Over the past couple of years, Apple has stirred up the notebook market with its M1 and M2 processors. While Apple’s new silicon has given its competitors a kick in the pants, there is still a lot of functionality to be desired from Apple’s new M1 and M2-based MacBooks.
In my last comparison of the MacBook Air M2, I compared it to the Surface Laptop 4 at the same price point, with as similar as possible specifications. I highly recommend reading the comparison, which you can find here. I make a very similar argument in my comparison of the Surface Laptop 4 and the MacBook Air M2 and reference that comparison throughout this comparison, noting the upgrades and enhancements that the Microsoft Surface Team made to the Surface 5. While the MacBook Air M2 has disruptive raw performance and performance per watt (PPW) within the notebook market, it has inferior gaming, productivity, web, and connectivity experience to that of the Surface Laptop 5.
I have compared the MacBook Pro and the Surface Laptop Studio, the Surface Laptop 3 and the MacBook Air M1, the Surface Book and the MacBook Pro, and the Surface Laptop 4 and the MacBook Air M2. Now I want to compare the Surface Laptop 5 to the MacBook Air M2.
As I did with the other MacBook Air M2 comparison, I want to preface it with my expectations of the benchmarks and why I am comparing both systems.
I expect the Arm-based Apple Silicon to outperform the x86-based Intel Core processor and Intel Iris GPU benchmarks. Apple’s silicon is more efficient and is based on a smaller node; the M2 is based on “5nm” technology, while the 12th Gen Core processor is based on Intel 7. However, Intel has changed the architecture of its 12th Gen Intel Core processors to have performance (P-cores) and efficiency (E-cores) cores. In my comparison of the Zephyrus m16 and the Apple MacBook Pro, the Zephyrus m16 with an 11th Gen Intel Core processor and NVIDIA graphics won out in raw performance but tanked in battery life. The change in architecture should offer a better PPW, especially for these small, ultra-mobile form factors. Although I do not do a battery test and none of my benchmarks are on battery, I believe the Surface Laptop 5 has sufficient enough battery life to make a battery test unnecessary.
Both devices target a very similar audience. The Surface Laptop 5 and MacBook Air M2 target folks who want a smaller form factor laptop, don’t need a significant amount of performance, and are looking to be productive. I would even throw in some light gaming as I believe both devices have the raw performance to play the most strenuous games.
My goal with these benchmarks is for my testing and experience to be easily replicable with benchmarks that are easily accessible. Benchmarking is not my full-time job and is used to show each device’s performance under specific parameters—whether that is API, GPU or CPU, etc. Each benchmark has been chosen to express the capabilities of each device.
My methodology has not changed. Both systems have default settings and are exactly how they came out of the box, except for sleep settings. Both systems are run on AC and connected to their out-of-the-box charges rather than running on DC and using the battery. Both systems were also in the same room running the same test side-by-side, and each benchmark was run three times back-to-back. The results are the average of those three runs.
I have split this comparison into three parts—Creativity, productivity, and gaming. I believe all three categories are important to the target audience of these systems.
