Flat Panel Display recommendations

[wesley]

Finally getting rid of the CRTs on our workstations on both PCs and Macs. Looking for suggestions on decent flat panel displays from 20 - 24". I have found for the most part, a standard LCD will display much better on a Mac. The PCs with assorted video cards have a HUGE difference in quality. We are MUCH more picky than the standard office worker in that we need reasonably accurate color throughout the screen and, (here is the important part) the color needs to be somewhat close Regardless of a small change in viewing angle. It seems several of the flat panels we have tried will look completely different from the top of the screen to the bottom. A square of PMS 300 blue will look like cyan at the bottom, but then turn much darker as you lower your viewing angle. The same monitor will look and act much better on a mac, but still nothing like a CRT. I know this will be a learning curve, but there must be some interaction of video driver cards and monitors on a PC that is causing this. Just cannot find any good info about it. I would love to find a flat panel and video card for a PC that behaves like the display on the new iMacs. We also need dual head cards for sidekick monitors for our pallets. I suppose I could buy one of the Eizo ColorEdge CG monitors for $1600, but I just assume for that price they are fantastic. Is there anything that is pretty good that is reasonably less expensive? Thanks

 

[UncleSam]

All you need is Eye-One Display 2 to calibrate your PC/Mac Monitor so that they can display the color more accurately. It is selling $219.99 at Newegg.com. For the monitor, I recommend you to look at Samsung P2370 ($269), it's a very thin 23" LCD 1080p monitor. Also, Dell G2410 24" LED 1080p Monitor which is only $259 from Dell.

 

[maxon]

First thing is to look for a good quality LCD. Most panels even so called graphic series are TN technology, you should stay away from these. Look for S-IPS or P-MVA panels, a bit on the expensive side but at least they give you good uniformity, less bleeding on edges and much better color reproduction quality. Could be Dell, LaCie, even Apple etc. We use LaCie which are not the best in class probably, but good enough, with their factory colorsync profile loaded. From that point onwards, if you need super accurate colours you should think of monitor calibration to bring colours into specs.
A no-name (read TN) display is usually unable to provide anything close to sRGB .. you can of course calibrate it but that calibration in the center won't be the same as corners etc. The ColorEdge may be expensive for a reason.

 

[Shawn]

Hi Wesley

I agree with Maxon about avoiding the TN based panels. Avoid them at all costs for color critical work. They are 6-bit instead of 8-bit and can only natively display around 200,000 colors and achieve the rest by dithering. They also have a very narrow viewing angle, the other types of panels will reduce or avoid the viewing angle problem. TN panels are cheaper to produce than the other types, so unfortunately as everyday users have become more interested in larger screens, TN panels have moved upstream into larger monitors that used to be based on the better but more expensive S-IPS or PVA panels. Any 24" display selling for around $200-300 is probably using a TN panel.

This site attempts to keep a database of what panels are used by various manufacturers. Try entering a monitor here before purchasing to check if the panel is suitable for higher end work. They also have some good, in-depth reviews of various models.
www.flatpanelshd.com - Your guide to flat panel monitors and TVs - Panel Search

My personal recommendation for color critical work, if you can afford it would be for one of the Eizo models. You get what you pay for. For some cheaper options, we've also had some of the older matte finish Apple Cinema Displays that worked well. We've tried some of the higher end Dell monitors but found that the quality varied quite a bit, even within the same model, depending on the batch. I just set up a Dell 2408WFP for general use that is very accurate after calibration, but your milage may vary. Samsung has some good screens, the XL series in particular, but they cost in the same range as an Eizo.

Some of the above units will come with their own colorimeter for calibration. If you're looking to buy it separately, the Eye-One display will work well. Also consider the DTP94 ColorEyes bundle from Integrated Color Corp. Home of ColorEyes. It runs around $300 but gives excellent results. We've compared the ColorEyes software with the Eye-One on the same hardware and preferred the profile from ColorEyes. If you already have a compatible colorimeter, the ColorEyes software runs around $150.

You don't mention the type of video card in your PCs. If you're still using the save VGA connection you used with a CRT you'll get a much worse image. You'd be much better off switching to a native digital DVI connection. The internal software that converts from analog VGA back to digital in many monitors will introduce a softness to the image as well as the potential for color shifts.

Shawn

 

[tmiller_iluvprinting]

Check out the LED flat panels from NEC. You will want to order NEC's calibration bundle. Their Eye-One Display is setup to properly calibrate the LED screen, something I don't believe you can do with your everyday Eye-One Display. I am using an Eizo ColorEdge presently, my next screen will be the NEC LED.
Best regards,
Todd

 

[Lukas Engqvist]

I saw the Quato screens at Drupa, I was wowed. Wish I had the budget for one of those. They have some optimised for ISO coated giving 100% of print gamut..Â*unfortunately you have to pay for it The IP260excel on my wishlist, but christmas is a long way away.

 

[David Ingram]

2 cents

2 cents

Is it also still true that Macs still use a higher default gamma (1.3) vs 1.0? could that be the reason? Also how does the new apple LED display stack up? I noticed a long time ago that web pages i made with mac were always darker than expected when viewed on PCs.

 

[fotografx]

LCD displays brightness problem

LCD displays brightness problem

If you all were involved in digital photography you might be aware of the "prints too dark" problem. The last I checked Google a search on that phrase will obtain 100's of millions of reports. The source of the problem is the fact LCD display brightness is much greater than that of CRT's which had a white luminance of about 90.0 CD/m2, while LCD's at default can be two to three times as bright or more. This causes a perceptual error in digital image photo file brightness adjustment. The final result is when this file goes to a printer the brightness information assumes printing paper brightness is the same, but in fact it is at best the equivalent of CRT luminance.

I have just finished My final report on this problem and what the solutions are that need to be applied to the use of LCD displays that should be published this summer. But have been writing bits about it in my blog at Shutterbug: David B. Brooks Blog

My first article on this "prints too dark" problem was published in December and can be read at Shutterbug: Prints Don’t Match The Image On Screen

If you want to get involved in the search for solutions to the LCD display problems I can be reached at: [email protected]

David B. Brooks

 

[Lukas Engqvist]

The gamma on the mac is 1.8 and PC 2.2 (not 1.0). What normally is the problem in print is that there is no gamma adjustment at print, and many just use the photoshop defaults.

Looking at the Europe defaults at least the dotgain for grey is set to 15% whereas the dotgain if extracted from the K channel of the CMYK profile (FOGRA 27) is 17%

The luminance setting does have effect, but one must bear in mind the working environment. A CRT has luminance better suited for normal office space whereas a CRT @ 90Cd/m2 you will need to work in a dimmed environment

 

[fotografx]

First off Apple now has the same OS defaults as a PC D65 and 2.2 gamma.

The difference in print density between 1.8 and 22 gamma if the setting mistake is made is very small compared to the difference in luminance white point of a current new iMac even turned down to the lowest brightness the white luminance can be 220.0 CD/m2 and a Sony Artisan CRT at 100% gain which was only 97.0 to 99.00 CD/m2 at best.

Dark output results have been reported by literally thousands on-line in a variety of forums, and reports in one-on-one from e-mails from digital photographers I have received here in California from as far away as New Zealand and Moscow. and the one consistent factor is user have adjusted RGB photo image files on a bright LCD display and then that file information transferred to output rasterization results in very dark results because the brightness of the display used to make the image adjustments should match paper white which is the equivalent of between 80.0 and 90.0 CD/m2 white luminance in display setting.

Sorry, I have been researching this "prints too dark" problem for months now and investigating the factors involved in my lab, as well as consulting with some of the best technical people in color management, and the main culprit of unexpected very dark print output is an LCD display that is set much too bright. This is essentially what I have been doing professionally since 1975, testing and evaluating imaging products and publishing reports on my results in national magazines.

 

[Lukas Engqvist]

Sry yes i did know that apple had changed to 2.2, and 6500, my point was it wasn't gamma 1 (unix used to have 1.1 I think) 6500K being the same white point as AdobeRGB and sRGB, but ISO for DTP viewing is 5000K.

I'm sorry to counter but there is no brightness of paper unless you have a light source.
To even state that paper white is between 80 and 90 CD/m2 doesn't make sense unless you have a light source, and a standardised one at that. Paper can be measured in reflectance, and possibly as a specified brightness in a given environment. I don't really see how quantity of research articles can change that. If you will give me an argument instead of a quantive number of references there will be a possiblitity to come to terms.

I do agree however that a monitor can be set too bright for the environment that it is in. If a photographer switches monitor and has a habit of working in a dimmed room (ISO 12646 was defined for CRT @80-120 cd/m2 with an ambient light of 32 lux, wheras a normal office is 500-600 lux wich means that the CRT would not appear white but the LCD would be more appropriate, for a full gamut on printed paper we would need light of 1500-2500 lux).

I still think the dark greys are much more likeley to be a workflow issue than a monitor issue. But we are getting a little off topic since it is about displays and not about colour management. I would say if you are concerned about dark prints then it is tha colour management discussions you want to go (well worth a visit) but if you have an issue with monitor and print not matching I think it is not a given that the monitor is wrong.

Calibrating and proofing does takes care of remapping gamma, isn't that the whole point?

From this knd of discussion on another forum I would also say check that you don't get banding, especially not coloured banding on a grey gradient before deciding on a monitor.

 

[fotografx]

So tell me there isn't a standard luminance for viewing booths? Tell me why in the Eizo Nanao Color Navigator Software for adjusting, calibrating and profiling an Eizo ColorEdge LCD display there is a utility to match the brightness of a display to paper white as it is measured in the viewing booth used?

Does it make any sense to set the monitor/display according to whatever the work environment is? Shouldn't the work ambient illumination be set to support an ideal monitor display condition that
matches output?

Displays are the perceptual environment that is a user/operators keystone for everything that is reproduced, what adjustments are made on-screen become the factors in the file that is reproduced in output, if you can't see it accurately you can't control output accurately. And I believe anyone truly familiar with color management should realize that the display and its calibration and profiling is the accuracy hub of any color managed system.

But lightness/darkness in output results are not addressed by color management, there is no brightness/density information in an ICC profile, just color information. Brightness/density correlate information is in the image file data as the result of perceptual adjustment of the image on-screen. What's on screen is recorded as brightness, contrast and curve information in the image file which results in density of ink relative to the reflectance value of the substrate.

Finally gamma problems can cause output problems, but to a large degree that is the result of using discrete gamma values in setup. This can be resolved at the display monitor end by calibrating with L* gamma.

As for banding in displays it was eliminated in professional models with now 12 or 14 bit display LUT's.

 

[Lukas Engqvist]

We are slighting to an off topic since this has to do with the calibration of the monitor not the monitor itself. during the calibration process there is a setting of white and black points. If these are not set taking viewing booth and ambient lighting into acount, or even more if once the profile is made the monitor is adjusted by the user because "it was too dark" then ofcourse there will be errors between screen and output, but this i would again classify as a workflow issue, not a monitor issue.

I think we are not in such great disagreement as it may seem. Yes it would make sense to set the working environment to the monitor ideal…Â*but who does? There is very few people who have the ability to control their working and viewing environments in every detail (I have seen many workplaces where the ambient light varies with the time of day). The important part is to understand what factors are important, and then use a common sense approach to see wich you need to adjust. I have not tested as many monitor as you have, fotografx, but if you measure the paper in a given viewing booth you may be able to get a cd/m2 value.
I though that was the purpose of the XYZ tables. Could you explain what you mean there is no lightness/darkness in colour management? Are you talking about perceptual rendering? I though that was the whole point of using proof colours in photoshop, with simulate paper white and ink black? This is why I say it is a workflow issue rather than a monitor issue. I think it would be more profitable to get a good discussion by opening a discussion on "why prints are too dark compared to monitor" in the colour management section. There are those much more knowlegable than me.
Let me also say I have seen quite a variation of results from light to dark with the same monitor and different settings in the calibration program, and different software. (I personally prefer MonacoOptix where I can also get a verification of the profile.)

The banding issue I take up because many of the people I deal with consider a monitor good enough and buy it, then a couple of days later comes the question about banding.