• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

SpectraView Users, Settings Screenshot

The questions are these:

  1. Do these settings make sense?
  2. What are your settings?

My hope is that the SpectraView users that read this will upload screenshots of these settings A) post-calibration Summary, B) pre-calibration Target and C) software Preferences. I hope this will make a great reference and discussion.

MY CALIBRATION GOALS
In addition to accurate monitor colors, my goal is to use settings that will enable me to create reasonably pleasing prints from a commercial printer (such as Cosco), most of the time, without soft-proofing. My understanding is that one of the keys to this is keeping the luminosity and contrast ratio somewhat lower than what the monitor is capable of.

All that said, here are my current setting screenshots:

TARGET SETTINGS
spectraview-tar.jpg


POST-CALIBRATION SUMMARY
spectraview-sum.jpg


PREFERENCES
spectraview-pref.jpg



Thoughts?
 
Understood. Where does your contrast ratio end up?

The Eye-One Match software I use doesn't report that as such. It reports the display luminance (Black level, which I can set on my display, and the maximum output level) expressed in candelas (cd/m2). When I set a target of 120 and a blackpoint of 0.5 then my contast would be 240:1.

I don't know if you can manually set the blackpoint on the LG monitor, but if you can then it depends on you viewing conditions at what level you should set it. It might need a higher BP in a normally lit room or else you won't be able to judge shadows. In a very dark room it can probably be dopped as far as the monitor allows. However, I suggest trying a few settings for your specific work environment, and look at shadow definition.

The contrast is not a goal for me, I just want to be able and judge shadow detail and have a display brightness that allows me to previsualize how dark my prints will become.

Bart
 

Rod Witten

pro member
Hi Ed,

A target intensity of 150 cd/m2 seems a bit high for my taste. I target something like 110 or 120 cd/m2 (on LCD, 100 cd/m2 on CRT).

Bart

Edward,

I agree with Bart about the luminosity - my 2690 is calibrated at 100 lum with contrast targeted for 400. The actual contrast reading was 238 after the calibration. I ran the calibration in a fairly dark room.

Rod
 

Cem_Usakligil

Well-known member
Edward,

I agree with Bart about the luminosity - my 2690 is calibrated at 100 lum with contrast targeted for 400. The actual contrast reading was 238 after the calibration. I ran the calibration in a fairly dark room.

Rod
While I agree with both Bart and Rod, I do not think that 100 cd/m2 is realistic for modern LCD monitors. I do not have the NEC but with my Dell monitors, I cannot go much lower than 120-130. To do that, I have to set the brightness of the monitor to 0 and also pull down the individual RGB channels to lower values. I have discovered that this causes color saturation and fidelity issues later. So I now stick to calibrating at 140 cd/m2 which works just fine for me. If I calibrate to 150 cd/m2, then the prints come out a bit darker than what I see on the monitor (as expected). YMMV.

Cheers,
 

Andrew Rodney

New member
I'm using basically the same settings and 150cd/m2 is fine for me in terms of matching to my GTI Lightbox (dimmer set to 50%). The values is totally up to you in terms of matching the nearby print, YMMV.

As to the contrast ratio, I've seen in the past, with older versions (on Mac) that I don't always exactly hit the values but I get close. I'm working with a new build that should be available any day now, if not already, it seems to hit the target closer.
 
Interesting questions, Ed

(I tried posting images but couldn't get them to work.)

Calibration Target Configuration: I have 120cd/m2 and Monitor Default contract ratio (I think the latter was the default).

Calibration info: My white point target is 6506K, calibrated is 6598K and DeltaE 1.46 (That DeltaE is probably a little high and may indicate scope for improvement).
Display: Target 120; Calibrated 0.24/ 116.1​
Contrast Ratio (Monitor Default) comes out as 484.1​
The manual seems to imply that you only reduce contrast for special purposes:​
"In some applications, such as simulating low contrast images such as in newsprint, it is useful to have a lower contrast ratio than the display’s native value. Contrast Ratios in the range of 50:1 to 500:1 can be selected." (p15)​

Preferences: Average low light measurements unchecked
Calibration & profile steps: 32 (default)​
Calibration Priority: Maximise Contrast ratio (Default - would "best greyscale colour tracking" be where your primary purpose is monochrome printing?)​

I should at least test whether I get a better profile by checking "Average low light measurements " or increasing the number of steps.

