[stalepie]

When considering the BenQ EW277HDR again and looking at your review, I was wondering if 93% coverage of DCI-P3 meant over-saturation of colors when viewing “regular” (sRGB) content online and whether it’s not better to go for a monitor that sticks strictly to the sRGB standard. Is it likely that DCI-P3 and Adobe RGB gamuts will become the new norm down the line, for regular content (websites, online videos, home movie watching, photos, games, etc.) and this is an transitional stage at this point? (similar to how HDR has been slow to roll out)? That is, is it all on the road toward REC 2020 and beyond, which will eventually trickle down to the cheapest displays people use?

I don’t like the idea of having oversaturated colors for most content, but then switching to emulation mode to see it normally. It’s hard to understand why a wider gamut wouldn’t simply reproduce the sRGB content as a 100% sRGB (or 72% NTSC) monitor would, but have the ability to go further for content that needed it. Similar to how a color TV can accurately display a black and white movie without introducing shades of color somewhere on screen.

[PCM2]

Yes, as the review explores having a wide colour gamut increases saturation of sRGB content, so it becomes oversaturated. HDR is not going away and neither is the DCI-P3 colour gamut which it targets. In fact the longer term goal is Rec. 2020 which is a significantly broader colour gamut. Many users actually like the extra saturation provided when it’s provided in a balanced way (wide colour gamut in all regions), as explained in the review. The sRGB colour space is extremely restrictive and encompasses a tiny portion of the shades you see in the real world. So content developers are quite happy to be able to use larger colour spaces and many consumers are as well. Your average consumer is used to looking at mobile phone screens and TVs with extension well beyond sRGB these days as well, but again it’s personal preference.

If you don’t like this sort of look or need higher levels of accuracy within the sRGB colour space then that’s also understandable. And that’s why many models with an extended colour space offer some sort of sRGB emulation mode to cut down on saturation and deliver arguably more appropriate output within the sRGB colour space. This includes the BenQ, but as is often the case you get some undercoverage of sRGB as well. It is often the case that a monitor manufacturer offers separate HDR presets that use the full gamut and they are automatically activated when HDR content is displayed. It is up to the user whether they use an sRGB emulation mode or the full native colour gamut outside of that.

At least a little extension beyond sRGB is often pleasant and I wouldn’t say the EW277HDR is at all extreme with its native colour gamut. I agree with the assessment that sRGB is too restrictive, it could be considered a dull and lifeless representation of colours that does not properly reflect the variety in the real world. You should consider the EW2775H as a suitable alternative if you would prefer closer sRGB tracking and just a little extension beyond.

[stalepie]

Thanks so much for the answers. I really appreciate it.

[stalepie]

I have a few other questions, if you don’t mind.

I bought the EW277HDR monitor and do find things to look a little garish or oversaturated. Actually my issue is more that colors don’t look accurate – that is, yellows look orangeish, and oranges look red (where that leaves red, I don’t know, but reds do seem to pop unnaturally, which is obvious with computer icons or symbols that use it, like the YouTube symbol).

However what I noticed is that this is the same problem I had with a previous BenQ monitor (actually two) which I didn’t think supported much in the range of DCI-P3. They were the GW2270 and GW2280. I would say the EW277HDR is more extreme, so perhaps the color gamut simply extends further beyond sRGB than those two standard office/home-type monitors. But I did wonder if it was behavior peculiar to BenQ or the types of AU Optronic VA panels they source.

I have an older Sony TV that is surely using a VA panel but doesn’t have these color characteristics.

I guess what I’m wondering is why DCI-P3 (and presumably AdobeRGB) shows regular sRGB content in an oversaturated fashion to begin with. Why wouldn’t it be like a color TV that displays an old black and white movie properly? It does not need special metadata from the B&W film, and it doesn’t need to be set into a special B&W emulation mode (like sRGB or REC 709 emulation).

