ZBrushCentral

Making Time For Art #3 - "Clay Time"

optimized density=more polys were more curvature is needed and less where its flat or straight opposed to evenly spaced quads everywhere.

Agreed gasoil.
There’s a problem though, a zb particular issue. It uses vertex painting system not a direct texture painting. This method prefers evenly subdivided meshes -tris and/or quads. Was never a good friend of optimized density. You can’t decently paint on very sharp tris or stretched quads.
A trap LOL.

So, it’s more complicated now. A request for better retopology tools, another one for direct painting on maps. (3dcoat, even sculptris, blender MB and other decent expensive massive packages-apps)
BTW the old projection master is broken in zb4, have you tried it?

T Minus Nine Days And Counting.

I wonder if Mahlikus would loan me that nifty new computer… gee that would sure be spiffy.

GRIN!!!

Not true. Z makes texture too.

She’s tasty! Should be here Friday! It’s almost as exciting as the new Z!

Not true. Z makes texture too.

I can’t see how, except of baking in UV maps or using projection master for direct painting. (but as I already mentioned, the last one is broken)

My apologies.
Ran some tests and found some things broken (not the way they used to work).
That or I am remembering the process wrong.

Not sure whats going on but we only have 9 days left before we work with the new version so c’est la vie.

@Michalis

I agree that retopo and “direct” Uv painting will be nice or fix what is broken…so i definetely second that.

This said just keep in mind that it is prefectly possible to vertexpaint tris (as seen in Sculptris for isntance)
Also this is not so important but keep in mind that painting on UV is actually “indircet” painting as pixel painting between UV coord implies a interpolation…also painting on UVS is Indirect because u cant paint on later displaced shapes causing smearing …also when painting on a “low poly” cage this can cause subdivison strteching as well…as incomplete or not so production friendly as vertex painting might appear vertex painting relies exclusively on what exists allowing you things like dry brush techniques since What you see is what you get and not a on rendering displacemnet effect.

When you paint on vertex you are painting on the limit surface no cheating, no guessing and no compromise
As you might be aware UV coords is already a half assed attempt to establish a relationship between a 3d volume and some 2d Map…
Again teh way i see things is…we can have retopo and UV painting and have what other softs have…or we can have what NO ONE else has…teh uniqueness of Zbrush approach is what makes it so powerful…and also keep in mind that sooner or later you will ahve to jump to another soft (for rendering, rigging or even for texture painting)
Bottom line is that vertex painting is truely amazing and without compromise and i would imagine rather the competition demanding to have a vertex painting option since there are hundreds or apps out there form free to expensive that do UV painting etc…

Again im more than happy to see Pixo improving this but i am quite happy that yet again they deliver never seen before techinques and paradigms rather than just another flavor of the same thing

If you had Ptex you wouldn’t have those problems that uv’s have with stretching and smearing…

it seems that Ptex would be easy enough to implement since zbrush uses quads…

What about this?

Im not sure Ptex allows dry brush techniques as in color being aware of deformation but what i am sure is that Ptex does not tolerate any changes to the topology after you have painted the Ptex files wich on teh other hand destroys all chance of editng or updating your mesh after the fact so not really ideal at the concept phase at all…
In practical terms you cannot add resolution in teh mesh or add a limb or any other detail in your mesh after Ptex while using vertex paint you can sculpt and paint at any stage without worrying of breaking anything also notice that Ptex only allows 3d painting with proprietary tools and you wont be able to bake to UVS for further Photoshop editing (so u need a full 3d painting and editing solution to use it as well)

in real life situation in Zbrush u can start sculpting then painting then sculpt some more…u can chaneg the topology totally (remesh or Dynamesh or Goz) and the vertex paint will still be there (with the eventual touch up in teh areas of new topo) then u can create a new new topology and reproject the painting back into thsi one too at any level or any density and any flow…u can also redo your uvs at any point and if you dont have enough painting/editing tools u can bake to UVS and paint in photoshop and back again and reproject and it still will be there…Also notice taht you can locally subdivide (and tehrefore have more vertices to paint on( locally wich will be teh equivalent of selecting faces in Ptex and upresing locally…

