Discord screen sharing a game causes massive frame drops with motion

Brenvh · 9 July 2025 16:45

I can’t figure this one out, I’m new to Garuda Linux, this issue definitely isn’t a Garuda specific issue, I’ve had this on a few different distros (Ubuntu, Fedora) When I try streaming a game the encoder just drops frames when there is motion, the faster the motion the worse the frame drops, then it becomes a slideshow. I’ve tried different versions of discord, native packages, flatpaks but all of them yield the same results. I’ve tried screen capture instead of capturing the game itself, still the same. Using the vesktop client helps as it’s then using my CPU to encode, but that comes at a sizeable cost in performance unfortunately. I’ve tried making sure the native discord client is actually using my GPU to encode by monitoring in NVTOP and using the watch command watch -n 0.5 nvidia-smi --query-gpu=utilization.gpu,utilization.memory,utilization.encoder,utilization.decoder,temperature.gpu,power.draw,clocks.gr,clocks.sm,clocks.mem --format=csv,noheader doesn’t help as they don’t show any encoder specific usage, and in watch both decode and encode are just stuck at 0% which can’t be true because my GPU draws more power when I actually encode or decode.

System:
Kernel: 6.15.4-2-cachyos arch: x86_64 bits: 64 compiler: gcc v: 15.1.1
clocksource: tsc avail: hpet,acpi_pm
parameters: BOOT_IMAGE=/@/boot/vmlinuz-linux-cachyos
root=UUID=c69ffc36-5de9-4028-ab3f-988e79b3c35c rw rootflags=subvol=@
quiet loglevel=3 ibt=off
Desktop: KDE Plasma v: 6.4.2 tk: Qt v: N/A wm: kwin_wayland dm: SDDM
Distro: Garuda base: Arch Linux
Machine:
Type: Desktop Mobo: Gigabyte model: B550 AORUS ELITE V2 v: x.x serial: N/A
uuid: 035e02d8-04d3-05d4-b006-c80700080009 UEFI: American Megatrends LLC.
v: F18g date: 03/11/2025
CPU:
Info: model: AMD Ryzen 7 5700X socket: AM4 bits: 64 type: MT MCP
arch: Zen 3+ gen: 3 level: v3 note: check built: 2022 process: TSMC n6 (7nm)
family: 0x19 (25) model-id: 0x21 (33) stepping: 2 microcode: 0xA201213
Topology: cpus: 1x dies: 1 clusters: 1 cores: 8 threads: 16 tpc: 2
smt: enabled cache: L1: 512 KiB desc: d-8x32 KiB; i-8x32 KiB L2: 4 MiB
desc: 8x512 KiB L3: 32 MiB desc: 1x32 MiB
Speed (MHz): avg: 4092 min/max: 562/4666 boost: enabled
base/boost: 3400/4650 scaling: driver: amd-pstate-epp governor: powersave
volts: 1.1 V ext-clock: 100 MHz cores: 1: 4092 2: 4092 3: 4092 4: 4092
5: 4092 6: 4092 7: 4092 8: 4092 9: 4092 10: 4092 11: 4092 12: 4092
13: 4092 14: 4092 15: 4092 16: 4092 bogomips: 108594
Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3 svm
Vulnerabilities: <filter>
Graphics:
Device-1: NVIDIA GA104 [GeForce RTX 3070 Ti] vendor: CardExpert
driver: nvidia v: 575.64.03 alternate: nouveau,nvidia_drm
non-free: 550-570.xx+ status: current (as of 2025-04; EOL~2026-12-xx)
arch: Ampere code: GAxxx process: TSMC n7 (7nm) built: 2020-2023 pcie:
gen: 4 speed: 16 GT/s lanes: 16 ports: active: none off: DP-1,DP-2
empty: DP-3,HDMI-A-1 bus-ID: 05:00.0 chip-ID: 10de:2482 class-ID: 0300
Device-2: SN0002 2K Web Camera
driver: cdc_acm,hid-generic,snd-usb-audio,usbhid,uvcvideo type: USB rev: 2.0
speed: 480 Mb/s lanes: 1 mode: 2.0 bus-ID: 3-1:2 chip-ID: 2bdf:028a
class-ID: 0300 serial: <filter>
Display: unspecified server: X.Org v: 24.1.8 with: Xwayland v: 24.1.8
compositor: kwin_wayland driver: X: loaded: nvidia unloaded: modesetting
alternate: fbdev,nouveau,nv,vesa gpu: nvidia,nvidia-nvswitch
display-ID: :1 screens: 1
Screen-1: 0 s-res: 4480x1440 s-dpi: 96 s-size: 1185x381mm (46.65x15.00")
s-diag: 1245mm (49.01")
Monitor-1: DP-1 note: disabled pos: primary,top-right model: WV27QHDML1S
serial: <filter> built: 2025 res: mode: 2560x1440 hz: 300 scale: 100% (1)
dpi: 109 gamma: 1.2 size: 597x336mm (23.5x13.23") diag: 685mm (27")
ratio: 16:9 modes: max: 2560x1440 min: 640x480
Monitor-2: DP-2 note: disabled pos: bottom-l model: ASUS VG249
serial: <filter> built: 2019 res: mode: 1920x1080 hz: 144 scale: 100% (1)
dpi: 93 gamma: 1.2 size: 527x296mm (20.75x11.65") diag: 604mm (23.8")
ratio: 16:9 modes: max: 1920x1080 min: 640x480
API: EGL v: 1.5 hw: drv: nvidia platforms: device: 0 drv: nvidia device: 2
