- Get link
- Other Apps
- Get link
- Other Apps
Gigabyte Aero 17 Detail Review
Let’s find out what you get with this new larger version of Aero 17 and if it’s a laptop you should consider.
Aero 17 Specs:
Starting with the specs I’ve got the HDR XA model, so there’s an Intel i9-9980HK CPU, 90 watt Nvidia RTX 2070 Max-Q graphics, 16gb of memory in dual channel, 17.3” HDR
4K 60Hz IPS screen, and a 512gb M.2 NVMe SSD in one of the two available slots.
For network connectivity, it’s got gigabit ethernet, 802.11ax WiFi and Bluetooth 5.
There are a few different configurations available, the HDR version that I’ve got is available.
On the matte black metal lid, we’ve got the Aero logo, with a subtle brushed finish towards the bottom.
The interior is similar to most of the lid, all matte black but with a plastic outer trim, and overall it felt well built with no sharp edges or corners.
Gigabyte list the weight as around 2.5kg or 5.5lb, and mine was just over 2.6kg. Once we add the 230-watt power brick and cables the total rises to 3.5kg.
The dimensions of the Aero 17 are 39.6cm in width, 27cm in-depth, and 2.1cm in height, so on not too big for a 17-inch machine. The smaller footprint gives us thin bezels.
Gigabyte list them at 3mm however from edge to the actual screen I measured them at 8mm, giving it an 89% screen to body ratio.
The box and looks excellent, noticeably better compared to most 1080p panels I regularly test with.
I’ve measured the color gamut using the Spyder 5 Pro and got 100% of sRGB and 98% of AdobeRGB, very impressive results. At 100% brightness, I measured the panel at 570 nits in the center.
Which is way above the usual 300 average I typically see, and with a 1020:1 contrast ratio, again better than most others I’ve tested.
Backlight bleed looked fine, it was evenly lit and I didn’t notice any issues while viewing darker content, but this will vary between laptop and panel.
It’s also DisplayHDR 400 certified, and while nothing amazing when compared to other
HDR monitors I’ve tested, it’s still a step up over nothing, though I haven’t tested many HDR laptops yet.
Without an HDR camera and you having an HDR monitor I can’t fully capture the differences, but to me it did look better viewing HDR content than standard.
There was minimal screen flex, it felt quite solid. The hinges are found out towards the
corners and felt much stiffer compared to what I noted in the Aero 15 OLED.
No problems at all opening it up with one finger, it felt a little back heavy
but no problems using it on my lap.
Like other Aero laptops, the screen bezel is too thin so the 720p camera is found down
the bottom, and it’s also got a physically sliding privacy cover.
The camera looks about average and the microphone sounds pretty good. Fingers sort of get in the way when you type, and here's what it sounds like when you put the fan on maximum speed.
So it gets pretty loud and you can't hear me, so it's not good at isolating my voice. The keyboard worked well and I had no issues typing with it.
It’s got individual key RGB backlighting with quite a few effects built-in, however, unfortunately, the secondary key functions do not get lit up.
Here’s how typing sounds to give you an idea of what to expect.
There was minor keyboard flex while pushing down hard, overall the metal body was quite sturdy.
The precision touchpad was extremely smooth to the touch and worked very well. It presses down anywhere to click, supports the usual gestures, and has a fingerprint scanner in the top left corner.
Which means that the top corner has no touchpad functionality. I was never limited by this and preferred the addition of the fingerprint scanner, which seemed quite fast for unlocking the machine.
Fingerprints show up easily but are kind of masked by the matte finish, and as a smooth surface, they were easy to clean.
On the left from the back, there’s an air exhaust vent, gigabit ethernet port which
is facing the preferred way.
No lifting the machine to get the cable out, UHS-II SD card slot, USB 3.1 Gen1 Type-A port, 3.5mm mic and headphone jacks, and second USB 3.1 Gen1 Type-A port.
On the right from the front, there’s a third USB 3.1 Gen1 Type-A port, two Type-C ports.
The first of which is Thunderbolt 3 and second is USB 3.1 with DisplayPort 1.4 output, HDMI 2.0 output, power input, and another air exhaust vent.
On the back there are air exhausts towards the corners, and the Aero logo in the center
does not get lit up here, on the Aero 15 it had its light.
Meanwhile, the front is just plastic with a groove for your finger to open the lid.
The Aero logo on the lid lights up white from the backlight of the screen, so it cannot be adjusted.
Underneath there’s heaps of air ventilation towards the back half of the machine.
The two speakers are found towards the front left and right corners, they sound alright, clear at higher volumes with a little bass present.
Here’s what we’re looking at while playing music at maximum volume, and the Latencymon results could have been better.
The bottom panel can be easily removed by taking out 14 TR6 screws.
Once inside from left to right we’ve got an M.2 slot for NVMe or SATA storage, the battery, WiFi 6 card and two memory slots above this, followed by the second M.2 slot.
Powering the laptop is the large 94-watt hour battery the Aero series is well known for.
I’ve tested it with the screen brightness at 50%, background apps disabled, and all RGB lighting off.
While just watching YouTube videos it lasted for 6 hours and 32 minutes, and it was using the Intel integrated graphics due to Nvidia Optimus.
While playing the Witcher 3 with medium settings and Nvidia’s battery boost set to 30 FPS the battery lasted for 1 hour and 41 minutes in total.
However, after the first hour and 19 minutes at 17% charge remaining the frame rate dipped to 6 FPS and was no longer usable.
This puts battery life around average for gaming in terms of the period the game was playable, while the battery life outside of gaming made this one of the best machines I’ve tested.
The 230-watt power brick that Gigabyte includes with the Aero 17 appears to be adequate for these specs.
I was seeing some drain during my testing but it would stop at around the 95% mark which is pretty standard behavior, it never dropped further.
It’s also worth noting the AI feature doesn’t work on battery power, and you can’t charge it over Type-C.
The Aero 17 is an Nvidia Studio certified laptop, and as such it comes with the Nvidia Studio drivers installed out of the box.