As another point of comparison for me is my iPhone 6, which I understand was well-reviewed at the time of release for its color accuracy. For instance, on this page:

https://www.anandtech.com/show/8554/the-iphone-6-review/8

It says something interesting to me, which is:

“This means that we use the sRGB gamut and 2.2 gamma, which most content is adapted to. While AdobeRGB and other gamuts exist, these are for limited use cases and only applicable to operating systems that are aware of multiple gamuts and can dynamically switch between them depending upon the metadata of the content.”

Granted, this was written four years ago, but what is this about “only applicable to operating systems that are aware of multiple gamuts and can dynamically switch between them based on the metadata of the content?

Is it the case then that older content, meant for sRGB or REC 709, should somehow be marked by the operating system (Windows) as needing these gamuts? Would a monitor that displays DCI-P3 or AdobeRGB be able to show both sRGB and DCI-P3 content simultaneously (such as in two windows or in two pictures on the same webpage)?

This again is what I don’t understand: it seems to me that these monitors (and TVs) should be designed to display the max color gamut as required for any images, but without switching or needing metadata, also display lesser gamut content too, just as a color TV can display black and white content without introducing color artifacts into it.

[PCM2]

The sort of oversaturation you describe is typical when observing sRGB content on monitors with significant extension beyond sRGB. The monitor and game does not intelligently map things to more appropriate locations on the wider gamut. Everything is essentially stretched out instead. A shade which a developer intends to look moderately saturated (let’s say a yellowish orange flame) under the sRGB colour space may appear perhaps half way between the centre of the gamut and the edge. When the gamut is expanded, the edge is brought further out (the gamut is larger) – yet the shade will still appear half way between the centre and edge of the gamut. It will therefore be more saturated – if the gamut is sufficiently wide, or depending on the shade in question, it could well appear beyond the boundary (edge) of the sRGB gamut. I usually use fire as an example when I talk about this, indeed I did in the colour reproduction section of the EW277HDR review with reference to Battlefield 1. Yellows verging on orange, oranges appearing more saturated than they should and some saturated oranges appearing with a red hue. And that reddish brown hue in some earths and sands and some skin tones can become exaggerated by the gamut in the same way.

I agree this isn’t how things should work, but it’s how things do work. Wikipedia has a reasonable summary of how colour management is implemented on the OS and application level in their article on the topic. Windows itself is not “colour aware”, so to speak. So it ignores any metadata or specific gamut information for an image or other content. If you’re in a colour managed environment (some web browsers are ‘colour aware’, for example) then any metadata or tags related to the gamut can be read, but Windows won’t just do this automatically. It will require you’re using a custom ICC profile, ideally one that has been specifically made for your monitor using a colorimeter. In my experience colour management in web browsers can be quite hit and miss. It gives a far more consistent experience if you calibrate the monitor correctly (using an sRGB emulation mode, if the monitor has a good one) in the OSD, apply an ICC profile on top but don’t rely on it to be used correctly.

Back to unmanaged environments, you can observe a degree of oversaturation without expanding the colour gamut for many reasons as well. Obviously the boundary of the gamut is not expanded, so you don’t reach the same saturation levels (it’s physically impossible) but depending on gamma setup or even colour channel balance you could get yellows that appear too orange etc. And a final note – I appreciate the donation and your support, I’m grateful for that. And also for creating this topic, which will be of interest to many users I’m sure.

P.S. Have you tried the ‘sRGB’ or ‘Rec709’ presets on the EW277HDR? They restrict the colour gamut as noted in the review. But they give some under-coverage and there are those saturation losses towards the edge of the screen (VA panel issue) as well, so you might feel things look too ‘washed out’ when using these presets.

[stalepie]

Yes, the sRGB and Rec 709 modes make it look much closer to the iPhone 6 or other displays I’m used to. I often like to go back to classic video games and feel I know “by heart” how they should look, after seeing them on so many TVs and in print as well; I’ve kind of averaged together in my head how they should look, so something will often look instantly wrong to me then when I see it incorrectly displayed with an expanded gamut. The same is true of photos and movies, though. A normal face of a guy may look like he’s wearing lipstick or other makeup when the colors are “expanded.”

I’m not sure saturation is the right word then. If it is, is it the same thing as adjusting the “saturation” in the monitor UI? Changing it from 50 to 60 up to 100? I assume not, although dropping it down to 45 or 40 actually makes it look closer to the Rec709 and sRGB modes (making me wonder if that is what is happening software-side on the BenQ monitor when you select these modes).

It’s sort of a chicken-and-egg problem or cart-before-horse, because how can artists be expected to use enhanced color gamuts (at least when working purely digitally) if so few can correctly see it, and they have to scale down their color, output to sRGB and double check each time. Of course, a traditional artist working in analog mediums doesn’t have to worry about this, but I do have to wonder why more attention isn’t brought to the subject, and why, if we’ve known for so long about the extent of human vision (such as CIE1931), why even our best specification, REC2020, is only around 50% of what the eye can see.

There’s probably tons and tons of stuff I can’t understand without being a color scientist or extremely well read on the subject. It apparently gets rather technical very fast when trying to read up on it.

I suppose it may come down to data limits. It requires more bandwidth to send wirelessly when using more shades of color. Interesting that such things put forth by the… United Nations? https://en.wikipedia.org/wiki/International_Telecommunication_Union

Anyway, back on topic, the Rec. 709 and sRGB modes seem a little lacking in intensity or saturation, in comparison to a native sRGB display. Could be wrong. I don’t have one side by side at the moment. The iPhone wouldn’t be the best comparison…

I know it’s not meant as an art or photography / professional monitor, just one for entertainment. My “issue” is more my curiosity into the subject and sampling different products sometimes, to try to understand how it works.

So a good TV of the 90s or 80s, a CRT, would be capable of something close to REC 709? or even worse? how high of static (?) contrast ratio did consumer CRT televisions have?

I personally am not sure I notice a huge difference in contrast ratio. I thought this would be important to me, but maybe the general glow that edge-lit displays have makes this self-defeating. I still see blacks as gray-ish, but probably I did with CRTs as well. Just my eyes, I guess. I have a PlayStation VR which has an AMOLED screen in its mask, as well as a Samsung Galaxy phone in the house, so sometimes those make for interesting comparisons… but I think I’ve found that the 1000:1 contrast ratio standard for normal displays (IPS, TN, etc.) probably exists for a reason: that, at least in normal viewing conditions (with ambient light on in the room), it provides enough contrast to see most things, to feel one sees “true black”..

Gosh, I’m rambling way too much. Guess I’m trying to get out a few thoughts before some relatives come over for Thanksgiving week. I had some other ideas about this, but I’ve forgotten them now. Just seems to be a shame that it isn’t easy to make displays show more colors without, at the same time, incorrectly mapping older content (~sRGB). This means A LOT of older content (anything before this century?) cannot be correctly shown on displays moving forward without an emulation mode… all those movies of the 20th century… it’s a lot of content. They’d have to be remastered, all of them re-released into “HDR” or something, just so they can look normal again.

[PCM2]

It is certainly possible to restrict the colour gamut by adjusting saturation levels. Be that on the monitor OSD or using software. I should have been more explicit, I’ve slightly edited what I said because it made it sound like it wasn’t possible to restrict the gamut in this way – which isn’t true. The problem is that it is extremely difficult – essentially impossible – to maintain accurate restriction of the gamut and appropriate colour accuracy at the same time. What you’ll find is that you can certainly restrict the gamut by lowering saturation, but it sort of just pinches in for some regions but remains stronger for others. Some models offer 6-axis saturation adjustment which allows for more precision, but even then you’ll find that you are satisfied with some shades but not others. You reach a point where the saturation of some shades is just about right, then suddenly things will become dull or undersaturated elsewhere. It’s very messy and not an accurate way of doing things.

A proper sRGB emulation setting is much more accurate, but even then there can be underlying differences vs. a proper sRGB reference due to the spectral differences of the backlight. Colour perception is a complex thing, so complex in fact that colorimeters can’t truly evaluate or correct for some of these differences either! For colour managed programs the use of a colorimeter or similar device, which will account for the colour gamut and apply corrections at the GPU level based on that, can also work well.

So yes, the transition to wider colour spaces is not a particularly smooth one and has its issues. But the best manufacturers can do is to offer a good sRGB emulation setting. And that’s, unfortunately, not something you can take for granted.

[stalepie]

I notice here, on the red block that hides the WebKit logo on sRGB displays…

https://webkit.org/blog/6682/improving-color-on-the-web/

that my iPhone 6 correctly shows just a solid red block (best I can tell), while the EW277HDR still shows the logo no matter which mode I set it to. (The color tone does match the phone better on sRGB and Rec709 modes, though).

[stalepie]

Sorry I see your new response after I posted that last one – yes, the emulation mode, at least on this monitor, seems imperfect, but better than nothing. I guess I should have gone with the older version that was more sRGB limited or another one of theirs. I’d be happy with DCI-P3 or AdobeRGB monitors if it just treated smaller gamut content properly, instead of stretching it to the wrong shades. In the abstract it doesn’t seem that important, but when you realize fire rarely looks “red,” or water should look light blue instead of deep blue, it becomes obvious that everything is wrong.

I would think software could be programmed to recognize the original gamut of images by counting the number of colors used? Would not work for everything, especially simple pictures, but a movie for instance would normally have a lot of colors sometime during its 90 minute+ duration, enough for the software to deduce that it’s dealing with non-HDR content and act accordingly (at least in full screen mode), effectively putting in an auto-switch between sRGB or REC 709 emulation and then back to wide gamut when exiting the program.

(I don’t know, maybe that is impractical. Seems like it would be even more so with streaming content. Would have to sample multiple sections of the movie if it starts out dark with just opening titles). I’m just thinking… requiring everything to use metadata requires everyone to get on board, even with old content (updating old files). For instance, I read that YouTube treats videos as BT709 if it’s not specified in the metadata. It’s one thing to expect major movie companies to get on board with this (and they probably have!) but there’s so much other stuff out there.

I’ll look into the Windows color spaces thing now and the other stuff you mentioned. Reading about Color Sync Utility now on MacOS, but that seems printing related.

[stalepie]

This iPad review page is helping me catch up to these topics…
https://www.anandtech.com/show/10265/understanding-the-97-ipad-pros-true-tone-display

feel free to ignore some of my questions – I am just a beginner to these topics and starting to learn & still searching on the web about it

[PCM2]

It’s alright, asking questions and reading the sorts of articles you are is a great way to learn. It isn’t a simple topic, as you’ve found out, and I totally agree that things should be handled better than they are at the moment with respect to sRGB content on wide gamut displays. I’m not a programmer or engineer myself so I’m not really aware of what would or wouldn’t technically be possible or practical to implement. 🙂

[Mihle]

Ok, but something I dont totally understand in your review/in general. (Last questions, I hope)

When it comes to colour and colourspaces. As it cant do 100% Adobe RGB (or whatever the other name on the wider thing was), If I edit a photo once in that and I try to do something that is outside of that, will it just stop getting greener sort of like, 100% and 90% green is the same because it cant display more?

Also, if I try to edit sRGB if the monitor that it set to something wider, will it still look correctly on another monitor that is just sRGB, or will it then look muted colours there because space is smaller?

Anyway, is phones sRGB still or are they become wider too now?

Also, can you calibrate it to sRGB if I wanted that? (not decided at all, answer on those two above matters)

[PCM2]

Mihle – I’ve moved your reply over to this thread as it’s an appropriate place and some of your questions are addressed here.

Your first question isn’t something I have personal experience with as I don’t work within the Adobe RGB colour space, but this is how I understand it would work. If you’re viewing content which was created to look correct in the Adobe RGB colour space on a device with a narrower colour space and you don’t convert the image, things are simply going to look undersaturated for some shades. The degree of undersaturation and how noticeable it would be depends on the image itself and how close to the edge of the Adobe RGB gamut (or how far beyond the sRGB gamut) the shades you’ve used sit. Remember that the Adobe RGB colour space coincides with sRGB in the red and blue region of the gamut but extends massively further in the green region. If you’re using shades that are deep into that extended green region (this doesn’t just include pure green shade, but also shades where green is mixed in) then they will look much less saturated than they should. They will also be crushed into shades that are technically at the sRGB boundary in the original image. Again, this isn’t something I can confirm through experience and just my understanding of how it would work.

To edit accurately in the sRGB colour space, you need to profile the monitor using a colorimeter or similar device and be using colour-managed applications. Alternatively, you need the colour gamut of the monitor to coincide quite closely to sRGB. As explored earlier in this thread, that means an sRGB emulation mode. Which is usually a setting on the monitor or can be achieved through the AMD graphics drivers on some displays including the 328E1CA (as covered in the review). This model offers some extension beyond sRGB but doesn’t reach Adobe RGB. So it just gives a somewhat more saturated representation of shades within the sRGB colour space. Profiling can tame this for colour-managed workflows, however.

Smartphones these days often have rather generous colour gamuts. Good DCI-P3 coverage, but not necessarily Adobe RGB. I know Samsung phones have an sRGB emulation mode of sorts in their display settings and some users do actually use their smartphone as a reference device when editing on a monitor with a wider colour gamut than sRGB.

[ssss]

I plan on buying the Asus PG329Q because it ticks everything I want (IPS, 32″, 1440p, high refresh). The monitor has a coverage of “160% sRGB / DCI-P3 98%” as advertised. xxrsm (Chinese reviewer) got the following measurements:

https://i.gyazo.com/30dba884acb17ff783cf0318566b1ace.png
(yeah, it’s pretty high for a “gaming” monitor).

I’m worried that it might be too saturated since I also plan on doing photo editing on it. The monitor has a sRGB mode but it’s often terrible on Asus monitors (my XG279Q‘s sRGB mode limits the brightness to almost 500 nits which is, quite frankly, unusable).

What’s your opinion on this? I do have a colorimeter and use a Nvidia GPU if this helps.

[PCM2]

If you’ve got a colorimeter then you’ve got nothing to worry about with respect to photo editing in a colour-managed environment. You’ll have the monitor profiled as described in my previous reply. That wouldn’t apply more broadly – for example to games, the desktop or some content on the internet (Google Chrome is your best bet for ensuring the profile is used properly). Acer often, but not always, have more flexible sRGB emulation settings, so you might want to consider the XB323U GP as an alternative.

[PCM2]

Our latest article looks at sRGB emulation. How you can tame the wide gamut on the monitor, some associated shortcomings and some alternatives in the graphics driver. AMD only, currently, but hopefully something Nvidia can get on board with. The article ends with a wishlist of things we hope to see implemented by monitor and GPU manufacturers.

[Meda]

Hello, first of all, thank you for all the information on this page, it was very useful.

Secondly, I just wanna know how much sRGB coverage is the 328E1CA you reviewed and mentioned when replying to Mihle

You said it was more than the sRGB coverage, less than the Adobe RGB coverage. and I think I would like my next monitor in between both just like the 328E1CA.

But if it’s more than sRGB can you tell in percentage how more than 100% it is ? so I can compare that sRGB percentage with monitors available in my country and hopefully choose the right one for me.

Thanks in advance.

[PCM2]

Hi Meda,

The colour gamut is shown in the review, where 85% Adobe RGB was recorded. Our colorimeter software doesn’t report relative coverage values so none of the values will exceed 100%. We compare to both Adobe RGB and DCI-P3 in more recent reviews to help give some comparative figures. Philips specifies 122% sRGB which matches well with the gamut shown in the review.

[Pillo]

I am extremely sensitive to oversaturation. I’ve been lucky enough to test various monitors (not mine) to get a feel for it, and VAs have always impressed me with good sRGB gamut coverage, unlike IPS, but I needed an IPS gaming monitor. So when I bought the Benq EX2780Q I still wasn’t sure how good its Rec.709 emulation was, because despite Adam’s thorough review, experiencing it in real life is different. But now i’m very happy with it, and i use it exclusively in Rec.709 mode, setting just the backlight intensity when needed. I dont care so much about the possible slight “weakness” of certain color tones accuracy, what matters to me is the gamut does not bother my eyes, and in games this is great because it seems to me “real” compared to the display gamut out of the box

[Author]

Posts