The Surface Laptop 5 and MacBook Air configurations have been chosen based on price and model. Both systems are priced at around $1,699.99, but the Surface Laptop 5 is currently on sale for $1,399.99, saving $300. Both devices are configured to have their highest respective processors and 16GB of memory, and 512GB of storage. Below are the full specifications of both devices:
Apple MacBook Air M2 2022
- Apple M2 chip with 8-core CPU, 8-core GPU, 16-core Neural Engine
- 16GB of unified memory
- 512GB of SSD storage
- 13.6-inch Liquid Retina display
- 2560 x 1664 resolution (224 PPI)
- 1080p Face Time HD Camera
- Two Thunderbolt 4 (TB4) ports, MagSafe charging port, 3.5mm headphone jack
- Touch ID
- macOS Ventura 13.01
- Microsoft Surface Laptop 5
- Intel Evo 12th Gen Core i7-1255U
- Intel Iris Xe Graphics
- 16GB of memory
- 512GB of removable SSD storage
- 13.5-inch PixelSense touch display
- 2256 x 1504 resolution (201 PPI)
- 720 HD camera
- 1 USB-C with USB 4.0/Thunderbolt 4, 1 USB-A 3.1, 3.5mm headphone jack, Surface Connect port
- Windows Hello
- Windows 11 Pro 22H2
If there is one category that the MacBook excels at, it is content creation. Most people buy Mac devices for their ability to create content, and Apple has made it clear that content creation is its target market. However, Apple targets a very niche category of content creation and a large chunk of content creation, like CAD and product design tools are left out of the question. In a study done this year (April 2022) by Principled Technology, “the Windows-based Surface Laptop Studio models we tested were compatible with all 31 Rhino®, Autodesk, Dassault Systemes, and Siemens professional CAD apps and the macOS-based MacBook Pro models we tested were only compatible with 10.”
While the MacBook Air targets a very mobile audience, I believe Apple is still pulling toward its audience of video and graphic design with the MacBook Air M2. It definitely has the performance.
I have chosen the Cinebench 23 benchmark to show off the 3D rendering capabilities of the Surface Laptop 5 and the MacBook Air M2. Cinebench 23 has two benchmarks, a multi-thread (MT) benchmark and a single-thread (ST) benchmark. According to Maxon’s website, “Cinebench is a real-world cross-platform test suite… providing a more accurate measurement of Cinema 4D’s ability to take advantage of multiple CPU cores and modern processor features available to the average user.”
I believe Cinebench 23 should give us a good idea of how well these systems handle 3D computer animation, modeling, simulation, and rendering using the CPU of both systems. I chose it over other considerable benchmarks like Blender to get an idea of CPU performance rather than GPU performance.
It is worth noting that the 12th Gen Core i7-1255U has 2 P-cores and 8 E-cores. Hyper-Threading is only available on the P-cores, making for a total of 12 threads. The M2 chip has 4 performance cores and 4 efficiency cores with one thread per core. As we will see with the benchmarks, I believe this is why the Surface Laptop 5 scored lower than the MacBook Air in multi-threading. Although it has more threads, it has half of the performance cores.
The Surface Laptop 5 wins out for ST performance, while the MacBook Air keeps its distance for MT performance. While the MacBook Air is still more powerful than the Surface Laptop 5, this shows that the M2 chip is not out of reach of its competitors like many would think.
As I said in the previous comparison of the MacBook Air, I believe this is at the lower end of importance because people who need a laptop for video editing and rendering will look for more powerful and performant devices. Content creation categories that I believe users will use these systems for would be for podcasts, video calling events, and writing. These categories appeal more to a matter of preference—Microsoft 365 or iWork—than to device performance.
I do believe versatility plays a considerable role in choosing these devices. Most content creators would be doing this light content creation on the go, where versatile tools like a touch screen and would be more beneficial than slightly more multi-threaded performance.
On top of having a touch screen, the Surface Laptop 5 has support for the Surface Slim Pen and Pen2. While Pen usage is slightly more awkward than on a Surface Pro or Surface Studio device, it works flawlessly. I would not call it a replacement for a Surface Pro or Studio, especially considering it does not have the G6 processor and haptic feedback. For the MacBook Air M2, its most redeeming qualities with respect to versatility is that it is slightly lighter, has more battery, and performs better than the Surface Laptop 5.
For the productivity category, I wanted to measure each device’s ability to get things done efficiently and productively. Not only are we looking to get things done, but we are also looking to see the quality and quickness of the outcome. Anyone can write a page essay, but not everyone can do it productively and efficiently. Like content creation, where small form-factor notebooks rely on versatility as much as performance, productivity relies on the quality and versatility of the tools provided. With that said, I want to compare how both systems handle web browsing performance and productivity performance and then look at each system’s productivity and multitasking experience.
I used Web XRPT4 on two different browsers to measure web browsing performance. WebXRPT 4 is a benchmark created by the folks at Principled Technology (PT). In the words of PT:
I ran both systems on their respective default browsers and then both on Chrome. The MacBook Air uses Safari 16.1, and the Surface Laptop 5 uses Edge (Chromium) 108.0.1462.42.
The MacBook Air M2 maintained its lead over the Surface Laptop 5, but the difference could be considered nominal and, in most cases, are unnoticeable to the everyday user. It is no surprise that the Surface Laptop 5 works equally well on Chrome as it does Edge. Chrome and Edge are built on the same Chromium browser. Out of the three browsers, Chrome, Safari, and Edge. I believe Edge offers more to the productivity experience.
My assessment of the Edge browser holds true as it did about two years ago and then some. Since that assessment, Microsoft has added vertical tabs, Collections, and, recently Workspaces for a collaborative web browsing experience. While I was initially hesitant to use vertical tabs, it is a feature I cannot live without and is a complete game changer.
Yes, the Edge browser is available on macOS, but vertical tabs are not as useful on the MacBook Air M2 since it cannot snap the browser window next to my Word document. A core value proposition for the vertical tabs feature is that it saves room while giving a better place to manage tabs. This lack of window snapping brings me to my next point: Windows is so much better at multitasking hands-down.
I will quote myself from my last comparison to relieve the tension of what meaningful multitasking consists of:
“Multitasking on a laptop is where the device is multitasking and providing a service for the user that makes the person’s task more productive or efficient.”
While the MacBook Air M2 is marginally more performant in the web browser, its productivity and multitasking experience is lacking compared to the Surface Laptop 5.
If Stage Manager is Apple’s solution for those who want better multitasking on macOS, it should stick to iPhones. Not only would Stage Manager be more useful for a touch screen device since you could grip the display and select the group, it is also inefficient. Aesthetically, the thumbnails look great, but the thumbnails take up a significant amount of screen real estate for a task that Apple users already used with a three-finger touchpad gesture. Yes, the thumbnails contract when the window is close to it, but it adds little value for multitasking. Another aspect of Stage Manager is that it must be turned on from the settings, adding an extra step to multitasking that I believe is unnecessary. It shows that the MacBook’s ability to multitask is not a center point (surprise) to the MacBook when it should be.
CrossMark is another benchmark that I used to measure the productivity of each device. Business Applications Performance Corporation (BAPco), the creator of the benchmark, says, “CrossMark measures the overall system performance and system responsiveness using models of real-world applications.” It measures a system in three areas: productivity, creativity, and responsiveness. I believe this is a sufficient benchmark to show how each system is handling a workflow with word processing, document editing, reading and writing files, and editing photos and videos. I will give the overall CrossMark score and highlight the Productivity score which looks at office-specific workloads.
The MacBook Air M2 scores higher than the Surface Laptop 5 in both the overall and productivity scores. Again, I was not expecting the Surface Laptop 5 to surpass the MacBook Air M2 in any performance benchmarks, but I believe this shows that the Surface Laptop 5 can keep up with the MacBook Air M2 by a margin of about 10%. Where the Surface Laptop 5 lacks in performance, I believe it makes up for with versatile features.
The Surface Laptop 5’s multi-touch display with pen support allows it to do more on the go than the MacBook Air M2, which lacks touch. The pen support enables the user to sign documents directly from the Surface Laptop 5 without needing to switch devices. It also has more functionality with Android apps than the MacBook Air with iOS and iPadOS apps because of its touch support. The touch screen is also more useful in scenarios with less room, like an airplane. There have been countless times when I use the touch display on my notebook because it is easier to use in my limited space on an airplane.
The Surface Laptop 5 is more versatile in the office as well. Its Thunderbolt 4 support makes it great for hot desking. Rather than connecting multiple dongles, the Surface Laptop 5 is able to connect up to two 4K monitors at 60Hz. The Surface Laptop 5 TB4 is also backward compatible with earlier generations of Thunderbolt devices, making it an optimal upgrade for existing setups. On the other hand, the MacBook Air M2 only has Thunderbolt 3 and has a limited number of supported Thunderbolt docks, none of which support external multi-monitor setups. Apple says the MacBook Air supports “one external display with up to 6K resolution at 60Hz.”
Another feature that I would like to compare for both systems is how quickly each device is able to unlock. In my testing, Windows Hello was able to unlock the Surface Laptop 5 faster than the TouchID on the MacBook Air M2. I found this to be the case consistently and reliably, even in dim and unlit environments. Opening up both systems takes less time for Windows Hello to recognize me and unlock the device than for me to put on my finger on TouchID.
For organizations that are still considering the MacBook Air M2, keep in mind that storage is soldered on and, therefore, not upgradeable. The Surface Laptop 5 has upgradeable storage post-purchase, and the ability to remove it is for security purposes when being discarded.
While gaming is not as important as productivity for me, it is a very considerable category. Like content creators, I doubt the target audience will be looking for ultra-portable notebooks for AAA gaming. However, both these laptops have the performance to run AAA games if the desire is there, but only one is capable of playing most, if not all, AAA games. Before I jump into compatibility, let’s run through some gaming benchmarks. Keep in mind the MacBook Air M2 is equipped with the upgraded 10-core GPU, not the base 8-core GPU.
The first game is Sid Meier’s Civilization 6 (Civ6). I have chosen Civ 6 because it has an Artificial intelligence benchmark that measures the time it takes for the NPCs of the game to take calculated turns. Although it may not be as graphically intensive of a game compared to other AAA games, it is very fitting for this type of laptop. I do not expect either laptop to run something like Cyberpunk 2022 very well for a long period.
I decided to opt-out of including the Civ 6 graphics benchmark since it is not a graphically intensive game and so that I do not overwhelm the reader with redundant benchmarks.
The benchmarks were close for both systems, but the Surface Laptop 5 outperformed the MacBook Air on average by two milliseconds. Interestingly, I found that the Surface Laptop 4 actually performs about the same as the Surface Laptop 5, which is why I included it in the last comparison.
While this difference is not enough for any gamer to notice, it still follows what I said in my Surface Laptop 4 and MacBook Air M2 comparison.
“The M2 has a 14-core NPU for specializing in AI processing. I am not certain whether Civ 6’s AI uses the 14- core NPU, but if it does, it is quite underwhelming, and if it doesn’t, then it shows the lack of support for M2. If the lack of AAA games for Apple silicon did not tell you how Apple doesn’t care about gaming, this should.”
The next game I chose was Counter Strike: Global Offensive (CS:GO). Like Civ 6, CS:GO runs very well on the MacBook Air M2 through Rosetta 2, and I believe it shows the superior graphics of the 10-core M2 chip over the Intel Iris Xe. CS:GO is played competitively, so having the right frames per second (fps) is an important aspect of the game. Anything under 60 fps is not a passing score but remember that these systems should not be considered gaming replacements.
The MacBook Air M2 performs well over 60 fps and doubles the fps of the Surface Laptop 5 at 49 fps on average. I believe this leaves a lot to be desired from the Surface Laptop 5 and shows the potential of gaming on the MacBook.
These next two benchmarks show what I believe to be the state of gaming for the MacBook and how the Microsoft and the Surface team have little to worry about in terms of gaming, no matter how good Apple’s silicon gets.
I tested the Surface Laptop 5 and MacBook Air M2 in 3D Mark’s Wild Life Extreme Unlimited cross-platform benchmark. There is not a 3D Mark macOS application, and so the MacBook Air is only limited to the 3D Mark benchmarks that are available through iPadOS and iOS, the 3D Mark Wild Life Extreme Unlimited benchmark. It has a graphics benchmark and a stress test to measure how well each system maintains its performance over time. It is important to have a stable gaming device considering video game sessions last over five hours on average.
The MacBook Air M2 had trouble maintaining its performance over time. I believe this is due to the thermal throttling issues that the MacBook Air M2 has faced since its launch. While the Surface Laptop 5 did not pass the 97% threshold for 3D Mark to consider it a passing stress test, it was much closer than the MacBook Air M2.
The MacBook Air M2 doubles the score of the Surface Laptop 5 in the 3D mark Wild Life Extreme Unlimited benchmark. While the 3D Mark Wild Life Extreme Unlimited test looks amazing for the MacBook Air M2, I believe it is only half of the answer. While the MacBook Air M2 is able to achieve such a higher score in the confines of the 3D Mark benchmark, these results are not as practical for the MacBook Air M2. The 3D Mark benchmark gives an anchor.
The 3D Mark benchmark has a calculation of estimated gaming performance by game. It gives an estimated gaming performance for Apex Legends, Battlefield V, GTA V, and Red Dead Redemption 2. Out of these four games, which are very popular games and franchises, only three are playable through Crossover and Parallels, and only one is considered “perfect,” according to AppleGameWiki.com. 3D Mark estimates that the Surface Laptop 5 is capable of reaching 45+ fps on 1080p. The MacBook Air M2 then should be capable of reaching well over 45+ fps and double that if the scores are linear.
However, like most of the games on the Apple Game Wiki website, it is played using either CrossOver, which creates Wine bottles, or Parallels, which creates a Windows virtual machine. While these methods allow the user to play the game, it comes at a cost. Firstly, both Parallels ($99) and CrossOver ($74) cost money. Secondly, these methods come at a loss in performance. I benchmarked both systems using the same Battlefield sequence that Notebookcheck.net uses here. For the MacBook Air, I used Parallels which allowed me to also use FRAPS to record the fps average.
Battlefield V struggled on the MacBook Air M2 to stay consistently above 30 fps while consistently having lag spikes and rarely reaching above 50 fps. The Surface Laptop 5, which runs the game natively on the Direct X11 and Direct X12 API, reached 45 fps, meeting the estimates of 3D Mark.
The Surface Laptop 5 on the left and the MacBook Air on the right showing the explosion scene from the story Tirailleur in Battlefield V. Source Jacob Freyman
On top of this, there were countless graphics inaccuracies. In the example above a massive explosion occurred, and the MacBook Air M2 could not render the lighting correctly, making the scene unwatchable. Yes, Battlefield V is only one example and it is because of the hoops we have to jump through in Parallel. However it sets the expectation for franchise games and the most popular games from publishers like EA, Microsoft, Activision Blizzard, and Epic Games.
Out of the consistently updated list of M1 and M2 compatible games master list on Applegamingwiki.com, only 52% of the 1269 games listed are playable either natively or using Rosetta. There is also little incentive for game developers to port over there games, especially older games, to Apple’s new silicon.
Apple has deprecated OpenCL in favor of its Metal API. This means that developers would have to port there games to Metal, adding time and money to the development of the game. Not to mention, game developers may be hesitant to support macOS when Apple is actively threatening Epic games, the creators of Unreal Engine.
I did not expect the Surface Laptop 5 to outperform the MacBook Air M2 in any of those benchmarks, but it did hold its ground and created less distance with respect to raw performance. I do believe the MacBook Air M2 offers unrivaled battery efficiency and an unbeatable PPW. However, the Surface Laptop 5 has shown itself to offer arguably a better experience for content creation, productivity, and gaming with its small form factor audience.
The Surface Laptop 5’s touch display and pen support allow it to be versatile in more scenarios and environments. The touch display offers a better experience for content creators on the go with the Surface Slim Pen and allows users to sign documents without switching devices. It also provides a better Android app experience with its touch display than Apple with its iOS and iPadOS apps.
The Surface Laptop 5 also has a better productivity experience with preloading Office 365 apps and using Window Snaps on Windows 11 22H2. I believe Windows has a better productivity and multitasking experience than macOS Ventura with Stage Manager.
