• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Not complaining about Windows 10 20H2 vs. 2004 so far (game performance)

Joined
Feb 22, 2009
Messages
725 (0.16/day)
System Name Artas1984
Processor Xeon E5-2699 V3 3.6GHz
Motherboard Asrock X99X Fatal1ty Killer
Cooling Noctua NH-D14
Memory 48 GB DDR4 ECC 2133 MHz
Video Card(s) MSI Radeon R9 290 1 GHz
Storage Plextor M9PeG 512 GB
Display(s) Dell S2415H
Case Fractal Design Define XL Titanium
Audio Device(s) SSL2
Power Supply Evga 750 GQ
Mouse Razer DeathAdder 2013
Keyboard Roccat Ryos MK Glow
Software Windows 10 Enterprice 20H2
Benchmark Scores i've got a shitload of them in 12 years of TPU membership
Just updated Win10 to from 2004 to 20H2 on my main desktop PC. I've read quite a few contradicting statements from people which Windows 10 version cripples gaming performance more - 2004 or 20H2. I will take a deeper personal interest in this by benchmarking something like 20 games once again. After all my NVIDIA gimping myth benchmark on GTX 780 was done over two years ago. But so far i am happy! Just tested two of my recently played games - Metro and Soma. If anything FPS have improved! I did a benchmark on 1080p, lowest settings to extract as much FPS difference as possible, since i am dusting on R9 290 now...

Metro Exodus DX11 FPS improved from 206/105/68/61 to 213/112/71/64 (max/med/min/1 %)
Metro Exodus DX12 FPS remained the same.
Soma DX11 FPS improved from 140/132/127/105 to 146/136/130/101 (max/med/min/1 %)

This is nowhere enough for a statement of course. Does anyone here have made more conclusive performance tests?
 
Last edited:
Joined
Jul 13, 2016
Messages
1,128 (0.64/day)
Processor Ryzen 5800X
Motherboard ASRock X570 Taichi
Cooling Le Grand Macho
Memory 32GB DDR4 3600 CL16
Video Card(s) EVGA 1080 Ti
Storage Too much
Display(s) Acer 144Hz 1440p IPS 27"
Case Thermaltake Core X9
Audio Device(s) JDS labs The Element II, Dan Clark Audio Aeon II
Power Supply EVGA 850w P2
Mouse G305
Keyboard iGK64 w/ 30n optical switches
I'm currently on 19041. Would you like me to test on this version and 20H2?
 
Joined
Feb 22, 2009
Messages
725 (0.16/day)
System Name Artas1984
Processor Xeon E5-2699 V3 3.6GHz
Motherboard Asrock X99X Fatal1ty Killer
Cooling Noctua NH-D14
Memory 48 GB DDR4 ECC 2133 MHz
Video Card(s) MSI Radeon R9 290 1 GHz
Storage Plextor M9PeG 512 GB
Display(s) Dell S2415H
Case Fractal Design Define XL Titanium
Audio Device(s) SSL2
Power Supply Evga 750 GQ
Mouse Razer DeathAdder 2013
Keyboard Roccat Ryos MK Glow
Software Windows 10 Enterprice 20H2
Benchmark Scores i've got a shitload of them in 12 years of TPU membership
How many games can you test? I see you have GTX 1080 Ti - that will be great!
 
Joined
Jul 13, 2016
Messages
1,128 (0.64/day)
Processor Ryzen 5800X
Motherboard ASRock X570 Taichi
Cooling Le Grand Macho
Memory 32GB DDR4 3600 CL16
Video Card(s) EVGA 1080 Ti
Storage Too much
Display(s) Acer 144Hz 1440p IPS 27"
Case Thermaltake Core X9
Audio Device(s) JDS labs The Element II, Dan Clark Audio Aeon II
Power Supply EVGA 850w P2
Mouse G305
Keyboard iGK64 w/ 30n optical switches
How many games can you test? I see you have GTX 1080 Ti - that will be great!

20 max

I have:

Baulders Gate 3
BioShock infinite
Borderlands 2 / 3
Borderlands prequel
Cyberpunk 2077
DAO
DBZ Xenoverse
Deeprock
Deus Ex Human revolution / MKD
Disco Elysium
Divinity Original sin 1 / 2
Dynasty warrior 8
Fall guys
Fallout 4 GOTY
FF 7, 8, 13 (1,2,3)
Grim dawn
Hades
Life is strange (all)
My time at portia
No man's sky
Pillars of eternity
Planetary annihilation
PUBG
Risk of Rain 2
Rocket League
Overwatch
Shadow Warrior 2
Civ VI
Sleeping dogs
Star wars battlefront 2
starcraft
Subnautica
The outer worlds
the suicide of rachel foster
The witcher 3
Tomb raider
Tota war shogun 2
Valheim
WFTO
Xcom2
Yakuza 0
Skyrim SE

(I've excluded a bunch of games that I don't think are of any interest. I also have many titles on HDD that are not installed)

I also have a bunch of VR titles:

Half-Life Alyx
Hellblade
Superhot
Until you fall
The walking dead
Thrill of the fight
Robo recall
The morrigan
Paranormal activity the lost soul
Arizon sunshine
The climb
and lots more....

Oh by the way, just upgraded my CPU to a 5800X. Updated my specs to reflect that.
 
Joined
Feb 22, 2009
Messages
725 (0.16/day)
System Name Artas1984
Processor Xeon E5-2699 V3 3.6GHz
Motherboard Asrock X99X Fatal1ty Killer
Cooling Noctua NH-D14
Memory 48 GB DDR4 ECC 2133 MHz
Video Card(s) MSI Radeon R9 290 1 GHz
Storage Plextor M9PeG 512 GB
Display(s) Dell S2415H
Case Fractal Design Define XL Titanium
Audio Device(s) SSL2
Power Supply Evga 750 GQ
Mouse Razer DeathAdder 2013
Keyboard Roccat Ryos MK Glow
Software Windows 10 Enterprice 20H2
Benchmark Scores i've got a shitload of them in 12 years of TPU membership
20 max

I have:

Baulders Gate 3
BioShock infinite
Borderlands 2 / 3
Borderlands prequel
Cyberpunk 2077
DAO
DBZ Xenoverse
Deeprock
Deus Ex Human revolution / MKD
Disco Elysium
Divinity Original sin 1 / 2
Dynasty warrior 8
Fall guys
Fallout 4 GOTY
FF 7, 8, 13 (1,2,3)
Grim dawn
Hades
Life is strange (all)
My time at portia
No man's sky
Pillars of eternity
Planetary annihilation
PUBG
Risk of Rain 2
Rocket League
Overwatch
Shadow Warrior 2
Civ VI
Sleeping dogs
Star wars battlefront 2
starcraft
Subnautica
The outer worlds
the suicide of rachel foster
The witcher 3
Tomb raider
Tota war shogun 2
Valheim
WFTO
Xcom2
Yakuza 0
Skyrim SE

(I've excluded a bunch of games that I don't think are of any interest. I also have many titles on HDD that are not installed)

I also have a bunch of VR titles:

Half-Life Alyx
Hellblade
Superhot
Until you fall
The walking dead
Thrill of the fight
Robo recall
The morrigan
Paranormal activity the lost soul
Arizon sunshine
The climb
and lots more....
Ok, this would be great if you really commit yourself to this cause! You will do us (TPU) a big favor! Personally I think 20 games should ''do proper''. Others might be satisfied with less. Just pick up the 20 most demanding titles and use the same scenes for repeated benchmarking (do not use multiplayer games). Make sure you do 3 runs in a row for a single game for proper validation and benchmark a scene for at least 15 - 30 seconds with MSI Afterburner's build-in benchmark (perhaps you know something better). Disable Windows game mode before bechmarking. I would also recommend you use low settings - even though GTX 1080 Ti is still a beast, with low settings you might find what you would not with high settings. And ye, make sure you benchmark Starcraft from 1998 ;-D LMAO...
 
Last edited:
Joined
Jul 13, 2016
Messages
1,128 (0.64/day)
Processor Ryzen 5800X
Motherboard ASRock X570 Taichi
Cooling Le Grand Macho
Memory 32GB DDR4 3600 CL16
Video Card(s) EVGA 1080 Ti
Storage Too much
Display(s) Acer 144Hz 1440p IPS 27"
Case Thermaltake Core X9
Audio Device(s) JDS labs The Element II, Dan Clark Audio Aeon II
Power Supply EVGA 850w P2
Mouse G305
Keyboard iGK64 w/ 30n optical switches
Ok, this would be great if you really commit yourself to this cause! You will do us (TPU) a big favor! Personally I think 20 games should ''do proper''. Others might be satisfied with less. Just pick up the 20 most demanding titles and use the same scenes for repeated benchmarking (do not use multiplayer games). Make sure you do 3 runs in a row for a single game for proper validation and benchmark a scene for at least 15 - 30 seconds with MSI Afterburner's build-in benchmark (perhaps you know something better). Disable Windows game mode before bechmarking. I would also recommend you use low settings - even though GTX 1080 Ti is still a beast, with low settings you might find what you would not with high settings. And ye, make sure you benchmark Starcraft from 1998 ;-D LMAO...

Starting right now.
 
Joined
Feb 22, 2009
Messages
725 (0.16/day)
System Name Artas1984
Processor Xeon E5-2699 V3 3.6GHz
Motherboard Asrock X99X Fatal1ty Killer
Cooling Noctua NH-D14
Memory 48 GB DDR4 ECC 2133 MHz
Video Card(s) MSI Radeon R9 290 1 GHz
Storage Plextor M9PeG 512 GB
Display(s) Dell S2415H
Case Fractal Design Define XL Titanium
Audio Device(s) SSL2
Power Supply Evga 750 GQ
Mouse Razer DeathAdder 2013
Keyboard Roccat Ryos MK Glow
Software Windows 10 Enterprice 20H2
Benchmark Scores i've got a shitload of them in 12 years of TPU membership
Starting right now.
Take your time man! This is not an urgent job required :) Can you please specially more what version and edition of Windows 10 you have?
 
Joined
Jul 13, 2016
Messages
1,128 (0.64/day)
Processor Ryzen 5800X
Motherboard ASRock X570 Taichi
Cooling Le Grand Macho
Memory 32GB DDR4 3600 CL16
Video Card(s) EVGA 1080 Ti
Storage Too much
Display(s) Acer 144Hz 1440p IPS 27"
Case Thermaltake Core X9
Audio Device(s) JDS labs The Element II, Dan Clark Audio Aeon II
Power Supply EVGA 850w P2
Mouse G305
Keyboard iGK64 w/ 30n optical switches
Windows 10 Pro version 19041

I'll be disabling any extra programs I usually have running. I typically have a DisplayCal color profiler, IDM, and directory opus running. These will all be turned off for the benchmarking.

Of course windows game mode is disabled as is hardware scheduler.

Nvidia driver version 461.72

All 3 benchmark runs will follow the same route. The game will be restarted after each run.

All games are being run off a HDD.
 
Joined
Feb 22, 2009
Messages
725 (0.16/day)
System Name Artas1984
Processor Xeon E5-2699 V3 3.6GHz
Motherboard Asrock X99X Fatal1ty Killer
Cooling Noctua NH-D14
Memory 48 GB DDR4 ECC 2133 MHz
Video Card(s) MSI Radeon R9 290 1 GHz
Storage Plextor M9PeG 512 GB
Display(s) Dell S2415H
Case Fractal Design Define XL Titanium
Audio Device(s) SSL2
Power Supply Evga 750 GQ
Mouse Razer DeathAdder 2013
Keyboard Roccat Ryos MK Glow
Software Windows 10 Enterprice 20H2
Benchmark Scores i've got a shitload of them in 12 years of TPU membership
Of course windows game mode is disabled as is hardware scheduler.
This is just in case, you know, apples to apples, because i will be benchmarking too in my laptop without game mode (within the next couple of months). Funny enough, i have seen comments of people stating that this ''game mode'' actually increases FPS, but i disable anything that Windows offers. I also turn off a lot of Windows services after a major update.
 
Joined
Jul 13, 2016
Messages
1,128 (0.64/day)
Processor Ryzen 5800X
Motherboard ASRock X570 Taichi
Cooling Le Grand Macho
Memory 32GB DDR4 3600 CL16
Video Card(s) EVGA 1080 Ti
Storage Too much
Display(s) Acer 144Hz 1440p IPS 27"
Case Thermaltake Core X9
Audio Device(s) JDS labs The Element II, Dan Clark Audio Aeon II
Power Supply EVGA 850w P2
Mouse G305
Keyboard iGK64 w/ 30n optical switches
This is just in case, you know, apples to apples, because i will be benchmarking too in my laptop without game mode (within the next couple of months). Funny enough, i have seen comments of people stating that this ''game mode'' actually increases FPS, but i disable anything that Windows offers. I also turn off a lot of Windows services after a major update.

I've only had problems with windows game mode myself so I leave it off.
 
Joined
Feb 22, 2009
Messages
725 (0.16/day)
System Name Artas1984
Processor Xeon E5-2699 V3 3.6GHz
Motherboard Asrock X99X Fatal1ty Killer
Cooling Noctua NH-D14
Memory 48 GB DDR4 ECC 2133 MHz
Video Card(s) MSI Radeon R9 290 1 GHz
Storage Plextor M9PeG 512 GB
Display(s) Dell S2415H
Case Fractal Design Define XL Titanium
Audio Device(s) SSL2
Power Supply Evga 750 GQ
Mouse Razer DeathAdder 2013
Keyboard Roccat Ryos MK Glow
Software Windows 10 Enterprice 20H2
Benchmark Scores i've got a shitload of them in 12 years of TPU membership
Will you be upgrading to Windows 10 20H2, November 2020 update or the new 21H2, March 2021 update??

I think it would be more relevant to test the newer 21H2... What do you think?
 
Joined
Jul 13, 2016
Messages
1,128 (0.64/day)
Processor Ryzen 5800X
Motherboard ASRock X570 Taichi
Cooling Le Grand Macho
Memory 32GB DDR4 3600 CL16
Video Card(s) EVGA 1080 Ti
Storage Too much
Display(s) Acer 144Hz 1440p IPS 27"
Case Thermaltake Core X9
Audio Device(s) JDS labs The Element II, Dan Clark Audio Aeon II
Power Supply EVGA 850w P2
Mouse G305
Keyboard iGK64 w/ 30n optical switches
Will you be upgrading to Windows 10 20H2, November 2020 update or the new 21H2, March 2021 update??

I think it would be more relevant to test the newer 21H2... What do you think?

Yes, I'll first be testing the games on my current version and then again on the new version.
 
Joined
Feb 22, 2009
Messages
725 (0.16/day)
System Name Artas1984
Processor Xeon E5-2699 V3 3.6GHz
Motherboard Asrock X99X Fatal1ty Killer
Cooling Noctua NH-D14
Memory 48 GB DDR4 ECC 2133 MHz
Video Card(s) MSI Radeon R9 290 1 GHz
Storage Plextor M9PeG 512 GB
Display(s) Dell S2415H
Case Fractal Design Define XL Titanium
Audio Device(s) SSL2
Power Supply Evga 750 GQ
Mouse Razer DeathAdder 2013
Keyboard Roccat Ryos MK Glow
Software Windows 10 Enterprice 20H2
Benchmark Scores i've got a shitload of them in 12 years of TPU membership
Ok then, once you are ready, present your results in your own thread, we will be eagerly awaiting them. I am just not sure where do such results belong, in ''software'' subforum or in ''benchmakrs'' or in ''video cards''...
 
Joined
Jul 13, 2016
Messages
1,128 (0.64/day)
Processor Ryzen 5800X
Motherboard ASRock X570 Taichi
Cooling Le Grand Macho
Memory 32GB DDR4 3600 CL16
Video Card(s) EVGA 1080 Ti
Storage Too much
Display(s) Acer 144Hz 1440p IPS 27"
Case Thermaltake Core X9
Audio Device(s) JDS labs The Element II, Dan Clark Audio Aeon II
Power Supply EVGA 850w P2
Mouse G305
Keyboard iGK64 w/ 30n optical switches
Ok then, once you are ready, present your results in your own thread, we will be eagerly awaiting them. I am just not sure where do such results belong, in ''software'' subforum or in ''benchmakrs'' or in ''video cards''...

Definitely benchmarks

I've decided that because I'm going through each run twice for the before and after, I'm going to restrict the game list to 12. Each individual run out of 3 per game will be 80 second to 260 seconds. I'd rather sink more time to ensure I am getting a large enough sample size of each game then to instead insert more.

The games I've selected are:

Bioshock Infinite
The Witcher 3
Fallout 4
No Man's Sky
Borderlands 3
The Outer Worlds
Cyberpunk 2077
XCOM2
Starcraft
GTAV
Deus Ex Mankind Divided
Overwatch (going to use the replay system to ensure that runs are 100% repeatable)

I've already benchmarked 5 of the titles on windows version 1904. Fallout 4 for example was extremely consistent. The average FPS changed but the 1% lows and 0.1% lows are exactly the same down to the first decimal place. Of course, this is with FPS uncapped (which requires an ini edit for this game).
 
Joined
Jul 13, 2016
Messages
1,128 (0.64/day)
Processor Ryzen 5800X
Motherboard ASRock X570 Taichi
Cooling Le Grand Macho
Memory 32GB DDR4 3600 CL16
Video Card(s) EVGA 1080 Ti
Storage Too much
Display(s) Acer 144Hz 1440p IPS 27"
Case Thermaltake Core X9
Audio Device(s) JDS labs The Element II, Dan Clark Audio Aeon II
Power Supply EVGA 850w P2
Mouse G305
Keyboard iGK64 w/ 30n optical switches
Update, swapped out starcraft for Divinity Original Sin 2. Blizzard launcher was giving me issues no matter what I'd try to do.

Ok then, once you are ready, present your results in your own thread, we will be eagerly awaiting them. I am just not sure where do such results belong, in ''software'' subforum or in ''benchmakrs'' or in ''video cards''...

It was a ton of work but here are the results: https://www.techpowerup.com/forums/threads/windows-20h1-vs-20h2-12-game-benchmark.281125/
 
Top