2015 VM Benchmarks: Parallels 11 vs. Fusion 8 vs. VirtualBox 5

Posted by Jim Tanous on September 4, 2015
fusion parallels virtual box

fusion 8 benchmarks 3dmark header


Futuremark’s 3DMark is the latest in a long line of industry-standard gaming benchmark suites. 3DMark offers a range of tests that can evaluate the performance of everything from a low-power tablet to a $15,000 quad-SLI gaming PC, and it looks at a CPU’s ability to handle gaming-related physics calculations in addition to pure GPU performance.

3DMark offers multiple tests of increasing complexity that support DirectX 9, DirectX 10, and DirectX 11. However, Fusion 8 and Parallels 11 only support up to DirectX 10, allowing us to run only the entry-level and mid-range 3DMark tests. Of note, VirtualBox technically supports up to DirectX 11, and its graphics driver reports that compatibility to the guest OS, enabling access to the higher-end tests. Graphics performance, however, is so poor in VirtualBox at these higher levels that it is unusable for any practical purpose.

We’ll start with Ice Storm, the lowest-end 3DMark test that uses DirectX 9 APIs.

parallels vs fusion benchmarks 3dmark ice storm

As we noted in our Fusion 8 benchmarks earlier this week, VMware made graphics performance a major goal of this new release, and that work pays off with Fusion 8 enjoying a significant lead over Parallels 11 in GPU-focused portions of the Ice Storm test. The CPU-bound physics test, however, reveals the much closer competition between these two products, with Parallels narrowly beating Fusion and both approaching near-native performance.

The next test, Ice Storm Extreme, is another DirectX 9 benchmark that runs the same basic script as the standard Ice Storm test above, but increases the resolution to 1080p and utilizes higher quality textures and lighting effects, making it harder overall on the GPU.

parallels vs fusion benchmarks 3dmark ice storm extreme

When confronted with a more demanding DirectX 9 scenario, Fusion 8 still easily wins, but sees its margin over Parallels 11 fall from about 46 percent to just under 25 percent. Again, we see that Parallels has a slight edge over Fusion when it comes to CPU-focused physics calculations.

The Cloud Gate benchmark is a DirectX 10 test, and the most advanced 3DMark benchmark that’s compatible with our virtualization software.

parallels vs fusion benchmarks 3dmark cloud gate

As expected, neither platform offers anywhere near native performance when it comes to the still-developing DirectX 10 APIs, but Parallels’ implementation of the feature gives it about a 3 percent lead over Fusion 8.


The introduction of DirectX 10 support in Fusion 8 means that we no longer need to rely on older benchmarks in order to have cross-compatibility between the various Parallels and Fusion updates. Since we included it in our individual looks at Fusion 8 and Parallels 11, however, here’s how the two compare in 3DMark06 performance, although this will likely be the last year that this benchmark is included in our testing suite.

parallels vs fusion benchmarks 3dmark06

The performance margins are much closer in 3DMark06 than they were in the Ice Storm tests, with Fusion performing better in terms of pure graphics power and Parallels taking the lead in CPU-focused tests. Of note, our VirtualBox VM would not complete the 3DMark06 test, and so it is omitted from the results.

Table of Contents

[one_half padding=”0 5px 20px 0″]
1. Introduction
2. Test Setup & Methodology
3. Geekbench
4. 3DMark
5. FurMark OpenGL
6. Cinebench R15
7. PCMark 8
8. Passmark PerformanceTest

[one_half_last padding=”0 0px 20px 5px”]
9. Video Encoding
10. File Transfers
11. USB 3.0 Speed
12. Virtual Machine Management
13. Battery Life
14. Mac Pro: Gaming
15. Mac Pro: CPU
16. Conclusions

28 thoughts on “2015 VM Benchmarks: Parallels 11 vs. Fusion 8 vs. VirtualBox 5”

s says:
Hello, can I use the 2015 VM Benchmark Showdown picture in my Sr. Design project documentation? I wanted to compare different virtual machine throughput.
Anton Iakimov says:
Would be nice to see also
* ESXi with macOS guest
Janto Lenherr says:
i like how you go into detail, but it lacks audio software, that is what i would be interested about.
Michael M says:
I am late to this game. I do agree that Parallels offers a more “friendly” interface than Fusion but beyond that, they seem to function pretty much the same except for one nagging item that I have noticed,

Has anyone else notices that Fusion runs a bit “hotter” than Parallels? I find that opening up Fusion to Win 7 or Win 10 automatically gets the temp up at least 20 degrees (F) or so on my rMBP (2015) set for 2 of 4 cores and 8 of 16 gigs of the RAM allocated. This is very disappointing. I have zero issue with the VMware offering (used it years ago) save for this need to use more energy than Parallels. I only use Windows for a couple of apps but I would be worried about adding a more CPU intensive app if heat is generated like this at a greater scale.

Any thoughts or confirmation of Fusion running hotter?

Paul Bartlett says:
Excellent and thorough review, many thanks for this. It seems to mostly reflect what I have experienced. I have been a VMWare Fusion fan for many years and it has never let me down. Recently I tried Virtual Box and it is much better that it used to be, but too many problems, such as weird mouse pointers on Win10/retina display. I have not tried Parallels for several years. Last time I tried it also had a bunch of annoying issues. I am a developer often interfacing with USB devices so for me one of the most important things is USB support and Fusion has been great there. Several years ago Parallels did not have such good USB support either and it was quite expensive. I must try Parallels again one day as this review makes it look like it is up there with VMware, which it was not when I last used it.
Simone says:
When Parallels 12 vs Fusion 8.5 comparison?
niico100 says:
A very common use case is Visual Studio / SQL Server developers. So if you could do any kind of development benchmark that would be useful. (I have found on a retina 12″ MacBook VMWare runs significantly smoother – and Parallels has been locking up in retina mode).
Paul Bartlett says:
me too, would love to see Visual Studio and even scanning and printing. This benchmark shows ‘perfect world’ whereas I use Fusion because its the only one that doesn’t have beyond annoying issues.
Robert says:
Hoping to see a comparison of the newer Vmware Fusion 8.5.0 free upgrade and Parallels 12 and and the latest VirtualBox.
Paul Charlesworth says:
For me, using Parallels over Fusion has always come down to the fact that Fusion requires a password every time I want to access the bootcamp partition. Parallels also has Parallels Access, although the integration between their two products could be better.
RW says:
Excellent and exhaustive review, however I typically run linux guest appliances (primarily headless). It would have been help to see some linux guest performance comparisons in the future. Thank you.
Ksix says:
Do you recommend one over the other for running ArcGIS software?
Stephen says:
I am really interested in understanding the performance for disk access speed to the file system when compared between virtualized machines and boot camp. I am almost certain I will get some improvement by switching from Parallels to Boot Camp, but I don’t want to go through the effort/force a change in my workflow without solid numbers.
On my current setup it seems to take an eternity for the memory on my Windows VM to be filled with all the data it needs (I need to have a SQL server and a related application running on my Windows OS). Are there any good metrics on how long it takes for the OS to load up files from the drive into memory?
PS: Hard drive is a traditional drive, system report describes as APPLE HDD ST1000DM003
daniael says:
i own both vmware 8 pro and parallels 11 pro, parallels is faster by far. small files, large files, medium high res, real life work (compiling, resource editing, various ide(s), git, office, etc) there is no comparison between the two, in my experience. run them in 2 machines, a macbook pro i7 2.7ghz, 1tb ssd and a mac pro 24 cores 3.33ghz, also 1tb ssd. i am not game-ing nor watching videos in vm, i do that in host machine, when i do (not often)
Tien Dinh says:
Me too. I use Parallel desktop for years and recently switched to VMWare 8 Pro due to this article. I found the performance of VMWare 8 Pro to be unbearable. After wasting tons of time trying to determine the causes, i found that the disk performace is extremely slow compare to Parallel desktop or native boot camp speed. I’m swiching back to Parallel Desktop now.
DanialThom says:
This would be more useful if you outlined the tuning settings. You don’t even mention the number of cpus in the “multicore” test
TekRevue says:
The configuration options for the virtual machines are explicitly mentioned on the “Test Setup & Methodology” page. Quoting directly from that page:

Each of our Parallels and Fusion Windows 10 virtual machines was configured for maximum performance, with 8 assigned virtual CPUs, 12GB of RAM (the maximum recommended amount in order to ensure that enough is reserved for OS X), and 1GB of graphics memory configured for each platform’s most advanced DirectX and OpenGL graphics support. The VirtualBox VM was also configured to its maximum supported hardware level, with matching CPU and RAM, but with only 256MB of graphics memory, the most it allows. For all platforms, all features that could possibly impact performance, such as error logging or an expanding virtual disk, were disabled.

Samir Abdo says:
Can you use the Mac hot corners and gestures in Fusion 8?
daniael says:
1gb graphics memory is NOT the recommended size for parallels. since you mentioned 12gb ram, then you most likely run it on a modest macbook pro with 16gb ram and mediocre video card. when it comes to 24 core cpu (2 physical processors, 6 cores each with enough L2 cache) 64+GB RAM and super fast ssd(s) and video cards with 4gb and 1500 cores, fusion is way inferior (cannot even compare). most professional places use similar machines. this does not apply to home end-users or non technical enterprises…
TekRevue says:
You’re a bit confused, Dan. Let me clarify for you:

1) The exact models and specifications for the testing hardware are listed in the article, along with the methodology used for the tests. You’ll find this information on the ironically named “Test Setup & Methodology” page.

2) We submitted our results and methodology to both Parallels and VMware to give each company a chance to respond. Neither company claimed that our testing procedure or the configuration of our virtual machines were incorrect or unfair.

3) Other than providing licenses for both Fusion 8 and Parallels Desktop 11 for our tests, TekRevue was not remunerated by either company in any way, nor were the tests conducted or guided by anyone outside of TekRevue.

4) These products, and our review, are primarily targeted at consumers. That’s why we used both a MacBook Pro and mid-level Mac Pro in the tests. If you’d like us to conduct future tests on your custom Mac Pro (I say “custom” because the Mac Pro you describe in your other comment — 2 x 6-core hyperthreaded 3.33GHz — is not a configuration Apple ever shipped), then please send it to us and we’ll be happy to do the additional testing. In general, however, I hope you’ll agree that your specific configuration isn’t applicable to ~99% of users.

daniael says:
they probably ran it with default settings on a modest macbook pro
Mike S. says:
Thought I’d try out both on my Mac Pro 2013 (8-core CPU, 32gb RAM, 1tb PCIe SSD) and while both perform admirably, Parallels bested Fusion on Windows boot/startup times (as this articles mentions). For me, boot time is top priority, so Parallels was the winner for me.
I didn’t try any hard-core games, as I don’t use the machine for that purpose. However, multiple version of Windows (XP, 7, 10) ran faster and smoother overall in Parallels for general use in my experience.
Having said that, I had some trouble with Parallels recognizing certain USB devices that Fusion did not have trouble with. Complained about it to Parallels, and they’ve recently pushed out a new release that addressed and fix those problems, so I appreciate their customer support.
Good article, thanks!
NativeCalifornian says:
I opted for VMware Fusion and ran into a problem or 2. When I tried to upgrade my W7P VM to W10 it said the graphics driver was out dated. When I tried to run SoftPlan 2016 in my W7P it wouldn’t because it got errors installing DirectX 11. I think all things being equal Fusion is a better buy especially because I don’t have to buy a new copy for each machine in my house.
Ryh says:
I’m Parallels user for 6 years but I think I’ll switch to Fusion. With every free OS X upgrade I’m forced to upgrade Parallels for $50…
ceeioo says:
It’s the main reason i don’t choose Parallels
Simon says:
still confused which is better, according to parallels comparison http://www.parallels.com/landingpage/pd/fusion-compete/ parallels is the best, though it would be strange to see smth. else here 🙂 if you check vmware tests https://www.vmware.com/products/fusion/compare you will see another picture, but the strange thing is that both companies has different digits for the same features/options, eg. check VM capabilities, other comparisons show better results for VMware Fusion http://www.softocoupon.com/comparisons/parallels-desktop-11-for-mac-vs-vmware-fusion-8.php so which is the best?

VM Capabilities
VM Capabilities

wimver says:
I’ve been a Parallels 9 user and was about to upgrade to VMWare Fusion 8 after reading several reviews. Also because I did not like it I had to pay to upgrade my Parallels license if I wanted to upgrade Mac OS to OS X El Captain… I don’t like this kind of “you have to”‘s. So I was holding my credit card ready in my hand, ready to pay for Fusion 8. Especially with their Parallels to VMWare upgrade offering. But I thought: let’s try it fist; so I installed the trial of Fusion 8… This was so disappointing for me… First I imported the existing Parallels virtual machine and converted it which went really smooth. But then… Starting up time: disappointing by all means. I first thought: OK, maybe VMWare tools have to be installed first; but no: even after rebooting several times: boot-up time kept on being disappointing. Then… opening a Visual Studio project: disappointing. I’m not even talking about compiling some code: very disappointing. It just took so much much much longer time then it did in Parallels 9! So now; I just upgrade to Parallels 11: I love it! Boot-up time, Visual Studio performance, … I love it. It’s even better than Parallels 9. The only thing I don’t love is that I had to upgrade. But for my point of view it’s obvious: Parallels beats VMWare Fusion.
simmerkatt says:
I think it’s unfair to include VirtualBox with Fusion and Parallels.

You are forgetting one major thing: Fusion and Parallels are designed to integrate Windows with OS X. VirtualBox is designed to virtualise an entire machine as if it were 100% separate, no integration at all.

Max Coplan says:
I think BootCamp and VirtualBox are the two extremes in this comparison. VirtualBox virtualizes everything, without any integration. And BootCamp has zero virtualization at all, aka complete integration.
Marco Schirrmeister says:
Great post. I am a Fusion user since day 1 and really like it, but now that VMware has doubled the price I will stay with 7 as long as I can.
No Name says:
Great review. Thanks for writeup.
Br. Bill says:
Another thing to consider is the licensing. Yes, Virtualbox is free. Fusion gives you licenses for 3 hosts. Parallels gives you a license for one machine only. So Parallels effectively costs 3x what Fusion does, if you want to host VMs on multiple machines.
JBacTee says:
Any chance you could gain access to a multi-processor (read: 2 physical processor chips) Mac Pro for a similar comparison? Virtualbox has a blatant issue ( https://www.virtualbox.org/ticket/12742 ) negotiating multicore setups in a multiple CPU host (experienced this myself with the dual-CPU 2008 Mac Pro), where the virtualized guest suffers a huge performance hit if you’ve set it up with more than one virtual core. Both Windows 7 & attempts at running the Windows 10 preview were so terrible I gave up trying to run them with any more than a single allocated core on the Mac Pro (whereas my i7 MacBook Pro with a single-chip 8-thread CPU can easily virtualize up to 8 cores with no evident performance hit).

It would be worth seeing how Parallels & Fusion (or even Virtualbox, but I assume it’ll look absolutely terrible as far as performance goes) similarly act under a multi-chip host environment to see how they negotiate allocations between multiple physical chips, as opposed to threads all being sent to the same chip.

James Nguyen says:
I had a similar problem with my MPB mid 2009. Disgusting performance when virtualizing windows 7 or 10. Did you find a solution to this problem?
Fair game says:
Awesome! Review is great. Fusion is great for biz, PD 11 is good for personal/home use, use bootcamp if your windows friends bug you but you don’t need virtualization, use VitrualBox if you just need to visit the other worlds (Win/Linux), if you want something reliable go for fusion and hey if you have extra money, throw that shit at Parallel they deserve it, all good products but hey Oracle has screwed they entire industry, just like CUNYfirst, but trust me they don’t need your petty cash! Fair ‘n square.
Henrique Tavares says:
Really well made review, thank you for the excelent work!
Enrico G. Lupo says:
Many thanks for this full review.
I’m an enthusiast of fusion 8 and a user fro version 5. This last release is a great goal!
Zero1 says:
Really nice review. The first that I have seen online of such sorts.
Overall content on Tekrevue is very solid. I enjoy reading OS X tips mostly.
Thanks guys!:)

Leave a Reply

Your email address will not be published. Required fields are marked *

Disclaimer: Some pages on this site may include an affiliate link. This does not effect our editorial in any way.