Forums

And now something new has arrived. BREP and NURBS are computationally costly and unstable. Thus we now have nTopology which is based on Signed Distance Fields, SDF. https://www.ntop.com/resources/blog/how-implicits-succeed-where-b-reps-fail/ https://www.ntop.com/resources/blog/implicit-modeling-for-mechanical-design/ For B-rep modelers, answering the containment question means checking a point against every piece of topology in the model. A common approach is to create a ray that begins with the point in question and travels in any direction. If the number of intersections is odd, the point is inside the shape, otherwise it is outside. Such a calculation is expensive and error-prone, as faces can get missed or double counted if struck near their edges or near tangent to edges or faces. Although new techniques mitigate this problem for mesh modelers, B-rep and mesh modelers still struggle with this basic test. Implicits use a sign convention to designate inside versus outside, so containment is a trivial test. Values closer to zero are closer to the boundary, so one can establish if a point is within tolerance of the boundary. In both accuracy and performance, implicits are better at the containment test.   Though they continue to be the most widely used and successful representation for geometry, B-rep systems have some significant flaws that make them unsuitable for computational design. First, their architecture has remained mostly unchanged since the 1980s, so they are not suited to parallel computing, especially on GPUs. In a typical B-rep system, the GPU is relegated to rendering triangles spit out from a geometry kernel running single-threaded on one core of a CPU. More importantly, B-rep calculations can fail for a variety of reasons. Heroic efforts over the past four decades have improved reliability, but it’s still unsatisfactory, and it’s not likely to get much better. In implicit modeling, the approach is completely different: the shape of an object is described by a mathematical function that returns the distance to the closest point on its surface. The function is constructed so that it is negative inside the object, positive outside, and zero on its surface, so it's called a Signed Distance Function (SDF). Am I asking the devs here to do this? No I'm not but this is very interesting stuff...worth having a look at for sure. By the way, Gemini AI told me that buyers have reported that nTop pro license sells for $31,000.00 or so. https://www.ntop.com   I had a conversation with Gemini AI about this;   https://i.imgur.com/xy9aN5C.jpeg  

Bien sûr, de nombreuses ressources sont disponibles pour apprendre à utiliser 3DCoat, allant des tutoriels officiels aux cours en ligne spécialisés.  Voici où vous pouvez trouver des tutoriels pour débuter avec 3DCoat :   Ressources Officielles Le site officiel de 3DCoat propose une section dédiée aux tutoriels vidéo, idéale pour commencer.  Tutoriels vidéo 3DCoat officiels : Vous pouvez consulter ces leçons directement sur le site officiel de 3DCoat. Ces vidéos couvrent les bases et les différentes fonctionnalités du logiciel. Documentation : Une documentation complète est également disponible pour référence, offrant des explications détaillées sur les menus et les espaces de travail.    Chaînes YouTube Plusieurs chaînes sur YouTube se spécialisent dans les formations 3D, dont certaines sont dédiées à 3DCoat.  Chaîne de formation 3DCoat (officielle) : Le logiciel dispose d'une chaîne de formation officielle sur YouTube (3DCoat Training Channel) qui propose de nombreux guides pour les débutants. Playlists de tutoriels : D'autres playlists gérées par des utilisateurs ou des formateurs indépendants, comme celle-ci ou celle-ci, peuvent être très utiles.    Cours en Ligne Pour un apprentissage plus structuré et approfondi, vous pouvez opter pour des plateformes de cours payants, souvent avec des formateurs certifiés. Udemy : Des cours complets, tels que "Maîtriser 3DCoat" de Formajeux, sont disponibles sur Udemy. Ces cours sont souvent conçus pour les débutants absolus et couvrent tous les aspects, de la sculpture à la peinture de textures. Skillshare : Cette plateforme propose également des guides complets pour les nouveaux utilisateurs souhaitant apprendre la sculpture et la peinture dans 3DCoat.  En tant que débutant, il est recommandé de commencer par les tutoriels vidéo officiels pour comprendre les bases de l'interface et des outils. Bon apprentissage !

I'll try to find out from the developers, since your workspace file works fine for me.

bonjour, je m'appel jacques nouveau sur 3d coat, quelqu'un connait ou je peut trouver des tutoriel pour apprendre à utiliser 3dcoat svp.

