NEC Monitor Set @ 4850/2.2/72.5?

[Photolandscape]

I have an NEC 2690 monitor, a Mac G5, CS3, an Epson 3800 printer with all Epson inks. I view my prints in a small room, next to my monitor, lit with 4700K halogen bulbs. I've used "canned" paper profiles, paid for custom profiles, and this weekend used my employer's very pricey Eye-One equipment to profile the 5-6 papers I generally stick to. I use NEC's Spectraview profiling software to profile my monitor. It's a great monitor capable of transmitting an image almost too beautifully--far beyond what a reflective print can offer.

For the past several years I have been unable to achieve an acceptable color match between monitor and print output. I have read repeatedly that when profiling an LCD monitor, you're best off setting it at 6500/2.2/120 (or so) luminance.

Since the beginning, these customary settings haven't allowed me to get anywhere close to matching my monitor and my print. So for now, at least, I decided to say the hell with 6500/2.2/120 and started experimenting.

For the first time in longer than I can remember, I am achieving a very, very close match between print output and monitor. What did I do to achieve this match--simple, I experimented, factored in the ambient lighting color temperature, chucked all the customary settings and came up with my own settings of 4850K/2.2/72.5. Of course the image on the monitor isn't terribly bright, and the color balance is as you'd expect it to be at 4850K. While it could use a couple of tiny tweaks, it really works.

So here's my question: since the color temperature and brightness of the light in your viewing environment make such a tremendous difference, at least in this case, why are we so adamant about 6500K/2.2/120 or 140 settings on an LCD monitor? I guess I am too literal minded, but sticking with this formula has cost me a lot of time and money, not to mention frustration.

I'd really be interested in hearing other peoples' thoughts on this. Thanks.

[DavidJ]

My understanding was that D50 is much nearer the print profiles and it is the recommended setting in the Spectraview manual for graphics work and anything ICC managed. Since switchng to D50 from D65 I find a good colour match between screen and print. Looks as though you have arrived at nearly the same point. My calibration using Spectraview software is D50 L* 140 cdm and this is giving a good match to my prints. (using 16 bit LUT)

David

[bill t.]

For my 2690, some of the snappier papers like Epson Premium Luster give me a much better screen-to-print if I put the SprectraviewII numbers at 6500K, 90 or 100 cdm, gamma 2.3 or 2.35.  For most matte papers, gamma 2.2 is about right on my Vista64 system.

Your 4850K setting seems odd, I'm getting a great color matches on Luster at 6500K, basically perfect except for the gamut barriers.  Off course the matte paper and art papers and general each has its own little twist, colorwise.

Whatever else, the richness of these high gamut monitors takes some getting used to for us old inkjet jocks.   I have also experimented with creating a top level curve layer with a "compensation curve" for certain difficult papers, the curve adjusted empirically to create a good screen-print match only while editing...disable the compensation curve for printing.

[Czornyj]

Your 4850K setting seems odd, I'm getting a great color matches on Luster at 6500K

Photolandscape is using Solux 4700K lightbulbs. Maybe you're just using a different light source?

[Tim Lookingbill]

I can give you only an assumptive answer based on several years of observation calibrating and matching color seen mostly on close to sRGB gamut displays on both CRT's and LCD's to inkjet and minilab prints under 5000K fluorescent tubes and outdoor daylight.

The newer, higher gamut and brighter displays have an issue with saturation levels out of the box due to their extreme brightness. You're mentioning the NEC being more beautiful than a print can render seems to support this. Not sure because I've never seen an NEC, yet and I don't know if you're using Soft Proofing.

Since saturation appearance is tied to brightness levels AND color temp appearance you are basically trying to reduce the dynamics and gamut of your display to match your warmish looking dim light source and ink on paper instead of letting soft proofing do it for you.

However, I also question if you're calibration may be messed up or corrupted by the system or software integration which can screw up the appearance of hue and/or saturation levels by corrupting the color tables that control this. The final calibration and subsequent profile is suppose to render color as it is suppose to be seen and not as more beautiful than a print can render. I mean your image viewed in Photoshop on your calibrated and profiled NEC should look the same on my calibrated and profiled 2004 G5 iMac. It shouldn't look more beautiful whatever that means.