The Aero series is typically meant more for content creation than gaming, something made evident by the excellent 4K HDR screen, but don’t worry you can of course still play games and we’ll test these out later.
For now, though, let's get an idea of how the machine works for content-creation tasks.
I’ll just be comparing between the Aero 17, and my older Aero 15x which has the i7-8750H CPU and GTX 1070 Max-Q graphics.
I’ve tested the Aero 17 both at stock, and with the following customizations in place to try and further boost performance as it does have an unlocked CPU.
Starting with Adobe Premiere there wasn’t much difference, while the extra two CPU cores from the 9980HK in the Aero 17 are helping, both laptops are using quick sync which seems to be taking most of the work.
This is my first time using the Puget Systems photoshop benchmark, so I don’t have any other data to compare against.
The Aero 17 is scoring 13% better in this test, with a little boost possible with CPU overclocking.
The blender benchmark only makes use of the CPU, and we can see in this test the two extra cores and higher clock speeds from the Aero 17 are making quite a large difference when compared to my 15x, and the overclock to the 17 made a nice improvement.
I’ve also used Handbrake to convert a 4k video to 1080p, and then a separate video
file from 1080p to 720p.
This is another task that benefits from additional CPU cores, and we’re seeing nice improvements to this workload with the Aero 17.
I’ve used the LuxMark benchmark which uses OpenCL to render various scenes, and for this test, we’re only looking at Nvidia GPU performance.
So we’re just seeing the differences between 2070 Max-Q and 1070 Max-Q here.
SPECviewperf measures graphics performance based on professional applications that run on both OpenGL and Direct X.
While I haven’t got numbers in this test from other laptops at the moment, this one is quite popular in the professional space so should let you see how it stacks up against others.
Let me know if you’d be interested in seeing me test this on future machines.
Thermal Testing:
Let’s move onto the thermal testing. Just to recap, there’s a big mesh area on the bottom for air intake, and it’s also pulled in above the keyboard and through the keyboard.
Air is exhausted through the vents on the back, and the vents on either side up the back.
In terms of heat pipes, we’ve got two shared between the CPU and GPU, and the design looks the same as the new 15-inch model.
However, the heatsinks look a bit thicker due to the extra space available with the larger 17-inch chassis.
Gigabyte have stated that they’re using Thermal Grizzly hydronaut thermal paste, so that should be better compared to many
others out there.
The Gigabyte control center software allows us to set the CPU between 6 different levels of power, and the GPU between two levels, as defined here.
Throughout this testing, I’ve tested the stock settings with the CPU at level 2 and GPU at level 0 with the fans on the normal profile.
Here’s what the default fan curve of the normal profile looks like, so we’re not even getting to 70% max speed and expect it to run warmer but quieter as a result.
The gaming profile boosts the fan speed a bit, and we also have the option of customizing the curve for optimum performance.
You can also quickly set the fan speed to maximum by pressing the function and escape keys, and when I’ve tested the fan at max speed.
I’ve also boosted the CPU to level 5 and GPU to level 1 for highest levels of performance, so up to 80 watts on the CPU
and 90 watts on the GPU, thermals permitting of course.
Thermal testing was completed in an ambient room temperature of 21 degrees Celsius, so expect different results in different environments.
At idle it was a little warm, however in quiet mode the fans were completely silent so this is expected.
The rest of the results are from combined CPU and GPU workloads and are meant to represent worst-case scenarios as I ran them for extended periods.
Dogs 2, as I find it to use a good combination of processor and graphics.
The stress test results shown on the lower half of the graph are from running the Aida64 CPU stress test with only the stress CPU option checked.
The Heaven GPU benchmark at max settings at the same time to fully load the system.
I’ll start by noting that the CPU was thermal throttling at 90 degrees Celsius,
which we can see happening in many of these tests.
I think this is a good thing,
Gigabyte is capping the thermal limit to prevent it from getting too hot, and as we’ll
see in the next graph performance is still great.
There was some thermal throttling on the GPU while gaming at stock when it averaged 86 degrees, but that was the only instance and was tested with the normal fan profile, simply swapping to gaming or above would remove it.
By setting the fan speed to maximum we’re able to improve the GPU temperature despite the GPU power limit raising from 80 to 90 watts, however this action alone wasn’t enough to remove the thermal throttling on the CPU.
Once we undervolt the CPU no changes were seen in terms of temperatures while under stress test, however this was enough to remove the thermal throttling while playing this particular game.
As thermal throttling was the limitation while under stress test I added in a cooling pad, and we saw this drop temperatures by 6 degrees and remove thermal throttling.
The thermals were now under control so we’re in a place to attempt some CPU overclocking, and as this requires more power more heat is expected.
So the overclock took us back to thermal throttling while under stress test.
While gaming the overclock rose the CPU temperature back up by 7 degrees Celsius, however it wasn’t thermal throttling under this workload.
These are the average clock speeds for the same tests just shown. We’re seeing the
lowest clock speeds with the normal fan profile as thermals were the key limitation at this point.
By increasing the fan speed we have more thermal headroom, which is why I also
boosted the CPU and GPU power limits here, and this is why we’re getting higher clock speeds.
The CPU undervolt improved performance further, though more so for the stress test
result compared to gaming.
The cooling pad made no major difference in terms of performance, while the 9980HK has an all-core turbo boost speed of 4.2GHz I found mine would cap out at 4.1GHz, and this was hit with the cooling pad both under stress test and while gaming.
By overclocking to 4.4GHz we’re almost averaging this while gaming, so I probably could have pushed even further there, an excellent result.
While under stress test the thermal throttling was back with the overclock, however I think a 4.25GHz average on 8 CPU cores is an excellent result.
These are the average TDP values reported by hardware info during these same tests.
We can see the GPU power limit raise from 80 to 90 watts by going from GPU level 0 to level 1, as per the green bars.
While the CPU is at level 5 it has an 80-watt power limit, however we never got close to this under these combined CPU and GPU tests due to thermal throttling.
