BenQ XL2540 vertical pixel inversion: A panel lottery?

[PedanticFool]

I recently had the opportunity to purchase a BenQ ZOWIE XL2540 (monitor #1) and compare it to another XL2540 owned by a friend (monitor #2).

My monitor (monitor #1) has significantly worse vertical pixel inversion lines.

Testing was done using the following link after resetting both monitors to factory defaults:

https://www.testufo.com/ghosting#backgr ... &pursuit=1

We captured a video which demonstrates the vertical inversion artifacts on both monitors:

https://drive.google.com/file/d/1YUEqHL ... sp=sharing

The first monitor seen in the video is monitor #2. It has very minor vertical pixel inversion lines.

The second monitor seen is mine, monitor #1. Vertical pixel inversion lines are much more visible.

I captured stills from the video which illustrate the issue very well:

Edit: Reuploaded captures to Google Drive as the screenshotcomaparison link went down.

Monitor #1: https://drive.google.com/file/d/1dyre36 ... sp=sharing
Monitor #2: https://drive.google.com/file/d/1FmIn2W ... sp=sharing

Testing was performed using two DisplayPort outputs, two DisplayPort cables and two power cables. The issue followed monitor #1 every time regardless of cable and port configuration.

I've only included footage of a single test but this effect is visible during other tests as well as gaming, watching videos etc. Note it's not always visible, only during certain types of motion on certain colors. But it's always worse on monitor #1.

Using a lower refresh rate will amplify the vertical pixel inversion but for this testing I ensured both monitors were set to 240hz. Display was set to duplicate in Windows 10.

I'll be contacting BenQ and asking them about an RMA and the chances of receiving another unit with significant vertical pixel inversion lines. I'll keep the thread updated.

What does everyone think? Pixel lottery/expected variance, or is there a larger issue with monitor #1?

[PedanticFool]

I've received communication from the BenQ Australia service team.

I initially sent them the information in the OP along with the these questions:

What exactly causes this effect?

Is the significantly worse vertical pixel inversion on monitor #1 expected? I.e. do you consider this an acceptable standard variance between two panels?

Would you be willing to perform an RMA and swap the monitor?

What are the chances of receiving another monitor which is just as bad?

They responded with the following:

The vertical pixel inversion issue may cause by voltage control (prevent burn in problem) on higher refresh rate monitor models and there are having complex parts to cause the issue.

There are a lot of thread discussing on the forum about the issue on all high end models, not only BenQ, but also other brands.

Normally, it may happen on different panel types, refresh rate setting, g-sync activates or monitor turn on time/temperature.

I would like to suggest the user to reset the monitor to default setting first and burn the monitor for over 30 minutes that may solve the problem or improve the performance.

But no guarantee to solve the issue 100% perfectly and swap another monitor may find the issue as well.

A rather disappointing generic reply, they touched on a few of my questions but in little detail. They asked me to reset to factory defaults and test even though I'd already stated I'd done that. They said some competitor products can have similar issues but nothing with regards to why these two XL2540s are so different. They did not provide a clear answer to my queries asking if this is expected variance and if a replacement is likely to be just as bad. I proceeded to respond with the following:

Thanks very much for the quick response. As mentioned in my initial email all testing was already after resetting to factory defaults. The testing procedure took multiple hours too, it certainly had 30 minutes+ to burn in during that time.

I do agree there is plenty of discussion about this issue affecting other monitors which use the same AU Optronics panel. However as you can clearly see from the video and pictures there is a huge difference between just these two BenQ monitors which are the same model. I believe you are implying this difference is expected variance between different XL2540 monitors but I would appreciate if you could clarify there.

So what are the options now? Can we try a swap in hopes I get a panel with less noticeable vertical pixel inversion? Would you perhaps be able to open and check the replacement to ensure the vertical pixel inversion is minimal before sending it off to me?

I quickly received this reply:

It’s best to take the unit back to place of purchase for replacement/refund

Even more disappointing. They completely dodged the question about panel variance and simply directed me to the reseller without providing any further information at all. I am obviously already aware that I can seek a refund or replacement through the reseller but I'm hoping to keep that as a last resort. The reseller won't be able to provide the same level of guarantees and checks as BenQ potentially could before providing a replacement, the reseller will also be unable to answer questions regarding acceptable panel variance and other such technical queries. I also fear the reseller will try to force me to pay return shipping.

I plan to call the BenQ Australia warranty team tomorrow and attempt to get this escalated. I hadn't dealt with them until now but was hoping they might be willing to go above and beyond to help me get a replacement panel which more closely resembles monitor #2. Realistically I was expecting average service, that they would answer my questions and offer to ship a replacement but without any guarantees that it would be an improvement. Instead they've avoided my questions twice and responded with a one line reply which simply directs me to the reseller, I consider that quite poor.

I'll continue to update the post as I explore this further with BenQ and the reseller. I really hope I'm able to get it resolved with BenQ directly.

[PedanticFool]

I escalated and spoke to a BenQ Australia product manager today and am now feeling much more comfortable about this situation.

The product manager advised that the service technician I exchanged emails with was operating from a standard script and apologized that I wasn't satisfied with the communication.

He stated that due to the mass production process used there will always be slight variances between each panel as we expect. Based on the photos and videos provided though he was unable to confirm if this is expected panel variance or if monitor #1 is defective. He offered for me to bring my monitor in to the BenQ office so he could see it and we could sort it out immediately. Unfortunately I'm in another city so this isn't an option or it would have been perfect.

We discussed how panels can look different to different people, an artifact or effect obvious to one person may be barely noticeable to someone else. Based on this he advised that he would be unable to guarantee 100% that I would be satisfied with the quality of a replacement panel should he send one. I think this is reasonable but enquired further asking if he could do some testing.

After more discussion he generously agreed to open and inspect a number of these XL2540 monitors in their head office. They'll compare both brand new and demo stock to see how much the vertical pixel inversion lines vary and report back to me. They'll also send me a testufo clip to demonstrate this effect on the monitors they test. I'm quite happy with this offer, it goes beyond what I expected. I hope they'll find a monitor closer to #2 (with very little vertical pixel inversion lines) and offer me that as a replacement.

I expect to hear more from BenQ soon, I'll keep this thread updated.

[RealNC]

To tell you the truth, I would have told you "take it or leave it."

[Chief Blur Buster]

Interestingly, I recently posted about inversion artifacts in other threads, most recently this thread.


(From strobed TN, odd pixel step. They can be horizontal lines, vertical lines, or checkerboards)

Inversion (the art of balancing the voltage on an LCD panel to prevent burn in too) -- creating inversion artifacts usually as fine-patterned lines/checkerboards -- is much more visible on TN panels than IPS panels. I can induce a very minor amount on faster IPS, but I always cannot avoid at least a faint visibility of inversion effects on any TN panels I've ever used.

The cloudy-crystal-ball issue of inversion artifact badness / intensity necessary to trigger an RMA:

It is normal but the threshold of intensity is the problem: How bad does inversion get before it becomes RMA? That can become a very good question, since there are no published standardization AFAIK for the minimum-acceptable-quality inversion for a TN LCD panel.

Dead-pixel policies are much simpler than inversion-artifact policies. But maybe it's time for the industry to create an inversion-artifacts-intensity benchmark. Blur Busters in the future, might begin looking into creating a reliable benchmark test so that the worst 10% of panels can be much more easily flagged by bloggers/reviewers.

I think I may have mentally invented a way to make it much easier to assign a benchmark number to inversion artifacts. (Most of my TestUFO tests come from mental simulations & inventions -- I'm able to accurately predict the appearance of a yet-unmade TestUFO test in my head! Even the Eye Tracking test and Pesistence test, as well as the MPRT-versus-GtG test.). I'll queue a brand new TestUFO inversion-intensity measurement benchmark sometime in the future, perhaps by late 2018 for common inversion patterns. Priorities first, though.

Temperatures makes a big difference in inversion artifacts with many panels. BenQ/ZOWIE is correct about temperature. Warm up a panel for 30 minutes.

Meanwhile, in increasing order of known inversion artifact badness:
-- Default Hz, IPS (rarely seen)
-- Default Hz, TN (more invisible)
-- Low Hz TN (worse at 30Hz during variable refresh rate, can look like a faint flicker)
-- Strobed TN (especially at odd-pixel-step motion, worst inversion artifacts, even at high Hz)

There are definitely panel-to-panel variances.
And there are definitely cases where badness (e.g. the worst TN panels) can trigger an RMA.
But the problem is there's currently no easy user-friendly standard of measuring badness of inversion artifacts.

For those who really, really hate inversion artifacts.... Get an IPS panel. They have more motion blur and strobe crosstalk but it's all a matter of end-user preference (input lag, color quality, motion blur quality, strobing quality, etc). For me, I don't mind inversion artifacts when they are faint enough, since motion clarity often greatly outweighs this.