How to Improve Virtual Machine Performance in Parallels Desktop for Mac M1

C

CarlaH1

Bit poster
Hi everyone,
I'm currently using Parallels Desktop on my Mac with an M1 chip, and I'm experiencing some performance lag when running Windows 11 in a virtual machine. I primarily use the VM for light development work, browsing, and a few other applications. However, I've noticed that certain tasks like running multiple browser tabs or using IDEs like Visual Studio can cause noticeable slowdowns.
I've already tried the following troubleshooting steps:

Allocating more CPU cores and RAM to the VM.
Switching the graphics setting from "Auto" to "Best for Retina".
Enabling virtualization in macOS settings.
Keeping both Parallels and macOS updated to the latest versions.
Despite these efforts, the performance is still not as smooth as I would like. Does anyone have any additional tips or settings tweaks that could help improve the VM's responsiveness, especially on an M1-based Mac?

Also, are there any performance trade-offs between the different virtualization options in Parallels (like Hypervisor vs. Apple Virtualization)? Any insights on how to optimize for the M1 chip would be greatly appreciated!

Thanks in advance for your help!
 
I'll give it a shot...

While there is not much you can do to improve CPU performance apart from upgrading to a more powerful system (which is cost prohibitive... usually!), you can tweak the memory settings to get the most out of Parallels.

Firstly, you can give Parallels as much memory you need to run your guest OS, but no more than half of the system memory (not storage) that you have on your Mac. On an M1 Mac, you shouldn't use more memory, since the M1 requires some memory for the GPU cores, which you cannot directly control. Pay attention to the applications you use alongside Parallels, as it all depends on the Mac applications what the GPUs are used for, and how much memory they use. While this may not be the best indicator, you can try using Activity Monitor to see how much memory applications use with the CPU--usually, they often correlate with how much memory is devoted to the GPU by the system, but again, this depends on what the applications do, and how graphically intensive these applications are. If you can, try to make it so that Parallels Desktop is the only app running on your system apart from things like system utilities and tools like those apps that come with iCloud, as an example. Monitoring your system's memory use as you use it is the first step towards observing bottlenecks that may appear as you run Parallels Desktop alongside other Mac apps.

Keep an eye on the amount of page-ins and page-outs your Mac is performing... this is determined in Activity Montior by looking at your Mac's memory pressure. When the memory pressure of your Mac starts to appear amber, your Mac is exhausting the ability to compress pages, and is starting to swap pages of memory between system memory and storage. Memory pressure that appears red indicates that there is an excessive amount of page-ins and page-outs occurring, slowing down your Mac.

And lastly, for Parallels Desktop running on laptops, try to use the built-in display--access to the video hardware is fastest with the built-in video interface of most Macs. External monitors that are usually attached to your Mac via the HDMI or Thunderbolt-4 (DisplayPort) ports will slow down graphics performance since the bandwidth isn't has high as the internal video hardware of a MacBook Air, MacBook Pro or an iMac. The only computer that has fast(er) external video through HDMI or Thunderbolt-4 (DisplayPort) ports is the Mac Pro, thanks to the PCIe Gen 3 slot and the Apple I/O Card.

If you have to use an external monitor (because you want to, or because of those Macs that require one), you can:
-- use the native (default) resolution of your primary monitor if it uses Retina Display capabilities, or
-- use the scaled resolution of your primary monitor to reduce the bandwidth that Parallels has to push out to your Mac's graphics hardware. Good settings may be scaled resolutions of 1920 by 1080 or 1600 by 900. If you have Parallels Tools installed, you can use the Switch Resolution tool to select a low-resolution mode that you can use in Parallels scaled video mode.

I hope ths helps! These techniques have helped me push Parallels to peak performance on Intel Macs with much smaller GPUs! I have taken care to consider the M-series Macs, but the above should still apply to Apple's latest computers.

Good luck!
 
Hey

This is pretty common on M1 Macs with Windows 11 ARM in Parallels.

A few things you can still try:

First, don't over-allocate CPU/RAM. It sounds strange, but giving too much can actually hurt performance. For your use case, 4 cores + 8-12GB RAM is usually the sweet spot.

Second, check your storage. If your Mac disk is nearly full, the VM will feel slow no matter what.

Third, make sure you're using the Apple Virtualization framework instead of the legacy hypervisor. On M1, Apple's framework is usually smoother and more efficient.

Fourth, disable unnecessary Windows background apps. Especially startup apps and indexing services.

Fifth, in Parallels settings:

  • Turn off shared clipboard if you don't need it constantly

  • Disable Coherence mode if you want a slight performance boost

  • Set Power Mode in Windows to "Best Performance"
Also, Visual Studio is heavy even on native hardware. If possible, try lighter alternatives or move some builds outside the VM.

If it's still lagging, the limitation might simply be Windows 11 ARM + M1 combo under heavier IDE workloads.

If you want, tell me your exact RAM + chip model and I can tune settings more precisely.
 
You must log in or register to reply here.
Back
Top