Artful Physics
Building soft worlds out of hard math.
A tool for the real-time design and simulation of virtual clothing.
Real-time simulation and modeling of clothing is made possible by taking full advantage of the power offered by modern GPUs.
This speed enables a level of interactivity that would otherwise be impossible.
Tailor lets you grab your clothing models and change their look or fit with virtually instant feedback. Starting from a simple base you can build a wide variety of clothing by adding a variety of cut cone shapes for sleeves, collars, cuffs and even skirts and ruffles.
Note: Intuitive and fun is very much a work in progress for the current release. See "Bleeding Edge" below for more about this.
Tailor ships with some patterns built in for a t-shirt, sweater, frilly dress and more. These patterns should adapt themselves to fit reasonably on most figures.
If you don't like the fit you can alter the patterns until you get them how you want. You can even make more radical changes to create new patterns all built off of the same base.
Bleeding edge technology is a category of technologies so new that they could have a high risk of being unreliable and lead adopters to incur greater expense in order to make use of them. The term bleeding edge was formed as an allusion to the similar terms "leading edge" and "cutting edge". It tends to imply even greater advancement, albeit at an increased risk because of the unreliability of the software or hardware. - Wikipedia 12-2017
Working with early stage technology can be both great fun and terribly frustrating. If you are looking for a polished and bug-free tool then you should wait for more maturity.
At this stage, I am looking for people who are excited by the chance to build things with this technology and are willing to suffer some frustration to get to use it now. More than that, I'm looking for people who would like to join a community that can help shape this technology moving forwards into something truly great.
If you want to try and crash this software you will not find it to be much of a challenge. I highly recommend saving any work that you do early and often.
I know that there are parts of the UI in this initial release that are annoying to me and I'd be surprised if they didn't bug others as well. There are other parts of the UI that I find brilliant and intuitive and that I hope will also delight others. However, part of the point of early access is to find out where and how badly I've got those parts wrong and make them better.
There are a lot of limitations in this initial software just because there hasn't been time yet to implemnent the feature - or often just because there hasn't been time to put a reasonable UI on features that are already implemented in the core simulator. All patterns must be built off of a fairly simple tank-top base. Only the most basic cloth physical properties can be changed and they must be changed for the whole garment at the same time. All coloring and texturing of the garment must be done in outside tools...
This first release has at least one very severe platform limitation and several others that I don't like. Unfortunately, living with these kinds of restrictions is the price of shipping.
There are two ways that I see Artful Physics being used in the future. One is working with existing 3d modeling and rendering engines to provide clothing and other soft physics support. The other is more of a stand-alone tool that can deliver unique interactive and real-time experiences both on the desktop and through VR/AR.
I chose Daz Studio and the Genesis 3 based figures as the first existing tool to integrate with. The fact that Daz Studio is available for free is certainly one reason for this choice. I am also a fan of the quality, customizability and strong user community built around the Genesis figures. It was important to pick a single base figure model to focus on for an initial release, and I couldn't come up with a better choice.
The integration with Daz Studio is a small collection of scripts whose source code is included. Most of the interchange is based around standard 3d file formats. Simulated cloth is exported as an .obj file that can be used with any 3d tool and animation frames are exported as a sequence of obj files to capture point positions for the frames. If Daz Studio had decent support for pc2 files, that would be used instead.
I'm 99% sure that this particular limitation will be one of the first ones to be removed. However, there are so many different directions to go from here that I'm looking to the explicit and implicit feedback from early patrons to help me figure out which is the most promising next step to take.
Primary development is done with a GTX 1080 Ti and this software is designed to take advantage of as much GPU power as you can give it.
The recommended system would have at least a Radeon 580 or GTX 1060 card or better. A lot of parameters in the software are tuned for this range of high-end GPU. It could be adapted to work better with lower speed systems by using lower resolution cloth, less accurate simulations, or simply running at a much lower simulation frame rate - but that is currently not the focus.
Lower-end video cards will run the software, but they will run it more slowly. See the video below for examples of running with a GTX 660 Ti. This is roughly equivalent to the current generation GTX 1050 Ti and somewhere in between the RX 560 and RX 570. Personally, I am irritated working at these slower speeds, but I can still use it. Your perceptions may vary.
This first release is only built to run on Windows and has and will only be tested on Windows 10. It currently depends on DirectX 11, but that will likely move to DirectX 12 in the future.
The software is built using the Unity engine and should “just work” on Windows, Mac and Linux. I have tested the Mac build on my old Macbook Pro laptop and found that it was extremely slow, but still just barely useable. Part of the problem right now is that there are very few Mac systems with sufficiently high-end video cards to make them a good fit. I remember the days when a Mac was the only system you would pick for high-end media development (if you couldn't afford an SGI Reality Engine). It makes me a little bit sad to see how far they have fallen in that respect.
It is really important to me that this software can run on a wide variety of operating systems. Platform lock-in is deeply unappealing to me. However, in the current state of the world, Windows really shines as a platform for high-end GPU development work and that is where I intend to focus my efforts unless convinced otherwise. However, I will continue testing builds on Macs and possibly Linux machines because the state-of-the-art is always changing and I want to be ready to go wherever it makes sense.
Do you want to work with this bleeding edge software and join the community to help make it great? Early access releases are being provided in return for a pledge of support on Patreon. Patreon will also contain links to the url for community discussions.
