!dSLljtUMNpdwwnkjGq:matrix.org

Portable Graphics in Rust

362 Members
https://github.com/gfx-rs/gfx/discussions/376834 Servers

Load older messages


SenderMessageTime
27 Apr 2022
@sajattack:matrix.orgsajattack changed their profile picture.02:42:21
@jaankaup:matrix.orgjaankaup joined the room.06:31:12
@jaankaup:matrix.orgjaankaup Hi. This code doesn't work when I compile it to wasm (https://github.com/jaankaup/jaankaup-engine/blob/e7076dff8be1c3c1a754604f2766d36e5dd5b5ca/jaankaup_core/src/buffer.rs#L32). I guess the solution would be to use .await (https://github.com/gfx-rs/wgpu/discussions/2593) but I have no idea how to modify my code. This function is called inside winit loop. I guess it's not possible to block the winit thread. I tried to use the spawner (spawn_local) but I just can't get it to work.07:02:15
@kvark:matrix.orgkvark hi jaankaup ! I suggest asking this in #wgpu-users:matrix.org 07:30:43
1 May 2022
@wybpip:matrix.org@wybpip:matrix.org joined the room.16:06:48
@wybpip:matrix.org@wybpip:matrix.org left the room.16:06:51
2 May 2022
@kimapr:tchncs.de@kimapr:tchncs.de changed their profile picture.11:04:47
3 May 2022
@probablykevin:matrix.org@probablykevin:matrix.org joined the room.00:34:26
8 May 2022
@theli7:matrix.orgthlnz changed their display name from theli7 to thlnz.23:12:38
9 May 2022
@kimapr:tchncs.de@kimapr:tchncs.de changed their profile picture.07:49:10
@niklaskorz:matrix.org@niklaskorz:matrix.org left the room.12:18:36
12 May 2022
@stanzabird:matrix.org@stanzabird:matrix.org joined the room.20:39:42
13 May 2022
@stanzabird:matrix.org@stanzabird:matrix.org left the room.12:05:22
14 May 2022
@exellian:matrix.orgexellianHi08:20:02
@exellian:matrix.orgexellianis there currently a way in wgpu to preallocating framebuffer textures08:20:22
@exellian:matrix.orgexellianso essentially what i want to achieve is to reduce the calls to surface.configure08:20:54
@exellian:matrix.orgexellianby configuring it for a bigger size and just reconfiguring it if it passed a bigger size08:22:19
@cwfitzgerald:matrix.orgcwfitzgeraldNo not really08:23:25
@cwfitzgerald:matrix.orgcwfitzgeraldSome of the underlying apiz require that the swapchain and the window be the same size08:24:10
@cwfitzgerald:matrix.orgcwfitzgeraldSo you must reconfigure every time08:24:26
@exellian:matrix.orgexellianhm thats bad because the resize time is around 5ms which is alot (gtx1060). Which apis do you mean? 08:26:18
@cwfitzgerald:matrix.orgcwfitzgeraldVulkan does, dxgi (dx12) will stretch it for you08:27:16
@exellian:matrix.orgexellianhmm but in vulkan I thought this is easily possible, by just allocating the swapchain images of a bigger size, but telling the swapchain just to use the window size. Then you still would have to resize the swapchain every time but you wouldn't have to reallocate the swapchain images every time08:29:36
@cwfitzgerald:matrix.orgcwfitzgeraldI mean I guess, but have you measured to see what the actual bottleneck here is anyway08:33:37
@cwfitzgerald:matrix.orgcwfitzgerald (also we should probably move over to #wgpu:matrix.org) 08:33:54
@cwfitzgerald:matrix.orgcwfitzgerald* (also we should probably move over to #wgpu-users:matrix.org 08:34:08
@exellian:matrix.orgexellianok i posted my new answer in wgpu-users08:40:57
@kvark:matrix.orgkvark
In reply to @exellian:matrix.org
hmm but in vulkan I thought this is easily possible, by just allocating the swapchain images of a bigger size, but telling the swapchain just to use the window size. Then you still would have to resize the swapchain every time but you wouldn't have to reallocate the swapchain images every time
Resizing swapchain == reallocating images in vulkan
22:49:41
16 May 2022
@idahotokens:matrix.org@idahotokens:matrix.org joined the room.11:12:30
@idahotokens:matrix.org@idahotokens:matrix.org left the room.16:11:35

Show newer messages


Back to Room ListRoom Version: 5