Budget 1440p 160Hz Ultrawide – XG34UWQ Gaming Monitor Review
I really like gaming on ultrawide monitors – something about the aspect ratio filling my peripheral vision really aids in immersion regardless of the genre of game. Sadly, most ultrawides are either heinously expensive or painfully bad, but can this UK based X= brand best the competition and offer both one of the cheaper 1440p high refresh rate 34” ultrawides on the market, and one of the best? Let’s test it and find out.
First, the specs. This is the XG34UWQ, a 34” 3440×1440 160Hz Curved VA panel with adaptive sync, backlight strobing and 3 overdrive modes not including off. It quotes an 8ms GtG response time, 400 nits of peak brightness and a 4000:1 contrast ratio. I/O wise it sports two DisplayPort 1.4 ports, one HDMI 2.0, DC in, audio out and a USB port for updating the firmware.
Physically, it’s definitely on the cheaper side. While the monitor does attach to the stand tool-free, the mounting mechanism has an unsettling amount of play, meaning whenever you try to adjust the panel’s tilt angle, the lower edge of the mount pulls away from the monitor itself. To be clear, the mount is secure, I have no worries about it falling off the stand at all but it speaks to the build quality. In fact, adjusting the monitor at all tells you a whole lot about the sacrifices you make in buying one of the cheapest options on the market, you can see and hopefully even hear how much play the stand has itself, and see how much the entire monitor flexes when I’m pretty gently pushing on the stand. I’m not putting enough force in to actually change the position of the stand, and yet the whole panel is flexing and even lifting off the table on one corner.
When upright, the base can also cause instability, both from it’s rather shallow design meaning you have to hold it down to be able to tilt or swivel the monitor, and because it is almost perfectly flat across quite a long surface, meaning any bowing or imperfections in your desk surface can make it rest on a singular point making it unstable and prone to wobbling.
When it comes to the OSD controls, those are the old-school, downward facing, individual buttons that are labelled… On the bottom face. I would have much preferred a joystick style switch most other manufacturers have adopted on the back, which ends up being much more ergonomic and easy to use. The menu itself is also pretty retro, with three black and white columns for each layer of settings. The main options you’ll frequent here are “Picture Quality Setting” and “Other”, with the former holding the “Response Time” setting, aka overdrive. Here you have 4 total options, off, low, middle and high. Down in the “Other” option you’ll find the Adaptive Sync control, HDR mode, an option labelled “MPRT” which is just backlight strobing – basically just turning the display off for all but 1ms per frame to hypothetically improve the motion performance although in practice doesn’t do anything to improve the pixel transition time.
It also has an option to enable or disable “flicker free”, which has confused the hell out of me. Flicker Free as a branding normally refers to the monitor using DC backlight control – changing the voltage going to the LEDs – rather than a PWM (pulse width modulated) backlight controller, as the latter is basically just turning the backlight off hundreds or thousands of times per second which can cause “flicker”. Having the option to enable or disable “flicker free” seems impossible as DC control and PWM are two completely different hardware designs, so the only way that would make sense is if it is still using a PWM backlight but is using a very high frequency (say 10kHz or 10,000 times per second) and disabling the setting changes the frequency to a much lower value (which itself would require more advanced hardware than is normally included) – but even then it still doesn’t actually make sense.
Further to that, testing with my open source response time tool, I found functionally no difference between the setting enabled or disabled – in fact setting it to “off” actually reduced the noise ever-so-slightly. Like I said, I’m very, very confused.
Since we are talking about the panel and it’s pixel response times… That 8ms GtG is being very, very optimistic. Even on its maximum overdrive setting, the best initial response time I got averages out to 9.5ms, and the perceived response time – the time including any overshoot – is more like 13ms. With overdrive off it’s more like 21ms for both, which is just.. Awful.
That is represented in the UFO test where you get 5 to 7 ghosted frames on screen at the start of a new frame draw, and even to the eye is very, very visible. Even with overdrive on high you still end up with 4 to 5 ghosted frames on screen at once which is pretty lacklustre – albeit typical for vertical alignment (VA) panels. You might be wondering just how a 21ms average response time equates to up to 7 frames on screen at any one time, I mean at 160Hz each new frame is drawn every 6.25ms, which would imply more like 3 frames at a time… Well, this is why.
This is the light level captured during the RGB 0 to RGB 51 transition. The X axis is time, 150ms all in, and the Y axis is light level. For some context I’ll add in a transition from my most recent monitor review, the BenQ EX2710Q – it won’t be perfectly lined up but you should get the idea. The orange BenQ line completes its transition in around 3ms, but our X= here? Even if you are charitable and subtract the antiquated 10% of the light level tolerance, it still takes just shy of 50ms, and if you are strict about how long it actually, fully takes, it’s 84ms. 84ms!!! Even the clipped 50ms time is painful, and nicely lines up with what you experience on screen.
Normally that sort of issue is meant to be solved by overdrive, and sure with overdrive on high the complete response time is down to more like 65ms and the 10% time is more like 36ms, but that’s still really, really bad. From looking at hundreds of these graphs just for this monitor, I’ve been drawn to the conclusion that the overdrive profiles are not tuned or validated, possibly at all. The darker transitions have literally zero overshoot, meaning the pixels aren’t being driven all that hard, but middling transitions like this RGB 0 to RGB 153 have horrific peaks to 40% higher than the end light level. Much like a car’s fuel or boost map, monitor scalers hold a similar table that it looks through to find how hard to overdrive any given pixel and for how long to set it to that overdriven state. This monitor doesn’t appear to have had that table actively calibrated and fine tuned to offer the best experience for you the user.
This ends up being pretty visible to the eye, as even in the UFO test the undershooting of the bright red to the middling green makes the trail rather noticeable. In games, even fast paced games, with overdrive set to its maximum setting, I was mostly able to ignore the slow panel and focus on clicking on heads. I mean bot’s heads, but that’s about the limit of my “talent” so I’ll take that. While any sort of fast motion made it rather difficult to be able to make out which of the 4 to 8 copies of the enemy I was meant to be aiming at at any given moment, for anything where motion blur wouldn’t be a genuine hindrance I could see some less ghosting-sensitive players being mostly fine with this. With overdrive off it was pretty unplayable, with a jelly-like, almost tube-train-turning feel to it – verging on nauseating for me personally. Definitely keep this on the high response time setting.
Input lag seemed a little hit-or-miss. My time sleuth reported around 5ms which isn’t exactly perfect – many other monitors offer 1ms or lower – but I’d take this result with a pinch of salt as the time sleuth caps at 1080p60 thanks to the FPGA that runs everything being underpowered for anything higher, which means the monitor’s scaler is having to take the 16:9 1080p signal and upscale it to 3440×1440 which would add some delay. Luckily I also test the total system or click to photon latency using NVIDIA’s LDAT tool, which also had some mixed results. Multiple sets ran in the 30-40ms range which is pretty poor – and spikes up to the mid 60’s – but some runs averaged much more respectably at around 22ms. This was with adaptive sync enabled while playing in CSGO, so again could be something to address in a firmware update.
When it comes to colours and brightness, the good news is the 400 nits of peak brightness is accurate, as I saw anywhere between 415 and 425 nits in my testing. Strangely though, the gamma curve seems off. Like, really off. For comparison’s sake, here it is up against that BenQ panel I compared to previously. The BenQ panel is much flatter down in the dark areas, which if I add in the actual Gamma 2.2 curve you can see is exactly what you want. The offset here is because these monitors can’t get to full black as the backlight is still on, it’s just the blinds are closed but some light will still leak through the cracks. The X= panel outputs significantly more light at say RGB 50 than the BenQ, which is something I noticed while using the monitor during my testing. What should be almost black areas still stand out as a dark grey, much more so than you’d see on most other monitors.
Colours wise, you get around about 100% coverage of the sRGB spectrum – it’s not too bad, but competitors like the AOC CU34G2X which is also a curved VA panel offer a little wider range. Not too significantly, but even to the eyes this panel does look a little more muted and dull compared to the usual IPS panels I both use day-to-day and review. Oh, and the quoted 4000:1 contrast ratio appears to be more like 2800:1 – not horrific for sure but not quite meeting the claim.
Now you could argue I’ve been somewhat hard on this monitor, considering it’s meant to be a budget option. If it costs like £250 or £300 then I could pretty happily forgive almost all of my grievances with it, that would be a pretty great deal even if the monitor itself is far from perfect. So, how much is it? Well it’s currently listed for £400. That’s only £40 less than the Gigabyte G34WQC, an equally curved 1440p high refresh rate VA panel based monitor that is likely to have much better build quality, overdrive tuning and gamma curves. Or another tenner nets you the AOC CU34G2X, a very similar panel again with better build quality, overdrive tuning and gamma curves, or get a flat IPS version from iiyama and their GB3461WQSU which is a legitimately impressive option I’ve reviewed in the past.
On the whole, I’m not sure I can recommend this monitor, at least for now. Most of the problems I’ve spoken about are one solid firmware update away from being either solved completely or at least dramatically improved. If that were to happen, if they can invest some time into tweaking the overdrive look up tables, to correcting the gamma curve, and ideally knocking a touch more off the price – I think £380 would be perfect – I could happily excuse the build quality and still relatively slow response time performance.
-
TechteamGB Score
