I have a two-month-old NEC 2690 LCD monitor with Spectraview II and the NEC/Gretag calibration device, a Mac G5 with OSX 10.4.9, printing to an Epson 3800. I view my prints under 4700K Solux halogen bulbs. Printing these days on Epson Premium Luster, using Epson's furnished profile. Also tried the BookSmartStudio Premium Luster profile, but found it to be virtually identical to the Epson profile.

Monitor settings are D65/2.2/95. Why 95? I find that even at 95, the screen image is slightly brighter, slightly contrastier, and slightly bluer in the shadows.

I'd appreciate any suggestions for getting a better match. I'm still at a point where what I see on the monitor isn't what I get off the printer (which is operating flawlessly), and I'm wasting ink and paper compensating for the difference between monitor and print.

I thought about trying the Native White Point Setting, but using it will not allow you to control the brightness manually--Spectraview II sets it for you at about 400--way too bright. There is another setting called Native Monitor Gamma, I believe. Would that help in any way?

I'm pretty clueless on what to try next.

Thanks.

I have the 2690 also. Screen to print density match is working the same for me. Image on screen is brighter than prints. I'm using the Solux 4700's also. That has definitely made a difference but still not perfect. I mean just how much light are you supposed to shine on a print to match the screen? There is a limit. After all, normal viewing of a print is done in light not nearly as bright. I am beginning to think that you just have to get used to compensating by sight how bright the image should look on screen to match a printer. There are other threads on LL about this. 90 cd is too low for the monitor. If you check the other threads its pretty much been established that 120-150 is where you should be. I don't really have an answer. This is also my first lcd.

First, no screen is going to be an exact match when it comes to density. The screen is a transmissive device and the paper is reflective. You also need to take into account paper whiteness, this varies greatly from paper to paper. Second, the Solux viewing that you are using is not as accurate compared to a professional print viewer or light booth. 3rd, just tweak the profile to your taste and/or make slight adjustments in PS via adjustment layers that you can turn off or discard to more closely match if you're really that concerned/far off. Finally, most people today are using D50 rather than D65 for at home printing - think about the color temp differences you are viewing 6500 vs... Finally, I'd suggest viewing the print on a bright, overcast day and see if you find your results acceptable or not, at least to compare and contrast. There is so much subjectivity and so many different light sources that the print may or may not be viewed in.

Except for the ability to turn up or down the intensity I think Solux lights are actually more color accurate than a viewing booth with fluorescent bulbs. At least thats the impression I get here on LL from other posts. I'm going to try a D50 calibration to see how it looks.

Thanks. I agree and understand every point you are making. I may try D50 instead of D65. What I find frustrating is that I once had an Epson 2200 and a Mitsubishi 22" CRT monitor, and on several different papers I consistently achieved virtually dead-on matches between screen and print. Sometimes I wonder whether as everything has become more sophisticated and we've added so many bells and whistles and can do a million different things, we can't necessarily do something basic?

Besides brightness, the other thing I am seeing is more blue on the screen--particularly in what we in the offset printing industry refer to as the mid-tones and 3/4 tones--that is, where you would see a 30-60% cyan dot, and a 60-75% dot. In the deeper shadows and brighter areas/highlights, it's not a problem. I just hate to have to make layer adjustments for printing when I have what I want on the screen and want to see the same thing on paper.

> Except for the ability to turn up or down the intensity I think Solux lights are actually more color accurate than a viewing booth with fluorescent bulbs. At least thats the impression I get here on LL from other posts.

See also "Light under control" by Danny Pascale and Roger Breton:
http://www.babelcolor.com/download/Light_u..._2005-11-08.pdf

95cd/m2 is way, way too low for this unit, especially a new one. You'd be hard pressed to get 120 without letting it burn in for 100 hours or more.

The luminance should, if possible in this case, match the viewing of the print itself. Are your Solux lamps really far away? Get em closer. Also be sure you're setting up the soft proof correctly using the Paper White and Ink Black simulation.



Set luminance first, then play with white point, starting at D65 and moving higher or lower based on the print to screen match.

I did at least part of what you suggested tonight. Crept up to 110cd/m2. I have 6 35w Solux bulbs on a track about 5 feet above a small desktop easel where I study my prints--2 of the bulbs are aimed directly at the print, while the other six light up the white walls of the very small room I work in--6 x 12 feet.

Just for kicks, I did a little experiment. After profiling the monitor (D65/2.2/110). Made a print--but it looked fairly dim compared to the screen. So--I did a little manual tweak by adjusting the contrast using the adjustment buttons on the monitor--dropped it from 50% to 31%. The match is far closer than it was. So far I am encouraged, but is there some reason why I shouldn't go down this road?

Thanks for your help.

I also have the Spectraview 2690, I'm printing to both an Epson 4800 and 2100 (2200) but I must say that my old Barco that I have next to it looks more acurate to my prints... And the Barco I can hardly calibrate anymore.

So if the 2690 looks lighter than both my Barco and my prints how will my stuff look once I sent it to the printer for my jobs? When I just worked with my Barco I was most of the times satisfied with what came out of the printers, even if i didn't CMYK myself. But it's very rare that I felt that my images camew out too dark or too light. So now I'm worried with the 2690 if it's looks good at it will it still look good when my stuff gets printed???

