Performance using physical vhdx files compared with standard


jimbo45

Well-known member
Pro User
VIP
Local time
10:34 AM
Posts
3,905
Location
Hafnarfjörður IS
OS
Windows XP,7,10,11 Linux Arch Linux
Hi folks
Has anybody done any sensible comparrative tests of the relative performance of Windows 11 using Physical vhdx files (NOT a VM or running under HYPER-V) compared with using "Standard partitioning".

It seems to me (only intuitively) that while bootup might take a few microsecs longer (if you have several Windows installs and go via the basic Windows Blue boot menu) that things run just as smoothly -- I certainly can't tell any difference in normal use. Maybe those with nice "Geeky" measurements out there could answer this.

I'm curious.

Cheers
jimbo
 

My Computer

System One

  • OS
    Windows XP,7,10,11 Linux Arch Linux
    Computer type
    PC/Desktop
    CPU
    2 X Intel i7
Well, I haven't run any performance tests or benchmarks, but this is how I am currently running Windows 11 on my main desktop gaming machine. My main machine still runs Windows 10 and is rock solid, so I haven't felt the need to change anything. However, I wanted to test and see how Windows 11 worked with my hardware, so I created a .vhdx file and installed Windows 11 to that.

I haven't noticed any performance penalty in normal day to day tasks. It feels like native performance. But one could probably prove synthetically that it's different.
 

My Computers

System One System Two

  • OS
    Windows 11 Pro
    Computer type
    PC/Desktop
    Manufacturer/Model
    Beelink SEI8
    CPU
    Intel Core i5-8279u
    Motherboard
    AZW SEI
    Memory
    32GB DDR4 2666Mhz
    Graphics Card(s)
    Intel Iris Plus 655
    Sound Card
    Intel SST
    Monitor(s) Displays
    Asus ProArt PA278QV
    Screen Resolution
    2560x1440
    Hard Drives
    512GB NVMe
    PSU
    NA
    Case
    NA
    Cooling
    NA
    Keyboard
    NA
    Mouse
    NA
    Internet Speed
    500/50
    Browser
    Edge
    Antivirus
    Defender
    Other Info
    Mini PC used for testing Windows 11.
  • Operating System
    Windows 10 Pro
    Computer type
    PC/Desktop
    Manufacturer/Model
    Custom
    CPU
    Ryzen 9 5900x
    Motherboard
    Asus Rog Strix X570-E Gaming
    Memory
    64GB DDR4-3600
    Graphics card(s)
    EVGA GeForce 3080 FT3 Ultra
    Sound Card
    Onboard
    Monitor(s) Displays
    ASUS TUF Gaming VG27AQ. ASUS ProArt Display PA278QV 27” WQHD
    Screen Resolution
    2560x1440
    Hard Drives
    2TB WD SN850 PCI-E Gen 4 NVMe
    2TB Sandisk Ultra 2.5" SATA SSD
    PSU
    Seasonic Focus 850
    Case
    Fractal Meshify S2 in White
    Cooling
    Dark Rock Pro CPU cooler, 3 x 140mm case fans
    Mouse
    Logitech G9 Laser Mouse
    Keyboard
    Corsiar K65 RGB Lux
    Internet Speed
    500/50
    Browser
    Chrome
    Antivirus
    Defender.
Hi folks
Has anybody done any sensible comparrative tests of the relative performance of Windows 11 using Physical vhdx files (NOT a VM or running under HYPER-V) compared with using "Standard partitioning".

It seems to me (only intuitively) that while bootup might take a few microsecs longer (if you have several Windows installs and go via the basic Windows Blue boot menu) that things run just as smoothly -- I certainly can't tell any difference in normal use. Maybe those with nice "Geeky" measurements out there could answer this.

I'm curious.

Cheers
jimbo
Well, it is not that difficult to test yourself.

I have run passmark in the past on native OS and VHDX files, and got maybe up to 2-3% difference but I have an mid range i7 withfast nvme drives.

It all comes down to speed of access to drives and cpu time to manage it.

However, with slower drives, difference may be more obvious. In the end, the difference is pretty unoticeable but the gaming brigade may notice it.

On a similar note, having Hyper-V enabled marginally slows down pc as well (Host OS actually runs as a virtual machine as it is on top of the hypervisor). The fastest pc will be Host OS with Hyper-V disabled.

I am prepared to "sacrifice" a bit of performance to run Hyper-V as the vms run much better than using type 2 hypervisors. However my main OS is always directly installed. I use vhdx files for Insiders where performance is not really an issue - I am just using them as test beds.
 

My Computer

System One

  • OS
    Windows 10 Pro + others in VHDs
    Computer type
    Laptop
    Manufacturer/Model
    ASUS Vivobook 14
    CPU
    I7
    Motherboard
    Yep, Laptop has one.
    Memory
    16 GB
    Graphics Card(s)
    Integrated Intel Iris XE
    Sound Card
    Realtek built in
    Monitor(s) Displays
    N/A
    Screen Resolution
    1920x1080
    Hard Drives
    1 TB Optane NVME SSD, 1 TB NVME SSD
    PSU
    Yep, got one
    Case
    Yep, got one
    Cooling
    Stella Artois
    Keyboard
    Built in
    Mouse
    Bluetooth , wired
    Internet Speed
    72 Mb/s :-(
    Browser
    Edge mostly
    Antivirus
    Defender
    Other Info
    TPM 2.0

Latest Support Threads

Back
Top Bottom