jedwards

More doodles. Just trying to keep things fast and loose.

Silo's GUI can be whatever you want it to be. You can make almost any command a button, nest it within a tab or float it onscreen. You can even make your own custom buttons so really as far as GUI goes Silo can look any way you want it to - even another app if you want to put the time in to customize it that far. Silo's real UI strength has nothing to do with that however. One of the biggest reasons I use it (aside from being a great organic modeler) is that you have total control over your keyboard binds as well as your mouse. You can embed any command to a mouse click or combination of ctrl, alt, or shift plus a mouse button. This makes it unique over any other software. My default GUI is largely untouched - I turn the command buttons off though to free up screenspace since most of how I work has been reconfigured to use the mouse and those 3 modifier keys. I've completely reconfigured it to work how I prefer it to work. It behaves nothing like Maya or Max or XSI or LW/Modo, blender etc... imo those apps just pale in comparison when it comes to the workflow I created for myself in Silo. I rarely have to break focus from what I'm working on in the viewport. That to me is what good UI should be all about.

Really impressed with how far you got with the old man using voxels! But I love that killbear!

Wow! This is an awesome insight into what can be done with the new tools! Great work all around!

Cool! What brush did you use for the hair?

Good start on it. This is what I see voxels being great for at the moment, unless we see a big performance boost with cuda and 64 bit optimizations. Did you find all strokes were slowed? I found making normal brush strokes a little better performance wise, but when smoothing with the same brush performance dropped by a lot. This happened with most of the brushes I tried. This also happens when painting too for me. default brush strokes are fine but smoothing kills it. Also, did you increase the resolution with that mesh or just let 3dcoat do it for you as you sculpted?

Yeah would love to see this thing detailed - and more gnarly!

I wanted to mess with some of the retopology and baking tools so I finished off the arms on the brute really quick. I don't think the UV tools in 3dc will make me want to drop UVLayout anytime soon, but they were definitely good enough to do a quick pass on the mesh so that I could try the baking/conversion from voxel to uv mesh. That was very impressive. I think my next test will be to see how I can export out my retopologized meshes for UV work elsewhere, bring them back in and try baking from the voxels. That would be my ideal approach. Still a bit iffy on using the regular sculpting tools at this point. They just don't feel as nice as in Zb or Mud yet... but I think that might have a lot to do with my own need to get more familiar with editing normal maps in 3d - that's really something I'm interested in too so I'm not giving up on it yet.

Yeah I used curves in some of the first alpha builds. It's pretty amazing how many different approaches volumetric sculpting gives you. I've also used the 2d plane to sketch out something like a hand with fingers and then just brush some volume into it and pose it from there. I'd also like to mess with the merge tool and see if it helps when building from a library of pre-built meshes. I've seen a few nice pieces done with that approach already. Note: with that human sketch I mostly used the build brush with the transpose tool to get what I needed in terms of shape and pose.

Ok, started a new sketch - with a bit more focus in mind, partly just for practice with voxel tools, but also to see what I have to work with in terms of resolution and performance at different stages of a sculpt. I didn't up the resolution at all this time and instead just sculpted and let 3dcoat add resolution as needed. This worked much better and gave me a very reasonable mesh in no time. There's plenty of resolution left to spare if I want to keep detailing too. I'm very happy with these new voxel tools overall - especially since it's all still first gen stuff. Very impressive for alpha. I've been slinging polygons for a long time so I'm fairly proficient with polygon modeling these days. I breathed a sigh of relief though when Zspheres were added to zbrush. They were a real time saver and allowed me to work more often in 3d, especially for concepting. But you still had to deal with the quirks of zsphere modeling, and its ultimate dependency on topology, which can be tricky to work with when posing or constructing. This voxel stuff just blows Zsphere modeling away. It is such a cool feeling to not think about topology anymore in the early stages and just add stuff where you want to - just totally disregard the technical stuff and focus on the subject. The retopology is literally the best I've used, and I've used all of them - not to mention it works right over top of the voxels (thanks to those who pointed that out to me). No more messing with mesh resolutions or separate programs for me for that stage anymore - that is all getting done in 3dc now.

Build 40, 64 bit DX. Transpose affects hidden voxels. It shouldn't affect anything that is hidden.

Thanks Jokermax, that helped a bit. I probably should have waited on upping the resolution, but I had no real plan for what this doodle was going to become and wanted to add more details to the head before I'd decided to give him arms. I suppose there are still workflow considerations to keep in mind when using voxels so I'll keep resolution in mind in the future - especially at such an early stage.