Regards,
Håkan

Had a Barco for years (you actually can run the software???). Anyway, even new, out of the box, you'd probably never see more than 100 cd/m2 out of a CRT. That's really tough for a new LCD to hit (way too low). So again, up the lighting on the prints. Or let that LCD run (with a screen saver) for a few months. Some of these pups can hit 300 cd/m2 out of the box. You need sun glasses.

Yepp I can run Calibrator Talk for Barco! I'm on my old G5 with 10.4.11. Wouldn't try it on my new MacPro... The 2690 is on my new MacPro. But the computer is in for repair since it dies all the time, they say it's the graphic card that Apple need to fix the driver for. But Apple isn't admitting that there's a problem until they have a solution for it... So I'm pissed at Apple right now! And as a stock owner I'm even more pissed how they handle matters like this!

So if the LCD is too bright I can't trust it visual? That's what I love with the Barco that I can judge more or less on the screen how it's coming out once it gets printed. Off course not exactly but as it seems closer than the 2690 if I haven'ät missunderstood everything...

When you calibrate a display, one of the targets is luminance. You need to match that to the viewing conditions. In the old days, CRT's didn't pump out much and we worked in dark caves. As to viewing the prints, the better solutions have digital dimmers (I have one on my GTI SOFV box). Solux are better in terms of light quality but worse in terms of control. Bottom line is, you want to set the luminance at calibration to match visually the viewing conditions. Now the one issue with new LCD's is they are so damn bright out of the box. Like I said, you can drive them at 200-250 cd/m2 all day long. 120cd/m2 isn't that high for such units and a new one may have difficulty hitting that (certainly something like the SpectraView). I'd try 120-150 to start. As it ages, it will be easier to lower that value (and it will last that much longer). But the easy solution is to raise the lighting intensity on the prints (assuming you can keep the color the same).

I have the GTI SOFV-2e, I ave it dimmed down to 50, I guess that's 50% since it goes up to 100 max.

My NEC 2690 I have at 150 cd/m2 calibrated after the receipe I got from you with both Eye-One Photo and i1 Display.

But since everything is lighter on my 2690 than my prints or on my old Barco that is at 80 cd/m2. So my questions now is if I always have to make prints and judge the darkness/lightness from that or could I ever like with my Barco just try to judge that on my screen on the LCD?

I don't want the printers or clients starting to complain about my images if I've tried to judge my images wrongly on my 2690...

First, you can't have both the Barco and NEC working for you due to the huge differences in luminance. It makes sense that the Barco and NEC don't match. And due to the huge differences, you can't up the Barco to the lower end settings on the NEC and vise versa. So concentrate on one or the other and adjust your light booth intensity and display to visually match.

I understand that I can't have both the Barco at the same time, once I have my new MacPro back and working with the NEC 2690, the old G5 with Barco is going to rest. Since I just got the MacPro and NEC setup off course I compare it to my old setup to learn the difference...

And I do understand that I should change the way I look at my prints to be able to judge them. But what worries me is since the LCD is so much lighter and I have to look on my prints under a brighter light. But the file will still be the same so that means that my files will be darker than the ones produced with the Barco on the old G5? SInce I'm compensating the lighter screen by making the images darker. So what will happen when I send those to the printer? Will all my stuff be darker now? I'm sure I must be missing something here...?

The RGB numbers haven't changed (which is good). You've increased the luminance output on the display and the proofing station as well. Numbers are the same, display and prints match. That's all good.

Going back to CRT. Numbers are the same, luminance of calibrated display is lower. Light box is then lowered. Same numbers, output and display match which is all good.

But since it will be so much brighter on the LCD I would think I would compensate for that making my images darker than they have to be?

If I have an image that I edited on the Barco/G5 and I'm satisfied with, if I open that one on my 2690/MacPro then it looks too light... So that would mean I would have to make it darker to compensate for that. I haven't tried the other way around yet. But as I wrote before once my images gets printed (offset) then it normaly looks good, not too light, not too dark.

I'm still sure I'm missing something here.........

Do they look too light and washed out? Do they match the print?

Compared to the Barco they look slightly too washed out. Compared to the print I can only say by memory since the MacPro is in the repair shop... But from memory it was slightly lighter than the print. The colors didn't match exactly but I think I have to make new profiles for my new computer.

Andrew,

In the best case scenario would it be better to have an LCD that calibrates down to 95-110 rather than 120-150 if print matching is important?

My CG19 Eizo seems to have no problem working at the lower numbers.

Should I assume that the the Eizo 24" CG241W would work better for print matching that the NEC 2690? Would the Eizo also keep a larger Gamut of colors than the NEC at the lower luminance settings?

It seems strange to me that the Nec 2690 is geared to imaging professionals but cannot be calibrated down to industry standard luminance levels.

Are people compromising in the purchase of a 2690 for a good price because it's "good enough"?

My instinct tells me to just spend the extra money on the CG241W, work at 110 lumens, and forget about it.

