Age | Commit message (Collapse) | Author | |
---|---|---|---|
2024-01-06 | replay ram info | dec05eba | |
2024-01-02 | Create directory to output file if not exists | dec05eba | |
2023-12-31 | Fix for opensuse: set unix domain socket mod 777 | dec05eba | |
Fixes screen capture on opensuse amd/intel or nvidia wayland | |||
2023-12-23 | Small changes who cares | dec05eba | |
2023-12-06 | Add logo to README | dec05eba | |
2023-12-06 | m | dec05eba | |
2023-12-06 | old info gone | dec05eba | |
2023-12-06 | Update readme with info about codecs | dec05eba | |
2023-11-20 | Remove mention of nixos package being out of date | dec05eba | |
2023-11-17 | Info about live streaming | dec05eba | |
2023-11-11 | Readme, info about old nvidia gpus | dec05eba | |
2023-11-05 | Update usage info about screen direct option | dec05eba | |
2023-10-21 | Attempt to fix amd/intel driver bug vram leak, force cfr when livestreaming, ↵ | dec05eba | |
replace std::filesystem with posix create replay directory automatically, recursively | |||
2023-10-15 | Install systemd service in install script, skip empty audio device | dec05eba | |
Stop systemd service with SIGINT | |||
2023-10-15 | Add the ability to save replays into folders by date | 4censord | |
This adds a new flag, '-mf' with a boolean argument If this is set, replays are organized in folders by date. With this, replays are saved like: $output_folder/2023-10-15/Replay_15-39-28.mp4 instead of all into the same folder. | |||
2023-10-07 | Improve dependencies section in readme | dec05eba | |
2023-10-04 | m | dec05eba | |
2023-10-04 | README replay on startup section | dec05eba | |
2023-10-01 | m | dec05eba | |
2023-10-01 | Issues readme section | dec05eba | |
2023-10-01 | Update readme with info about replay path stdout | dec05eba | |
2023-09-29 | Remove unused cursor code | dec05eba | |
2023-09-10 | Readme | dec05eba | |
2023-08-30 | readme | dec05eba | |
2023-08-18 | README password prompt | dec05eba | |
2023-08-17 | Unofficial package info | dec05eba | |
2023-08-09 | Remove coolbits script (document the nvidia-xconfig command to use instead) | dec05eba | |
2023-07-28 | Remove TODO items, improve readme replay service section | dec05eba | |
2023-07-28 | Nvidia readme | dec05eba | |
2023-07-28 | Create gpu-screen-recorder systemd service | dec05eba | |
2023-07-28 | Move additional scripts to 'extra' directory | dec05eba | |
2023-07-24 | README | dec05eba | |
2023-07-23 | readme | dec05eba | |
2023-07-23 | Update readme | dec05eba | |
2023-07-22 | Exit with exit code 10 if pkexec fails | dec05eba | |
2023-07-21 | vfr on wayland (even on nvidia), add uninstall script | dec05eba | |
2023-07-18 | readme typo | dec05eba | |
2023-07-18 | Experimental wayland support, test 1 | dec05eba | |
2023-07-16 | Wayland broken shit | dec05eba | |
2023-06-20 | Do not automatically install coolbits | dec05eba | |
2023-06-18 | Update readme | dec05eba | |
2023-06-07 | Mention that only h264/hevc is supported at the moment | dec05eba | |
2023-05-29 | Update readme about flatpak | dec05eba | |
2023-05-17 | Cleanup readme how to use section | dec05eba | |
2023-04-22 | kms_vaapi: add cursor capture for amd/intel monitor capture | dec05eba | |
Always find largest drm buf, to fix some cases when there are multiple planes and we accidentally capture a drm buf that isn't the target buf. We always want the full SCREEN drm buf. | |||
2023-04-21 | Add code to copy cursor image to opengl texture and update on event update ↵ | dec05eba | |
(not used yet) | |||
2023-04-21 | m | dec05eba | |
2023-04-21 | FAQ: amd/intel h264/hevc issue because of patent | dec05eba | |
2023-04-18 | Temporary issue description for VFR video | dec05eba | |
2023-04-15 | Fix AMD single monitor rotated display being rotated in recording | dec05eba | |
If there is only one monitor connected and it's rotated then the drm buf will also be rotated. This only the case with AMD and only when using one monitor! To fix this, we perform color conversion with an opengl shader which allows us to also rotate the texture. VAAPI supports rotation but it's not implemented by AMD at least. Performance seems to be the same as when using VAAPI, even when GPU usage is 100%. |