Hi Nemoid,
yes i agree with you…not saying it wouldnt be nice to have better retopology but just saying that universal and artist friendly shape creation of any sort without constraint is not yet solved today although we are quite close now…and this is why i rather see Pixo perfecting the tool rather than going in too many directions…
we need a few good accuracy tools now similar to what lets say Freeform can do when slicing with curves or projecting curves on voxels or filleting and things like that navigate trough this website and realize in the case studies how this product many many years before even Zbrush was designed started tryin to solve everydays modeling problems, notice how the ïdea of autoretopo existed before some in this forum even started a 3d program http://www.sensable.com/freeform-videos.htm…
notice how mesh difference and displacement extraction was used back then too …notice that some of those shaping effects are feasible today in Zbrush but require some workaround basically because of the surface nature of the soft…this is why clip curves for instance work only in certain cases and this is also why you have to preplan holes in your shape wich defeats a bit the concept of being totally able to concept model in Zbrush and this is also why we have things like Goz to sort of force you out of Zbrush and solve it…again the mesh on the fly approach of Sculptris solves poly surface stretching but there is no equivalent of the opposite wich will be poly erasing on the fly…its the 3d equivalent of a photoshop that would paint but not erase…a pretty damn basic limitation im sure you agree…
A funny fact is that while Zbrush is a modeling tool, you have to realize that Film and game might be the most glamourous and visible part of the modeling iceberg yet this industry is probably the smallest and definetely the least high end of what modeling can and offers to many many modelers in other fields.
To give you an idea of how limited our poly toolsets can be try doing something as simple as a perfect extrusion offset …it will fail even if the soft splash screen says MAYA 2011… .I think it takes more of a" bigger picture" approach rather than total revolution and this is where i become suspicious of our workflows…i fight production everyday just to realize that some people have been working in very closed pipelines for way too long…some very famous studio that i wont name here was still modeling using Nurbs a couple of years ago…i can guarantee you that other than they very proprietary tools using Nurbs surfaces there was no reason for this at all and this had a huge man hours impact in production inflating time and money costs artificially.
Bigger picture example and how digital artist are sometime alienated by branding, name dropping and trends…the concept of progressive rendering and instant feedback has been out for almost 10 years now with things like Fprime or even Modo…it is funny to see production people suddenly discovering how good that is when using this"new concept" in things like iRay for Mental Ray or Octane…i truely believe that Production is very often the mediocre approach and the safe approach because a lot of people are in the middle of production and have no time to have a look at other more recent options.Changing technologies is a great risk of course in the middle of a project so dont expect huge pipelines to start using things that hobbyist can today and right now.
For all this reasons i wouldnt take production standards as absolute standards…i still can remember how “modeling in Nurbs is the standard” from TDS not so long ago while everywhere else in High end modeling industries were trying to find a solution to solve their very time consuming stitching nurbs problem using things like voxels or solid modeling approaches …the truth is that there is nothing that is set in stone and retopologizing is after all a thing that can be solved by hiring a Junior artist, using a small specialized tool like topogun or even using the Maya toolset u can get away with a lot actually…
again concept modleing is part of the pipeline today…the client usually couldnt care less of what topology means nor the designers directors or art directors…flexibility in shape creation, good presentation are by far more important to me…today we are even rendering decimated meshes for static or non deformable objects…as u can see retopology is needed after all in a very specific moment of very specific assets…they are often hero assets i agree but the sheer volume and scale of environment assets, vehicles props, terrain etc and there complex , often non reusable shape nature makes them a huge bottleneck in production…do the math…for a movie like Avatar for example there were a few characters with reusable base meshes…lots of love for a few models and then everything else! plants, rocks choppers props, spaceships etc…i can guarantee you that there was much many man hours involved in MODELING those than the characters and probably more tools involved too.
as a conclusion yes better topology tools will be nice…this is needed yet a very low end tool after all you just need to snap to surface and build polys on top and there are many little tools doing just this like Topogun or 3dcoat or even Modo…on the other hand there are lots of modeling scenarios that arent solved and that are much harder to solve and that no one even dare to try to solve and i would like to see Pixo focused on that first…i can see that they have tried, i can also see that the idea is there with the voxel engine in teh backgorund for things like Shadowbox or remesh all…i just want them to make a bold statemnet and perfect that…they are leader already in their approach and this way they will really distance competition big time even more.
Wow, what a whiney bunch.
