Mint 20 Cinnamon with uhd 620 very sluggish

[sander29]

Had to create a new topic, but basically the same issues as described in this thread

It's hard to pinpoint what exactly is the problem, there are no several-second-long complete freezes, but the whole DE seems slow. It's like every action has 50-100ms delay.

Things I've tried:

• Removing the xserver-xorg-video-intel

• disabled all the bells and whistles I was able to find from the settings

• switched kernel from 5.4 to 5.8

• updated the Mesa video package (and reinstalled xserver-xorg-video-intel as well)

• increased the default video memory in bios (to 512MB)

• running the DE in software rendering mode

• tried various perfomance tweaks in the applications (disable smooth scrolling, enable/disable hardware acceleration etc)

So far nothing has helped. Running the DE in software rendering seems to improve the situation a bit, but compared to the results phd21(in the linked thread) was able to get from the same UHD 620 it seems nowhere near the performance it should be getting.

Code: Select all

$ inxi -Fxzd
System:
  Kernel: 5.8.0-29-generic x86_64 bits: 64 compiler: N/A 
  Desktop: Cinnamon 4.6.7 Distro: Linux Mint 20 Ulyana 
  base: Ubuntu 20.04 focal 
Machine:
  Type: Laptop System: HP product: HP EliteBook 840 G5 v: N/A 
  serial: <filter> 
  Mobo: HP model: 83B2 v: KBC Version 04.53.00 serial: <filter> UEFI: HP 
  v: Q78 Ver. 01.04.00 date: 09/12/2018 
Battery:
  ID-1: BAT0 charge: 42.5 Wh condition: 42.5/42.5 Wh (100%) 
  model: Hewlett-Packard Primary status: Full 
  ID-2: hidpp_battery_0 charge: N/A condition: N/A 
  model: Logitech G703 LIGHTSPEED Wireless Gaming Mouse w/ HERO 
  status: Discharging 
CPU:
  Topology: Quad Core model: Intel Core i7-8650U bits: 64 type: MT MCP 
  arch: Kaby Lake rev: A L2 cache: 8192 KiB 
  flags: avx avx2 lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx 
  bogomips: 33599 
  Speed: 1840 MHz min/max: 400/4200 MHz Core speeds (MHz): 1: 1954 2: 1890 
  3: 1948 4: 1939 5: 1935 6: 1911 7: 1920 8: 1916 
Graphics:
  Device-1: Intel UHD Graphics 620 vendor: Hewlett-Packard driver: i915 
  v: kernel bus ID: 00:02.0 
  Display: x11 server: X.Org 1.20.8 driver: modesetting unloaded: fbdev,vesa 
  resolution: 1920x1080~120Hz, 2560x1440~60Hz, 2560x1440~100Hz 
  OpenGL: renderer: Mesa Intel UHD Graphics 620 (KBL GT2) 
  v: 4.6 Mesa 20.2.3 - kisak-mesa PPA direct render: Yes 
Audio:
  Device-1: Intel Sunrise Point-LP HD Audio vendor: Hewlett-Packard 
  driver: snd_hda_intel v: kernel bus ID: 00:1f.3 
  Sound Server: ALSA v: k5.8.0-29-generic 
Network:
  Device-1: Intel Ethernet I219-LM vendor: Hewlett-Packard driver: e1000e 
  v: 3.2.6-k port: efa0 bus ID: 00:1f.6 
  IF: enp0s31f6 state: down mac: <filter> 
  Device-2: Intel Wireless 8265 / 8275 driver: iwlwifi v: kernel port: efa0 
  bus ID: 01:00.0 
  IF: wlp1s0 state: up mac: <filter> 
Drives:
  Local Storage: total: 2.29 TiB used: 431.50 GiB (18.4%) 
  ID-1: /dev/nvme0n1 vendor: Toshiba model: N/A size: 476.94 GiB 
  ID-2: /dev/sda type: USB vendor: Seagate model: Expansion size: 1.82 TiB 
  Message: No Optical or Floppy data was found. 
Partition:
  ID-1: / size: 466.30 GiB used: 69.61 GiB (14.9%) fs: ext4 dev: /dev/dm-1 
  ID-2: /boot size: 704.5 MiB used: 305.0 MiB (43.3%) fs: ext4 
  dev: /dev/nvme0n1p2 
  ID-3: swap-1 size: 980.0 MiB used: 0 KiB (0.0%) fs: swap dev: /dev/dm-2 
Sensors:
  System Temperatures: cpu: 70.0 C mobo: 57.5 C 
  Fan Speeds (RPM): N/A 
Info:
  Processes: 284 Uptime: 2m Memory: 15.03 GiB used: 1.43 GiB (9.5%) 
  Init: systemd runlevel: 5 Compilers: gcc: 9.3.0 Shell: bash v: 5.0.17 
  inxi: 3.0.38 

Code: Select all

$$ glxinfo -B
name of display: :0
display: :0  screen: 0
direct rendering: Yes
Extended renderer info (GLX_MESA_query_renderer):
    Vendor: Intel (0x8086)
    Device: Mesa Intel(R) UHD Graphics 620 (KBL GT2) (0x5917)
    Version: 20.2.3
    Accelerated: yes
    Video memory: 3072MB
    Unified memory: yes
    Preferred profile: core (0x1)
    Max core profile version: 4.6
    Max compat profile version: 4.6
    Max GLES1 profile version: 1.1
    Max GLES[23] profile version: 3.2
OpenGL vendor string: Intel
OpenGL renderer string: Mesa Intel(R) UHD Graphics 620 (KBL GT2)
OpenGL core profile version string: 4.6 (Core Profile) Mesa 20.2.3 - kisak-mesa PPA
OpenGL core profile shading language version string: 4.60
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile

OpenGL version string: 4.6 (Compatibility Profile) Mesa 20.2.3 - kisak-mesa PPA
OpenGL shading language version string: 4.60
OpenGL context flags: (none)
OpenGL profile mask: compatibility profile

OpenGL ES profile version string: OpenGL ES 3.2 Mesa 20.2.3 - kisak-mesa PPA
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20

 

And the glmark2 results:

Code: Select all

$ glmark2
=======================================================
    glmark2 2014.03+git20150611.fa71af2d
=======================================================
    OpenGL Information
    GL_VENDOR:     Intel
    GL_RENDERER:   Mesa Intel(R) UHD Graphics 620 (KBL GT2)
    GL_VERSION:    4.6 (Compatibility Profile) Mesa 20.0.8
=======================================================
[build] use-vbo=false: FPS: 111 FrameTime: 9.009 ms
[build] use-vbo=true: FPS: 88 FrameTime: 11.364 ms
[texture] texture-filter=nearest: FPS: 109 FrameTime: 9.174 ms
[texture] texture-filter=linear: FPS: 107 FrameTime: 9.346 ms
[texture] texture-filter=mipmap: FPS: 95 FrameTime: 10.526 ms
[shading] shading=gouraud: FPS: 99 FrameTime: 10.101 ms
[shading] shading=blinn-phong-inf: FPS: 109 FrameTime: 9.174 ms
[shading] shading=phong: FPS: 72 FrameTime: 13.889 ms
[shading] shading=cel: FPS: 82 FrameTime: 12.195 ms
[bump] bump-render=high-poly: FPS: 77 FrameTime: 12.987 ms
[bump] bump-render=normals: FPS: 92 FrameTime: 10.870 ms
[bump] bump-render=height: FPS: 94 FrameTime: 10.638 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 99 FrameTime: 10.101 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 74 FrameTime: 13.514 ms
[pulsar] light=false:quads=5:texture=false: FPS: 91 FrameTime: 10.989 ms
[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 79 FrameTime: 12.658 ms
[desktop] effect=shadow:windows=4: FPS: 75 FrameTime: 13.333 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 52 FrameTime: 19.231 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 57 FrameTime: 17.544 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 47 FrameTime: 21.277 ms
[ideas] speed=duration: FPS: 84 FrameTime: 11.905 ms
[jellyfish] <default>: FPS: 82 FrameTime: 12.195 ms
[terrain] <default>: FPS: 52 FrameTime: 19.231 ms
[shadow] <default>: FPS: 83 FrameTime: 12.048 ms
[refract] <default>: FPS: 67 FrameTime: 14.925 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 86 FrameTime: 11.628 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 94 FrameTime: 10.638 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 88 FrameTime: 11.364 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 87 FrameTime: 11.494 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 83 FrameTime: 12.048 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 85 FrameTime: 11.765 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 88 FrameTime: 11.364 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 96 FrameTime: 10.417 ms
=======================================================
                                  glmark2 Score: 84 
=======================================================
 

Running in software rendering mode increases the FPS to ~240, but compared to 1-2k fps the phd21 is/was getting on the same integrated graphics device clearly shows how subpar the performance is. Note that this is brand new installation and the issues were glaringly obvious from the start. Before trying out the mint, I was running debian buster with kde for a year and although I did not do any performance tests on it, I did not face such drastic issues.

Any help would be greatly appreciated

UPDATE:
Installed mate (sudo apt install mint-meta-mate) alongside and it's night and day, glmark2 shows similar results as in the linked thread and everything is running smooth as butter. Technically I could consider the issue resolved. However while I understand that Cinnamon requires better hardware and is more resource hungry, I refuse to believe that the intended performance difference is supposed to be ~20x. Even though I don't have anything strictly against Mate (or any strong feelings for Cinnamon for that matter), I would still like to have the choice to use either of them.

Code: Select all

$ glmark2 
=======================================================
    glmark2 2014.03+git20150611.fa71af2d
=======================================================
    OpenGL Information
    GL_VENDOR:     Intel
    GL_RENDERER:   Mesa Intel(R) UHD Graphics 620 (KBL GT2)
    GL_VERSION:    4.6 (Compatibility Profile) Mesa 20.2.3 - kisak-mesa PPA
=======================================================
[build] use-vbo=false: FPS: 2762 FrameTime: 0.362 ms
[build] use-vbo=true: FPS: 4022 FrameTime: 0.249 ms
[texture] texture-filter=nearest: FPS: 3368 FrameTime: 0.297 ms
[texture] texture-filter=linear: FPS: 3551 FrameTime: 0.282 ms
[texture] texture-filter=mipmap: FPS: 3383 FrameTime: 0.296 ms
[shading] shading=gouraud: FPS: 3262 FrameTime: 0.307 ms
[shading] shading=blinn-phong-inf: FPS: 3148 FrameTime: 0.318 ms
[shading] shading=phong: FPS: 2945 FrameTime: 0.340 ms
[shading] shading=cel: FPS: 2797 FrameTime: 0.358 ms
[bump] bump-render=high-poly: FPS: 1688 FrameTime: 0.592 ms
[bump] bump-render=normals: FPS: 3899 FrameTime: 0.256 ms
[bump] bump-render=height: FPS: 3760 FrameTime: 0.266 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 2199 FrameTime: 0.455 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 1153 FrameTime: 0.867 ms
[pulsar] light=false:quads=5:texture=false: FPS: 3337 FrameTime: 0.300 ms
[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 1129 FrameTime: 0.886 ms
[desktop] effect=shadow:windows=4: FPS: 1973 FrameTime: 0.507 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 619 FrameTime: 1.616 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 477 FrameTime: 2.096 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 605 FrameTime: 1.653 ms
[ideas] speed=duration: FPS: 2877 FrameTime: 0.348 ms
[jellyfish] <default>: FPS: 2024 FrameTime: 0.494 ms
[terrain] <default>: FPS: 190 FrameTime: 5.263 ms
[shadow] <default>: FPS: 1759 FrameTime: 0.569 ms
[refract] <default>: FPS: 403 FrameTime: 2.481 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 2723 FrameTime: 0.367 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 2706 FrameTime: 0.370 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 2721 FrameTime: 0.368 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 2713 FrameTime: 0.369 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 2729 FrameTime: 0.366 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 2709 FrameTime: 0.369 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 2717 FrameTime: 0.368 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 2721 FrameTime: 0.368 ms
=======================================================
                                  glmark2 Score: 2396 
=======================================================

[SMG]

Welcome to the forum, sander29.

The Q78 firmware is currently on version 1.14. Your system is only on 1.04. Software and drivers for HP EliteBook 840 G5 Notebook PC. A number of those fixes reference waking from power states which I expect are applicable to your system's processor and graphics.

Code: Select all

Machine:
  Type: Laptop System: HP product: HP EliteBook 840 G5 v: N/A 
  serial: <filter> 
  Mobo: HP model: 83B2 v: KBC Version 04.53.00 serial: <filter> UEFI: HP 
  v: Q78 Ver. 01.04.00 date: 09/12/2018 

The Mesa PPA mentioned by phd21 in the other thread was applicable to an LM19 install because that one did not yet have Mesa 20.0. The LM20 install does have it, so I'm not sure if pushing beyond what comes in LM20 is helping you. It may be causing issues, especially since that has an effects on windowing and the ability to move them around on the screen. Your comment

Running the DE in software rendering seems to improve the situation a bit..

seems to verify the fact it is not helping you. Software rendering means the "renderer" goes to software mode and does not use the listed renderer (which is all the info after the word renderer in your graphics section including Mesa).

Code: Select all

Graphics:
  Device-1: Intel UHD Graphics 620 vendor: Hewlett-Packard driver: i915 
  v: kernel bus ID: 00:02.0 
  Display: x11 server: X.Org 1.20.8 driver: modesetting unloaded: fbdev,vesa 
  resolution: 1920x1080~120Hz, 2560x1440~60Hz, 2560x1440~100Hz 
  OpenGL: renderer: Mesa Intel UHD Graphics 620 (KBL GT2) 
  v: 4.6 Mesa 20.2.3 - kisak-mesa PPA direct render: Yes 

I'm not sure when you ran the below data, but it shows the standard LM20 Mesa 20.0.8 and not the kisak-mesa PPA Mesa 20.2.3, so I'm not exactly sure what your system is running.

And the glmark2 results:

Code: Select all

$ glmark2
=======================================================
    glmark2 2014.03+git20150611.fa71af2d
=======================================================
    OpenGL Information
    GL_VENDOR:     Intel
    GL_RENDERER:   Mesa Intel(R) UHD Graphics 620 (KBL GT2)
    GL_VERSION:    4.6 (Compatibility Profile) Mesa 20.0.8
=======================================================

= = = = = = = = = =

Running in software rendering mode increases the FPS to ~240, but compared to 1-2k fps the phd21 is/was getting on the same integrated graphics device clearly shows how subpar the performance is.

phd21's data indicates his/her system is using updated BIOS/firmware (Feb 2020).

Why is this data using the kisak-mesa PPA while the other you posted in this thread was not?

UPDATE:
Installed mate (sudo apt install mint-meta-mate) alongside and it's night and day, glmark2 shows similar results as in the linked thread and everything is running smooth as butter.

Code: Select all

$ glmark2 
=======================================================
    glmark2 2014.03+git20150611.fa71af2d
=======================================================
    OpenGL Information
    GL_VENDOR:     Intel
    GL_RENDERER:   Mesa Intel(R) UHD Graphics 620 (KBL GT2)
    GL_VERSION:    4.6 (Compatibility Profile) Mesa 20.2.3 - kisak-mesa PPA
=======================================================

Cinnamon and MATE use different windowing systems windows compositors, but I would think there is something else at play given they are not both showing the same Mesa version being used.

I strongly suggest going through all the firmware updates to see what may apply to your system. There were a lot of fixes done in the past two years.

[sander29]

SMG thank you for the very detailed response on the matter.

Indeed the firmware is ancient, didn't even occur to me that I should look into it. Fortunately the modern systems are able to get their updates directly from BIOS and I did not need to go through the hassle of windows.

However after updating the firmware I can confirm that the issue is not resolved, updated info:

Code: Select all

$ inxi -Fxzd
System:
  Kernel: 5.8.0-29-generic x86_64 bits: 64 compiler: N/A 
  Desktop: Cinnamon 4.6.7 Distro: Linux Mint 20 Ulyana 
  base: Ubuntu 20.04 focal 
Machine:
  Type: Laptop System: HP product: HP EliteBook 840 G5 v: N/A 
  serial: <filter> 
  Mobo: HP model: 83B2 v: KBC Version 04.6A.00 serial: <filter> UEFI: HP 
  v: Q78 Ver. 01.14.01 date: 10/16/2020 
Battery:
  ID-1: BAT0 charge: 42.5 Wh condition: 42.5/50.2 Wh (85%) 
  model: Hewlett-Packard Primary status: Full 
  ID-2: hidpp_battery_0 charge: N/A condition: N/A 
  model: Logitech G703 LIGHTSPEED Wireless Gaming Mouse w/ HERO 
  status: Discharging 
CPU:
  Topology: Quad Core model: Intel Core i7-8650U bits: 64 type: MT MCP 
  arch: Kaby Lake rev: A L2 cache: 8192 KiB 
  flags: avx avx2 lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx 
  bogomips: 33599 
  Speed: 598 MHz min/max: 400/4200 MHz Core speeds (MHz): 1: 600 2: 600 
  3: 600 4: 600 5: 600 6: 600 7: 600 8: 600 
Graphics:
  Device-1: Intel UHD Graphics 620 vendor: Hewlett-Packard driver: i915 
  v: kernel bus ID: 00:02.0 
  Display: x11 server: X.Org 1.20.8 driver: modesetting unloaded: fbdev,vesa 
  resolution: 1920x1080~120Hz, 2560x1440~60Hz, 2560x1440~100Hz 
  OpenGL: renderer: Mesa Intel UHD Graphics 620 (KBL GT2) 
  v: 4.6 Mesa 20.2.3 - kisak-mesa PPA direct render: Yes 
Audio:
  Device-1: Intel Sunrise Point-LP HD Audio vendor: Hewlett-Packard 
  driver: snd_hda_intel v: kernel bus ID: 00:1f.3 
  Sound Server: ALSA v: k5.8.0-29-generic 
Network:
  Device-1: Intel Ethernet I219-LM vendor: Hewlett-Packard driver: e1000e 
  v: 3.2.6-k port: efa0 bus ID: 00:1f.6 
  IF: enp0s31f6 state: up speed: 1000 Mbps duplex: full mac: <filter> 
  Device-2: Intel Wireless 8265 / 8275 driver: iwlwifi v: kernel port: efa0 
  bus ID: 01:00.0 
  IF: wlp1s0 state: up mac: <filter> 
Drives:
  Local Storage: total: 476.94 GiB used: 70.51 GiB (14.8%) 
  ID-1: /dev/nvme0n1 vendor: Toshiba model: N/A size: 476.94 GiB 
  Message: No Optical or Floppy data was found. 
Partition:
  ID-1: / size: 466.30 GiB used: 70.16 GiB (15.0%) fs: ext4 dev: /dev/dm-1 
  ID-2: /boot size: 704.5 MiB used: 305.0 MiB (43.3%) fs: ext4 
  dev: /dev/nvme0n1p2 
  ID-3: swap-1 size: 980.0 MiB used: 0 KiB (0.0%) fs: swap dev: /dev/dm-2 
Sensors:
  System Temperatures: cpu: 75.0 C mobo: 61.5 C 
  Fan Speeds (RPM): N/A 
Info:
  Processes: 273 Uptime: 2m Memory: 15.02 GiB used: 1.53 GiB (10.2%) 
  Init: systemd runlevel: 5 Compilers: gcc: 9.3.0 Shell: bash v: 5.0.17 
  inxi: 3.0.38 

and the resulting benchmark results:

Code: Select all

$ glmark2 
=======================================================
    glmark2 2014.03+git20150611.fa71af2d
=======================================================
    OpenGL Information
    GL_VENDOR:     Intel
    GL_RENDERER:   Mesa Intel(R) UHD Graphics 620 (KBL GT2)
    GL_VERSION:    4.6 (Compatibility Profile) Mesa 20.2.3 - kisak-mesa PPA
=======================================================
[build] use-vbo=false: FPS: 154 FrameTime: 6.494 ms
[build] use-vbo=true: FPS: 176 FrameTime: 5.682 ms
[texture] texture-filter=nearest: FPS: 212 FrameTime: 4.717 ms
[texture] texture-filter=linear: FPS: 233 FrameTime: 4.292 ms
[texture] texture-filter=mipmap: FPS: 225 FrameTime: 4.444 ms
[shading] shading=gouraud: FPS: 302 FrameTime: 3.311 ms
[shading] shading=blinn-phong-inf: FPS: 222 FrameTime: 4.505 ms
[shading] shading=phong: FPS: 142 FrameTime: 7.042 ms
[shading] shading=cel: FPS: 136 FrameTime: 7.353 ms
[bump] bump-render=high-poly: FPS: 60 FrameTime: 16.667 ms
[bump] bump-render=normals: FPS: 169 FrameTime: 5.917 ms
[bump] bump-render=height: FPS: 186 FrameTime: 5.376 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 154 FrameTime: 6.494 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 69 FrameTime: 14.493 ms
[pulsar] light=false:quads=5:texture=false: FPS: 63 FrameTime: 15.873 ms
[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 61 FrameTime: 16.393 ms
[desktop] effect=shadow:windows=4: FPS: 69 FrameTime: 14.493 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 29 FrameTime: 34.483 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 53 FrameTime: 18.868 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 29 FrameTime: 34.483 ms
[ideas] speed=duration: FPS: 63 FrameTime: 15.873 ms
[jellyfish] <default>: FPS: 64 FrameTime: 15.625 ms
[terrain] <default>: FPS: 38 FrameTime: 26.316 ms
[shadow] <default>: FPS: 97 FrameTime: 10.309 ms
[refract] <default>: FPS: 53 FrameTime: 18.868 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 61 FrameTime: 16.393 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 64 FrameTime: 15.625 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 61 FrameTime: 16.393 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 67 FrameTime: 14.925 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 68 FrameTime: 14.706 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 67 FrameTime: 14.925 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 63 FrameTime: 15.873 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 52 FrameTime: 19.231 ms
=======================================================
                                  glmark2 Score: 107 
=======================================================]

The LM20 install does have it, so I'm not sure if pushing beyond what comes in LM20 is helping you. It may be causing issues, especially since that has an effects on windowing and the ability to move them around on the screen.

Indeed it appears that I have misread, however as stated in the initial post, the problems started straight after the clean install, so not arguing that the newer version of Mesa might not help, but I don't think it's causing the issues either.

I'm not sure when you ran the below data, but it shows the standard LM20 Mesa 20.0.8 and not the kisak-mesa PPA Mesa 20.2.3, so I'm not exactly sure what your system is running.

Ah yes, that was the output saved from earlier benchmark before updating Mesa, however with the newer version the results were the same, that is why I used the same output.

I strongly suggest going through all the firmware updates to see what may apply to your system. There were a lot of fixes done in the past two years.

Any other ideas? I will try switching back to kernel 5.4 as well, however I don't have much faith in that.

[SMG]

Any other ideas? I will try switching back to kernel 5.4 as well, however I don't have much faith in that.

Your computer is not that old, so I'd recommend staying with the 5.8 kernel.

Have you tried the combo of the 5.8 kernel without the kisak PPA?

If you have not already done so, I would recommend turning off the windows effects in Cinnamon (although that may be the bells and whistles you mentioned earlier). They can consume resources and you are running three monitors on an igpu, so I would think every little bit would help. The integrated graphics just aren't going to work the same as if you would have a separate gpu when the computer has three monitors.

I checked phd21's data and it's an i5-8250U with one 1920x1080~60Hz monitor (running the 5.9 kernel on LM19.3).

You have an i7-8650U with three 1920x1080~120Hz, 2560x1440~60Hz, 2560x1440~100Hz monitors. I would expect the extra monitors would make a difference, especially given their size and Hz rates (and whether or not you are connecting directly or through a dock). I do not know what tweaking options you might have with regards to the monitors, but that is the only other idea I have now.

[sander29]

Have you tried the combo of the 5.8 kernel without the kisak PPA?

Not yet, I will give it a go at some point.

If you have not already done so, I would recommend turning off the windows effects in Cinnamon (although that may be the bells and whistles you mentioned earlier).

The effects were disabled.

The integrated graphics just aren't going to work the same as if you would have a separate gpu when the computer has three monitors. ... I would expect the extra monitors would make a difference

This is understandable and reasonable assumption, however looking at the benchmark results between cinnamon and mate, the difference is tenfold with the same setup. So both KDE on Debian and Mate on LM20 were/are able to handle it just fine so the issue can't be that.

Regardless, I think I found the culprit - one of the screens is rotated and I believe that causes the issues in Cinnamon. As soon as I rotated the screen back, I started getting normal performance and the benchmark results confirm that it is on par with the results seen in Mate. Further testing showed that it really doesn't matter which monitor gets rotated (built-in, displayport or hdmi one) performance loss happens on every one, although the integrated display slightly better when rotated (~600fps with other monitors connected and ~800fps when other monitors are disconnected).

[SMG]

Regardless, I think I found the culprit - one of the screens is rotated and I believe that causes the issues in Cinnamon. As soon as I rotated the screen back, I started getting normal performance and the benchmark results confirm that it is on par with the results seen in Mate. Further testing showed that it really doesn't matter which monitor gets rotated (built-in, displayport or hdmi one) performance loss happens on every one, although the integrated display slightly better when rotated (~600fps with other monitors connected and ~800fps when other monitors are disconnected).

That's wild that you were able to determine that. Kudos to you.

Given the sizes of your displays, it makes me wonder if there is some kind of total width or height dimension which is being surpassed when you have all three monitors the way you want them. Then when you rotate one, the total width or height goes under a certain value? I don't really know what it might be or how it might be different between Cinnamon and MATE.