drv: swrast gbm: drv: nvidia surfaceless: drv: nvidia x11: drv: nvidia
inactive: wayland,device-1
API: OpenGL v: 4.6.0 compat-v: 4.5 vendor: nvidia mesa v: 575.64.03
glx-v: 1.4 direct-render: yes renderer: NVIDIA GeForce RTX 3070 Ti/PCIe/SSE2
memory: 7.81 GiB
API: Vulkan v: 1.4.313 layers: 3 device: 0 type: discrete-gpu name: NVIDIA
GeForce RTX 3070 Ti driver: nvidia v: 575.64.03 device-ID: 10de:2482
surfaces: N/A
Info: Tools: api: clinfo, eglinfo, glxinfo, vulkaninfo
de: kscreen-console,kscreen-doctor gpu: nvidia-settings,nvidia-smi
wl: wayland-info x11: xdpyinfo, xprop, xrandr
Audio:
Device-1: NVIDIA GA104 High Definition Audio vendor: CardExpert
driver: snd_hda_intel v: kernel pcie: gen: 4 speed: 16 GT/s lanes: 16
bus-ID: 05:00.1 chip-ID: 10de:228b class-ID: 0403
Device-2: Advanced Micro Devices [AMD] Starship/Matisse HD Audio
vendor: Gigabyte driver: snd_hda_intel v: kernel pcie: gen: 4 speed: 16 GT/s
lanes: 16 bus-ID: 07:00.4 chip-ID: 1022:1487 class-ID: 0403
Device-3: SN0002 2K Web Camera
driver: cdc_acm,hid-generic,snd-usb-audio,usbhid,uvcvideo type: USB rev: 2.0
speed: 480 Mb/s lanes: 1 mode: 2.0 bus-ID: 3-1:2 chip-ID: 2bdf:028a
class-ID: 0300 serial: <filter>
Device-4: fifine Microphone driver: hid-generic,snd-usb-audio,usbhid
type: USB rev: 1.1 speed: 12 Mb/s lanes: 1 mode: 1.1 bus-ID: 3-3:3
chip-ID: 3142:a008 class-ID: 0300
API: ALSA v: k6.15.4-2-cachyos status: kernel-api tools: N/A
Server-1: PipeWire v: 1.4.6 status: n/a (root, process) with:
1: pipewire-pulse status: active 2: wireplumber status: active
3: pipewire-alsa type: plugin 4: pw-jack type: plugin
tools: pactl,pw-cat,pw-cli,wpctl
Network:
Device-1: Realtek RTL8125 2.5GbE vendor: Gigabyte driver: r8169 v: kernel
pcie: gen: 2 speed: 5 GT/s lanes: 1 port: f000 bus-ID: 04:00.0
chip-ID: 10ec:8125 class-ID: 0200
IF: eno1 state: up speed: 1000 Mbps duplex: full mac: <filter>
Info: services: NetworkManager,systemd-timesyncd
Drives:
Local Storage: total: 1.86 TiB used: 250.09 GiB (13.1%)
ID-1: /dev/nvme0n1 maj-min: 259:0 vendor: Hikvision
model: HS-SSD-FUTURE 2048G size: 1.86 TiB block-size: physical: 512 B
logical: 512 B speed: 63.2 Gb/s lanes: 4 tech: SSD serial: <filter>
fw-rev: HS13683 temp: 43.9 C scheme: GPT
SMART: yes health: PASSED on: 214d 16h cycles: 659
read-units: 112,126,183 [57.4 TB] written-units: 94,994,729 [48.6 TB]
Partition:
ID-1: / raw-size: 115.19 GiB size: 115.19 GiB (100.00%)
used: 43.1 GiB (37.4%) fs: btrfs block-size: 4096 B dev: /dev/nvme0n1p5
maj-min: 259:5
ID-2: /boot/efi raw-size: 100 MiB size: 96 MiB (96.00%)
used: 54.8 MiB (57.1%) fs: vfat block-size: 512 B dev: /dev/nvme0n1p1
maj-min: 259:1
ID-3: /home raw-size: 115.19 GiB size: 115.19 GiB (100.00%)
used: 43.1 GiB (37.4%) fs: btrfs block-size: 4096 B dev: /dev/nvme0n1p5
maj-min: 259:5
ID-4: /var/log raw-size: 115.19 GiB size: 115.19 GiB (100.00%)
used: 43.1 GiB (37.4%) fs: btrfs block-size: 4096 B dev: /dev/nvme0n1p5
maj-min: 259:5
ID-5: /var/tmp raw-size: 115.19 GiB size: 115.19 GiB (100.00%)
used: 43.1 GiB (37.4%) fs: btrfs block-size: 4096 B dev: /dev/nvme0n1p5
maj-min: 259:5
Swap:
Kernel: swappiness: 133 (default 60) cache-pressure: 100 (default) zswap: no
ID-1: swap-1 type: zram size: 31.24 GiB used: 0 KiB (0.0%) priority: 100
comp: zstd avail: lzo-rle,lzo,lz4,lz4hc,deflate,842 dev: /dev/zram0
Sensors:
System Temperatures: cpu: 37.6 C mobo: N/A
Fan Speeds (rpm): N/A
Info:
Memory: total: 32 GiB available: 31.24 GiB used: 4.65 GiB (14.9%)
Processes: 412 Power: uptime: 1m states: freeze,mem,disk suspend: deep
avail: s2idle wakeups: 0 hibernate: platform avail: shutdown, reboot,
suspend, test_resume image: 12.41 GiB services: org_kde_powerdevil,
power-profiles-daemon, upowerd Init: systemd v: 257 default: graphical
tool: systemctl
Packages: pm: pacman pkgs: 1491 libs: 460 tools: octopi,paru Compilers:
clang: 20.1.7 gcc: 15.1.1 Shell: garuda-inxi (sudo) default: Bash v: 5.2.37
running-in: konsole inxi: 3.3.38
Garuda (2.7.3-1):
System install date:     2025-07-07
Last full system update: 2025-07-09
Is partially upgraded:   No
Relevant software:       snapper NetworkManager dracut nvidia-dkms
Windows dual boot:       Yes
Failed units:

filo · 9 July 2025 17:43

Hi there, welcome to the forum.
Have you looked in the Arch wiki?

You could try e.g. webcord, or an x11 session.

Brenvh · 10 July 2025 08:18

Hey, thank you : )

I’ve looked at the wiki, it’s mostly about screen sharing with audio which isn’t really an issue anymore, I tried Webcord, it just closes as soon as I try sharing my screen. I tried Vesktop as well, that works but only because it uses my CPU to encode. The X11 option isn’t on my lock screen anymore, not sure how to get to that anymore unfortunately.

nepti · 10 July 2025 09:12

Check whether you have installed kwin-x11 and plasma-x11-session. If not, then install the two packages. You should then be able to select the x11 session again in the login screen.

astrodarius · 10 July 2025 09:55

But you have a Nvidia RTX 3070 Ti:

I had similar issues with video.
I think you have the wrong drivers for your 30XX GPU. You need nvidia-open-dkms.

I’ll just tell you what others have told me, and this actually fixed most of the issues for me on a RTX3080 M:

First, remove incorrect drivers:

sudo pacman -Rdd garuda-nvidia-config nvidia-dkms nvidia-settings nvidia-utils opencl-nvidia libxnvctrl lib32-nvidia-utils lib32-opencl-nvidia gwe python-py3nvml nvidia-prime`

then (without reboot)

sudo pacman -S nvidia-open-dkms egl-wayland lib32-nvidia-utils lib32-opencl-nvidia nvidia-settings opencl-nvidia nvidia-utils nvidia-prime`

and type yes for all dependencies
then reboot

Then install vulkan and dxvk

sudo pacman -S dxvk vulkan-icd-loader vulkan-tools lib32-vulkan-icd-loader
paru -S lib32-libdxvk-gplasync

Then update kwin

kwriteconfig5 --file kwinrc --group Compositing --key Backend OpenGL3

Please see the topic I marked as solution here: Sluggish system on i9-11900H and RTX3080M - #52 by astrodarius

It’s a long topic about similar graphics issues, I was dropping a lot of frames (70%) when simply playing 4k videos.

