Vulkan 1.0 is out. Thoughts?
- Vous devez vous identifier ou créer un compte pour écrire des commentaires
https://www.phoronix.com/scan.php?page=news_item&px=Vulkan-1.0-Quick-Highlights
https://www.phoronix.com/vr.php?view=22837
https://www.phoronix.com/scan.php?page=news_item&px=Intel-Mesa-Vulkan-Published
It will be interesting to see if this will aid in better performing free software firmware/drivers or if we will still be locked out by AMD/Nvidia and have to use non-free firmware/drivers to take advantage of our hardware.
If someone does create a free software game under Vulkan, it will make porting to other platforms easier.
What does Vulkan mean for us libre users? Any improvement in graphics rendering?
https://www.phoronix.com/scan.php?page=news_item&px=Intel-Mesa-Vulkan-Published
I heard that Intel is pushing support into MESA and we will have to see if AMD and Nvidia eventually offer a libre option too. As for graphics rendering, you won't see an improvement as it is more of a developer thing. If you compare an OpenGL version of a game to the Vulkan one, the Vulkan version will show an improvement over the OpenGL one if optimized.
The non-free game Talos Principle recently released a beta Vulkan version, with some benchmarks at http://www.anandtech.com/show/10047/quick-look-vulkan-performance-on-the-talos-principle showing an improvement over OpenGL, but still behind DirectX 11.
The problem is, it'll take years for libre games using Vulkan to appear. Very few libre games run on something newer than OpenGL 2.1 already. Some libre game engines will likely get Vulkan support in the next 2 years, but there needs to be libre games using those for it to be really useful.
Not to mention support of Vulkan in free drivers other than Intel's…
In the short term, this won't affect us at all. In the long term this is HUGE. It means high performance, high fidelity games are not going to force you to go the DX12 route, which requires Windows 10 (the Facebook of operating systems). I hope Vulkan takes off and dominates. It also means that bringing games developed for it cross platform is a very small effort for developers. You can take your iOS or windows game, make a few minor changes and you have a fully featured Linux version.
It may fail miserably, but it's a great start. Just don't expect results in a short time period.
There's not much interest to having more proprietary games on GNU/Linux. Sure, it might help people migrate (which is why proprietary software ports are overall a good thing), but it's useless to the people of the free/libre world.
Instead, people should stop buying proprietary games and start playing libre games, it'd be a start.
- Vous devez vous identifier ou créer un compte pour écrire des commentaires