![]() If that's actually correct and ZSphere armatures aren't real animation rigs and can't be converted to one. It should be good to go? At the very least, someone must make a plugin that will convert the ZSpheres to a real rig, right? The docs say it does "auto-weighting," and I'll be honest I don't exactly know what "weighting" is, but I know it's an important part of an animation rig. Please tell me I'm finding out of date info or something. not much turns up, but a couple several year old forum posts I found say. Then I tried to look up how to export a model with its ZSphere rig and. ![]() I'd always heard rigging was some kind of nightmare, and this approach of building models from the rig up looked so easy and just made so much sense. So, when I first saw the intro videos showing building armatures with ZSpheres, I was super-excited. Good Luck.First, a small disclaimer: I am an absolute and total noob. Sorry for the fanboi response but I'm just a hobbyist and wanted you to get a feel from a non-professional how valuable Zbrush is for the types of things you want to do. Yet these guys still keep giving (livestreams etc). no one since Zbrush's first release in 1994 (I think) has ever paid for an upgrade. So if you're really serious then the financial hit won't feel so bad. No Live Booleans, not projection, no Zspheres, the list goes on. There is so much missing from core that I have found useful for 3D printing that Core just doesn't cut it. Search this subreddit for Core comparisons and look at comparisons at Pixologic. For that print I just started using Zsphere's for purpose built supports which are far better than auto or even adjustable supports in typical slicer software. Scaling, exporting for 3D printing is a no brainer. Just thickened up some ears on a Husky Buddha print. It's mesh manipulation power is outstanding so I've been able to readily tweak and fix sculpts so they can be printed with FDM. I haven't created much original yet but I have worked on many premade assets such as game rips, sculpts by people who have artist talent etc. ![]() I am little bit unsure if that would be a problem for me down the line? I am not opposed to putting down the 900$ for a full license of Zbrush, but I would like to avoid ending up with a tool that is no good for printing or vice versa a tool that has too many limitations and then requiring me to upgrade when buying Core instead (with would lose me 100$, essentially) And unlike "ZbrushCore", it does not seem to have dedicated features for 3D printing. The price for Zbrush 2020 as a hobbyist is a bit. Sculptis is just too stripped down and Blenders handling is just not for me to be "productive". I started with Sculptis and Blender, but feel that neither is what I need. I would also like to do some "proper" digital painting. My "goal", if you will, is to produce figurines/statues (similar to HotToys, Neca, etc) from the various IPs I enjoy as well as from my own ideas. Then, a few months ago I started to get curious about 3D printing, and have now bought my first printer. Haven't done it in a few years tho as I got distracted by other interests. I have recently decided to pick up digital art again I used to spend a lot of time drawing and painting (physical, not digital) as well as played around with 3DSMax and Blender.
0 Comments
Leave a Reply. |