Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

nvidia capture sdk support #1317

Closed
totaam opened this issue Sep 20, 2016 · 12 comments
Closed

nvidia capture sdk support #1317

totaam opened this issue Sep 20, 2016 · 12 comments

Comments

@totaam
Copy link
Collaborator

totaam commented Sep 20, 2016

Split from #365: the nvidia documentation is finally available: NVIDIA Capture SDK.

We could use this for high performance shadow servers or with stereo rendering, especially for win32 (#389) where this solves the biggest problem (efficient pixel capture).

@totaam
Copy link
Collaborator Author

totaam commented Feb 20, 2017

See also #1347 and #1308

@totaam
Copy link
Collaborator Author

totaam commented Mar 27, 2017

Problem is that the SDK states: NVIDIA Quadro 2000 class or higher, select Tesla including M6/M60/M10 and P40.
And sure enough, I tried on on both Linux and MS Windows 7, no go with a GTX 1070 / GTX 970. Looks like a purely commercial licensing restriction, again.

We probably need a Quadro Maxwell or newer to be able to do capture at 4k to HEVC, and those cards aren't cheap: M4000 costs ~$800.
The M2000 is a bit cheaper at $420, but for our purpose the performance will be similar to my old GTX 750 Ti... which costs $100, sigh.

For non-proprietary options, see #389 comment:19

A good explanation of the differences between NvFBC, NvIFR, NvENC:Steam : Explanation NvFBC, NvIFR, NvENC.

@totaam
Copy link
Collaborator Author

totaam commented Apr 28, 2017

Recorded some progress here: #389 comment:21.
Needs porting to Linux.

@totaam
Copy link
Collaborator Author

totaam commented May 2, 2017

Partial port to Linux done in r15787.
(CUDA variant still needs doing, will be useful for toying with GPU pre-compression)

@totaam
Copy link
Collaborator Author

totaam commented May 2, 2017

2017-05-02 18:36:45: antoine uploaded file nvfbc.pc-win32 (0.2 KiB)

pkgconfig file to use on MS Windows

@totaam
Copy link
Collaborator Author

totaam commented May 24, 2017

PM packaging fix in r15941 so users without the nvidia proprietary drivers can still install the package!

before:

$ rpm -qpR ./RPMS/x86_64/python2-xpra-2.1-0.fc26.x86_64.rpm  | grep -i nvidia
libnvidia-fbc.so.1()(64bit)

and after:

$ rpm -qpR ./RPMS/x86_64/python2-xpra-2.1-0.fc26.x86_64.rpm  | grep -i nvidia

@totaam
Copy link
Collaborator Author

totaam commented Jul 22, 2017

Important fix in r16457, see #1552#comment:4.

We now have zero-copy GPU transfers! See #365#comment:12

For Linux, as of r16479, the download-to-host-memory version works, the CUDA imagewrapper does not (memcpy_dtod fails).

@totaam
Copy link
Collaborator Author

totaam commented Jul 23, 2017

2017-07-23 17:22:06: antoine uploaded file nvfbc.pc (0.2 KiB)

pkg-config file used on Linux - fixes double-slash problem during rpmbuild: https://bugzilla.redhat.com/show_bug.cgi?id=304121

@totaam
Copy link
Collaborator Author

totaam commented Jul 24, 2017

Linux fixes in r16492, re-tested on win32 too. Closing at last.

Will follow up in #1597 + #1598.

@totaam totaam closed this as completed Jul 24, 2017
@totaam
Copy link
Collaborator Author

totaam commented Oct 3, 2017

r17078 adds support for loading API license keys.

@totaam
Copy link
Collaborator Author

totaam commented Jun 25, 2019

NVFBC WINDOWS 10 SUPPORT DEPRECATION.
The recommended alternative is Desktop Duplication API. (some details are already in #389)

@totaam
Copy link
Collaborator Author

totaam commented Dec 3, 2019

To enable this on win32 consumer hardware, see nvidia-patch Windows: NvFBC support

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant