Significant performance drops on linux?

[hayakiii]

I've tried out several linux distros so far and I like linux over windows due to how fast and responsive it is. But there's been one thing making me stick to windows which is performance drops when coming to linux, Although I've heard that amd hardware is great on linux and works out of the box which does aswell in my case but I don't get on par performance in linux with any games I've tried so far.. I'll list down my laptop specs aswell as games aswell as the tweaks I've tried to get atleast on par performance.

Laptop specs:

Ryzen 5 5625u
Vega 7 (apu)
16gb single channel ram

Games I've tried:

Life is strange 1, Before the storm
Gta 4
Tomb raider 2013
Mad Max
Batman Arkham Asylum
Sleeping dogs
Assassin Creed 4 blackflag

All these games I've tried above give me around less 15-20 fps and no matter what I do the fps difference just does not goes away.

Tweaks I've tried so far:
Using gamemode on launch options
Using high performance power plan
Playing with the laptop is on charging
Increasing VRAM from the bios
Using Different protons. (GE, Experimental, Cachyosproton, and have tried protons from the site protondb for those games.)

I've also made sure the settings are same and the resolution aswell but the fps difference is always there.

Distros I've tried so far.

Linux Mint
CachyOS
Bazzite
Nobara
Ubuntu

If anyone has any idea of why this happens lmk, My current guess is its due to the translation layer proton creates. Which can't be solved entirely and there will always be a performance tax but that's just something I've thought of.. I'm still not sure if that thing is true or just a myth with the performance tax..

 

[CaffeineAddict]

You need to tell us what software/setup you use to play games as well as which GPU driver do you used to play them.

And also provide detailed system info with:

sudo inxi -remzax

More details you provide the easier it will be to find a solution/problem.

 

[AlphaObeisance]

It's likely a combination of DXVK and your Single Channel RAM bottleneck. Linux+Proton kind of exacerbates the bottleneck of your hardware because it's constantly demanding bandwidth to translate graphics instructions in real time.

Your GPU relies entirely on system RAM, your RAM is single channel, which is effectively starving the GPU of bandwidth; toss that into the salad with Linux and Proton demands and it's even more memory bandwidth to translate windows games on Linux. It's kind of like a quicksand circumstance.

 

[hayakiii]

You need to tell us what software/setup you use to play games as well as which GPU driver do you used to play them. And also provide detailed system info with:

sudo inxi -remzax

More details you provide the easier it will be to find a solution/problem.

You need to tell us what software/setup you use to play games as well as which GPU driver do you used to play them.

And also provide detailed system info with:

sudo inxi -remzax

More details you provide the easier it will be to find a solution/problem.

Hey the command output this:

System:
Kernel: 7.0.9-1-cachyos arch: x86_64 bits: 64 compiler: clang v: 22.1.5
clocksource: tsc avail: hpet,acpi_pm
parameters: BOOT_IMAGE=/@/boot/vmlinuz-linux-cachyos
root=UUID=87cd8e6f-f25a-4679-a77b-5f15ebfabad6 rw rootflags=subvol=@
nowatchdog nvme_load=YES splash loglevel=3
Desktop: Cinnamon v: 6.6.8 tk: GTK v: 3.24.52 wm: Muffin v: 6.6.3 tools:
avail: cinnamon-screensaver dm: LightDM v: 1.32.0 Distro: CachyOS
base: Arch Linux
Machine:
Type: Laptop System: Acer product: Aspire Lite AL15-41 v: N/A
serial: <filter> Chassis: type: 10 serial: N/A
Mobo: Acer model: Aspire Lite AL15-41 serial: N/A
part-nu: 0000000000000000 uuid: 7e9fe9cd-2089-455d-be49-e9031a3887f2
Firmware: UEFI vendor: INSYDE v: 1.07.30RAC1 date: 02/11/2025
Battery:
ID-1: BAT0 charge: 27.3 Wh (79.4%) condition: 34.4/36.2 Wh (95.1%)
volts: 12.16 min: 11.4 model: Notebook BAT type: Li-ion serial: <filter>
charging: status: not charging cycles: 22
Memory:
System RAM: total: 16 GiB available: 14.46 GiB used: 2.28 GiB (15.8%)
Array-1: capacity: 64 GiB slots: 2 modules: 1 EC: None
max-module-size: 32 GiB note: est.
Device-1: Channel-A DIMM 0 type: no module installed
Device-2: Channel-B DIMM 0 type: DDR4 detail: synchronous unbuffered
(unregistered) size: 16 GiB speed: 3200 MT/s volts: curr: 1.2 min: 1.2
max: 1.2 width (bits): data: 64 total: 64 manufacturer: Corsair
part-no: CMB6-DHDA1BAR08D00 serial: <filter>
CPU:
Info: model: AMD Ryzen 5 5625U with Radeon Graphics socket: FP6 bits: 64
type: MT MCP arch: Zen 3 gen: 3 level: v3 note: check built: 2021-22
process: TSMC n7 (7nm) family: 0x19 (25) model-id: 0x50 (80) stepping: 0
microcode: 0xA500012
Topology: cpus: 1x dies: 1 clusters: 1 cores: 6 threads: 12 tpc: 2
smt: enabled cache: L1: 384 KiB desc: d-6x32 KiB; i-6x32 KiB L2: 3 MiB
desc: 6x512 KiB L3: 16 MiB desc: 1x16 MiB
Speed (MHz): avg: 1111 min/max: 423/4390 boost: enabled
base/boost: 2300/4350 scaling: driver: amd-pstate-epp governor: performance
volts: 1.2 V ext-clock: 100 MHz cores: 1: 1111 2: 1111 3: 1111 4: 1111
5: 1111 6: 1111 7: 1111 8: 1111 9: 1111 10: 1111 11: 1111 12: 1111
bogomips: 55092
Flags-basic: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a
ssse3 svm
Vulnerabilities:
Type: gather_data_sampling status: Not affected
Type: ghostwrite status: Not affected
Type: indirect_target_selection status: Not affected
Type: itlb_multihit status: Not affected
Type: l1tf status: Not affected
Type: mds status: Not affected
Type: meltdown status: Not affected
Type: mmio_stale_data status: Not affected
Type: old_microcode status: Not affected
Type: reg_file_data_sampling status: Not affected
Type: retbleed status: Not affected
Type: spec_rstack_overflow mitigation: Safe RET
Type: spec_store_bypass mitigation: Speculative Store Bypass disabled via
prctl
Type: spectre_v1 mitigation: usercopy/swapgs barriers and __user pointer
sanitization
Type: spectre_v2 mitigation: Retpolines; IBPB: conditional; IBRS_FW;
STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not
affected
Type: srbds status: Not affected
Type: tsa mitigation: Clear CPU buffers
Type: tsx_async_abort status: Not affected
Type: vmscape mitigation: IBPB before exit to userspace
Graphics:
Device-1: Advanced Micro Devices [AMD/ATI] Barcelo vendor: CLEVO/KAPOK
driver: amdgpu v: kernel arch: GCN-5 code: Vega process: GF 14nm
built: 2017-20 pcie: gen: 3 speed: 8 GT/s lanes: 16 ports: active: eDP-1
empty: DP-1,DP-2 bus-ID: 05:00.0 chip-ID: 1002:15e7 class-ID: 0300
temp: 53.0 C
Device-2: Bison BisonCam NB Pro driver: uvcvideo type: USB rev: 2.0
speed: 480 Mb/s lanes: 1 mode: 2.0 bus-ID: 3-4:3 chip-ID: 5986:9102
class-ID: 0e02
Display: x11 server: X.org v: 1.21.1.22 with: Xwayland v: 24.1.11 driver:
X: loaded: amdgpu unloaded: modesetting alternate: fbdev,vesa dri: radeonsi
gpu: amdgpu display-ID: :0 screens: 1
Screen-1: 0 s-res: 1920x1080 s-size: <missing: xdpyinfo>
Monitor-1: eDP-1 mapped: eDP model-id: HKC 0x023d built: 2023 res:
mode: 1920x1080 hz: 60 scale: 100% (1) dpi: 142 gamma: 1.2
size: 344x194mm (13.54x7.64") diag: 395mm (15.5") ratio: 16:9 modes:
max: 1920x1080 min: 640x480
API: EGL v: 1.5 hw: drv: amd radeonsi platforms: device: 0 drv: radeonsi
device: 1 drv: swrast gbm: drv: radeonsi surfaceless: drv: radeonsi x11:
drv: radeonsi inactive: wayland
API: OpenGL v: 4.6 vendor: amd mesa v: 26.1.1-arch2.1 glx-v: 1.4
direct-render: yes renderer: AMD Radeon Graphics (radeonsi renoir ACO DRM
3.64 7.0.9-1-cachyos) device-ID: 1002:15e7 memory: 1000 MiB unified: yes
API: Vulkan v: 1.4.350 layers: 5 device: 0 type: integrated-gpu name: AMD
Radeon Graphics (RADV RENOIR) driver: mesa radv v: 26.1.1-arch2.1
device-ID: 1002:15e7 surfaces: N/A
Info: Tools: api: eglinfo, glxinfo, vulkaninfo x11: xprop,xrandr
Audio:
Device-1: Advanced Micro Devices [AMD/ATI] Renoir/Cezanne HDMI/DP Audio
vendor: CLEVO/KAPOK driver: snd_hda_intel v: kernel pcie: gen: 3
speed: 8 GT/s lanes: 16 bus-ID: 05:00.1 chip-ID: 1002:1637 class-ID: 0403
Device-2: Advanced Micro Devices [AMD] Audio Coprocessor
vendor: CLEVO/KAPOK driver: N/A alternate: snd_pci_acp3x, snd_rn_pci_acp3x,
snd_pci_acp5x, snd_pci_acp6x, snd_acp_pci, snd_rpl_pci_acp6x, snd_pci_ps,
snd_sof_amd_renoir, snd_sof_amd_rembrandt, snd_sof_amd_vangogh,
snd_sof_amd_acp63, snd_sof_amd_acp70 pcie: gen: 3 speed: 8 GT/s lanes: 16
bus-ID: 05:00.5 chip-ID: 1022:15e2 class-ID: 0480
Device-3: Advanced Micro Devices [AMD] Ryzen HD Audio vendor: CLEVO/KAPOK
driver: snd_hda_intel v: kernel pcie: gen: 3 speed: 8 GT/s lanes: 16
bus-ID: 05:00.6 chip-ID: 1022:15e3 class-ID: 0403
API: ALSA v: k7.0.9-1-cachyos status: kernel-api
tools: alsactl,alsamixer,amixer
Server-1: sndiod v: N/A status: off tools: aucat,midicat,sndioctl
Server-2: JACK v: 1.9.22 status: off tools: N/A
Server-3: PipeWire v: 1.6.5 status: n/a (root, process) with:
1: pipewire-pulse status: active 2: wireplumber status: active
3: pipewire-alsa type: plugin tools: pactl,pw-cat,pw-cli,wpctl
Network:
Device-1: Realtek RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet
vendor: CLEVO/KAPOK driver: r8169 v: kernel pcie: gen: 1 speed: 2.5 GT/s
lanes: 1 port: 3000 bus-ID: 02:00.0 chip-ID: 10ec:8168 class-ID: 0200
IF: enp2s0 state: down mac: <filter>
Device-2: Realtek RTL8851BE PCIe 802.11ax Wireless Network
vendor: AzureWave driver: rtw89_8851be v: kernel pcie: gen: 1
speed: 2.5 GT/s lanes: 1 port: 2000 bus-ID: 03:00.0 chip-ID: 10ec:b851
class-ID: 0280
IF: wlan0 state: up mac: <filter>
Info: services: NetworkManager, systemd-timesyncd, wpa_supplicant
Bluetooth:
Device-1: IMC Networks Bluetooth Radio driver: btusb v: 0.8 type: USB
rev: 1.0 speed: 12 Mb/s lanes: 1 mode: 1.1 bus-ID: 3-3:2 chip-ID: 13d3:3601
class-ID: e001 serial: <filter>
Report: btmgmt ID: hci0 rfk-id: 0 state: up address: <filter> bt-v: 5.3
lmp-v: 12 status: discoverable: no pairing: no class-ID: 6c010c
Drives:
Local Storage: total: 505.59 GiB used: 34.8 GiB (6.9%)
ID-1: /dev/nvme0n1 maj-min: 259:0 vendor: BIWIN
model: AP823 CE980TB8900-512G size: 476.94 GiB block-size: physical: 512 B
logical: 512 B speed: 63.2 Gb/s lanes: 4 tech: SSD serial: <filter>
fw-rev: K.3.6.01 temp: 37.9 C scheme: GPT
SMART: yes health: PASSED on: 23d 3h cycles: 1,061
read-units: 45,235,834 [23.1 TB] written-units: 27,365,390 [14.0 TB]
ID-2: /dev/sda maj-min: 8:0 vendor: SanDisk model: Cruzer Blade
size: 28.65 GiB block-size: physical: 512 B logical: 512 B type: USB
rev: 2.0 spd: 480 Mb/s lanes: 1 mode: 2.0 tech: N/A serial: <filter>
fw-rev: 1.00 scheme: MBR
SMART Message: Unknown USB bridge. Flash drive/Unsupported enclosure?
Partition:
ID-1: / raw-size: 97.16 GiB size: 97.16 GiB (100.00%)
used: 10.44 GiB (10.7%) fs: btrfs block-size: 4096 B dev: /dev/nvme0n1p6
maj-min: 259:6
ID-2: /boot/efi raw-size: 512 MiB size: 511 MiB (99.80%)
used: 664 KiB (0.1%) fs: vfat block-size: 512 B dev: /dev/nvme0n1p5
maj-min: 259:5
ID-3: /home raw-size: 97.16 GiB size: 97.16 GiB (100.00%)
used: 10.44 GiB (10.7%) fs: btrfs block-size: 4096 B dev: /dev/nvme0n1p6
maj-min: 259:6
ID-4: /var/log raw-size: 97.16 GiB size: 97.16 GiB (100.00%)
used: 10.44 GiB (10.7%) fs: btrfs block-size: 4096 B dev: /dev/nvme0n1p6
maj-min: 259:6
ID-5: /var/tmp raw-size: 97.16 GiB size: 97.16 GiB (100.00%)
used: 10.44 GiB (10.7%) fs: btrfs block-size: 4096 B dev: /dev/nvme0n1p6
maj-min: 259:6
Swap:
Kernel: swappiness: 150 (default 60) cache-pressure: 50 (default 100)
zswap: no
ID-1: swap-1 type: zram size: 14.46 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: 63.9 C mobo: N/A gpu: amdgpu temp: 54.0 C
Fan Speeds (rpm): N/A
Repos:
Packages: pm: pacman pkgs: 1177 libs: 374 tools: paru pm: flatpak pkgs: 0
Active pacman repo servers in: /etc/pacman.d/cachyos-mirrorlist
1: https://mirror5.krfoss.org/cachyos/repo/$arch/$repo
2: https://mirror.krfoss.org/cachyos/repo/$arch/$repo
3: https://cachyos.next-works.it/repo/$arch/$repo
4: https://mirror.hb9hil.org/cachyos/repo/$arch/$repo
5: https://cdn77.cachyos.org/repo/$arch/$repo
6: https://us-mnz.soulharsh007.dev/cachyos/repo/$arch/$repo
7: https://mirror.zyner.org/mirror/cachyos/repo/$arch/$repo
8: https://mirror2.keiminem.com/cachyos/repo/$arch/$repo
9: https://cachyos-mirror.m1k.cloud/repo/$arch/$repo
10: https://cdn.lansing2600.org/cachyos/repo/$arch/$repo
11: https://nl.mirror.cx/cachyos/repo/$arch/$repo
12: https://mirror.keiminem.com/cachyos/repo/$arch/$repo
13: https://at.cachyos.org/repo/$arch/$repo
14: https://cachyos.doridian.net/repo/$arch/$repo
15: https://mirrors.lansing2600.org/cachyos/repo/$arch/$repo
Active pacman repo servers in: /etc/pacman.d/cachyos-v3-mirrorlist
1: https://mirror5.krfoss.org/cachyos/repo/$arch_v3/$repo
2: https://mirror.krfoss.org/cachyos/repo/$arch_v3/$repo
3: https://cachyos.next-works.it/repo/$arch_v3/$repo
4: https://mirror.hb9hil.org/cachyos/repo/$arch_v3/$repo
5: https://cdn77.cachyos.org/repo/$arch_v3/$repo
6: https://us-mnz.soulharsh007.dev/cachyos/repo/$arch_v3/$repo
7: https://mirror.zyner.org/mirror/cachyos/repo/$arch_v3/$repo
8: https://mirror2.keiminem.com/cachyos/repo/$arch_v3/$repo
9: https://cachyos-mirror.m1k.cloud/repo/$arch_v3/$repo
10: https://cdn.lansing2600.org/cachyos/repo/$arch_v3/$repo
11: https://nl.mirror.cx/cachyos/repo/$arch_v3/$repo
12: https://mirror.keiminem.com/cachyos/repo/$arch_v3/$repo
13: https://at.cachyos.org/repo/$arch_v3/$repo
14: https://cachyos.doridian.net/repo/$arch_v3/$repo
15: https://mirrors.lansing2600.org/cachyos/repo/$arch_v3/$repo
Active pacman repo servers in: /etc/pacman.d/mirrorlist
1: https://archlinux.cachyos.org/repo/$repo/os/$arch
2: https://fastly.mirror.pkgbuild.com/$repo/os/$arch
3: https://sg.arch.niranjan.co/$repo/os/$arch
4: https://mirrors.saswata.cc/archlinux/$repo/os/$arch
5: https://singapore.mirror.pkgbuild.com/$repo/os/$arch
6: https://archlinux.interhost.co.il/$repo/os/$arch
7: https://at.arch.niranjan.co/$repo/os/$arch
8: https://mirror.keiminem.com/archlinux/$repo/os/$arch
9: https://mirror.neuf.no/archlinux/$repo/os/$arch
10: https://mirrors.aliyun.com/archlinux/$repo/os/$arch
11: https://ro.arch.niranjan.co/$repo/os/$arch
12: https://ftp.ek-cer.hu/pub/mirrors/ftp.archlinux.org/$repo/os/$arch
13: https://mirror.xeonbd.com/archlinux/$repo/os/$arch
14: https://mirrors.nguyenhoang.cloud/archlinux/$repo/os/$arch
15: https://mirror.limda.net/archlinux/$repo/os/$arch
16: https://tr.arch.niranjan.co/$repo/os/$arch
17: https://mirrors.uni-plovdiv.net/archlinux/$repo/os/$arch
18: https://arch.mirror.constant.com/$repo/os/$arch
19: https://gsl-syd.mm.fcix.net/archlinux/$repo/os/$arch
20: https://mirror.timtal.com.tr/archlinux/$repo/os/$arch
21: https://mk.mirrors.mk/archlinux/$repo/os/$arch
22: https://mirror.archlinux.si/$repo/os/$arch
23: https://ams.nl.mirrors.bjg.at/arch/$repo/os/$arch
24: https://taipei.mirror.pkgbuild.com/$repo/os/$arch
25: https://arch-mirror.wtako.net/$repo/os/$arch
26: https://arch.yhtez.xyz/$repo/os/$arch
27: https://mirror.mangohost.net/archlinux/$repo/os/$arch
28: https://mirrors.pidginhost.com/arch/$repo/os/$arch
29: https://theswissbay.ch/archlinux/$repo/os/$arch
30: https://mirrors.xtom.ee/archlinux/$repo/os/$arch
31: https://mirrors.huongnguyen.dev/arch/$repo/os/$arch
32: https://archlinux.mirror-services.net/archlinux/$repo/os/$arch
33: https://mirror.1ago.be/archlinux/$repo/os/$arch
34: https://mirrors.dotsrc.org/archlinux/$repo/os/$arch
35: https://mirror.tux.si/arch/$repo/os/$arch
36: https://au.arch.niranjan.co/$repo/os/$arch
37: https://repo.br.ee/arch/$repo/os/$arch
38: https://uk.arch.niranjan.co/$repo/os/$arch
39: https://md.arch.niranjan.co/$repo/os/$arch
40: https://mirror.ourhost.az/archlinux/$repo/os/$arch
41: https://mirror.maeen.sa/arch-mirror/$repo/os/$arch
42: https://mirror.dkm.cz/archlinux/$repo/os/$arch
43: https://mirror.aria-on-the-planet.es/archlinux/$repo/os/$arch
44: https://repository.su/archlinux/$repo/os/$arch
45: https://hu.mirror.frigyes.dev/archlinux/$repo/os/$arch
46: https://archlinux.grena.ge/$repo/os/$arch
47: https://mirror.quantum5.ca/archlinux/$repo/os/$arch
48: https://archlinux.koyanet.lv/archlinux/$repo/os/$arch
49: https://mirror.kku.ac.th/archlinux/$repo/os/$arch
50: https://ftp.linux.edu.lv/archlinux/$repo/os/$arch
51: https://us.arch.niranjan.co/$repo/os/$arch
52: https://mirror.raiolanetworks.com/archlinux/$repo/os/$arch
53: https://at.arch.mirror.kescher.at/$repo/os/$arch
54: https://mirror.osbeck.com/archlinux/$repo/os/$arch
55: https://mirror.twds.com.tw/archlinux/$repo/os/$arch
56: https://no.mirror.cx/archlinux/$repo/os/$arch
57: https://mirror.franscorack.com/archlinux/$repo/os/$arch
58: https://mirror.trapmaid.org/archlinux/$repo/os/$arch
59: https://ftpmirror.infania.net/mirror/archlinux/$repo/os/$arch
Info:
Processes: 346 Power: uptime: 16m states: freeze,mem,disk suspend: s2idle
wakeups: 0 hibernate: platform avail: shutdown, reboot, suspend, test_resume
image: 5.72 GiB services: csd-power, power-profiles-daemon, upowerd
Init: systemd v: 260 default: graphical tool: systemctl
Compilers: clang: 22.1.5 gcc: 16.1.1 Shell: Sudo (sudo) v: 1.9.17p2
default: Bash v: 5.3.9 running-in: gnome-terminal inxi: 3.3.40

~
❯

 

[kibasnowpaw]

I don’t think Proton/DXVK is the main villain here, at least not by itself.

Your biggest problem is probably the Vega 7 iGPU + single-channel RAM. The Ryzen 5 5625U uses system RAM as GPU memory, and your inxi output shows only one 16 GB DDR4-3200 stick installed, with the other channel empty. That means the iGPU is running with much less memory bandwidth than it would in dual channel. For an integrated GPU, that can be a massive bottleneck.

Proton can add some overhead, especially in older DirectX games, but it is not normally a fixed “Linux is always 15–20 FPS slower” tax. Plenty of games run close to Windows, and some even run better, but weak iGPUs are much more sensitive to memory bandwidth, driver path, shader translation, and compositor overhead.

The first thing I would do is add a second matching RAM stick so the laptop runs dual channel. That will probably help more than changing distro or Proton version again.

A few things I would also check:

• Make sure 32-bit Vulkan/Mesa packages are installed, since many older Windows games need them through Proton.
• Test with MangoHud and check whether the GPU is pinned at 99% or whether CPU clocks/power are dropping.
• Try disabling Cinnamon’s compositor while gaming, or test a lighter session/desktop.
• Do not expect increasing VRAM in BIOS to fix this. Vega uses shared system memory anyway, and bandwidth matters more than the VRAM number.
• Keep the laptop plugged in and check that it is not stuck in a low-power thermal or firmware state.

So yes, Proton/DXVK can cost performance in some cases, but with this hardware I would blame single-channel RAM first. A Vega iGPU without dual-channel memory is basically being starved.