Vulkan and OpenCL #653
Comments
|
I may be able to look at it...
…On Fri, Sep 30, 2022 at 7:34 PM darkjacky ***@***.***> wrote:
Would you mind adding support for Vulkan and OpenCL decoders?
Cuda decode benchmark:
video:12265kB audio:199549kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
bench: utime=3.781s stime=2.547s rtime=24.493s
bench: maxrss=139480kB
OpenCL decode benchmark:
video:12265kB audio:199549kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
bench: utime=27.750s stime=1.438s rtime=6.049s
bench: maxrss=128976kB
Vulkan decode benchmark:
video:12265kB audio:199549kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
bench: utime=26.078s stime=1.672s rtime=6.418s
bench: maxrss=208604kB
They are roughly 4x faster than hardware decoders (by using the videocard
3D chip) and they have a fallback in case there is no hardware decoder
making them a really good addition as you don't have to check anything.
Just -hwaccel opencl and it will figure it out.
—
Reply to this email directly, view it on GitHub
<#653>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAADBUGMEEF5IQ7OAYRY5Y3WA6IJFANCNFSM6AAAAAAQ2HJOSY>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
|
But I'm not super familiar with the options or libraries required?
…On Fri, Sep 30, 2022 at 9:27 PM Roger Pack ***@***.***> wrote:
I may be able to look at it...
On Fri, Sep 30, 2022 at 7:34 PM darkjacky ***@***.***>
wrote:
> Would you mind adding support for Vulkan and OpenCL decoders?
>
> Cuda decode benchmark:
>
> video:12265kB audio:199549kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
> bench: utime=3.781s stime=2.547s rtime=24.493s
> bench: maxrss=139480kB
>
> OpenCL decode benchmark:
>
> video:12265kB audio:199549kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
> bench: utime=27.750s stime=1.438s rtime=6.049s
> bench: maxrss=128976kB
>
> Vulkan decode benchmark:
>
> video:12265kB audio:199549kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
> bench: utime=26.078s stime=1.672s rtime=6.418s
> bench: maxrss=208604kB
>
> They are roughly 4x faster than hardware decoders (by using the videocard
> 3D chip) and they have a fallback in case there is no hardware decoder
> making them a really good addition as you don't have to check anything.
> Just -hwaccel opencl and it will figure it out.
>
> —
> Reply to this email directly, view it on GitHub
> <#653>, or
> unsubscribe
> <https://github.com/notifications/unsubscribe-auth/AAADBUGMEEF5IQ7OAYRY5Y3WA6IJFANCNFSM6AAAAAAQ2HJOSY>
> .
> You are receiving this because you are subscribed to this thread.Message
> ID: ***@***.***>
>
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Would you mind adding support for Vulkan and OpenCL decoders?
Cuda decode benchmark:
OpenCL decode benchmark:
Vulkan decode benchmark:
They are roughly 4x faster than hardware decoders (by using the videocard 3D chip) and they have a fallback in case there is no hardware decoder making them a really good addition as you don't have to check anything. Just -hwaccel opencl and it will figure it out.
The text was updated successfully, but these errors were encountered: