Skip to content

Releases: wol101/GaitSym2019

v1.14

24 Feb 14:24

Choose a tag to compare

Just a few bug fixes in this version and some support for outputing GaitSym5 files. Currently there is a Windows installer and a portable Windows version.

v1.13

25 Oct 13:16

Choose a tag to compare

The big change here is some NVIDIA Omniverse integration. GaitSym can now write the animation out as a series of Pixar USD files and these can be rendered very quickly and to extremely high quality using the bundled Omniverse extension. They can also be read into Blender and include a lot more information than raw OBJ files (things like the camera and light positions). If you want to create a nice ray-traced final rendering of your simulation for publication then this is the way to go. The only down side is that the USD files can be big, and this means that you need a fair bit of storage and even with Omniverse, the final rendering will be slower than ideal (but still much quicker than the alternatives). Also to use Omniverse you will need an NVIDIA RTX graphics card on Windows or Linux. On other configurations you will need to use alternative rendering options but USD files are a very standard interchange and there are lots of options. The RTX code was supported by an NVIDIA Academic Hardware Grant entitled Omniverse Dinosaur Musculoskeletal Simulation.

v1.12

09 Mar 15:57

Choose a tag to compare

Just a few fixes to allow the FluidSac element to be more stable when used in Incompressible mode. Windows installer and portable versions, and an Ubuntu cqtdeployer package currently available. I've also added a Ventura mac build which is hopefully a fat binary.

v1.11

05 Feb 14:00

Choose a tag to compare

Added a few new features to the interface and of course some more bug fixes. The Windows 11 version is now statically compiled with Qt6.4.2 and available as an installable and a portable version.

v1.10

08 Aug 17:18

Choose a tag to compare

This release contains a number of bug fixes that were needed to get the chimpanzee paper working properly and to help produce the figures. There is a portable Windows 64 version, a Windows 64 installer, a Ubuntu installer and a Mac version.

v1.9.1

30 Jun 13:57

Choose a tag to compare

Adds a force sensor to the data target scalar option.

v1.9

22 Jun 12:43

Choose a tag to compare

This version has a new TwoJointDriver to support the latest models we have created. It also fixes a few bugs in the GUI relating to editing some of the components. There are some less important changes under the hood too. There is a Windows 64 bit setup file and a Windows portable 64 bit binary, a Linux installer created on Ubuntu, and an ARM/Intel Mac version. There is also a Windows 64 coimmand line version which is useful for machine learning applications and the command line client that can communicate with AsynchronousGA2022.

v1.8

14 May 14:53

Choose a tag to compare

Update to include basic python bindings and an example scipy script to allow python-based optimisation to work rather more seamlessly. Also various bug fixes and a new driver to allow direct control of marker positions. The release includes a Windows 64 bit installer, a Windows portable binary, a Ubuntu 64 bit installer, and a MacOSX disk image (Intel). The Windows installer is created using Inno Setup and the Ubuntu installer is created using cqtdeployer.

v1.7

09 Mar 16:56

Choose a tag to compare

This is a bugfix release. The major fix relates to joint limits on ball joints but there are a few other tweaks about the place. There is a Windows 64 bit installer, a Windows 64 bit portable version, a Mac Intel version and a Ubuntu version. There is also a Windows command line version (GaitSym2019CL.exe) which is useful if you are using some sort of machine learning to optimise drivers (e.g. with the scipy_optimize.py script).

v1.6

13 Feb 11:01

Choose a tag to compare

This release fixes a memory leak that causes the graphical interface to slow down over time. You should see a big improvement in the visualisation speed for models with a lot of muscles. There is a Windows installer, a Windows portable version, a distribution kit compiled on Ubuntu and an Intel Mac version. Nothing is signed so you will have to ignore the various warnings if your machine wants signed binaries.