Yes, we could nitpick a few areas in each example, but it seems clear to me that Quadremsher is more focused on strictly even quads, where 3DCoat's Auto-Retopo algorithm focuses on more optimized meshes. You can see this in the area you called stretched. It looks that way when you expect all evenly spaced quads. But, under further examination, 3DCoat is simply prioritizing spacing and placing more polygons where they are needed. The "stretched" area is actually just a flat surface area that doesn't need so many polygons. I prefer that, rather than forcing evenly spaced quads. Adding stroke guides was actually my idea, back in 2009, when Auto-Retopo was first being introduced into 3DCoat. It was the first Auto-Retopo in the industry and ZBrush copied it with Z-Remesher a few years later. I am sorry to disagree with you, but the ability to dictate the topology with stroke guides IS NOT a fix. It's something you use from the very first attempt on some models where you know you want and need a very specific topological flow...such as the face/head of a character that is intended for animation.   This example I made personally, in the 4.7 promo video (the end starting at the 2:27min mark), delivered as good, if not better results than I could ever expect from Z Remesher or Quad Remesher, and the algorithm has since been improved. I am not buying the notion that 3DCoat's Auto-Retopo is significantly inferior to Quad Remesher. Yes, there were some models where I wasn't totally pleased with the result (too many spiraling loops, and I would like to see that eliminated in the near future), but more often than not, Auto-Retopo...when used correctly...does a bang up job and quite frankly, I would take the mesh it created over the one Quad Remesher made, even though both are good results IMHO. If I felt I needed the "stretched" (optimized actually) polygons to be more uniform, that would only take 10 seconds or so to fix. I could simply select the vertices/polys in that area > RELAX a few times. If I had to choose between the two for Hard Surface objects, yes, I would probably choose Quad/Z Remesher because it seems to be very strict with Hard Edges in placing edgeloops along them. That is something I hope Andrew can improve soon, too. These last 2 videos demonstrate stroke guides being needed and used with Z-Remesher, and also that it too sometimes has issues with spiraling edgeloops, so Z-Remesher is not immune to issues that require fixes, cleanup or re-attempts. It's an Auto-Retopo toolset. They just simply call it something else. An Auto-Retopo feature is by nature a REMESHER. It's generating a new mesh that is generally lower in number of polygons and a cleaner overall topology.      

Maybe I am overlooking something, but that comparison, at a glance, makes the Quadremesher result vastly superior to me. In the green coat one, you can see the uneven poly distribution and stretching, which is not present in the Quadremesher.  I struggle to see how this is a better result. Sure, you COULD improve the result in 3D Coat. However, the whole point of Autopo or Remeshing for me is that I need to quickly have workable meshes. I would not ever use the first green result, not even as a base to reproject my high poly onto during the sculpting phase. Because the uneven poly distribution will inevitably show and cause problems, very quickly. Think of a Photoshop Canvas where you have a higher density of pixels in certain spot and a lower density in others. It is simply not workable. 3D Coat's Autopo, even with Curves, can cost a lot of time to get usable. For me, that kind of defeats the point. Of course, even in ZBrush, you can improve the result via curves. But it is just that, an improvement, not a "fix", like it often feels like in 3D Coat (if one is lucky). For me, as someone who uses 3D Coat in production, both commercially and privately, the current algo is not sufficient to work with as an Autoretpo. There is simply no time to waste fixing a Remesh by trying to fiddle with curves and hoping it works. Exoside Quadremesher/Zremesher is far more reliable. In 3D Coat, I lose precious time when trying to remesh something. Don't get me wrong, I would heavily cheer if 3D Coat's Autopo were able to do an equivalent job; I would really wish that. But it becomes apparent in the above comparison pics that it is not. And that is exactly congruent with my experience. And of course, 3D Coat Autopo is still better to what Blender has to offer natively, no question. But it is not the same solid, reliable, workable quality as Exoside deilvers with Quadremsher and Zremesher.  

Thanks! Here's the .3dcpack. Workspace4Test.3dcpack

Please send me your .3dcpack file so I can test it.

When I open 3D Coat the workspace is always as I left it the last work session, but if I try to open a project or start a new one, it reverts to the default workspace. I tried to reset defaults, restore Windows, uninstall and reinstall 3D Coat, boot it as administrator, uncheck and check Store Window State but nothing works and each time I have to Windows->Restore Workspace. It's not a big issue, but it's a bit annoying to have to fix the workspace everytime I open the software. How do I fix it?

Thanks for the tip. I just wanted to share my experience and I was pleasantly surprised by 3DCoat's Auto-Retopo result, without any tweaking (although I did later do some tests with Stroke guides because it was a tool that was available in 3DCoat and thus it would be fair to use it and compare with Quad Remesher's best result). I did acknowledge that Quad Remesher is probably more strict with its Hard Edge detection and forces edgeloops to stick to those edges better than 3DCoat. That makes it a better option for Hard Surfaces than 3DCoat's. However, for organic meshes, I find that 3DCoat's AutoRetopo often delivers results that exceed my expectations and are surprisingly good. As you can see from the comparison screen captures, 3DCoat performed as well...if not better...than Quad Remesher, in this example. 3DCoat's Auto-Retopo is the 1st image, in green, while Quad Remesher is the 2nd image, in orange. With some stroke guides, I think the 3DCoat result would be improved.