Its not better or worse. Again, you set luminance based on the ambient light around the display and how you'll view the prints. You want the display to be the lightest object in view so, if the ambient light is higher, you need a higher luminance. CRT's can't produce high luminance so we had to work in dark caves. Now we don't (although darker is always better). So instead of say 16-25 lux (very dim) for a CRT producing roughly 90 cd/m2, we can work in an office environment of much higher luminance since the LCD's can produce much higher values than that CRT.



There's probably no difference in what the minimal luminance of both devices, new out of the box can reasonably hit. And the older the display, the 'easier' its going to be to hit a lower value.

120 cd/m2 is just at the lower limit of what the (brand new) display can adjust itself to using the backlight alone. If you can go higher, to say 150 or so, it means the display can compensate for when the brightness drifts up instead of down (for example when you first turn it on). But you CAN calibrate to 120 cd/m2. Its just not a big deal to up that to 150, not live in as dark a cave and up the box to match if necessary.

Thanks Andrew,

I'm no expert, but had thought that using a higher setting like 150 cd/m2 would give the monitor a higher apparent contrast than working at a lower setting like 110 cd/m2. The result would be flat looking prints. I had made up in my head that the commonly accepted values of 90-120 cd/m2 were as much about controlling the perception of contrast in the images as it was to match the lighting in the room.

Just this week I'm retouching a bunch of images on black plexi. With the room lights on and the monitor set to 150 cd/m2 I find it difficult to see into the black values. With the room lights fairly dim and the monitor set to 110 cd/m2 it is much easier to control the black values in the images.

I must say I don't understand the concept of adjusting the print viewing lighting to match the monitor. It seems to me that in many if not most cases, prints are going to be viewed in a variety of lighting over which the photographer/printmaker will have no control.

And I've had no trouble calibrating my NEC 2090uxi to 95 cd/m² since the day it came out of the box. In fact, I'm thinking of dropping it to 90.

Nill
~~
www.toulme.net

Having tested stacks of different monitors there are clear differences in the lower luminance targets they can hit with accuracy and stability. Many eizo's are inherently darker because of the uniformity function they apply across the screen. From our tests the 2690 performs better at higher luminances, meaning hitting a specific white point and gray balance from white to black improves the brighter you are above 120. Again this is not true of all lcds but specifically the 2690.
As for viewing conditions we need to understand that we are forced to work with a sliding scale here. New monitors will inherently be brighter. Having two monitors of substantially different luminances will only screw this up. You need to pick a monitor as your standard and use a light source that you can adjust so that you can match the monitor luminance and viewing luminance. After that the best thing to do is get comfortable. Comparing old tools with new conditions just muddies the waters. Sticking with the new condition will allow you to adapt.

Is that similar at all to NEC's "ColorComp" function? (I think I'm remembering the name right.) Note that function comes turned off by default.

Nill
~~
www.toulme.net

honestly I'm not familiar with their colorcomp function. Eizo however applies an eveness adjustment by default which has the additional effect of making the max luminance of the monitor considerably darker than it would otherwise be.

It sounds similar. As I recall, overall brightness went down a bit when I enabled ColorComp on my NEC.

Nill
~~
www.toulme.net

What Andrew says is critical to understand...with CRTs you had both brightness and gain as controls to adjust the black level of the display. With LCDs, you ONLY have luminance output. Adjusting for black is a matter of adjusting the display environment. There are ISO standards (it's copyrighted so I can't just post them here) but you need to adjust the ambient lighting of the environment in order to create an optimal viewing environment. It's less important to pick an actual luminance output to the display than it is to adjust the surrounding lighting to make sure you are seeing the range of tones on the display correctly.

I'm running a NEC 2690 at 200 cd/m²...yet the black level is determined by the ambient lighting-which because the display is running so high is much, much brighter (I haven't done an actual reading so I don't know how bright it is at the moment).

One way to check this is to do a step wedge that goes from 0, 0, 0, to 15, 15, 15, in 1-2 units/swatch. (I say 1-2 because depending on how you profile, you may not be able to see a 1 unit change). Open the image in Photoshop and go to the full screen black and tab to hide Photoshop. You should be able to see almost nothing. Then, using the display profile you've made at the luminance target you've made, look at the image. You should NOT be able to see any difference between the surrounding black screen and the 0, 0, 0, swatches. If the 0, 0, 0, is lighter than the black surround, then you're driving the luminance too high. Either lower the display output or raise the ambient lighting. Note, you should NEVER have direct spectral light fall on the display...if need be, add a shroud around the display to shield it. Also note that the surrounding color should be neutral and ideally, the ambient lighting should be in the range of matching D50-D65. I use GE Chroma 50 fluorescent aimed up into the ceiling to bounce D50 lighting at a low'ish level in my imaging area.

The display is set to 90º to my GTI lightbox...when I put an image up on the display to softproof (and use a full screen black with no Photoshop UI visible) and put a print on the lightbox, I can see a very close match...when I turn soft proofing off, I actually don't see a close match because the dynamic range of the display (the NEC) is WAY beyond what the print can show. But that's ok...it's when softproofing that I really care.