Quoting Diego Biurrun (2016-05-11 17:40:12)
> On Tue, May 03, 2016 at 08:10:55AM +0200, Anton Khirnov wrote:
> > --- a/configure
> > +++ b/configure
> > @@ -242,6 +226,20 @@ External library support:
> >    --enable-x11grab           X11 grabbing through xlib (legacy, use xcb 
> > instead)
> >    --enable-zlib              compression [autodetect]
> >  
> > +  The following libraries provide various hardware acceleration features:
> > +    --enable-cuda    Nvidia CUDA (dynamically linked)
> > +    --enable-d3d11va Microsoft Direct3D 11 video acceleration [auto]
> > +    --enable-dxva2   Microsoft DirectX 9 video acceleration [auto]
> > +    --enable-libmfx  Intel MediaSDK (AKA Quick Sync Video)
> > +    --enable-libnpp  Nvidia CUDA processing
> > +    --enable-mmal    Broadcom Multi-Media Abstraction Layer (Raspberry Pi)
> > +    --enable-nvenc   Nvidia video encoding
> > +    --enable-omx     OpenMAX IL
> > +    --enable-omx-rpi OpenMAX IL for Raspberry Pi
> > +    --enable-vaapi   Video Acceleration API (mainly Unix/Intel)
> > +    --enable-vda     Apple Video Decode Acceleration [auto]
> > +    --enable-vdpau   Nvidia Video Decode and Presentation API for Unix 
> > [auto]
> > +
> >  Toolchain options:
> >    --arch=ARCH              select architecture [$arch]
> >    --cpu=CPU                select the minimum required CPU (affects
> 
> I'd maintain the columns.  What's the difference between -cuda and
> -libnpp?

libcuda is the lowlevel api for interacting with nvidia hardware. libnpp
is a collection of processing functions that work on the GPU (CUDA)
objects.

-- 
Anton Khirnov
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to