This means that power limit throttling was never an issue in these tests, so there’s that.
The highest I saw was a 69-watt average once overclocked and undervolted, higher was not possible due to thermal throttling, at least in these workloads.
Here are some CPU only benchmarks from Cinebench. Just for reference, your typical i7-9750H seems to be around 2800 in a good machine, while I managed to push the i7 in the Aero 15 OLED to around 3100 points for comparison.
This means we’re seeing a 35% higher score in this test with the Aero 17 and 8 core overclocked i9 CPU compared to the i7 in the Aero 15, and I already consider the Aero 15 score as one of the highest I’ve ever gotten from the i7.
The single-core clock speed was higher in the Aero 17 as the i9 can go right up to 5GHz at stock while under single-core workload, thermals permitting.
As for the external temperatures where you’ll be putting your hand, at idle it was in the mid-30s in the center, just a little above average.
While gaming with the fan on the normal profile it’s getting to the mid-40s in the center, so on the warmer side, and up to 60 right up the back which was hot to the touch.
Simply by setting the fan to maximum speed the hot spots drop by around 10 degrees and the keyboard was much cooler
to the touch.
Here’s what a combined CPU and GPU stress test looks like with the fans at maximum, but this time with the CPU and GPU power limits boosted, and it felt fine
due to the fan speed.
As for the fan noise produced by the laptop, I’ll let you have a listen to some of these tests.
At idle with the quiet profile it was completely silent however I could hear some subtle coil whine, though that vanished with any level of fan speed.
With the normal fan profile, it’s actually fairly quiet compared to most other laptops I’ve tested, so you do have the option of running quieter for less performance, and then the gaming profile rose that up a bit, but still lower than most other gaming laptops I’ve used.
With the fans at max speed it is quite loud, however I think this is a good thing. As Gigabyte let us adjust the fan curve with high levels of granularity it means you have the option of customizing it to find a sweet spot that works best for you.
Overall I’m quite impressed with these results, keeping in mind these are essentially worst-case scenarios that we’re looking at.
While this is the first time I’ve ever tested
the i9-9980HK CPU, the temperatures were always kept under control regardless of my modifications as Gigabyte is capping the CPU temperature to 90 degrees Celsius.
They could, of course, do what Dell did in the G5 and allow it to run up to 100 degrees, and while this would improve performance a little I’m personally happy to take the hit so temperatures don’t go above 90.
Even at stock simply by raising the fan speed we’re able to get pretty good clock speeds, especially considering these were over 8 cores.
While some simple undervolting improved things further.
In my tests just by undervolting with raised fan speed even under combined CPU and GPU stress test, we’re almost at 4.1GHz on all 8 cores.
Which I think is quite impressive when compared against most of the 6 core i7 machines I’ve tested.
As you’d expect, with 8 cores and high power limits thermals were the limitation, so using a cooling pad was able to help out a fair bit.
While I haven’t tested out the i7 in the Aero
17, based on these results I’m assuming it should handle it well.
I’ll also note that the BIOS was pretty locked down, there were no options to undervolt or even otherwise adjust CPU clock speed for overclocking, so that’ll all need to be done through software.
Gigabyte also have their AI software which automatically controls the power limits that
I’ve manually tested here.
The idea is that you just enable the AI and over time it learns what you do with the machine and then you don’t need to go through and constantly change these settings around yourself.
Games Testing On Gigabyte Aero 17:
Next let’s take a look at some gaming benchmarks. I’ve tested these with the CPU set to level 5 and GPU set to level 1.
So the highest power limits available, though this does not apply any overclocks or undervolting, we’ll look at those afterward.
While the Aero 17 came with Nvidia Studio drivers, for the game tests I installed game ready drivers for best performance.
I’ve run most of these games at 1080p for better performance, which scales down perfectly on the 4K screen, but I have also tested some at 4K too.
Battlefield 5 Testing:
Battlefield 5 was tested in campaign mode rather than multiplayer. I’ve tested with RTX off in the purple bars, and RTX on in the green bars.
Even with RTX on it was still playing well enough at high settings, ultra wasn’t too terrible either due to the high 1% low performance.
I’d rather play at ultra with RTX off, as it both looks good and runs well.
I’ve also tested this game at 4K just to show what it’s capable of, and while playable as the 1% lows were quite good, we weren’t even averaging 60 FPS at low settings.
Apex Legends Testing:
Apex Legends was tested with either all settings at maximum, or all settings on the lowest possible values, as it doesn’t have predefined setting presets.
It was still playing well with maximum settings, however with minimum settings we could boost average FPS by 36%, though I didn’t see much of a difference in this test compared to other i7 laptops with similar graphics.
Shadow of the Tomb Raider Testing:
Shadow of the Tomb Raider was tested with the built-in benchmark, the results from this test we're looking pretty great at lower settings due to the high CPU power limit, and the results were still better than most 2070 Max-Q machines at higher settings.
Though it was close, but we’ll see how this game and others compare later.
Far Cry New DawnDawn Testing:
Far Cry New Dawn was tested with the built-in benchmark.
This game seems to be fairly CPU heavy, however it was one where I didn’t find that much difference when compared
to other i7 laptops.
Though I have noted before in my desktop CPU comparisons that Far Cry tends to not work too well with increased core counts.
Fortnite Testing:
Fortnite was tested with the replay feature, and epic settings was still working just fine
with 130 FPS averages and relatively high 1% low performance, with much higher frame rates possible at the lower setting levels.
This was another game where I tested 4K, however this makes a bit more sense as it already performs very well at 1080p, and medium and low settings were still playing quite well.
Overwatch is another well-optimized game and was tested in the practice range. The 300
FPS frame cap was hit at low settings, while even max settings was still averaging well above 100 FPS for the 1% low and was playing very well.
Again I also gave 4K a shot here as it was running so well at 1080p, and for the most
part it wasn’t too bad at max settings, though even at ultra the 1% low was above
the refresh rate of the display.