Using Ptex forces you to absoultely lock your topology so it is definetely a final production model tool (and very final)

This said Ptex is an awesome tool while dealing with productionpipeline i.e Maya/Renderman in wich case yes we rather have it than not…what id like to see is more Texture artists being able to use vertex painting as it has a nice potential

Polypainting isn’t perfect either, and I can think of several limitations it suffers from off the top of my head. For what it’s worth, PUV/GUV/AUV Tiles in Zbrush aren’t the same as ptex at all. As far as I know, Mudbox has no trouble dealing with subdivision changes after the mesh has been painted (ptex was designed with this in mind). Don’t know about major mesh changes though. Whatever the case, I strongly suspect we’ll see ptex support in all of the major 3d apps long before we’ll see any of them supporting Zbrush’s polypainting, if ever. Oh, and dry brushing isn’t a limitation of ptex, but rather the paint engine being used (if I understood correctly).

Don’t get me wrong. The system of painting in Zbrush, even though it’s still not quite perfect yet, is quite good and a step in the right direction. If you’re the type of user who has no plans for rendering/animating in a third party app or game engine, then the current implementation of polypainting is probably all you’ll ever need. Working outside Zbrush requires more adaptability though. Ptex, UV’s, polypainting… they all have limitations and it is up to the artist to decided from the start which method will best suit their project. That said, I don’t think it would hurt Pixologic to implement one more. It keeps Zbrush competitive in a market that is clearly getting harder to dominate in.

Hi zeddicus

I totally agree that better have Ptex than not for sure!!
Never said PUV or any uv tiling in zbrush is like Ptex don’t know where u got that :slight_smile:
Ptex won’t break with subdiv levels (x4) catmull iterations but will break if arbitrary topology is added like dynamesh or shadowbox (remesh) or more everydays situation like let’s say your favorite rigger decided to change the flow of something last minute in which case u can start your painting again
Again having Ptex will be great but all this fantastic mesh creation engine that
Zbrush is is mainly focus on concept and free sculpting which requires on the fly solutions like vertex painting until someone comes with another solution
Basically at this point is kinda paradoxal … You can’t ask zbrush to break new ground with tri dynamic tessellation or dynamesh in one hand and then ask why is not playing nice with outdated tools (not talking about Ptex here)
Also keep in mind that many meshes contain for the sake of simplicity and optim the occasional tri yet not sure Ptex will like that
Last but not least remember that the game industry which is today probably the biggest client for a sculpting soft has little if no use at all for Ptex as it currently exists (game engines)
The only reason I think we should get Ptex is deffo for the sake of having another option for those who would need it nothing wrong with that…keep in mind that we will need an entire dedicated pixel painting core with it too which is not a small request.

One more week! This is a weird form of torture! :stuck_out_tongue:

I know! this will be the longest week so far this year for me. :lol:

Oh, happy day! I just scored an Intuos4 small for $140! That and the new laptop will be here the 19th! Thank you Univrse! You kick ass! Look out Z, Big Papa’s coming to play hard! :smiley: Longest…wait…ever!

You misunderstood me in a few places, and not all of my comments were directed specifically at you.

Mahlikus posted a link to the UV mapping page of the Zbrush wiki which talks primarily about textures and UV tiles. A lot of folks don’t fully understand what PTex is, so it often gets confused with the tiling in Zbrush which isn’t really the same. With Ptex, each polygon gets its own texture map, of which each can have its own resolution that can be changed at any time. All the resulting texture maps are stored in a single .ptx file. PTex also does some nifty filtering which prevents seams from becoming visible. The way Zbrush does tiling is to break each polygon into a rectangular shaped traditional UV, with all of the individual UV rectangles fitting on one single texture. It is possible to run out of resolution this way, and it doesn’t render very well outside of Zbrush (especially displacements).

Regarding game engines, I was talking about the needs of artists who work outside of Zbrush versus those who don’t. I know that PTex isn’t really suited for real-time rendering, at least until graphics technology improves. Zbrush is being used in 3D workflows everywhere these days (cinema & television, video game assets, printing, etc), and not just as a simple illustration tool. This is why we’ve seen so many export-centric additions since ZB 2.0. As I said before, if you’re not straying outside of Zbrush, polypainting is all you should ever need. For those who use it as part of a larger workflow, UV’s are an important aspect regardless of type, outdated or not. The idea behind PTex is to do away with traditional UV’s and all their inherit problems. And yes, it works with non-quad geometry too (with subdivision).

Polypainting is basically vertex painting, is it not? Zbrush is already doing some behind the scenes calculations/tracking to make sure all that paint stays where it’s meant to, which is especially important when making mesh modifications. The same thing can be done for Ptex as well. The only limitation is the amount of coding work the developer is willing to do. There is really no reason why an artist cannot make major modifications without screwing up their various paint layers when using Ptex (normal, specularity, diffuse, reflection, and so on). At least that is the impression I got from what I’ve read thus far. I have no idea how big of an undertaking this would be for Pixologic, but the competition doesn’t seem to be having too much difficulty. PTex, which is open source, is still under development as far as I can tell, meaning it should continue to improve as time passes.

This page has some interesting information posted by someone at Pixar. I should point out that the “can’t paint them flat” comment they made isn’t 100% accurate. I recently watched a video showing Mudbox passing Ptex texturing off to Photoshop and then back again, very similar to what ZAppLink does. Not technically flat, but close enough and better than nothing. You can find more interesting info on this page as well, which talks a little bit about its goal and capabilities. Finally, there is this site where you can talk with Disney employee Brent Burley, the developer of Ptex. Some of the other comments (mostly by other developers) are worth a read too.

In all honesty, polypainting and Ptex are two sides of the same coin. Whereas polypainting works best with high poly meshes, Ptex works best with low to mid density meshes (though from what I’ve read, million poly meshes might be workable too). As I said before, we’ll see Ptex support in all the major 3D apps long before we see something akin to the polypainting Zbrush does. When it becomes feasible to animate and render multiple million poly meshes in programs like Maya and 3ds Max, then we’ll probably see vertex painting take off. For now it is only useful to Zbrush. Unless you’re baking textures of course, in which case mapping of some sort will still be required, bringing us back to square one. Semi-automatic unwrapping algorithms (i.e. UV Master et al) are great, but all are still far from perfect. PTex solves all of that very simply, plus it truly is a huge time saver. Endless hours of tweaking UV’s can now be spent creating art and producing assets. Who won’t be grateful for that? :wink:

Got that right man. I’m losing sleep actually.

OK wait what is the difference from D-Mesh and just taking a sphere and creating a sculpt off of that? All I see you doing is simply taking a pre generated primitive like a sphere and using the clay brush…

Easily my most favorite brush for roughing out shapes btw I <3 play brush!

So how is this new?

Besides an attempt to make it feel even more like real clay, the primary benefit will be the prevention of stretched polygons while sculpting, also known as running out of polys. I’m sure all of us have run into that at one time or another. We won’t really know if dynamesh is the same as any other ztool, or if it’s in a class all by itself, until we can get our hands on it come Tuesday. I suspect it is the same, and all pressing the dynamesh button does is turn on dynamic retopology, the price of which being an ever increasing polycount as you sculpt much like Sculptris. That is probably why we see the cut tool used in this video; to get rid of the excess polygons that have built up during the sculpting process.

I agree and have been there many a times while sculpting in my free time. You want to add that one row of rivets and some tears near them but find “Hey I need to subdivinde yet AGAIN hope I have not hit my cap yet…”

You hit Divide… Zbrush stuters… and Down goes your project file and ac rash report to just rub it in your face. :mad:

Yep been there felt that hate it and why I try and keep a eye constantly on my mesh count before I hit my divide button. It looks like a tone of fun and I am looking forward to using it but I am a bit tiered of having 30 tools to scan though and only using maybe 5 of them 90% of the time then maybe only the other in spot moments…

It would be nice if you could customize the tools that are brushes as a drag drop onto the interface so you can swap between them quickly with out having to populate a giant menu.

the amount of artistic freedom this provides is just insane. pixologic is gonna have a hard time topping this :wink: