Everything about displays and monitors. 120Hz, 144Hz, 240Hz, 4K, 1440p, input lag, display shopping, monitor purchase decisions, compare, versus, debate, and more. Questions? Just ask!
-
Discorz
- VIP Member
- Posts: 999
- Joined: 06 Sep 2019, 02:39
- Location: Europe, Croatia
-
Contact:
Post
by Discorz » 28 Oct 2021, 15:44
GIGABYTE M32Q Review
Table of Contents
- Specifications
- Responsiveness
- Sample-and-Hold Performance
- TestUFO - Overdrive 'Balance', 60-170 FPS
- TestUFO - Overdrive 'Picture Quality', 60-170 FPS
- TestUFO - Overdrive 'Smart', 60-170 FPS
- TestUFO - Overdrive 'Speed', 60-170 FPS
- Overdrive 'OFF'
- Response Times
- Blur Reduction - Aim Stabilizer
- Hints
- TestUFO - Middle Screen, Overdrive 'Balance', 100-170Hz
- TestUFO - Middle Screen, Overdrive 'Picture Quality', 100-170Hz
- TestUFO - Strobe Crosstalk, Overdrive 'Picture Quality', 100-170Hz
- Duty Cycle
- Animation - Aim Stabilizer Waveform Estimate 170-100Hz
- Red KSF Phosphor
- VRR Blur Reduction - Aim Stabilizer Sync
- Hints
- Smooth Frog - Middle Screen, Overdrive 'Picture Quality', 170 FPS
- Smooth Frog - Middle Screen, Overdrive 'Picture Quality', 144 FPS
- Smooth Frog - Middle Screen, Overdrive 'Picture Quality', 120 FPS
- Smooth Frog - Middle Screen, Overdrive 'Picture Quality', 100 FPS
- Smooth Frog - Middle Screen, Overdrive 'Picture Quality', 85 FPS
- Smooth Frog - Middle Screen, Overdrive 'Picture Quality', 60 FPS
- Smooth Frog - Strobe Crosstalk, Overdrive 'Picture Quality', 85-170 FPS
- Animation - Single Pixel Line Panning Across the Screen, Strobe Crosstalk
- Duty Cycle
- Animation - Aim Stabilizer Sync Waveform Estimate, @170Hz 170-60 FPS
- Animation - Aim Stabilizer Sync Waveform Estimate, @120Hz 170-60 FPS
- Animation - Aim Stabilizer Sync Waveform Estimate, @85Hz 170-60 FPS
- Animation - Aim Stabilizer/Sync at Fixed Refresh Rates, @170, 120, 75Hz
- Input Lag
- Other
- Uniformity
- Colors
- Boot Times
- Factory menu
This post will be edited!
Last edited by
Discorz on 02 Apr 2022, 14:50, edited 3 times in total.
-
Discorz
- VIP Member
- Posts: 999
- Joined: 06 Sep 2019, 02:39
- Location: Europe, Croatia
-
Contact:
Post
by Discorz » 28 Oct 2021, 15:46
1. Specifications
- Size: 31.5"
- Resolution: 2560x1440 (16:9)
- Pixel Density: 93 ppi (0.273 mm pitch)
- Refresh Rate: 165 Hz (170 Hz OC)
- Panel Type: IPS (Innolux M315KCA-E7B)
- Bit Depth: 8+FRC
- Gamut: P3
- Adaptive Sync: AMD FreeSync Premium (G-SYNC Compatible also works)
- Blur Reduction: Aim Stabilizer, Aim Stabilizer Sync
- Inputs: HDMI 2.0 x2, Display port 1.2 x1 (capable for HDR), USB Type-C x1
- Price: $360-500 US (October 2021)
Last edited by
Discorz on 28 Oct 2021, 15:56, edited 2 times in total.
-
Discorz
- VIP Member
- Posts: 999
- Joined: 06 Sep 2019, 02:39
- Location: Europe, Croatia
-
Contact:
Post
by Discorz » 28 Oct 2021, 15:47
2. Responsiveness
- Sample-and-hold Performance
- Default VRR Range: 48-170Hz
- Overdrive: 5 levels (OFF, Smart OD, Picture Quality, Balance, Speed)
Pursuits bellow are taken with FreeSync enabled. Fixed, VRR ON, VRR OFF refresh rates have the same overdrive behaviour! I used Custom Resolution Utility (CRU) for custom resolutions, automatic (PC) timing preset.
View images in full resolution for detail! (3345 x 2250)
Overdrive Picture Quality
TestUFO at ∼960pps, Picture Quality
HUB 3-97% gamma corrected
Heatmaps by HUB: 170Hz, 144Hz, 120Hz, 100Hz, 85Hz, 60Hz
Heatmaps by RTINGS: 170Hz, 120Hz, 60Hz
'Picture Quality' is set and forget option for full 48-170 fps range with no overshoot. Compared to 'Balance' it only fits 13% of transitions within refresh window. But there is no much ghosting. 'Picture Quality' and 'Balance' are the only two reasonable overdrives on this monitor, other ones are basically useless. Something in between those two would be ideal.
Overdrive Balance
TestUFO at ∼960pps, Balance
3-97% gamma corrected
Heatmaps by HUB: 170Hz, 144Hz, 120Hz, 100Hz, 85Hz, 60Hz
Heatmaps by RTINGS: 170Hz, 120Hz, 60Hz
Hardware Unboxed measured 16% of overshoot ranked yellow, but for this panel 16% is not bad at all as you can see from pursuit shouts. Overshoot is subtle and it won't be noticed in games. Also it nicely fits most of the transitions (92%) within refresh window. 'Balance' is my recommendation for 120-170 fps gaming. So if u game at low frame rates consider other overdrive.
M32Q uses Low Framerate Compensation (LFC). So one trick you do could here to tame the overshoot at lower refresh rates is modify VRR range from default 48-170 to something like 85-170. With this modification as soon as fps dips bellow 85 the display driver starts doubling the refresh rate and it stays in sync. Meaning internal panel refresh stays high and overshoot gets cut in half. You can do this with Custom Resolution Utility (CRU by ToastyX).
Overdrive Speed
TestUFO at ∼960pps, Speed
HUB 3-97% gamma corrected
Heatmaps by HUB: 170Hz
Heatmaps by RTINGS: 170Hz, 120Hz, 60Hz
Gigabyte decided to include one setting with extreme amount of overshoot just like majority of gaming monitors on the market. If coronas bother you avoid this one.
Overdrive OFF
There is also OFF option which actually doesn't fully disable overdrive. I noticed some very small amounts undershoot. But you have to try really, I mean really hard to notice it. So we don't get to see panel's native performance. For those who are interested how pursuit shots look it's extremely similar to 'Picture Quality'.
HUB 3-97% gamma corrected
Heatmaps by HUB: 170Hz
Heatmaps by RTINGS: 170Hz, 120Hz, 60Hz
Overdrive Smart OD
TestUFO at ∼960pps, Smart OD
HUB 3-97% gamma corrected, Smart OD
Heatmaps by HUB: 170Hz, 144Hz (Balance), 100Hz, 60Hz (Balance)
Heatmaps by RTINGS: 170Hz, 120Hz, 60Hz
Gigabyte seem to be the first to include Smart OD that is supposed to be variable overdrive without using nvidia's overpriced g-sync module. Unfortunately they didn't do it right for M32Q. 'Smart OD' is just choosing between existing overdrive modes. For 170-120 fps range it uses 'Balance', for 120-75 fps 'Speed', 75-50 fps 'Balance', 48-37 'Speed'. This means if fps is limited to 120 you will see overdrive flickering because the display is trying to switch between one light and one aggressive level of overdrive. Obviously this is very distracting and for that reason I don't recommend using this. I will give credit to Gigabyte for implementing such thing. Smart OD has great potential, but it needs to be fixed!
My fix would look something like this: slightly less aggressive Balance for 145-170Hz, same Picture Quality for 48-85Hz and 2 new well tuned overdrives in-between instead of uneffective OFF and Speed modes.
Smart OD fix
Ideally a really good implementation of this type of variable overdrive requires much more levels of overdrive, preferably in 0-100 increments (gain) so that transitions turn out silky smooth.
Response Times
All data you see here is taken from Hardware Unboxed and arranged a bit different.
M32Q response times and overdrive comparison throughout full refresh range, data taken from Hardware Unboxed, method 3-97% RGB gamma corrected
Random monitors for comparison.
Response times and overdrive comparison between multiple monitors, data taken from Hardware Unboxed, method 3-97% RGB gamma corrected
If you'd like to know how these numbers look in real life use my TestUFO Collection for demonstration or comparison.
This next bar graph shows Persistence + Average Response Times + Overshoot.
Best overdrive at max refresh rate and few non-native 60Hz, data taken from Hardware Unboxed, method 3-97% RGB gamma corrected
Last edited by
Discorz on 21 Apr 2022, 10:37, edited 9 times in total.
-
Discorz
- VIP Member
- Posts: 999
- Joined: 06 Sep 2019, 02:39
- Location: Europe, Croatia
-
Contact:
Post
by Discorz » 28 Oct 2021, 15:47
- Aim Stabilizer
- Phase Adjustment: NO
- Pulse Width Adjustment: NO
- Maximum Brightness: ∼100 nits (OSD brightness ∼15) | Brightness Adjustment: NO
- Overdrive Adjustment: 5 levels
- Fixed Refresh Rate Range: 100-170Hz
View images in full resolution for detail! (3345 x 2250)
Overdrive Balance
TestUFO at ∼960pps, Aim Stabilizer at middle screen, overdrive 'Balance'
Overdrive Picture Quality
TestUFO at ∼960pps, Aim Stabilizer at middle screen, overdrive 'Picture Quality'
Strobe Crosstalk
TestUFO at ∼960pps, Aim Stabilizer fullscreen strobe crosstalk, overdrive 'Picture Quality'
- Duty Cycle:
- Baclight ON for 1.0-1.6 ms, main pulse
- Backlight OFF, KSF phosphor decay
M32Q Aim Stabilizer pulse estimated
I can't be sure if the yellow timing line is correct here.
Notice how primary pulse gets a little wider as fixed refresh rate decreases. In theory this would mean it would get much darker at lower refresh rate but it doesn't, stays perfectly consistent. I'm not sure why is that.
- Red KSF Phosphor
So how does the slow KSF phosphor look in games? Well it kinda looks like faint red overshoot, not very pleasant, but it's not too bad either. It's just unnecessary. Sometimes I can notice it when I blink or look around the screen or when powering monitor on/off. Here is high speed camera footage of strobed KSF display at "960" fps. I think it should be less distracting on higher refresh rates like 240Hz+ since phosphor decay is shorter due to faster update. Note that slow phosphor is only visible with blur reduction enabled (Aim Stabilizer/Sync). Because of this whole image will have a reddish tint which can be partially removed with few tweaks.
Slow red phosphor with Aim Stabilizer/Sync at @170Hz 170 fps, middle screen, Call of Duty 2 in Smooth Frog, pan right-left at ∼1920pps
Interestingly Aim Stabilizer turns on instantly, I mean no black screens whatsoever, fast as a toggle. But enabling Aim Stabilizer Sync (FreeSync enabled) takes about 2 sec. To make both toggle fast use Custom picture profiles instead. One profile with enabled and other with disabled Aim Stabilizer Sync. Add picture profiles to joystick shortcuts and switch between from there. Or even better use Gigabyte's OSD Side Kick with customizable keyboard skortcuts.
Last edited by
Discorz on 06 Apr 2022, 05:52, edited 7 times in total.
-
Discorz
- VIP Member
- Posts: 999
- Joined: 06 Sep 2019, 02:39
- Location: Europe, Croatia
-
Contact:
Post
by Discorz » 28 Oct 2021, 15:47
- Aim Stabilizer Sync
Hardware Unboxed comment on Aim Stabilizer Sync in their M32Q review got me real excited. Although they said it's not perfect, it was still the main reason why I bought it.
Quote Hardware Unboxed:
"Both Aim Stabilizer and Aim Stabilizer Sync appear to deliver similar results. This is the first monitor that appear to use the same strobing technique at a fixed or variable refresh rate...
It does work with adaptive sync enabled and does strobe appropriate to dynamic refresh rate...
However it's not without it's flaws. There is still some strobe crosstalk at pretty much all refresh rates leading to a faint double image and small amount of blur...
It's not a perfect implementation..."
They also warned about red KSF phosphor.
- Phase Adjustment: NO
- Pulse Width Adjustment: NO
- Maximum Brightness: ∼100 nits (OSD brightness ∼15) | Brightness Adjustment: NO
- Overdrive Adjustment: 5 levels
- Variable Refresh Rate Range: 75-170Hz
- Fixed Refresh Rate Range: 75-170Hz (60Hz in bug mode)
Since TestUFO doesn't work with adaptive sync we can't use it for VRR testing. We need an actual game or better something like Smooth Frog for pursuit shots.
View images in full resolution for detail! (3345 x 2250)
170 FPS
Smooth Frog at ∼1440pps, Aim Stabilizer Sync middle screen, @170Hz 170Hz/170FPS, overdrive Picture Quality
144 FPS
Smooth Frog at ∼1440pps, middle screen, @170Hz 144Hz/144FPS, overdrive Picture Quality
120 FPS
Smooth Frog at ∼1440pps, Aim Stabilizer Sync middle screen, @170Hz 120Hz/120FPS, overdrive Picture Quality
100 FPS
Smooth Frog at ∼1440pps, Aim Stabilizer Sync middle screen, @170Hz 100Hz/100FPS, overdrive Picture Quality
85 FPS
Smooth Frog at ∼1440pps, Aim Stabilizer Sync middle screen, @170Hz 85Hz/85FPS, overdrive Picture Quality
60 FPS
Smooth Frog at ∼1440pps, Aim Stabilizer Sync middle screen, @170Hz 60Hz/60FPS, overdrive Picture Quality
Below minimum range of 75 fps Gigabyte decided to disable strobing. It goes back to VRR sample-and-hold mode while keeping the brightness consistent. This toggle shift can be noticed in games. I assume they did this to avoid flickering at such low framerates.
Strobe Crosstalk
Smooth Frog at ∼1440pps, Aim Stabilizer Sync fullscreen strobe crosstalk, @170Hz 85-170Hz/FPS, overdrive 'Picture Quality', 6690 x 4500
By inspecting pursuit shots of single pixel white line panning across the screen we can see what is Gigabyte really doing here.
Frame by Frame
M32Q Aim Stabilizer Sync fullscreen strobe crosstalk, overdrive OFF
single pixel line panning across the screen at ∼1440pps (right-left), RGB 63-255-63,
And with a little help of our monitor specialist Ashun we managed to estimate what's the pulse logic behind Gigabyte's Aim Stabilizer Sync.
Quote Ashun (from Aperture Grille):
"Gigabyte is doing something very similar to ASUS’s first attempt with ELMB-Sync like on the VG27AQ, but they’re doing it backwards. They begin with a very high frequency PWM backlight fill that expands/contracts based on the current framerate, and then they add in the “primary” or “real” backlight pulse, which has the same duration for any framerate. The PWM fill is just there to keep the brightness consistent from frame to frame. Otherwise the screen brightness would be constantly changing. This strategy might work OK-ish for really high framerates, where you might not notice the PWM fill, but at lower framerates, it becomes very visually distracting and unacceptable."
What this means is Aim Stabilizer Sync is a single strobe, but when its combined with high frequency PWM backlight fill it looks just as bad as double strobe. And on top of all this you get the annoying red KSF phosphor.
- Duty Cycle for 170Hz:
- Variable high frequency PWM backlight fill (0-6.6 ms)
- Backlight OFF for 1.7 ms, KSF phosphor decay
- Baclight ON for 1.0 ms, main pulse
- Backlight OFF for 3.2, KSF phosphor decay
@170Hz
M32Q estimated Aim Stabilizer Sync pulse @170Hz
@120Hz
M32Q estimated Aim Stabilizer Sync pulse @120Hz
@85Hz
M32Q estimated Aim Stabilizer Sync pulse @85Hz
Quote Ashun (from Aperture Grille):
"But a good question is: why is the red fringing in front of the pulse? Think of the KSF phosphor as a secondary backlight that stays on and keeps emitting even after the backlight strobes are finished. So at the beginning of where the blue LCD response starts rising, the glowing red KSF phosphor begins to show. That’s why it leads the main image!"
Single pixel line panning across the screen (right-left) at ∼1440pps, @170Hz 120Hz/120FPS, Aim Stabilizer Sync middle screen, overdrive OFF
M32Q can do single strobe @75Hz minimum, but this is achievable only with fixed Aim Stabilizer Sync refresh rates. Basic Aim Stabilizer will do down to only 100Hz. If you use this and your fps drops bellow 75 Aim Stabilizer will disable it self permanently and stick since it is already bellow minimum (75 fps) value, this seems to be a bug. Current (fixed) refresh rates you set your monitor to will always be single strobed.
This leads us back to Hardware Unboxed confirming it strobes appropriate to dynamic refresh rate. They must of tested this with TestUFO which can't work with adaptive-sync, only with fixed refresh rates.
Single Pixel Line at ∼1440pps, Aim Stabilizer/Sync Middle Screen, Overdrive OFF
It can do 60Hz single strobe in bug mode.
Notice how primary pulse gets a little wider as fixed refresh rate decreases. I measured around 1.0ms @170Hz, 1.2ms @144Hz, 1.3ms @120Hz, 1.5ms @ 100Hz, 1.6ms @85Hz and 1.7ms @75Hz.
Last edited by
Discorz on 02 Apr 2022, 13:41, edited 4 times in total.
-
Discorz
- VIP Member
- Posts: 999
- Joined: 06 Sep 2019, 02:39
- Location: Europe, Croatia
-
Contact:
Post
by Discorz » 28 Oct 2021, 15:48
- Input Lag
Since I don't have any devices to measure input lag I've used humanbenchmark reaction time test to see what numbers say. I got surprised by how good input lag is. Very close to VG279QM, they are in the same category. These are an average number of many many tests, around 50-60 tests per result, so it should be accurate enough. Here is how that looks. For me ∼200 ms is bad, ∼150 ms is best it can get.
Last edited by
Discorz on 28 Oct 2021, 15:59, edited 1 time in total.
-
Discorz
- VIP Member
- Posts: 999
- Joined: 06 Sep 2019, 02:39
- Location: Europe, Croatia
-
Contact:
-
Verneclover
- Posts: 3
- Joined: 14 Nov 2021, 18:57
Post
by Verneclover » 14 Nov 2021, 19:08
Hello, nice review!
Did you notice some problems with inversion artifacts? I am considering this monitor to buy, but I have worries about inverse artifacts problems, because I have seen post of M32Q owner on Reddit about inverse artifacts problem.
-
Discorz
- VIP Member
- Posts: 999
- Joined: 06 Sep 2019, 02:39
- Location: Europe, Croatia
-
Contact:
Post
by Discorz » 16 Nov 2021, 12:14
Hi
From which monitor are you upgrading?
TN panels are much more prone to inversion, but you shouldn't be worried too much about this with IPS. To notice you really need to look for it. For example refresh rate would have to be at anything bellow ∼120Hz and mouse movement needs to be extra slow (1ppf) to start noticing inversion in games.
I tried to capture it with camera but couldn't so I simulated exact amount for you. It actually turned out spot on.
Note that this is worst case scenario at 60Hz, as refresh rate increases it disappeares. At 120 its already gone plus its only visible at parts of the screen, mainly middle.
Watch in full resolution for true 1:1 pixel
- ps 60hz inversion simulation m32q crop.png (864.62 KiB) Viewed 16528 times
-
Verneclover
- Posts: 3
- Joined: 14 Nov 2021, 18:57
Post
by Verneclover » 16 Nov 2021, 20:54
Discorz wrote: ↑16 Nov 2021, 12:14
Hi
From which monitor are you upgrading?
TN panels are much more prone to inversion, but you shouldn't be worried too much about this with IPS. To notice you really need to look for it. For example refresh rate would have to be at anything bellow ∼120Hz and mouse movement needs to be extra slow for you to start noticing inversion in games.
I tried to capture it with camera but couldn't so I simulated exact amount for you. It actually turned out spot on.
Note that this is worst case scenario at 60Hz, as refresh rate increases it disappeares. At 120 its already gone plus its only visible at parts of the screen, mainly middle.
Watch in full resolution for true 1:1 pixel
ps 60hz inversion simulation m32q crop.png
Hi! Thank you for your answer with the excellent illustration. I am updating from AOC q3279vwfd8 (it is also 31.5", 1440p, IPS, but only 75hz) - it doesn't have inverse artifacts. Do these lines appear only during movement or u can see them on static image too? And if u can notice them on static image, do they have dynamic or static nature? I mean , if you watch on static image, could u notice that these lines have some movement/walk ?