Once more, the Surface Laptop 5 is shown to be better than the MacBook Air M2 in AAA gaming. Although the Surface Laptop 5 cannot handle every game thrown at it, it can run demanding games at reasonable graphics settings with support for most, if not all, AAA titles. Net-net, the Surface Laptop 5 has more to offer than the MacBook Air M2, even though the M2 chip outperforms Intel’s processors in most benchmarks.
Note: Moor Insights & Strategy co-op Jacob Freyman contributed to this article.
Moor Insights & Strategy, like all research and tech industry analyst firms, provides or has provided paid services to technology companies. These services include research, analysis, advising, consulting, benchmarking, acquisition matchmaking, and speaking sponsorships. The company has had or currently has paid business relationships with 8×8, Accenture, A10 Networks, Advanced Micro Devices, Amazon, Amazon Web Services, Ambient Scientific, Anuta Networks, Applied Brain Research, Applied Micro, Apstra, Arm, Aruba Networks (now HPE), Atom Computing, AT&T, Aura, Automation Anywhere, AWS, A-10 Strategies, Bitfusion, Blaize, Box, Broadcom, C3.AI, Calix, Campfire, Cisco Systems, Clear Software, Cloudera, Clumio, Cognitive Systems, CompuCom, Cradlepoint, CyberArk, Dell, Dell EMC, Dell Technologies, Diablo Technologies, Dialogue Group, Digital Optics, Dreamium Labs, D-Wave, Echelon, Ericsson, Extreme Networks, Five9, Flex, Foundries.io, Foxconn, Frame (now VMware), Fujitsu, Gen Z Consortium, Glue Networks, GlobalFoundries, Revolve (now Google), Google Cloud, Graphcore, Groq, Hiregenics, Hotwire Global, HP Inc., Hewlett Packard Enterprise, Honeywell, Huawei Technologies, IBM, Infinidat, Infosys, Inseego, IonQ, IonVR, Inseego, Infosys, Infiot, Intel, Interdigital, Jabil Circuit, Keysight, Konica Minolta, Lattice Semiconductor, Lenovo, Linux Foundation, Lightbits Labs, LogicMonitor, Luminar, MapBox, Marvell Technology, Mavenir, Marseille Inc, Mayfair Equity, Meraki (Cisco), Merck KGaA, Mesophere, Micron Technology, Microsoft, MiTEL, Mojo Networks, MongoDB, MulteFire Alliance, National Instruments, Neat, NetApp, Nightwatch, NOKIA (Alcatel-Lucent), Nortek, Novumind, NVIDIA, Nutanix, Nuvia (now Qualcomm), onsemi, ONUG, OpenStack Foundation, Oracle, Palo Alto Networks, Panasas, Peraso, Pexip, Pixelworks, Plume Design, PlusAI, Poly (formerly Plantronics), Portworx, Pure Storage, Qualcomm, Quantinuum, Rackspace, Rambus, Rayvolt E-Bikes, Red Hat, Renesas, Residio, Samsung Electronics, Samsung Semi, SAP, SAS, Scale Computing, Schneider Electric, SiFive, Silver Peak (now Aruba-HPE), SkyWorks, SONY Optical Storage, Splunk, Springpath (now Cisco), Spirent, Splunk, Sprint (now T-Mobile), Stratus Technologies, Symantec, Synaptics, Syniverse, Synopsys, Tanium, Telesign,TE Connectivity, TensTorrent, Tobii Technology, Teradata,T-Mobile, Treasure Data, Twitter, Unity Technologies, UiPath, Verizon Communications, VAST Data, Ventana Micro Systems, Vidyo, VMware, Wave Computing, Wellsmith, Xilinx, Zayo, Zebra, Zededa, Zendesk, Zoho, Zoom, and Zscaler.
Moor Insights & Strategy founder, CEO, and Chief Analyst Patrick Moorhead is an investor in dMY Technology Group Inc. VI, Dreamium Labs, Groq, Luminar Technologies, MemryX, and Movand