The MSI Stealth 15M is a thinner and lighter gaming laptop that still manages to pack in some good specs despite its smaller size. So what sort of compromises have been made to get away with this? I’ll tell you in this detailed review.
Design, Size & Weight
It’s got a clean all-aluminum design, I’ve got the carbon gray finish here, but there’s also a pure white version too. The Stealth weighs around 1.65kg or 3.6lb, or 2.12kg 4.7 lbs with the 150-watt power brick and cables included. It’s on the slimmer side for a 15-inch gaming laptop with these specs under 1.6cm thick, making it quite portable.
Screen (Response Time, Color, Brightness & Bleed)
Considering that this thing is advertised as a gaming laptop, we must check out the screen. My model has a 1080p 144Hz screen, but there are also 60Hz and 240Hz options too. There’s no G-Sync, no way of disabling Optimus, and no dynamic boost here. The laptop’s management software, MSI Dragon Center, gives us the option of enabling or disabling panel overdrive. I wasn’t actually seeing a difference in response time with it enabled or disabled though. I’ve measured the average gray to gray screen response time at 6.2 ms, so below the 6.9 that we’d like to see a 144Hz panel, a good result, though there was some overshoot in some transitions.
Here’s how it compares against others, so not super impressive, but I’ve definitely had plenty of slower 144Hz laptops too. I’ve tested the screen with the Spyder 5 and got 61% of sRGB, 43% of NTSC, 45% of Adobe RGB, and 45% of DCI-P3, so on the low side, I’d want more for content creation. Brightness wasn’t great either, below the 300 nits or so that I like to see with a lower 750:1 contrast ratio. Backlight bleed also had some issues, with a glowing patch down the bottom left in my unit, but this will always vary between units.
There’s a 720p camera above the screen in the center, no Windows Hello face unlock support though. The camera and microphone look and sounds great.
Keyboard & Touchpad
The keyboard has a single zone of RGB backlighting with limited effects, and you can cycle through them all or disable lighting with the F8 shortcut key. All keys and secondary functions are illuminated, but there’s no key brightness control. I had no problems typing with the keyboard.
The power button is part of the keyboard on the top right, but accidental presses don’t put it to sleep, you have to hold it for anything to happen. There are also some air vents behind the keyboard just below the screen. The precision touchpad clicks down anywhere and works fine, though it felt a little small. MSI usually uses wider touchpads, but that’s not the case here. Maybe I’ve just been spoilt. Fingerprints and dirt aren’t too obvious on the matte interior, but it’s easy to clean with a microfiber cloth as it’s a smooth surface.
On the left, there’s an air exhaust vent, the power input, MicroSD card slot, USB 3.2 Gen1 Type-A port, and 3.5mm audio combo jack. The right has a Type-C Thunderbolt 4 port with DisplayPort output, second USB 3.2 Gen1 Type-A port, HDMI 2.0 output, and there’s an air exhaust vent over on this side too. Both the HDMI and Type-C ports connect directly to the Intel Xe and not Nvidia, so VR isn’t possible. The Type-C port can be used to charge the laptop.
Screen Flip, Chassis Flex
The back has air exhaust vents towards the corners, while the front just has an indent in the center for getting your finger in to open the lid. Opening the lid with one finger is easy, and the screen goes the full 180 degrees back for sharing, which is further aided by the F12 shortcut key which can be used to flip the screen around. There’s some flex to both the chassis and lid, but it’s not too bad considering the thinner size and not something I noticed during normal use.
Getting Inside + Internals
Underneath has plenty of air ventilation holes towards the back half. Getting inside was fairly easy after removing 13 Phillips head screws. Once inside we can see that the motherboard is upside down, as some of MSI’s other models such as the GS66. We’ve only got access to the single M.2 storage slot, WiFi card, and battery down the front. If you want to access the two memory slots you’ll need to disconnect the motherboard and take it out, a more involved process.
The speakers are found on the left and right sides towards the front. I thought they sounded average, minimal bass, a bit tinny, and some wrist rest vibration at full volume. My measurement tool wasn’t reporting them as super loud, but my ears were hurting at max volume, and the latency on results wasn’t great.