CS:GO Testing:
CS:GO was tested using the Ulletical FPS benchmark, and as a game that is pretty CPU bound we’re seeing above average results compared to most i7-9750H laptops that I’ve tested.
Rainbow Six Siege:
Rainbow Six Siege was tested with the built-in benchmark, and the results weren’t too
different in this test compared to similarly specced machines.
Don’t get me wrong, still excellent performance, but it doesn’t seem that the extra CPU cores and slight clock speed difference helps here compared to an i7.
Assassin’s Creed Odyssey:
Assassin’s Creed Odyssey was tested with the built-in benchmark, and from my experience seems to be a fairly CPU heavy test.
There’s not much of a boost over most other laptops I’ve tested though, still fairly similar average FPS, however I was seeing higher 1% lows due to the CPU compared to say the Y740 with same GPU but i7 CPU.
Dota 2:
Dota 2 was tested playing in the middle lane with an average amount of action going on, and is yet another game that I’ve found to prefer higher CPU power.
Which explains why the average frame rates at lower settings are higher than the 9750H machines I usually test, granted not by enough to make a practical difference as we’re already around 200 FPS.
Watch Dogs 2:
Watch Dogs 2 is a resource-heavy game, but in my opinion, is still playable with a stable 30 FPS.
So even at ultra settings with our higher than 60 FPS average, it’s playing very well without any issues, and I don’t often see 100 FPS being reached at all in this title when it comes to laptops, so that seems to be due to the i9.
The Witcher 3:
The Witcher 3 was playing well with hair works disabled, I’ve found this to be a somewhat more GPU demanding title.
For quick comparison, the average frame rates are close to the Y740 I tested with 2070 Max-Q, however the 1% lows with the i9 are 10 to 20 FPS higher than the i7.
If you’re after more gaming benchmarks check the card in the top right corner where I’ve tested 20 games in total.
Let’s also take a look at how this config of the new Aero 17 compares with other laptops to see how it stacks up, use these results as a rough guide only as they were tested at different times with different drivers.
Battlefield 5 Tested:
In Battlefield 5 I’ve got the Aero 17 highlighted in red near similarly specced machines, including the new Aero 15 OLED just below it. In terms of average FPS.
The result is right in line with the Aero 15 and Lenovo Y740, both of which are machines with the same GPU but i7 CPU.
The average from the Razer Blade Pro 17 with the same GPU is significantly higher, as that machine undervolted the CPU, overclocks the graphics, and boosts the GPU power limit higher.
I’m only testing these machines at their best stock settings, so with some tweaks, we could also improve the Aero 17.
Which we’ll see soon, however even at stock due to the superior i9 CPU with higher clock speeds and more cores we’re getting the highest 1% low result out of all machines tested.
This 1% result is ahead of the average FPS in the Scar II with RTX 2060, just for comparison.
Far Cry 5 Tested:
Here are the results from Far Cry 5 with ultra settings in the built-in benchmark.
I’ve found this to be a pretty CPU heavy test, though unlike the last game we’re not seeing results too different from the other similarly specced machines.
The average FPS was again in line with most of the other 2070 Max-Q machines tested, though the 1% low was a little ahead of the others, again most likely due to the higher CPU performance, the difference is just much smaller in this title.
Shadow of the Tomb raider Tested:
These are the results from Shadow of the Tomb Raider with the built-in benchmark at the highest settings.
Once more the average frame rates are very similar to the other 2070 Max-Q machines
tested, again except for the Blade Pro as that’s just tuned very well out of the box, still though the Aero 17 holds up nicely here.
Overall the new Gigabyte Aero 17 laptop is providing high levels of performance, though realistically not too different from an i7 laptop.
So if you’re purely using it for gaming you could save money by looking at a non-HDR version with 1080p 144Hz screen and i7 CPU.
Now let’s see how much we can improve performance if we make some changes, undervolting seems required in most cases if you want a shot at overclocking to higher CPU clock speeds.
In battlefield 5 at maximum settings, we’re seeing a 6% improvement to both average FPS and 1% low performance with these changes in place.
In Shadow of the Tomb Raider with the built-in benchmark at highest settings, there was
a larger difference, almost 9% higher average frame rate was happening with these tweaks.
In Far Cry 5 with the built-in benchmark at ultra settings, there was almost a 7% improvement to average FPS with the changes in place, though there was less of a difference when it came to 1% low performance.
Storage Testing of Gigabyte Aero 17:
I’ve used Crystal Disk Mark to test the storage, and the 512gb NVMe M.2 SSD was performing well enough, though a fair bit lower on the writes compared to reads.
While the UHS-II SD card slot was performing very well with my V90 card for the reads, but again much lower for the writes.
Let’s conclude by covering the good and bad aspects of the new Gigabyte Aero 17 laptop.
The Aero 17 laptop is a nice addition to the Aero lineup for those that have been after
a larger version. For the most part, you’re getting the same machine, but bigger.
There’s no option of OLED here as there aren’t any panel manufacturers offering 17-inch versions yet.
However, the 4K HDR panel in my unit was quite good, though there are also 1080p 144Hz options which may be preferable for gamers.
The CPU performance I was seeing from the i9-9980HK was very impressive for a laptop, and while it did get warm, as expected, we could make quite a few improvements to both performance and thermals.
In the end I’m quite happy with the results, being able to average 4.25GHz over all 8 cores while under combined CPU and GPU stress test is a great result.
The Aero 17 is a great option for content creators. As mentioned the raw power is there, and the 4K screen looks great with good color gamut.
The highest brightness I’ve ever measured on a laptop, and as a result DisplayHDR 400.
Combined with the great selection of I/O such as the UHS-II SD card slot and Thunderbolt 3 and I think the Aero 17 is in a great position for creators compared to most others I’ve tested so far.
I’d still prefer the Aero 15, but that’s simply because I don’t need a larger screen and I’m happy with a more portable machine.
If I was after a 17-inch laptop purely for content creation though, this is the best I’ve tested so far in that regard.
I didn’t like that the secondary key functions weren’t lit, and I know some people won’t be a fan of the camera placement, other than that I can’t think of much else I didn’t like.
Let me know what you thought about the new Aero 17 laptop down in the comments.
Let’s find out what you get with this new larger version of Aero 17 and if it’s a laptop you should consider.
Aero 17 Specs:
Starting with the specs I’ve got the HDR XA model, so there’s an Intel i9-9980HK CPU, 90 watt Nvidia RTX 2070 Max-Q graphics, 16gb of memory in dual channel, 17.3” HDR
4K 60Hz IPS screen, and a 512gb M.2 NVMe SSD in one of the two available slots.
For network connectivity, it’s got gigabit ethernet, 802.11ax WiFi and Bluetooth 5.
There are a few different configurations available, the HDR version that I’ve got is available.
On the matte black metal lid, we’ve got the Aero logo, with a subtle brushed finish towards the bottom.
The interior is similar to most of the lid, all matte black but with a plastic outer trim, and overall it felt well built with no sharp edges or corners.
Gigabyte list the weight as around 2.5kg or 5.5lb, and mine was just over 2.6kg. Once we add the 230-watt power brick and cables the total rises to 3.5kg.
The dimensions of the Aero 17 are 39.6cm in width, 27cm in-depth, and 2.1cm in height, so on not too big for a 17-inch machine. The smaller footprint gives us thin bezels.
Gigabyte list them at 3mm however from edge to the actual screen I measured them at 8mm, giving it an 89% screen to body ratio.
The box and looks excellent, noticeably better compared to most 1080p panels I regularly test with.
I’ve measured the color gamut using the Spyder 5 Pro and got 100% of sRGB and 98% of AdobeRGB, very impressive results. At 100% brightness, I measured the panel at 570 nits in the center.
Which is way above the usual 300 average I typically see, and with a 1020:1 contrast ratio, again better than most others I’ve tested.
Backlight bleed looked fine, it was evenly lit and I didn’t notice any issues while viewing darker content, but this will vary between laptop and panel.
It’s also DisplayHDR 400 certified, and while nothing amazing when compared to other
HDR monitors I’ve tested, it’s still a step up over nothing, though I haven’t tested many HDR laptops yet.
Without an HDR camera and you having an HDR monitor I can’t fully capture the differences, but to me it did look better viewing HDR content than standard.
There was minimal screen flex, it felt quite solid. The hinges are found out towards the
corners and felt much stiffer compared to what I noted in the Aero 15 OLED.
No problems at all opening it up with one finger, it felt a little back heavy
but no problems using it on my lap.
Like other Aero laptops, the screen bezel is too thin so the 720p camera is found down
the bottom, and it’s also got a physically sliding privacy cover.
The camera looks about average and the microphone sounds pretty good. Fingers sort of get in the way when you type, and here's what it sounds like when you put the fan on maximum speed.
So it gets pretty loud and you can't hear me, so it's not good at isolating my voice. The keyboard worked well and I had no issues typing with it.
It’s got individual key RGB backlighting with quite a few effects built-in, however, unfortunately, the secondary key functions do not get lit up.
Here’s how typing sounds to give you an idea of what to expect.
There was minor keyboard flex while pushing down hard, overall the metal body was quite sturdy.
The precision touchpad was extremely smooth to the touch and worked very well. It presses down anywhere to click, supports the usual gestures, and has a fingerprint scanner in the top left corner.
Which means that the top corner has no touchpad functionality. I was never limited by this and preferred the addition of the fingerprint scanner, which seemed quite fast for unlocking the machine.
Fingerprints show up easily but are kind of masked by the matte finish, and as a smooth surface, they were easy to clean.
On the left from the back, there’s an air exhaust vent, gigabit ethernet port which
is facing the preferred way.
No lifting the machine to get the cable out, UHS-II SD card slot, USB 3.1 Gen1 Type-A port, 3.5mm mic and headphone jacks, and second USB 3.1 Gen1 Type-A port.
On the right from the front, there’s a third USB 3.1 Gen1 Type-A port, two Type-C ports.
The first of which is Thunderbolt 3 and second is USB 3.1 with DisplayPort 1.4 output, HDMI 2.0 output, power input, and another air exhaust vent.
On the back there are air exhausts towards the corners, and the Aero logo in the center
does not get lit up here, on the Aero 15 it had its light.
Meanwhile, the front is just plastic with a groove for your finger to open the lid.
The Aero logo on the lid lights up white from the backlight of the screen, so it cannot be adjusted.
Underneath there’s heaps of air ventilation towards the back half of the machine.
The two speakers are found towards the front left and right corners, they sound alright, clear at higher volumes with a little bass present.
Here’s what we’re looking at while playing music at maximum volume, and the Latencymon results could have been better.
The bottom panel can be easily removed by taking out 14 TR6 screws.
Once inside from left to right we’ve got an M.2 slot for NVMe or SATA storage, the battery, WiFi 6 card and two memory slots above this, followed by the second M.2 slot.
Powering the laptop is the large 94-watt hour battery the Aero series is well known for.
I’ve tested it with the screen brightness at 50%, background apps disabled, and all RGB lighting off.
While just watching YouTube videos it lasted for 6 hours and 32 minutes, and it was using the Intel integrated graphics due to Nvidia Optimus.
While playing the Witcher 3 with medium settings and Nvidia’s battery boost set to 30 FPS the battery lasted for 1 hour and 41 minutes in total.
However, after the first hour and 19 minutes at 17% charge remaining the frame rate dipped to 6 FPS and was no longer usable.
This puts battery life around average for gaming in terms of the period the game was playable, while the battery life outside of gaming made this one of the best machines I’ve tested.
The 230-watt power brick that Gigabyte includes with the Aero 17 appears to be adequate for these specs.
I was seeing some drain during my testing but it would stop at around the 95% mark which is pretty standard behavior, it never dropped further.
It’s also worth noting the AI feature doesn’t work on battery power, and you can’t charge it over Type-C.
The Aero 17 is an Nvidia Studio certified laptop, and as such it comes with the Nvidia Studio drivers installed out of the box.
The Aero series is typically meant more for content creation than gaming, something made evident by the excellent 4K HDR screen, but don’t worry you can of course still play games and we’ll test these out later.
For now, though, let's get an idea of how the machine works for content-creation tasks.
I’ll just be comparing between the Aero 17, and my older Aero 15x which has the i7-8750H CPU and GTX 1070 Max-Q graphics.
I’ve tested the Aero 17 both at stock, and with the following customizations in place to try and further boost performance as it does have an unlocked CPU.
Starting with Adobe Premiere there wasn’t much difference, while the extra two CPU cores from the 9980HK in the Aero 17 are helping, both laptops are using quick sync which seems to be taking most of the work.
This is my first time using the Puget Systems photoshop benchmark, so I don’t have any other data to compare against.
The Aero 17 is scoring 13% better in this test, with a little boost possible with CPU overclocking.
The blender benchmark only makes use of the CPU, and we can see in this test the two extra cores and higher clock speeds from the Aero 17 are making quite a large difference when compared to my 15x, and the overclock to the 17 made a nice improvement.
I’ve also used Handbrake to convert a 4k video to 1080p, and then a separate video
file from 1080p to 720p.
This is another task that benefits from additional CPU cores, and we’re seeing nice improvements to this workload with the Aero 17.
I’ve used the LuxMark benchmark which uses OpenCL to render various scenes, and for this test, we’re only looking at Nvidia GPU performance.
So we’re just seeing the differences between 2070 Max-Q and 1070 Max-Q here.
SPECviewperf measures graphics performance based on professional applications that run on both OpenGL and Direct X.
While I haven’t got numbers in this test from other laptops at the moment, this one is quite popular in the professional space so should let you see how it stacks up against others.
Let me know if you’d be interested in seeing me test this on future machines.
Thermal Testing:
Let’s move onto the thermal testing. Just to recap, there’s a big mesh area on the bottom for air intake, and it’s also pulled in above the keyboard and through the keyboard.
Air is exhausted through the vents on the back, and the vents on either side up the back.
In terms of heat pipes, we’ve got two shared between the CPU and GPU, and the design looks the same as the new 15-inch model.
However, the heatsinks look a bit thicker due to the extra space available with the larger 17-inch chassis.
Gigabyte have stated that they’re using Thermal Grizzly hydronaut thermal paste, so that should be better compared to many
others out there.
The Gigabyte control center software allows us to set the CPU between 6 different levels of power, and the GPU between two levels, as defined here.
Throughout this testing, I’ve tested the stock settings with the CPU at level 2 and GPU at level 0 with the fans on the normal profile.
Here’s what the default fan curve of the normal profile looks like, so we’re not even getting to 70% max speed and expect it to run warmer but quieter as a result.
The gaming profile boosts the fan speed a bit, and we also have the option of customizing the curve for optimum performance.
You can also quickly set the fan speed to maximum by pressing the function and escape keys, and when I’ve tested the fan at max speed.
I’ve also boosted the CPU to level 5 and GPU to level 1 for highest levels of performance, so up to 80 watts on the CPU
and 90 watts on the GPU, thermals permitting of course.
At idle it was a little warm, however in quiet mode the fans were completely silent so this is expected.
The rest of the results are from combined CPU and GPU workloads and are meant to represent worst-case scenarios as I ran them for extended periods.
Dogs 2, as I find it to use a good combination of processor and graphics.
The stress test results shown on the lower half of the graph are from running the Aida64 CPU stress test with only the stress CPU option checked.
The Heaven GPU benchmark at max settings at the same time to fully load the system.
I’ll start by noting that the CPU was thermal throttling at 90 degrees Celsius,
which we can see happening in many of these tests.
I think this is a good thing,
Gigabyte is capping the thermal limit to prevent it from getting too hot, and as we’ll
see in the next graph performance is still great.
There was some thermal throttling on the GPU while gaming at stock when it averaged 86 degrees, but that was the only instance and was tested with the normal fan profile, simply swapping to gaming or above would remove it.
By setting the fan speed to maximum we’re able to improve the GPU temperature despite the GPU power limit raising from 80 to 90 watts, however this action alone wasn’t enough to remove the thermal throttling on the CPU.
Once we undervolt the CPU no changes were seen in terms of temperatures while under stress test, however this was enough to remove the thermal throttling while playing this particular game.
As thermal throttling was the limitation while under stress test I added in a cooling pad, and we saw this drop temperatures by 6 degrees and remove thermal throttling.
The thermals were now under control so we’re in a place to attempt some CPU overclocking, and as this requires more power more heat is expected.
So the overclock took us back to thermal throttling while under stress test.
While gaming the overclock rose the CPU temperature back up by 7 degrees Celsius, however it wasn’t thermal throttling under this workload.
These are the average clock speeds for the same tests just shown. We’re seeing the
lowest clock speeds with the normal fan profile as thermals were the key limitation at this point.
By increasing the fan speed we have more thermal headroom, which is why I also
boosted the CPU and GPU power limits here, and this is why we’re getting higher clock speeds.
The CPU undervolt improved performance further, though more so for the stress test
result compared to gaming.
The cooling pad made no major difference in terms of performance, while the 9980HK has an all-core turbo boost speed of 4.2GHz I found mine would cap out at 4.1GHz, and this was hit with the cooling pad both under stress test and while gaming.
By overclocking to 4.4GHz we’re almost averaging this while gaming, so I probably could have pushed even further there, an excellent result.
These are the average TDP values reported by hardware info during these same tests.
We can see the GPU power limit raise from 80 to 90 watts by going from GPU level 0 to level 1, as per the green bars.
While the CPU is at level 5 it has an 80-watt power limit, however we never got close to this under these combined CPU and GPU tests due to thermal throttling.
This means that power limit throttling was never an issue in these tests, so there’s that.
The highest I saw was a 69-watt average once overclocked and undervolted, higher was not possible due to thermal throttling, at least in these workloads.
Here are some CPU only benchmarks from Cinebench. Just for reference, your typical i7-9750H seems to be around 2800 in a good machine, while I managed to push the i7 in the Aero 15 OLED to around 3100 points for comparison.
This means we’re seeing a 35% higher score in this test with the Aero 17 and 8 core overclocked i9 CPU compared to the i7 in the Aero 15, and I already consider the Aero 15 score as one of the highest I’ve ever gotten from the i7.
The single-core clock speed was higher in the Aero 17 as the i9 can go right up to 5GHz at stock while under single-core workload, thermals permitting.
As for the external temperatures where you’ll be putting your hand, at idle it was in the mid-30s in the center, just a little above average.
While gaming with the fan on the normal profile it’s getting to the mid-40s in the center, so on the warmer side, and up to 60 right up the back which was hot to the touch.
Simply by setting the fan to maximum speed the hot spots drop by around 10 degrees and the keyboard was much cooler
to the touch.
Here’s what a combined CPU and GPU stress test looks like with the fans at maximum, but this time with the CPU and GPU power limits boosted, and it felt fine
due to the fan speed.
As for the fan noise produced by the laptop, I’ll let you have a listen to some of these tests.
At idle with the quiet profile it was completely silent however I could hear some subtle coil whine, though that vanished with any level of fan speed.
With the normal fan profile, it’s actually fairly quiet compared to most other laptops I’ve tested, so you do have the option of running quieter for less performance, and then the gaming profile rose that up a bit, but still lower than most other gaming laptops I’ve used.
With the fans at max speed it is quite loud, however I think this is a good thing. As Gigabyte let us adjust the fan curve with high levels of granularity it means you have the option of customizing it to find a sweet spot that works best for you.
Overall I’m quite impressed with these results, keeping in mind these are essentially worst-case scenarios that we’re looking at.
While this is the first time I’ve ever tested
the i9-9980HK CPU, the temperatures were always kept under control regardless of my modifications as Gigabyte is capping the CPU temperature to 90 degrees Celsius.
They could, of course, do what Dell did in the G5 and allow it to run up to 100 degrees, and while this would improve performance a little I’m personally happy to take the hit so temperatures don’t go above 90.
Even at stock simply by raising the fan speed we’re able to get pretty good clock speeds, especially considering these were over 8 cores.
While some simple undervolting improved things further.
In my tests just by undervolting with raised fan speed even under combined CPU and GPU stress test, we’re almost at 4.1GHz on all 8 cores.
Which I think is quite impressive when compared against most of the 6 core i7 machines I’ve tested.
As you’d expect, with 8 cores and high power limits thermals were the limitation, so using a cooling pad was able to help out a fair bit.
While I haven’t tested out the i7 in the Aero
17, based on these results I’m assuming it should handle it well.
I’ll also note that the BIOS was pretty locked down, there were no options to undervolt or even otherwise adjust CPU clock speed for overclocking, so that’ll all need to be done through software.
Gigabyte also have their AI software which automatically controls the power limits that
I’ve manually tested here.
The idea is that you just enable the AI and over time it learns what you do with the machine and then you don’t need to go through and constantly change these settings around yourself.
Games Testing On Gigabyte Aero 17:
Next let’s take a look at some gaming benchmarks. I’ve tested these with the CPU set to level 5 and GPU set to level 1.
So the highest power limits available, though this does not apply any overclocks or undervolting, we’ll look at those afterward.
While the Aero 17 came with Nvidia Studio drivers, for the game tests I installed game ready drivers for best performance.
I’ve run most of these games at 1080p for better performance, which scales down perfectly on the 4K screen, but I have also tested some at 4K too.
Battlefield 5 Testing:
Battlefield 5 was tested in campaign mode rather than multiplayer. I’ve tested with RTX off in the purple bars, and RTX on in the green bars.
Even with RTX on it was still playing well enough at high settings, ultra wasn’t too terrible either due to the high 1% low performance.
I’d rather play at ultra with RTX off, as it both looks good and runs well.
I’ve also tested this game at 4K just to show what it’s capable of, and while playable as the 1% lows were quite good, we weren’t even averaging 60 FPS at low settings.
Apex Legends Testing:
Apex Legends was tested with either all settings at maximum, or all settings on the lowest possible values, as it doesn’t have predefined setting presets.
It was still playing well with maximum settings, however with minimum settings we could boost average FPS by 36%, though I didn’t see much of a difference in this test compared to other i7 laptops with similar graphics.
Shadow of the Tomb Raider Testing:
Shadow of the Tomb Raider was tested with the built-in benchmark, the results from this test we're looking pretty great at lower settings due to the high CPU power limit, and the results were still better than most 2070 Max-Q machines at higher settings.
Though it was close, but we’ll see how this game and others compare later.
Far Cry New DawnDawn Testing:
Far Cry New Dawn was tested with the built-in benchmark.
This game seems to be fairly CPU heavy, however it was one where I didn’t find that much difference when compared
to other i7 laptops.
Though I have noted before in my desktop CPU comparisons that Far Cry tends to not work too well with increased core counts.
Fortnite Testing:
Fortnite was tested with the replay feature, and epic settings was still working just fine
with 130 FPS averages and relatively high 1% low performance, with much higher frame rates possible at the lower setting levels.
This was another game where I tested 4K, however this makes a bit more sense as it already performs very well at 1080p, and medium and low settings were still playing quite well.
Overwatch is another well-optimized game and was tested in the practice range. The 300
FPS frame cap was hit at low settings, while even max settings was still averaging well above 100 FPS for the 1% low and was playing very well.
Again I also gave 4K a shot here as it was running so well at 1080p, and for the most
part it wasn’t too bad at max settings, though even at ultra the 1% low was above
the refresh rate of the display.
CS:GO Testing:
CS:GO was tested using the Ulletical FPS benchmark, and as a game that is pretty CPU bound we’re seeing above average results compared to most i7-9750H laptops that I’ve tested.
Rainbow Six Siege:
Rainbow Six Siege was tested with the built-in benchmark, and the results weren’t too
different in this test compared to similarly specced machines.
Don’t get me wrong, still excellent performance, but it doesn’t seem that the extra CPU cores and slight clock speed difference helps here compared to an i7.
Assassin’s Creed Odyssey:
Assassin’s Creed Odyssey was tested with the built-in benchmark, and from my experience seems to be a fairly CPU heavy test.
There’s not much of a boost over most other laptops I’ve tested though, still fairly similar average FPS, however I was seeing higher 1% lows due to the CPU compared to say the Y740 with same GPU but i7 CPU.
Dota 2:
Dota 2 was tested playing in the middle lane with an average amount of action going on, and is yet another game that I’ve found to prefer higher CPU power.
Which explains why the average frame rates at lower settings are higher than the 9750H machines I usually test, granted not by enough to make a practical difference as we’re already around 200 FPS.
Watch Dogs 2:
Watch Dogs 2 is a resource-heavy game, but in my opinion, is still playable with a stable 30 FPS.
So even at ultra settings with our higher than 60 FPS average, it’s playing very well without any issues, and I don’t often see 100 FPS being reached at all in this title when it comes to laptops, so that seems to be due to the i9.
The Witcher 3:
The Witcher 3 was playing well with hair works disabled, I’ve found this to be a somewhat more GPU demanding title.
For quick comparison, the average frame rates are close to the Y740 I tested with 2070 Max-Q, however the 1% lows with the i9 are 10 to 20 FPS higher than the i7.
If you’re after more gaming benchmarks check the card in the top right corner where I’ve tested 20 games in total.
Let’s also take a look at how this config of the new Aero 17 compares with other laptops to see how it stacks up, use these results as a rough guide only as they were tested at different times with different drivers.
Battlefield 5 Tested:
In Battlefield 5 I’ve got the Aero 17 highlighted in red near similarly specced machines, including the new Aero 15 OLED just below it. In terms of average FPS.
The result is right in line with the Aero 15 and Lenovo Y740, both of which are machines with the same GPU but i7 CPU.
The average from the Razer Blade Pro 17 with the same GPU is significantly higher, as that machine undervolted the CPU, overclocks the graphics, and boosts the GPU power limit higher.
I’m only testing these machines at their best stock settings, so with some tweaks, we could also improve the Aero 17.
Which we’ll see soon, however even at stock due to the superior i9 CPU with higher clock speeds and more cores we’re getting the highest 1% low result out of all machines tested.
This 1% result is ahead of the average FPS in the Scar II with RTX 2060, just for comparison.
Far Cry 5 Tested:
Here are the results from Far Cry 5 with ultra settings in the built-in benchmark.
I’ve found this to be a pretty CPU heavy test, though unlike the last game we’re not seeing results too different from the other similarly specced machines.
The average FPS was again in line with most of the other 2070 Max-Q machines tested, though the 1% low was a little ahead of the others, again most likely due to the higher CPU performance, the difference is just much smaller in this title.
Shadow of the Tomb raider Tested:
These are the results from Shadow of the Tomb Raider with the built-in benchmark at the highest settings.
Once more the average frame rates are very similar to the other 2070 Max-Q machines
tested, again except for the Blade Pro as that’s just tuned very well out of the box, still though the Aero 17 holds up nicely here.
Overall the new Gigabyte Aero 17 laptop is providing high levels of performance, though realistically not too different from an i7 laptop.
So if you’re purely using it for gaming you could save money by looking at a non-HDR version with 1080p 144Hz screen and i7 CPU.
Now let’s see how much we can improve performance if we make some changes, undervolting seems required in most cases if you want a shot at overclocking to higher CPU clock speeds.
In battlefield 5 at maximum settings, we’re seeing a 6% improvement to both average FPS and 1% low performance with these changes in place.
In Shadow of the Tomb Raider with the built-in benchmark at highest settings, there was
a larger difference, almost 9% higher average frame rate was happening with these tweaks.
In Far Cry 5 with the built-in benchmark at ultra settings, there was almost a 7% improvement to average FPS with the changes in place, though there was less of a difference when it came to 1% low performance.
Storage Testing of Gigabyte Aero 17:
I’ve used Crystal Disk Mark to test the storage, and the 512gb NVMe M.2 SSD was performing well enough, though a fair bit lower on the writes compared to reads.
While the UHS-II SD card slot was performing very well with my V90 card for the reads, but again much lower for the writes.
Let’s conclude by covering the good and bad aspects of the new Gigabyte Aero 17 laptop.
The Aero 17 laptop is a nice addition to the Aero lineup for those that have been after
a larger version. For the most part, you’re getting the same machine, but bigger.
There’s no option of OLED here as there aren’t any panel manufacturers offering 17-inch versions yet.
However, the 4K HDR panel in my unit was quite good, though there are also 1080p 144Hz options which may be preferable for gamers.
The CPU performance I was seeing from the i9-9980HK was very impressive for a laptop, and while it did get warm, as expected, we could make quite a few improvements to both performance and thermals.
In the end I’m quite happy with the results, being able to average 4.25GHz over all 8 cores while under combined CPU and GPU stress test is a great result.
The Aero 17 is a great option for content creators. As mentioned the raw power is there, and the 4K screen looks great with good color gamut.
The highest brightness I’ve ever measured on a laptop, and as a result DisplayHDR 400.
Combined with the great selection of I/O such as the UHS-II SD card slot and Thunderbolt 3 and I think the Aero 17 is in a great position for creators compared to most others I’ve tested so far.
I’d still prefer the Aero 15, but that’s simply because I don’t need a larger screen and I’m happy with a more portable machine.
If I was after a 17-inch laptop purely for content creation though, this is the best I’ve tested so far in that regard.
I didn’t like that the secondary key functions weren’t lit, and I know some people won’t be a fan of the camera placement, other than that I can’t think of much else I didn’t like.
Let me know what you thought about the new Aero 17 laptop down in the comments.
- Get link
- Other Apps
Comments
Post a Comment