From this week's news :
- Agility SDK for DirectX 12 is now available
- New file format for compressed realistic 3D textures on any GPU
- Godot 3.3 released with focus on optimization and reliability
- The Blender Foundation unveils Cycles X
- Blender 2021 roadmap
- Rider for Unreal Engine 2021.1.1 with macOS support
- Visual Studio 2022 will deliver x64
- Full release of Steam Playtest took place
Interesting:
- Rural Australia Pack for UE4 Now Free
- Workshop on working with light in CRYENGINE
- An example of what Roblox can achieve
- Interesting prototype in VR
- High jump simulation. A new approach to recreating sports movements in 3D
- How weapons were created in Cyberpunk 2077
Updates / releases / news
Agility SDK for DirectX 12 is now available
This toolkit allows developers to take advantage of new DirectX features without the need for players to update Windows. Now game developers can assume that most Windows computers will run all the features of DirectX 12.
Games can use the Agility SDK on systems running Windows 10 November 2019 Update or newer. Developers can already check out the starter guide , download page .
New file format for compressed realistic 3D textures on any GPU
Khronos KTXTM 2.0, Basis Universal.
GPU. Basis Universal — , Binomial, , GPU .
Godot 3.3
7 . Godot 3.3 Godot 3.2.x, 3.2.x .
3.0, Godot 4.0 Godot, , Godot 3.x , Godot 4.x .
The Blender Foundation Cycles X
Cycles.
Blender 2021
- Blender 3.0 , , Blender 2021.
Blender 2.93 . 2.83, LTS , .
.
Rider Unreal Engine 2021.1.1 macOS
Rider Unreal Engine macOS uproject.
, Windows .
Visual Studio 2022 x64.
- .
- .NET 6, -, Windows, Mac, Azure.
- .
- C++ 20.
- Live Share.
- Git GitHub.
- .
Steam Playtest
, Playtest , .
///
Rural Australia UE4
Unreal Engine Rural Australia Environment Pack . Rural Australia — , , , .
/
Cyberpunk 2077
, CD PROJEKT RED, Cyberpunk 2077.
- CRYENGINE
, Roblox
Interesting prototype in VR
Both cool and unpleasant.
High jump simulation
A new approach to recreating sports movements in 3D.
The new work "Discovering Diverse Athletic Jumping Strategies" describes strategies that are implemented as physical character control policies. The combination of physics simulation and deep reinforcement learning (DRL) provides a suitable starting point for learning, the authors say.
The researchers say the new method offers more realistic results and greater flexibility. More details can be found here .