The Stealth 15M is powered by a 52Wh battery. I’ve tested it with keyboard lighting off, background apps disabled and screen at 50% brightness. It only lasted for almost 4 and a half hours in the YouTube playback test, a lower result, and then a little over an hour in gaming, somewhat average. Let’s check out thermals next.
The MSI Dragon Center software lets us pick between different performance modes, which from lowest to highest are silent, balanced, and extreme performance. Extreme performance mode lets you overclock the GPU, but no overclock was applied by default, so I haven’t tested with one. You can also enable cooler boost mode here, which sets the fan to max speed. We can also enable user mode for some fan customization. After just a few days of testing, cooler boost randomly decided to stop working. I’m not sure what the issue was, but I had to manually uninstall the dragon center software and reinstall it to fix it.
Thermals & Performance
Here’s what MSI shows the cooling solution as looking like, the idle temperatures down the bottom were fine. Stress tests were done with the Aida64 CPU stress test with stress CPU only checked and the Heaven GPU benchmark at max settings run at the same time, while gaming was tested playing Watch Dogs 2. The CPU was thermal throttling with the stress tests running, but only in the lower silent and balanced modes as the fans run quieter here. With extreme mode though, throttling was removed, and enabling cooler boost further lowered temperatures, but at the expense of fan noise, as you’ll hear shortly.
These are the clock speeds during the same tests just shown. Despite thermals improving for the CPU in extreme mode, interestingly the processor performance actually lowers now, however, the GPU clock speeds increase with extreme mode. All is revealed when looking at the power limits. The Nvidia RTX 2060 Max-Q seems to be capped to 30 watts to counter the slower fans in silent and balanced modes. As a result, there’s less performance, and many GPU-bound games were playing noticeably badly. With extreme mode the GPU can run up to 65 watts, however, with the stress test at least, it seems that the CPU power limit lowers from 35 watts to 28 watts with extreme mode.
Now it’s worth noting that MSI is about to refresh this gaming laptop already with Intel’s new 11th gen H35 processors and Nvidia RTX 3060 graphics. I don’t think there will be too much difference on the CPU side, the processors will just have a little higher power limit at 35 watts. Technically the 1185G7 in this machine has a 28-watt TDP on paper, but as we saw earlier in some workloads it could exceed this. Anyway considering that those temperatures could be brought under control with extreme mode, it does appear that there is thermal headroom available with those H35 processors, so there might be a little CPU boost owing to that.
Here’s how CPU-only performance looks in Cinebench with the GPU now idle. The results were basically the same regardless of the performance mode in use, both for single and multicore scores, so it would seem that the processor hasn’t limited to a CPU-only workload. This is how it stacks up against other laptops. What I found most interesting was that the multicore score was basically the same as the Ryzen 7 4700U in the Acer Swift 3 just below it, as that’s an 8 core 8 thread part. The 5700U is about to launch though, which will change this.
At idle the middle of the keyboard was in the mid-30s, so a little warmer than usual but not an issue you’ll notice. With the stress tests running in silence it gets around 10 degrees warmer in the center, so a little warm to the touch in the middle now. The balanced mode was much the same, which makes sense as the internals were basically the same. The extreme mode was warmer now despite the fans running louder and internals running cooler as a result of the higher GPU power limit.
The back is quite hot, but you shouldn’t need to touch it there, so it’s not a practical problem. With the fan at full speed it’s much cooler now, still warm, but a significant improvement. Let’s have a listen to the fan noise. The fan was still audible when idling in silent mode. It was louder with the stress tests running, but still not too loud, then just a little louder stepping up to balanced mode.
The extreme mode was a little louder still, then cooler boost or max fan was significantly worse. Personally, I think this is a great thing, as it gives you the option of setting the fan speed that you like. If you’re comfortable with more noise you can take advantage of lower temperatures. In any case, it’s much better than being forced one way or the other. User choice is always best.
Now let’s check out how well the Stealth 15M performs in games and see how it compares with other laptops. I’ve tested Battlefield 5 in campaign mode at ultra settings, and the Stealth is highlighted in red. The results are on the lower side, in part due to the 65 watt 2060 graphics, though the Zephyrus G15 with the same power GPU was closer to 70 FPS in the same test, but with a more powerful 8 core processor.
As we saw earlier, CPU performance is actually lost in extreme performance mode to boost the GPU, so that could be resulting in a lower 1% low. These are the results from Far Cry 5 with ultra settings in the built-in benchmark. The Stealth 15M wasn’t doing great here for a laptop with 2060 graphics, but it was still near 70 FPS even with the ultra setting preset, so it’s not like it’s not going to be playable or anything. That said, still, around 10 FPS slower than the Zephyrus G14 and G15 with the same tier GPU, likely due to the processor difference, as this is a game I’ve noticed to be more processor dependent than others.
Shadow of the Tomb Raider was also tested with the game’s benchmark tool with the highest setting preset. This test seems to be more GPU heavy, so the Stealth rises a few levels, and is now just 1 FPS ahead of the Zephyrus G15 with the same tier GPU, despite the processor difference. Now for the benchmarking tools, I’ve tested Heaven, Valley, and Superposition from Unigine, as well as Firestrike, Timespy, and Port Royal from 3DMark, just pause the video if you want a detailed look at these results.
Content Creation Testing
I’ve used Adobe Premiere to export one of my laptop review videos at 4K. Lower times are better here, and the Stealth 15M was on the lower side, perhaps due to the power-limited quad-core processor. I’ve also tested Premiere but with the Puget systems benchmark, and this tests for more things like live playback rather than just export times. Higher scores are better now, and again the Stealth 15M was still lower than most.
Adobe Photoshop was a bit better comparatively in this same selection of laptops, very close to the Prestige 14 just above it with the same processor, which makes sense as this is more of a processor-heavy test. DaVinci Resolve on the other hand tends to be more GPU heavy. That said, it wasn’t scoring as well as the 2060 Max-Q with the same wattage in the ASUS Zephyrus G14 at 561 points, though that does also have an 8 core processor. I’ve also tested SPECviewperf which tests out various professional 3D workloads.
I’ve used Crystal disk mark to test the storage. The 512GB NVMe M.2 SSD was doing alright for the reads, but half the sequential speed when it came to writes. As an Intel 11th gen laptop, PCIe Gen 4 should also be supported if you want to upgrade to a faster SSD. The MicroSD slot was doing great, basically maxing out my card, and the card clicks in and sits most of the way into the laptop.
Finally, let’s discuss price, At the time of writing the 2060 model with the same specs I’ve tested is USD 1500 on Newegg. You could definitely get a better-performing laptop with more power and a better screen for less money, say the Lenovo Legion 5 for example, so it really comes down to how much you value the smaller and thinner size. In Australia, we’re looking at AUD 2500, or $500 more for the newer H35 and RTX 3060 Max-Q model.
As MSI is about to upgrade this model to the Intel 11th gen H35 processors and Nvidia RTX 3060 graphics, performance results with that hardware could vary a little to what we’ve seen in this review, but pretty much everything else should still apply. So with all of that in mind, let’s conclude by looking at both the good and the bad to help you decide if the MSI Stealth 15M gaming laptop is worth considering.
The 11th gen processor is performing quite well, particularly in terms of single-core performance, as Tiger Lake does quite well there. Intel 11th gen also means we get PCIe gen 4 and Thunderbolt 4, so it’s good to have some nice modern standards. The RTX 2060 graphics does do alright, but it is limited by that lower 65-watt power limit, so it can get beaten by cheaper 1660 Ti machines. The response time of the screen was quite good for a 144Hz gaming laptop, but the brightness, color gamut, and backlight bleed could be better.
The battery life wasn’t great either, and the upside-down motherboard makes memory upgrades a more involved process. I’m guessing it just must be a necessity to get it as thin as it is. It’s a bit strange that there appears to be space for a second M.2 slot that’s wasted, that could have otherwise been a larger battery.
Overall, I think the Stealth 15M is a bit pricey. It’s really only worth considering if you really want a smaller and lighter machine. Perhaps for travel, if that’s ever allowed again. Let me know what you thought about MSI’s Stealth 15M gaming laptop down in the comments, and share this article if it was useful.