Disagree with you there. Texture painting on low poly objects is probably more important to me at this moment even than voxel sculpting. This is a feature that doesn't work as it should yet and for games work it is very important. 3dcoat could very well displace the need for bodypaint, modo, and mudbox if it actually gets proper texture painting. That means more customers and a solid push into studio use. As much as I like the voxel sculpting I'm still not convinced I want to use it in place of say zbrush yet. I want to see it replace zbrush too but I also want texture painting that actually works just as much, if not more so. It was promised for 3.0 and I purchased the software in good faith, under that promise. Zbrush polypaint isn't exactly the best solution for all texturing needs. It's pretty much useless when you need to make tweaks or changes on texture assets that are already in game. With 3dcoat being able to paint accurately in pixels AND edit normal maps there will be no other competition from 3d painting apps at any level. None of them can do this.

Andrew: You'll have a lot more than 700 if more industry people start looking this way. Word is trickling out. More and more people I know in the games industry are taking notice. I think 3.0 will be a big deciding factor for those on the fence. The biggest attraction I can see will be if you make the painting tools actually work with low resolution geometry and UVs, as well as normal map editing. Being able to tweak and touch up your normal maps (or any others) on game resolution 3d meshes will be a huge attraction. The volumetric stuff and the retopo tools will also be big selling points if the UI for these tools can be streamlined enough to impress others using standalone tools for those tasks.

I third it =] Agree with most of polyxo's requests in fact. More focus on fewer, reusable keybindings and modifier keys as well as the input device buttons. 2 ways you can go about this are the closed way where you map everything internally the way NEX does it with its modeling tools. Multiple functions bound into one tool that are triggered by various combinations of modifier key and mouse input. This is awesome when you learn them all and keeps focus very tight and within a small amount of bindings overall. I would be very happy to see retopology go this route, so that it is reduced to a couple of powerful tools that can be activated with a keybind or two and then largely controlled by modifier keys and the input device. Other way is to open it up and let the users customize not just keybinds but mouse/tablet bindings as well, like Silo. I enjoy using Silo much more simply because it doesn't need to be nestled within a big bloated host app like maya. With its mouse configs I can build a custom workflow that comes very close to how I work in NEX without having to filter out all the maya fluff. I would probably vote for the second way. Open up the bindings to the users and let us hash out some good configs that you can then include with the software. Save yourself some work. Having a few major modes of operation as mentioned previously so that we can store contextual keybinds etc would be good. Then we can separate and reuse important binds for major tasks like retopology or painting/sculpting. These are different enough tasks that they can logically be separated. I'm not sure I'd want painting to be separated from sculpting as a separate mode... since 3dcoat can do both simultaneously. Seems it would be better to keep them together as far as task grouping goes. Though I admit I usually turn off color and spec input when sculpting... I need more experience with the software to figure this one out for myself. Graphically I'm not a big fan of a lot of buttons, menus or tabs that cause me to constantly break focus on the viewport. First thing I do with any software is learn how to avoid doing this. If I can't escape it then I ditch the software. I list modo in this category as I find it visually way too bloated and dependent on input sliders, menus, buttons and tabs for accessing tools or switching tasks. It's frustrating and customizing it takes way too much effort. Looks pretty though. The only reason I keep it installed is because it can paint on low resolution meshes for games (for basic seam coverage) - something I'm still waiting for 3dcoat to be able to do. =] With a system like this combined with a GUI to satisfy other users you should be able to provide the best of both worlds. Some people like to click buttons and stuff to feel like they are getting stuff done. Some prefer to not lose focus on the subject matter and ignore GUI as much as possible. So long as we are free to choose and one doesn't impede the other everyone should be happy. This is where Silo really shines. You can completely re-skin the way it looks and behaves if you want, and you can also map almost any command in the program to your mouse or keyboard the way you want.

Yeah I understand - you really don't want to be posing the fingers on a hand say and have the palm or wrist getting deformed as well. Masking in Zbrush helps control this as rimasson says, and it definitely lets you get down to more precise control of deformation, but I can't help thinking there's got to be a less 'fiddly' way of doing it. A more accurate deformer that can separate finer deformations like a single finger from the other fingers or the hand or worse, even the rest of the body, which zbrush won't do if you don't mask off parts before using transpose. Something that is much more localized without the need to mask would be awesome.

I much prefer transpose as well. It is a much better final implementation than the zsphere rigs and far more versatile. I use it for everything from simple transformation of objects to complex posing of my characters, at any stage of the creation process, including high rez, detailed meshes. It is a true sculpting tool that enhances the process greatly. I've also rigged in all the major packages and would much prefer you try to avoid mimicking those weighted deformation solutions that are really rather distracting to the sculpting process. Having a deformation rig is not important to me when sculpting. Having a tool that allows me to achieve any pose I want without requiring me to build and weight skeletons is. I'm also not interested in skinned mesh compatibility with other programs. Rigging is best done in the application you animate in, and as far as compatibility with other programs go, those packages already support fbx and collada formats.