I don't understand how you can drop the luminance of your display lower than I have my 2004 G5 iMac which is set at 6100K-(current native), 2.2 and 90cdm/2. My images match close enough to prints from several different printers such as a Noritsu minilab from my local grocery store, Epson 2200 from a pro photographer in another state and Epson 1270 viewed under 5000K GE Sunshine T8 fluorescent tubes.

The balance of saturation, contrast and color temp created by the standard 6500K, 2.2, 120cdm/2 has a lot to do with retaining the full gamut and dynamics of displays that are already near sRGB, a space the majority of printers base their hues and density levels upon but vary neutrality. Saturation levels are subject to gamut which is controlled by brightness levels.

The luminance number can be adjusted to match your viewing environment. I would suggest keeping all your other settings as is but raise the luminance level because soft proofing can take care of the rest. I can't believe you work in a more dim environment than mine when it's not necessary. I have three 18" GE Sunshines tubes one on each side of my iMac and one under the table I only use to view prints up close and vary brightness by moving the print away from the light. The NEC's brightness level allows warmer color temps without sacrificing brightness unlike CRT's and low end LCD's.

Have you tried using the 6500K, 2.2, 120cdm/2 standard and see if soft proofing gets you closer?

But hey, if it ain't broke, don't fix it.

[jackbingham]

The problem is that the notion of a standard has been foisted on this list and many others and it does not exist. Each user has to find his or her own white point, color and luminance based on lighting conditions and perception. Following blindly along with some experts "standard" is just silly. Experiment if you don't have a match. You will find a group of setting that work for you. That is all there is too it.

[Tim Lookingbill]

So you think 71cdm/2 is light enough to do adequate editing in? I mean I consider myself as working in cave-like conditions, but that just seems way too dark to be able to see down into the darkest shadow detail.

But if it works for you, go for it.

[rdonson]

So here's my question: since the color temperature and brightness of the light in your viewing environment make such a tremendous difference, at least in this case, why are we so adamant about 6500K/2.2/120 or 140 settings on an LCD monitor? I guess I am too literal minded, but sticking with this formula has cost me a lot of time and money, not to mention frustration.

I'd really be interested in hearing other peoples' thoughts on this. Thanks.
[a href=\"index.php?act=findpost&pid=219232\"][{POST_SNAPBACK}][/a]

You don't mention your softproofing technique in Photoshop.  That's the only way I know of for a monitor to emulate what the output to paper will resemble.

[Nill Toulme]

... Each user has to find his or her own white point, color and luminance based on lighting conditions and perception. Following blindly along with some experts "standard" is just silly. Experiment if you don't have a match. You will find a group of setting that work for you. That is all there is too it.
[a href=\"index.php?act=findpost&pid=220047\"][{POST_SNAPBACK}][/a]

Well that's fine if you're only producing prints to look at by yourself in your own environment.  But you might have green walls and an orange carpet and low voltage warmish lighting and be half color blind.  Do you still want settings that "work for you" or something that's at least in the ballpark of, if not "standards," what most in the industry are using?

Nill

[jackbingham]

Actually Nill I think your possible criteria rule out serious monitor calibration all together. If you are looking for a good monitor to print match with bright green walls you probably are in the wrong business. So lets please look at this with real world conditions in mind. If you have 4800k bulbs that you use for viewing and your monitor is set at d65 there are many among us that will see the monitor as much bluer than the print. On the other hand if we set 4800k as the monitor color temperature there will be many among us who now see the monitor as too warm. The problem is no one will agree. So since the question was how do we match monitors to prints, we experiment with our personal preferences and perceptions. If we have green walls and are half colorblind perhaps we should be doing something else with our time.
I tend to think that d65 is fine if you are only producing prints for yourself. If you are selling imagery for reproduction and you are concerned with critical color matching you should throw away the "standard." Get yourself a daylight viewing booth and match the monitor color temperature to it.
If you find this match not quite to your liking make subtle adjustments to the monitor profile targets to get to what you think is a match. Or trust the instrument and adapt. Consistency is the most critical element

[Tim Lookingbill]

The question I have in making the appearance of the display's color temp match the appearance of the color temp of the viewing light is will the calibration software create the correct adaptation algorithm within the final profile which can have  a noticeable affect on the appearance of hue/saturation in CM images.

