NVIDIA has put out what they call 'VkHLF', a high level abstraction library on top of Vulkan. Looks like it's under a BSD-style license too.
It's described as so:
It's still in the early stages, but it sounds very interesting. It essentially makes working with certain parts of Vulkan easier without adding an overhead if used correctly.
I wonder how many more Vulkan layers we can expect to see?
You can find VkHLF on github.
It's described as so:
QuoteVkHLF is an experimental high level abstraction library on top of Vulkan. It adds features like transparent suballocation, resource tracking on the CPU & GPU and simplified resource creation while staying as close as possible to the original Vulkan API. In contrast to Vulkan-Hpp, which was carefully designed not to introduce any additional cost, this library can add a signifact cost depending on the usage pattern. Though, when used properly, it is possible to use this library with minimal additional cost while improper use can result in OpenGL performance levels.
It's still in the early stages, but it sounds very interesting. It essentially makes working with certain parts of Vulkan easier without adding an overhead if used correctly.
I wonder how many more Vulkan layers we can expect to see?
You can find VkHLF on github.
Some you may have missed, popular articles from the last month:
Quoting: tuubiQuoting: ShmerlVulkano is easier to use than raw Vulkan: https://github.com/tomaka/vulkanoYeah, but I don't think game devs will switch to Rust just to use it. Anyway, the more the better. There's never a single approach that works best for everybody.
Some already might. Rust can remove a lot of pains that exist in C++ despite all the latest advancement in the language. The only thing that can slow down such adoption would be middleware. Tons of stuff exists in C++ for gaming engines to use, and so far there is no interop with Rust (or with any other language for that matter really). Libraries in C are better in this sense - you can use them from Rust. SDL is a good example.
0 Likes
Quoting: ShmerlSure, but this only applies to projects that start from scratch. And besides, you're underestimating the effort and the risk of dumping existing code bases and know-how in any organization larger than a single developer.Quoting: tuubiQuoting: ShmerlVulkano is easier to use than raw Vulkan: https://github.com/tomaka/vulkanoYeah, but I don't think game devs will switch to Rust just to use it. Anyway, the more the better. There's never a single approach that works best for everybody.
Some already might. Rust can remove a lot of pains that exist in C++ despite all the latest advancement in the language. The only thing that can slow down such adoption would be middleware. Tons of stuff exists in C++ for gaming engines to use, and so far there is no interop with Rust (or with any other language for that matter really). Libraries in C are better in this sense - you can use them from Rust. SDL is a good example.
I don't doubt that Rust is a nice programming language though.
0 Likes
Quoting: tuubiSure, but this only applies to projects that start from scratch. And besides, you're underestimating the effort and the risk of dumping existing code bases and know-how in any organization larger than a single developer.
I don't doubt that Rust is a nice programming language though.
Knowledge usually isn't dumped, it's simply extended. I.e. they might be invested in C++, but if they are smart, they can start shifting things to Rust gradually.
0 Likes
Quoting: ShmerlKnowledge usually isn't dumped, it's simply extended. I.e. they might be invested in C++, but if they are smart, they can start shifting things to Rust gradually.Maybe Rust will replace C++ in the industry one day, who knows. I don't think this will happen quickly though.
In any case, the thread is about VkHLF, which is a C++ library. I think most games that would benefit from Vulkan are still developed in C++ so this library certainly seems useful.
We'll see if AMD and/or Intel deem it necessary to provide their own competing abstractions. They could always try to optimize them for their own hardware's strengths. Nvidia certainly might.
0 Likes
Yeah, as long as the higher level API doesn't dictate a global design (for instance how to use multithreading) and doesn't mandate stateful approach (like OpenGL), it should be useful.
0 Likes
See more from me