Added ability to record other X displays

[KarolS]

No description provided.

[ubuntuaddicted]

does this merge request allow the capture of another X server IF it's not being displayed? for example there's an X server running on tty8 and ssr is running on tty7, can ssr record what's on tty8 IF it's currently not active? obs-studio tried implementing this as well but found that unless the other X server was active, it wasn't recording anything but the last still image of what was on the other X server's display just prior to switching to the other X server.

[MaartenBaert]

It depends on your drivers. It's up to the driver to decide whether it wants to render things that are not visible. In some cases you can't even record partially hidden windows with OpenGL (e.g. with the NVidia drivers, parts of the window that are not on your screen are not rendered at all).

[ubuntuaddicted]

that makes sense. it's most likely not rendering anything on tty8 when i'm on tty7 then. otherwise i think my GPU would be showing more utilization if i had a game actually being rendered on tty8 while i was active on tty7.

along the same lines of what's rendered on screen, when i would bring up the steam overlay whilke playing borderlands and move my mouse onto my other monitor the game would auto minimize which would result in the capture program to record a red box because due to the game auto minimizing it wasn't being rendered anymore. to get around that luckily there's an SDL envornment variable in my ~/.profile which states to never minimize a game even if it loses focus, worked out nicely. but that all depends whether the game was coded with SDL or not.

so you never accepted this commit?

[StripedMonkey]

I believe he did or he already had something that added it in the making. I
know this because I can record my second screen.

On Wed, Nov 26, 2014 at 3:39 PM, Ubu the Tech Guru <notifications@github.com

wrote:

that makes sense. it's most likely not rendering anything on tty8 when i'm
on tty7 then. otherwise i think my GPU would be showing more utilization if
i had a game actually being rendered on tty8 while i was active on tty7.

along the same lines of what's rendered on screen, when i would bring up
the steam overlay whilke playing borderlands and move my mouse onto my
other monitor the game would auto minimize which would result in the
capture program to record a red box because due to the game auto minimizing
it wasn't being rendered anymore. to get around that luckily there's an SDL
envornment variable in my ~/.profile which states to never minimize a game
even if it loses focus, worked out nicely. but that all depends whether the
game was coded with SDL or not.

so you never accepted this commit?

—
Reply to this email directly or view it on GitHub
#74 (comment).

[MaartenBaert]

I didn't accept it mainly because I don't want to add a GUI option for it at the moment. It's just one more thing that users can accidentally mess up, and it's really only useful in some very specific situations. I wouldn't mind adding this to a more low-level interface like the command-line, but recording from the command-line isn't possible yet. In the long term I'm planning to redesign the GUI to create more space for advanced options, which would make this more practical. So I'm leaving this here until I can find a good way to expose the feature.

[ubuntuaddicted]

makes sense, thank you for explaining it to me. ;) since I have your attention would it be reasonable to assume that creating a video with multiple audio tracks isn't that difficult, it just has to be coded correctly to make the main audio track be the game and then a second audio track be another pulseaudio source like a microphone. this would allow for a user to extract the microphone audio track from the video for better manipulation during post editing. i'm considering learning c++ to enable this feature within ssr myself. any information about this or even feedback would be much appreciated.

[MaartenBaert]

The synchronizer needs to be generalized so it can handle any number of streams rather than just one video and audio stream. Unfortunately that doesn't really fit with the source/sink model that I'm using now, so it requires some changes. The bigger problem is that it will make the user interface far more complex, because you will essentially need the sound settings multiple times. This is something that I want to address in the GUI redesign.

[StripedMonkey]

What kind of GUI are you looking at? I really like the design that you have
now but if you add more features I can't imagine how you would keep the
cleanliness of the design.

On Sat, Nov 29, 2014 at 11:10 AM, MaartenBaert notifications@github.com
wrote:

The synchronizer needs to be generalized so it can handle any number of
streams rather than just one video and audio stream. Unfortunately that
doesn't really fit with the source/sink model that I'm using now, so it
requires some changes. The bigger problem is that it will make the user
interface far more complex, because you will essentially need the sound
settings multiple times. This is something that I want to address in the
GUI redesign.

—
Reply to this email directly or view it on GitHub
#74 (comment).