I've noticed using my i1 Display if I leave color temp at native which already looks neutral on my 2004 iMac, the software reads it at 6200K. If I choose a target of 6500K which is pretty close to native, the display takes on a noticeable but slight coolish magenta tint. I surmise from this that the appearance according to the number of the color temp is subjective even within software much like the color temp number/appearance in raw converters. If the software is this much off on something as slight as a 300K difference I wonder how it will fare with interpreting a color temp set by eye to something as far from standard as 4800K.

IOW, if by making a custom color temp to match a light source who's appearance is unknown to the calibration software, how much will this affect the software in arriving at proper adaptation encoding within the final profile?

[Nill Toulme]

I was exaggerating to make the point.  If you adjust your monitor and your output to match your particular viewing conditions, what guarantee do you have that those conditions match up with those at the print's ultimate destination?

Nill

[jackbingham]

I was exaggerating to make the point.  If you adjust your monitor and your output to match your particular viewing conditions, what guarantee do you have that those conditions match up with those at the print's ultimate destination?

Nill
[a href=\"index.php?act=findpost&pid=220453\"][{POST_SNAPBACK}][/a]

I'm a bit confused by your implication that somehow choosing one of these approaches over another, say using d65 as the "standard" vs matching light source and monitor profile,  will somehow better guarantee a match outside the building. Unless you know something about what's outside the building your only guessing anyway.
The problem I have with everyone saying d65 is the standard is that it's a bit like saying f8 is the aperture everyone should use. Your mileage will vary and these tools should give you what you need to experiment and find something that works for you.
In the end what is outside the building is irrelevant. If you want to match screen and print you need a print profile targeted to the illuminant your are viewing under and your monitor profile should match that too. If you have lighting that is too yellow, say 4500k and your monitor is set to d65 you will either thing the prit is too yellow or the screen is too blue. Now what do you adjust and how does this relate to what's outside the building?

[jackbingham]

You could also surmise that this Eye One has a problem

[Nill Toulme]

In the end what is outside the building is irrelevant. If you want to match screen and print you need a print profile targeted to the illuminant your are viewing under and your monitor profile should match that too. ...
[a href=\"index.php?act=findpost&pid=220551\"][{POST_SNAPBACK}][/a]

My point was that what you say is true only if "the illuminant your are viewing under" is the only way the print is going to be viewed, or if that condition is reasonably close to viewing conditions at the print's ultimate destination.

Nill

[jackbingham]

My point was that what you say is true only if "the illuminant your are viewing under" is the only way the print is going to be viewed, or if that condition is reasonably close to viewing conditions at the print's ultimate destination.

Nill
[a href=\"index.php?act=findpost&pid=220577\"][{POST_SNAPBACK}][/a]

And if that's not true................then what. There is no standard and assuming d65 is makes little sense. Are you saying d65 is a more flexible choice? That somehow things look better using d65 regardless of the viewing condition? Nowhere is it written that the ultimate destination is more likely to be d65. On the contrary most galleries probably have warmer light, offices, warm white or cool white fluorescent and commercial printers are generally at d50. None of these are anywhere near d65.
I think the point you are trying to make is that d65 is the best choice if you don't know where you are going. If that is the point, lets turn this interrogation around and have you explain why it is the right choice. If that is not your point, please clarify

[Nill Toulme]

I believe that is indeed my point, unless you are printing for a specific known lighting condition that dictates something different.

This forum is rife with discussions of why D65 is, if not truly a "standard" (which it of course is not), the best default choice.  Try a search.

Nill

[jackbingham]

I believe that is indeed my point, unless you are printing for a specific known lighting condition that dictates something different.

This forum is rife with discussions of why D65 is, if not truly a "standard" (which it of course is not), the best default choice.  Try a search.

Nill
[a href=\"index.php?act=findpost&pid=220597\"][{POST_SNAPBACK}][/a]

So I'm asking you why you think d65 will work better as a standard than anything else and you want me to go search?

[Nill Toulme]

Well, um, actually... yes.  Andrew Rodney has explained it so much better than I could ever hope to.  Search on posts by digitaldog.

Nill

[Tim Lookingbill]

What color hue is D65, D50, 5000K, 4800K, 6500K? Have the appearance of these numbers been established and agreed upon?

No one has yet showed me or told me. It's all interpretive.

So much for everyone being on the same page in regards to color management.