Official Acer [XB270HU] 27" 1440p 144Hz G-Sync IPS ULMB Monitor Thread

flod said:
i know. but you can't talk about its effect on "perceived black depth" without first ensuring that the compared monitors have the same actual blacks. in the pictures of the hp monitors it's clear that the actual black in the pictures are way different:



i think ncx is saying that the reason there's a grey splotch on the glossy bezel hp monitor is due to the glossy bezels. even if this is the case (which i highly doubt... more likely a case of accidental additional illumination in that picture from something else) it doesn't showcase the more subtle effect of the bezel's color on perception.
Click to expand...

OK, yeah. Looks like differences in AG rather, if the two sets were each taken at the same time and in the same position, respectively.

It does depict the effect in a way, I think, though it's not scientific. It's clear that having a darker surface surrounding the screen will affect perceived black levels on a given display, even if you can't compare different ones conclusively without some controls.
 
flod said:
i know. but you can't talk about its effect on "perceived black depth" without first ensuring that the compared monitors have the same actual blacks. in the pictures of the hp monitors it's clear that the actual black in the pictures are way different.
Click to expand...

At some point you will have to cease being in denial and think rationally. The monitor's brightness settings and room lighting are identical. The glossy black bezel points upward and downward, and we can clearly see the bright spot caused by the room light on the actual bezel in both pictures. It's simple logic, and a point which can be proven with Microsoft paint, but I'm sure you will continue to be in denial. I also made a triplet compilation with the monitors silver frames exposed.

By using a glossy black bezel, Acer has made it very easy for competitors to create a vastly superior product with identical parts by simply using a dark matte grey bezel. If Asus is smart they will do this (Eizo & NEC certainly would), especially since their products are far more popular.
 
Last edited:
I think I'll wait until the Asus MG279Q gets reviewed.
It's expected to go only up to 120Hz and of course will lack strobing, but it should support FreeSync, multi-input, and be more affordable.

I don't mind switching to a cheap R7 260X if that works with it (it's supposed to), since I don't need much GPU power at all.
My low profile 750 Ti would end up retired into that micro-atx rig i've been itching to build for a while.
 
phorkz said:
Wait, what? i didn't know this!.

So people run their games in windowed fullscreen to get the .icc profile to work?
Click to expand...

Yep. At least I do. Most games reset all the changes made in the gpu driver to the default. There's an instruction in directx called "reset gamma ramp", which resets everything upon entering fullscreen mode, developers have the choice to use it or not, unfortunately most of them do for an unknown reason.
There's some pc games, that already feature borderless mode, or don't use this dx call in fullscreen but these are few and far between.

And you can do nothing about it other than run a dll interceptor app to force windowed mode which potentially can get you banned from online games.

I have no idea why this issue doesn't get any exposure on these forums, It's like people don't tweak their graphic drivers to get decent black/white levels, gamma and white point when most monitors come with badly done factory calibration, let alone those who own actual hardware tools.

I believe this and hdmi mismatched range issue should be brought to attention so nvidia and amd get their shit together and do something about it. :D
 
Since windowed fullscreen breaks g-sync you shouldn't expect to read much about it in threads dedicated to g-sync panels :)

And the issue gets plenty of exposure actually, I think there's still a sticky thread in the nvidia forums (drivers section).

Factory calibration or powerful OSD settings are a way better solution anyway, since software calibration on nvidia cards introduces banding.
 
It just baffles me why can't they at least include some basic sliders for gamma and blacks?
All monitors have white point and white levels controls but not gamma and black levels.

Im not asking for a full blown 3d hardware lut, but these are just the most essential color controls.

Especially when some of these gaming monitors come with 1.8 gamma and it's just a lottery getting one with the right settings.
 
Any reason Gamma setting 2.0 wasn't tested or is 2.2 the lowest setting? Wouldn't 2.0 make for a more accurate 2.2 target than 2.2 showing as ~2.4?
 
MistaSparkul said:
Thanks. Yeah the price bump for including G Sync does suck but the main thing turning me away from the Asus MG279Q is no confirmation of ULMB as I consider it a must have for certain fast paced fps games :( otherwise if Asus manages to include some form of ULMB or whatever they wish to call it then I'm probably going to jump ship for that.
Click to expand...

ULMB is good (I have only tried the eizo foris version of it), but it has some problems, you lose contrast ratio (which is already low on IPS screens) / brightness and also you can't use Gsync / freesync... I think you can use ULMB through hacks, I don't know that much about it but I remember reading some people with 120hz TN screens using ULMB through some driver hack but I might be wrong. Even without ULMB a 5.5ms response time and 120hz will be a lot better than current IPS screens. Not sure how they managed to do that it is impressive as IPS screens have been stuck at about 7.5-10ms response time for 5+ years. Definately makes AHVA the best out of PLS/IPS/AHVA.. All Asus need now is a TW polarizer and it will be good (but they won't do that probably)
 
RPGWiZaRD said:
Any reason Gamma setting 2.0 wasn't tested or is 2.2 the lowest setting? Wouldn't 2.0 make for a more accurate 2.2 target than 2.2 showing as ~2.4?
Click to expand...

there isn't a gamma 2.0 mode on that screen. see the "Gamma and Colour Temperature" section. 1.9 is the lower setting, resulting in a 2.1 average gamma (6% out)
 
Nicholars said:
ULMB is good (I have only tried the eizo foris version of it), but it has some problems, you lose contrast ratio (which is already low on IPS screens) / brightness and also you can't use Gsync / freesync... I think you can use ULMB through hacks, I don't know that much about it but I remember reading some people with 120hz TN screens using ULMB through some driver hack but I might be wrong. Even without ULMB a 5.5ms response time and 120hz will be a lot better than current IPS screens. Not sure how they managed to do that it is impressive as IPS screens have been stuck at about 7.5-10ms response time for 5+ years. Definately makes AHVA the best out of PLS/IPS/AHVA.. All Asus need now is a TW polarizer and it will be good (but they won't do that probably)
Click to expand...

To use ULMB hack you need monitor compatible with nvidia 3D and lightboost.
 
Drags said:
there isn't a gamma 2.0 mode on that screen. see the "Gamma and Colour Temperature" section. 1.9 is the lower setting, resulting in a 2.1 average gamma (6% out)
Click to expand...

Hmm, things like these upset me, there's no logic behind it, 2.4, 2.2 but 1.9... xD Still 2.1 is closer than 2.4 though.
 
I'd rather have the gamma a bit too high than a bit too low though, as long as it doesn't crush blacks that is. IDK how this specific panel behaves with the stock 2.4 gamma but if it doesn't crush blacks then I'd just use that tbh since I do my gaming and watch films in a dark room.
 
using a single number to evaluate a monitor's luminance response is kind of silly
as is using color temperature to evaluate a display's white point and using response time to evaluate pixel transitions.

oh well
 
What I am beginning to think is happening is you are getting smear blur encompassing what would have been the ghost overshoot limit on the acer. Therefore slower response time (smearing blur) = "no overshoot". It may be smearing at least as wide as any ghost would have overshot were it 1ms.

The response times were slightly slower than the Asus ROG Swift PG278Q and BenQ XL2720Z, but the benefit of that was that there was no overshoot on the Acer
Click to expand...

So it may be that your are trading a smear blur width rather than worst case color transition tiny amount of overshoot on normal overdrive.

The swift would likely show a soften (fuzzy) blur more within the object's limits/"shadow mask", with an occasional faint or dark edge (slim) overshoot.
The acer would likely show a smear blur "outside of the lines" of the object, wider than the swift's overshoot, and likely blurring the detail within the object/smear object a lot more, but the question is how much at 5.5ms at 144hz?

Pursuit camera shots of both monitors would plainly show what is going on one way or another. Unfortunately we have no pursuit camera shots showing more of what your eyes would see of either the swift or the acer's blur amounts and overshoot.

These aren't pursuit camera shots but I put info from both reviews in one place for convenience.

rog-swift-acer-xb270hu_overshoot-comparison.jpg
 
Last edited:
elvn said:
What I am beginning to think is happening is you are getting smear blur encompassing what would have been the ghost overshoot limit on the acer. Therefore slower response time (smearing blur) = "no overshoot". It may be smearing at least as wide as any ghost would have overshot were it 1ms.



So it may be that your are trading a smear blur width rather than worst case color transition tiny amount of overshoot on normal overdrive.

The swift would likely show a soften (fuzzy) blur more within the object's limits/"shadow mask", with an occasional faint or dark edge (slim) overshoot.
The acer would likely show a smear blur "outside of the lines" of the object, wider than the swift's overshoot, and likely blurring the detail within the object/smear object a lot more, but the question is how much at 5.5ms at 144hz?

Pursuit camera shots of both monitors would plainly show what is going on one way or another. Unfortunately we have no pursuit camera shots showing more of what your eyes would see of either the swift or the acer's blur amounts and overshoot.

These aren't pursuit camera shots but I put info from both reviews in one place for convenience.

rog-swift-acer-xb270hu_overshoot-comparison.jpg
Click to expand...

Can explain the Rog swift settings from the pic you made? Which one is extreme normal and off I'm confused.
 
The screen was also tested using the chase test in PixPerAn for the following display comparisons. As a reminder, a series of pictures are taken on the highest shutter speed and compared, with the best case example shown on the left, and worst case example on the right. This should only be used as a rough guide to comparative responsiveness but is handy for a comparison between different screens and technologies as well as a means to compare those screens we tested before the introduction of our oscilloscope method.
Click to expand...
neither are off or extreme. In the acer review thread tft central decided to use a best/worst two picture presentation rather than off/normal/extreme. So they are showing the normal common amount overshoot paired with the worst level of overshoot (still on normal) on the right image of each pair. I just added another larger image of the normal (not worst normal) from the middle of the image "ribbon" from the original pg278q thread since it was available and put it below the regular/worst ones. So the "best" and the larger "normal" picture are essentially the same on the pg278q side, I just kept the smaller one so that they would be more easily comparable across the middle ribbon/gallery.


TFT central swift overdrive (not pursuit camera shots obviously):
rog-swift_tft-central-overdrive-normal.jpg


A side by side pursuit camera shot of each monitor would give a lot more insight than still shots of overshoot or reviewer opinion of the tradeoffs does. (Full in-game scene fast FoV rotation or fast left/right FoV panning pursuit camera shots of the screens would be even better). Otherwise you are taking a subjective opinion until you see it for yourself.

This is a pursuit camera shot used as an example of overdrive error from the blurbusters site. It is not the swift so do not take this as a representation of how far the overshoot is on the swift. I have no idea what monitor he used for this I just thought I'd post it to show blur + overshoot shown with a pursuit camera.
lcd-overshoot-blurbusters.jpg



I'm really surprised that the more thorough monitor review sites haven't taken a cue from blurbusters.com and started adding pursuit camera shots to their reviews. They are aware of blurbusters testing and often quote blurbusters in their reviews.
 
Last edited:
Cannot complain about the TFTcentral reviews tbh they are much better than any other reviewers and give you a good idea what to expect.
 
Just hoping they'll adopt it eventually along with other testing and presentations they incorporate like the SMTT tool, oscilloscope, and other advancements while still showing pixelperan ufo with a high shutter speed camera (just not a pursuit camera yet!). A pursuit camera would give you a better idea of what you would actually be seeing with your eyes rather than guessing or imagining what it would look like after being shown what a stationary camera sees. You are never seeing a crisp un-blurred cartoon ufo like that (with your eyes) at any decent speed that would cause overdrive overshoot on any lcd without backlight strobing enabled. It would be more like the blurbusters overshoot example and depending on the monitor model it would have more or less overshoot and possibly much more blur and blur width than this example.
lcd-overshoot-blurbusters.jpg
 
Last edited:
elvn said:
I'm really surprised that the more thorough monitor review sites haven't taken a cue from blurbusters.com and started adding pursuit camera shots to their reviews. They are aware of blurbusters testing and often quote blurbusters in their reviews.
Click to expand...

You do not even need a pursuit camera, the point of the blurbusters tests is to enable the user to make these kind of photos with a normal camera: "Blur Busters invented an inexpensive method of accurately using a consumer digital camera as a pursuit camera."
http://www.blurbusters.com/motion-tests/pursuit-camera/
you basically need to enable the pursuit camera track, then track the ufo with your normal camera and make trial and error shots until you get a good one (a camera rail would help immensely though)
it is also possible to capture ghosting on video, as I did here with a simple recording function of a digital camera:
http://www.youtube.com/watch?v=uwewRjbo0Aw
this also shows how the Acer can drive down the pixel response times at higher refresh rates without causing overshoot - because my screen does exactly the opposite.
The Overdrive stays always the same here, so the ghosting trail on dark transitions is elongated at 76hz, and gone at 50hz.
The little Overshoot there is in bright transitions is gone at 76hz and stronger at 50hz.
So there would be leeway at higher refresh rates to further accelerate the panel without causing overshoot and this is exactly what the XB270HU does.
 
yes the term pursuit camera could generically mean pursuing on a track via that method, I don't know what else you would want to call it - consumer camera tracking in motion? Pursuit camera seems a lot easier generically. I'm sure tftcentral would have a nice camera setup if they chose to incorporate pursuit camera testing though.

My point, and question is about the blur width and blur amount within that width - which a high shutter speed still (non-pursuing) shot doesn't show. Is the blur width of the acer (to our eyes) as wide or wider than any quoted slim edge of occasional overshoot on the swift, and does that blur object smear out definition more than the typical soften/fuzz blur that a 1ms 120hz-144hz is capable of (~60% blur reduction) to more of what we see in 60hz pursuit camera photos?

Is the smear blur at 5.5ms wider than any overshoot would have been in the first place and if so by how much? Pursuit camera tests (of normal and worst if you were to follow their high shutter speed pixelperan convention) would show it more clearly (no pun intended).

These questions are important to me because remember that while these tests are done on a simple single bitmap/cell-shaded-like ufo object, in 1st/3rd person gaming you are continually moving the entire game world around relative to your screen when you mouse-look and movement-key around - so the blur amounts regularly apply to the whole game world viewport in those type of games.
 
Last edited:
I guess if you made pursuit camera shots of Acer and ROG side by side then the UFO would have the same sharpness on both, but on the Swift followed by a weak corona and a faint, thin ghosting trail on the Acer.
None of those would be problematic in practice (during gameplay)
 
How can you be so sure with a 5.5ms response time, regarding my questions in my previous post about motion clarity/blur amounts? The pursuit camera shots should show it just like the 60hz 1ms and 120hz 1ms (50% reduction)example ones do (no 144hz 1ms 60% reduction example available), but specifically to the panel models.
My point, and question is about the blur width and blur amount within that width - which a high shutter speed still (non-pursuing) shot doesn't show. Is the blur width of the acer (to our eyes) as wide or wider than any quoted slim edge of occasional overshoot on the swift, and does that blur object smear out definition more than the typical soften/fuzz blur that a 1ms 120hz-144hz is capable of (~60% blur reduction) to more of what we see in 60hz pursuit camera photos?
Click to expand...
 
Last edited:
NCX said:
At some point you will have to cease being in denial and think rationally. The monitor's brightness settings and room lighting are identical. The glossy black bezel points upward and downward, and we can clearly see the bright spot caused by the room light on the actual bezel in both pictures. It's simple logic, and a point which can be proven with Microsoft paint, but I'm sure you will continue to be in denial. I also made a triplet compilation with the monitors silver frames exposed.

By using a glossy black bezel, Acer has made it very easy for competitors to create a vastly superior product with identical parts by simply using a dark matte grey bezel. If Asus is smart they will do this (Eizo & NEC certainly would), especially since their products are far more popular.
Click to expand...
no one's in denial. i'm not arguing the existence of the effect.

i'm just saying that those pictures are not an appropriate demonstration of the effect. your revised crossover pictures and that mspaint example are.

https://www.flickr.com/photos/97364704@N08/16595283006/
https://www.flickr.com/photos/97364704@N08/16413945037/

again, just look at the monitor's shadow on the wall.... it's hard to believe that you have identical lighting in the two pictures.
 
flod said:
tftcentral is better than most but prad.de is better imo
Click to expand...

TFTCentral is the best, prad is irrelevant these days. They haven't tested a display since April 2013.
 
Last edited:
Prad reviews can be either super-comprehensive or rushed. Their speed/color measurements could gain from being followed by more subjective opinions and comparisons to other displays (such comments curiously can be more developed in the shorter reviews) but more bizarre are their lag measurements, which sometimes can be completely different from other reviews like tft's, pcm, ncx or oc.ru
I don't speak German so it's a pain but after a while you get used to it even through shitty auto-translators.
 
I need to do some re-search on 3D vision, not sure what does it do, and what is it important for :// :)
 
NEK4TE said:
I need to do some re-search on 3D vision, not sure what does it do, and what is it important for :// :)
Click to expand...

Active 3D with shutter glasses.This one's probably still a bit too slow to do it without a lot of crossover so they left it out.
 
Prad is one of the better ones but the only reviews I am 100% confident in are the TFTcentral reviews... translating from german is annoying and often gets things wrong in the translation so it could be one thing or the opposite... What I think would be good in TFTcentral reviews is a few videos for things like motion blur and viewing angle etc. I would think a 5.5ms average response time with a worst response of 6.9ms should be very good at 144hz as there would never be more than 1 frame (at 144hz) of trailing and not a lot of blur (also overshoot looks 10x worse than blur). Would be good enough for me anyway as I am more interested in PQ as long as the blur is not obviously bad. Its funny how we are now paying £700 for a monitor which is approaching the clarity of a 15 year old CRT monitor but still has 1/10th of the contrast ratio etc. and no changeable refresh and resolution... They should just bring back CRT's and make them bigger (screen size) and thinner. If they released a 27" 1440p CRT I would buy it now and would not have to piss about with all these LCD's waiting 7 years for a decent upgrade.
 
Last edited:
Vega said:
TFTCentral is the best, prad is irrelevant these days. They have tested a display since April 2013.
Click to expand...

Overclockers.ru also have great reviews and they always post macro shots of anti glare coating :D
 
I've mentioned them with 'pcm' :D
Problem is they seem to get even fewer review samples than before, clearly fewer than tft and prad now. :(
 
You must log in or register to reply here.
Back
Top