Parallels Desktop, the Mac virtualization software (AKA run Windows on your Mac without rebooting) has hit version 10. As something of a yearly tradition here at CCE, I went through the new version 10 and compared its real-world gaming performance with its predecessor (and with Boot Camp). Will there be enough of a performance bump to justify upgrading to the new version? Let’s find out together.
We used the following configuration to run the tests:
- Mac Pro 2010 2.8 Quad-Core Intel Xeon (4 CPUs allocated to Parallels)
- 10 GB RAM (4 GB allocated to Parallels)
- NVidia GTX 670 (512 MB video RAM allocated to Parallels)
- DIY Fusion Drive (Mac OS)
- Combination of SSD and 7200 RPM disk drive (depending upon game) for Windows
- DirectX 10
- Vertical synchronisation disabled
- Full-screen (1920×1080) in all cases
- Windows 7 (Boot Camp configuration)
- Parallels Desktop 9.0.24237
- Parallels Desktop 10.0.27675
- All games tested had settings optimised to run smoothly in Windows rather than Parallels.
A couple of changes from last time around are that V-Sync was turned off for every game (and Parallels) where possible, for fear that it would cap frame rates in some cases; and that we actually ran two passes on each game, once to capture the stats, and again to record the YouTube videos (using FRAPS for both).
Some important notes before we begin:
For each of the tests, we loaded up a save game from the mid-point of the game and recorded playthroughs where we tried to perform the same actions each time. Slight variations in gameplay are therefore to be expected.
Parallels Desktop 10 Supported Platforms
One of the purported benefits of Parallels 9 over previous versions is compatibility with OS 10.10 “Yosemite”. Being that OS 10.10 has not been released yet, we stuck with 10.9.
We ran DirectX 10 versions of every game (where available), as that’s the latest supported DirectX version in Parallels. Though we did our best to ensure that none of the games tested used DirectX 11 when doing the Boot Camp versions, it’s possible that some of the games do this automatically without letting you know.
It’s also worth pointing out that where DirectX 11 versions of games are available, performance in Boot Camp may be better because of this, in addition to the differences seen in our benchmarking process.
As noted above, we optimised the settings to run in Boot Camp. Boot Camp results are only used as a baseline reference. Just because the overall frame rate appears low in any of these tests, doesn’t mean that you wouldn’t be able to change the settings to get a decent frame rate in either versions of Parallels. Frame rates recorded are therefore for the purpose of relative comparison only, and should not be taken as an indication of maximum real-world performance.
Likewise, having FRAPS record or measure gameplay itself has a significant impact on performance, so expect real-world performance to be better than the figures below on the same hardware.
Differences between Parallels Desktop versions
In terms of graphical configuration, there was very little difference between either version of Parallels (indeed, the Video Options tab is identical to version 9). The only features of interest to gamers in the marketing material was better battery life (if you’re gaming on a MacBook), extended support for 3+ button mice (so if you have a fancy gaming mouse, you can now map functions to those extra buttons in games running in Parallels) and a gaming optimization preset, that configures the virtual machine with the best settings for running games.
Assassin’s Creed IV: Black Flag Test
Despite Ubisoft’s recent embracing of the Mac as a gaming platform, we’ve yet to see a return of the ubiquitous Assassin’s Creed series to the Mac (the last one, Brotherhood, won’t run on anything newer than OS 10.6.3). While it runs in Boot Camp, is it playable in Parallels? Let’s find out.
Assassin’s Creed IV frame rate over time (higher is better)
Assassin’s Creed IV frames over time (lower is better)
Assassin’s Creed IV Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
That seems pretty conclusive, doesn’t it? It’s playable in both Parallels 9 and 10 (though clearly the graphics options will need tweaking for it to run smoothly), and there’s no discernable difference between either version, perhaps the first indication that we shouldn’t expect much from the latest version.
Age of Wonders III Test
Age of Wonders III is one of those games that it’s hard not to like. Combining a fantasy version of Civilization with decent tactical combat, it’s a game that “might” one day grace OS X. Until that time, there’s always the virtual machine.
Age of Wonders III frame rate over time (higher is better)
Age of Wonders III frames over time (lower is better)
Age of Wonders III Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
Those results seem to be even more conclusive than the AC4 results: Boot Camp is almost double the frame rate of Parallels, and both Parallels versions are on equal footing.
Brothers: A Tale of Two Sons Test
In many ways, Brothers is the perfect game. Gorgeous, moving, and ingenious, it made waves across the gaming community when it was released. You may not get a chance to play a Mac version, but maybe with Parallels that won’t matter.
Brothers: A Tale of Two Sons frame rate over time (higher is better)
Brothers: A Tale of Two Sons frames over time (lower is better)
Brothers: A Tale of Two Sons Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
Seeing a pattern here? The good news: performance is pretty good in Parallels (although it appears the frame rate is capped at 60 in Boot Camp). The bad news: you won’t get any improvement in said performance if you upgrade to version 10.
DMC: Devil May Cry Test
I’m trying to think of a single spectacle fighter that has a Mac version but I’m drawing a blank. Regardless, the Devil May Cry series is one of the most popular spectacle fighters around, but only on every other platform.
DMC: Devil May Cry frame rate over time (higher is better)
DMC: Devil May Cry frames over time (lower is better)
DMC: Devil May Cry Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
Slightly better performance with PD10 over PD9 there, but the performance is highly variable on both, so take that with a pinch of salt.
The Typing of the Dead Test
It seems like on-rails shooters might well be the best genre to use for benchmarking, as it provides a consistent set of results. Typing of the Dead doesn’t even need a mouse, so that’s a plus.
Typing of the Dead frame rate over time (higher is better)
Typing of the Dead frames over time (lower is better)
The Typing of the Dead Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
So the same story as DMC– slightly better performance in version 10, but so much variance that it’s hard to point to a definite improvement.
Just Cause 2 Test
I’ve actually tried to use open world shooter Just Cause 2 every year now, but the built-in benchmark never ran correctly. This time around, I tried to just bite the proverbial bullet and play the game inside Parallels instead.
Just Cause 2 frame rate over time (higher is better)
Just Cause 2 frames over time (lower is better)
Just Cause 2 Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
Perhaps this shouldn’t come as a surprise, but the performance in Parallels was just awful. There’s seemingly a marked decline in performance in version 10 over 9, but just look how poor the performance was– pretty lousy in both versions.
Metal Gear Rising: Revengeance Test
The second spectacle fighter in our test batch, the Metal Gear series is very strange, but very enjoyable. Become a cyborg ninja and slice up some data with us.
Metal Gear Rising: Revengeance frame rate over time (higher is better)
Metal Gear Rising: Revengeance frames over time (lower is better)
Metal Gear Rising Revengeance Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
We’re back to somewhat more sensible results now, with PD10 gaining a very slight edge over its predecessor.
One Finger Death Punch Test
Time for something a little different: One Finger Death Punch is a game with stick man artwork that’s controlled by just two buttons. It’s also one of my favourite games ever, so I tried it out in Parallels.
One Finger Death Punch frame rate over time (higher is better)
One Finger Death Punch frames over time (lower is better)
One Finger Death Punch Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
Sadly this test didn’t reveal much, we got almost a solid 60 FPS across all three tests. But hey, at least you can’t argue it runs well.
South Park: The Stick of Truth Test
Obsidian Entertainment’s next game, Pillars of Eternity will be available on Mac very soon. But if you want to get your hands on their past games, Windows is the only way. South Park, the game based on the TV series, is their most recent, and (depending upon your sense of humour) hilarious.
South Park: The Stick of Truth frame rate over time (higher is better)
South Park: The Stick of Truth frames over time (lower is better)
South Park: The Stick of Truth Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
As with One Finger Death Punch, the three versions are indistinguishable from each other by virtue of their consistent performance.
The venerable procedurally-generated platformer that’s brutally difficult, and available on seemingly every platform except Mac OS, but the real question is, can I survive a game long enough to gather some numbers?
Spelunky frame rate over time (higher is better)
Spelunky frames over time (lower is better)
Spelunky Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
Once again, no real difference between any version- go now and play Spelunky in Parallels, safe in the knowledge that it will run just fine.
State of Decay Test
State of Decay is a game featuring zombies (exclamation mark), with a twist- it’s open world and your decisions (and failures) have a permanent impact on the world.
State of Decay frame rate over time (higher is better)
State of Decay frames over time (lower is better)
State of Decay Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
There does seem to be an improvement with Parallels Desktop 10 here, albeit a small one. Overall the performance was poor compared to Boot Camp, but with the right tweaks it’s likely it could be made to run well.
Torchlight II Test
Okay, okay. I’m over it. Torchlight II will probably never be available for Mac OS. Doesn’t mean I can’t play it anyway.
Torchlight II frame rate over time (higher is better)
Torchlight II frames over time (lower is better)
Torchlight II Boot Camp/Parallels Desktop 9/Parallels Desktop 10 video
In our last test case, it seems like Parallels Desktop 10 actually performs slightly worse than version 9, though there’s rather a lot of variance, which might account for the difference between the two.
The following games were also tested in Boot Camp but failed to run in Parallels, and are listed for completeness:
- Company of Heroes 2
- Wolfenstein: The New Order
The results this time around are disappointing. There was no mention of graphical performance improvements in the new features for Parallels Desktop 10, and the results clearly indicate that you’ll be unlikely to see much difference after upgrading. We saw a very minor degree of improvement in some games, but nothing significant overall.
Performance aside, it seems that the same games experienced the same issues on both versions, so you probably don’t need to consider upgrading for better compatibility. Given that it was a similar story in the update from version 8 to version 9 last year, we just have to hope that version 11 will give gamers something to get excited about, but for now, it’s safe to pass on version 10 if you’re only interested in using it for gaming.