Vulkan vs DX12 FPS #2444
-
|
Hello! I'm referencing the tutorials from learn-wgpu and tweaking this basic texture example. I see ~600 FPS using Vulkan and ~1600 using DX12. I'm curious why:
This OpenGL example from 2 years ago runs an entire game simulation with multiple textures at well over 2k FPS. Removing the Scenario:
|
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 2 replies
-
|
This is quite common to see between GL and other apis. Vulkan in particular I've always noticed just runs kinda slow when doing nothing. This is very likely because of some component of the VK => Windows layer that nvidia provides just being a bit slower than going directly through DXGI or through the GL => Windows layer. You'll very quickly find that this is by no means indicative of performance as it scales up. |
Beta Was this translation helpful? Give feedback.
-
|
In other words, Vulkan wasn't invented for doing faster screen clears. The more complex API interaction you do, the more Vulkan can help. |
Beta Was this translation helpful? Give feedback.
-
|
10-4. Appreciate your insight. That makes sense. Similar to something like deferred rendering, there's a lower baseline out of the gate but increased throughput when the action begins. Thank you both |
Beta Was this translation helpful? Give feedback.
This is quite common to see between GL and other apis. Vulkan in particular I've always noticed just runs kinda slow when doing nothing. This is very likely because of some component of the VK => Windows layer that nvidia provides just being a bit slower than going directly through DXGI or through the GL => Windows layer.
You'll very quickly find that this is by no means indicative of performance as it scales up.