Brenvh · 10 July 2025 18:32

I installed everything in order, did everything, I’m running the open kernel drivers now instead of the closed, now all decoding in my browser (brave) is being done by the CPU instead, don’t have any GPU acceleration anymore. I also tried installing the libva-vdpau-driver but no luck.

gampel · 10 July 2025 19:22

Default va-api + brave browser = nogo
to read:

astrodarius · 11 July 2025 10:03

Helpful URLs in Brave:
brave://gpu
It should show Video Decode: Hardware Accelerated

brave://media-internals
This allows you to look at what decoder is used for each specific video that is opened in brave. It’s easier if you click “Hide Players” at the top, and then refresh the tab that you want to analyze.

I did some more reading and tests on this, and it seems that I’m still on software decoding as well.

This is VP9 on a 4K video. Has software decoding (VpxVideoDecoder)

This is AV1 on a 1080p video. Also software decoding (Dav1dVideoDecoder)

And with the h264ify extension, I get the avc1 codec on yt, and brave uses ffmpeg to decode it.

Not sure if this is hardware or software, but my integrated gpu is doing something.

You’ve probably seen in the above screenshots some unusual buttons at the bottom of the youtube player. Well that’s my way of getting hardware video decoding.
Each button opens the video in MPV with the specified quality.
Steps to get the buttons: GitHub - dariusmihai/userscript-youtube-play-in-mpv: Allows you to launch the current YouTube video in MPV directly from the video page with a single click.

I tend to agree with @gampel that it’s just not possible right now to get reliable hardware video in Brave on wayland
I’m still trying various flags though. If I ever find a way that works, I’ll post here.

Brenvh · 11 July 2025 10:21

I’ve tested a bit further as well, the same video at 8k which I know is very extreme, zen browser (firefox) manages to play it with 20-30% cpu usage and with brave it maxes my CPU out in the beginning then it sits at between 50-60%, with both of these browsers my GPU is still sleeping pretty much drawing about 25watts, I’d imagine if the GPU decoded 8k it would draw a bit more than idle. As for the main issue I was facing regarding discord streaming (encoding) I’m probably just going to use vesktop and let the CPU endode, also seems like getting proper hardware acceleration there isn’t possible, I have yet to try an x11 session like the mod mentioned, which I’ll do now. But thanks for your extensive testing, let me know if you find anything interesting, and feel free to do some extensive testing regarding discord streaming! (specifically the encoding part) Not sure if you play games, but the encoder basically starts dropping frames when I stream a game and move my mouse around too much

astrodarius · 12 July 2025 11:17

With Firefox (Librewolf in my case), I also saw lower cpu usage, but I noticed lower quality even for 4k videos.
So I looked at “Stats for nerds” on youtube I saw that the codec was AV01, it was forcing that even though VP09 was available.

In Brave I have an average of 30% CPU usage when playing 4k videos, but also a lot of GPU usage on the intel iHD.
Dropped frames are at acceptable levels:

(i9-11900H, 32Gb RAM, RTX3080 M)
It’s still a lot of cpu for just a video, which is why I made that open-in-mpv thing.

Regarding your original streaming issue, is it possible for you to use OBS?
There you can set hardware encoding and custom options like bitrate and many more in the “Advanced” section.
I’m using OBS but not for games, it works well.
One thing to note though is that the Flatpak version for me doesn’t have the NVENC encoder option. So I’ve installed the native obs with pacman.

󰛓 ❯ pacman -Qs obs-studio
local/obs-studio 31.0.4-1
Free, open source software for live streaming and recording

A guide I found on how to use OBS to stream on Discord:

Install OBS Studio
Set Up Your Sources in OBS : In OBS, add the sources you want to stream. This can include a game capture, screen capture, or window capture. You can also add your webcam and any overlays or other visual elements you want to include in your stream.
Install the OBS VirtualCam Plugin : To send the OBS Studio signal to Discord, you need to download and install the OBS VirtualCam plugin. This plugin creates a virtual webcam that can be used in Discord. Close OBS Studio, navigate to the OBS VirtualCam plugin page, and download the plugin.
Configure OBS VirtualCam : Once the plugin is installed, open OBS Studio again and go to the Tools menu. Select “VirtualCam” and click “Start” to activate the virtual camera. This will allow OBS to send its output to the virtual camera.
Set Up Discord : Open Discord and go to Settings > Voice & Video. Under the Video section, select “OBS-Camera” as your webcam. This will enable Discord to use the virtual camera feed from OBS.
Start Streaming : Join a voice channel in Discord and click the “Screen” button to start streaming. Select the OBS window or specific application window to share. Click “Go Live” to start streaming your OBS output to the Discord channel.
Audio Setup (Optional) : If you want to include audio in your stream, you may need to set up additional audio sources in OBS. You can use tools like VB-Audio CABLE to route audio from OBS to Discord. In OBS, go to Settings > Audio and select “CABLE Input” as your Monitoring Device. In Discord, select “CABLE Output” as your input device under Voice & Video settings.

system · 26 July 2025 11:18

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.