? Reference: The red cube has really tight edges, which results in a very tight bevel on the edge, which will look awful when baked. The blue cube has nice fat edge bevels, which will bake much cleaner. Quote Link to comment Share on other sites More sharing options...
Carlosan Posted August 4, 2017 Report Share Posted August 4, 2017 Suggested reading http://polycount.com/discussion/167576/normal-map-baking-hard-edge-problem http://polycount.com/discussion/107196/youre-making-me-hard-making-sense-of-hard-edges-uvs-normal-maps-and-vertex-counts Quote Link to comment Share on other sites More sharing options...
Member LucaFox Posted August 4, 2017 Author Member Report Share Posted August 4, 2017 -SOLVED- Hey my friends! I solved the problem. The lowpoly mesh is meh. The uv is meh. The baking can probably be done better (even if I spent the last 2 weeks studying the baking options and what they're about because of this problem). But I couldn't understand exactly what I was doing terribly wrong... because I wasn't doing anything so terribly wrong. I just tried to bake in 3DCoat... worked flawlessly. ****. 3DCoat +1 ... again. 1 Quote Link to comment Share on other sites More sharing options...
Carlosan Posted August 4, 2017 Report Share Posted August 4, 2017 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.