Regards,
Murray
 
Calibration Target Configuration: I have 120cd/m2 and Monitor Default contract ratio (I think the latter was the default).
When I use "Monitor Default", I end up with a post-calibration contrast ratio of something between 200:1 and 300:1

Calibration info: My white point target is 6506K, calibrated is 6598K and DeltaE 1.46 (That DeltaE is probably a little high and may indicate scope for improvement).
My DeltaE enters into this range when I move my target Intensity lower - this number seems to improve when I use intensity levels up around, say, 150.

Display: Target 120; Calibrated 0.24/ 116.1​
I find my post-calibration Intensity is closer to my Target settings when I use a higher number for the Calibration Steps

Contrast Ratio (Monitor Default) comes out as 484.1​
That is higher than what I experience, but I assume that depends on the monitor being profiled - I have the 2490, what model do you have?

"In some applications, such as simulating low contrast images such as in newsprint, it is useful to have a lower contrast ratio than the display’s native value. Contrast Ratios in the range of 50:1 to 500:1 can be selected." (p15)​
I thought 150:1 would be a good compromise, based on the DeltaE values I was seeing through trial and error and an assumed print contrast ratio of 80:1 ?

Also, I have another monitor that I calibrate but I can't adjust its blackpoint, so I usually end up with contrast ratios of 300:1. So, after reading comments from others, it looks like a contrast ratio of between 200:1 and 300:1 is fairly common amongst professionals?

Preferences: Average low light measurements unchecked
I'm not sure about this either. Hopefully someone who knows more about this setting can add more here.

Calibration Priority: Maximise Contrast ratio (Default - would "best greyscale colour tracking" be where your primary purpose is monochrome printing?)​
Again, I haven't had much experience with this, but I assume better greyscale tracking would carry over and be beneficial to your color images also... My experience is that when using this setting your post-calibration Contrast ratio could end up lower?
 
SpectraView - New Version (1.1.00 Build 090114)

If I read the build number right, it looks like the new software was released on January 14, 2009.

Andrew Rodney mentioned there was a new version of the SpectraView software so I downloaded it and re-calibrated my NEC 2490.

I found that some options have been added and some removed. I list these after the screenshot.

LATEST POST-CALIBRATION SUMMARY

spectraview-sum.3.jpg


I'm still not perfectly happy with the 344:1 contrast ratio. In the software, you can choose between better grayscale color tracking or a target contrast ratio. In other words when you choose a specific target contrast ratio, "better grayscale color tracking" is not used, even if its checked. From what I've gathered, it seems that many of those with more experience than me choose "better grayscale color tracking" over a target contrast ratio and live with the resulting contrast ratio?

PREFERENCE SCREEN CHANGES

Whereas the options were all on one screen, they have been broke out into separate tabs and each option is now given a much more thorough explanation. This should be a much more novice-friendly screen as a result. Here is a listing of changes on the screen:


REMOVED OPTION: For the Calibration and Profile steps, the "256" option has been eliminated.

CHANGED NAME: "Average low light measurements" changed to "Average low luminance measurements"

ADDED FEATURE: Extended luminance stabilization time

ADDED OPTION: Under the feature for setting the source of primary color chromaticites for the ICC Profile, the "Automatic" option was added. From my understanding, the sensors measurements are used unless the system detects a standard gamut sensor being used on a wide gamut display, in which case, it will use factory production measurements for your monitor.

If would love it if anyone could speak to whether they use "better grayscale color tracking" or a target contrast ratio. Also, does anyone use the "auto-luminance" feature?
 
That's very interesting, Ed. I'll have to download and try it. There's probably a new manual to download as well.

Murray Foote said:
Contrast Ratio (Monitor Default) comes out as 484.1
That is higher than what I experience, but I assume that depends on the monitor being profiled - I have the 2490, what model do you have?

I have a 2690 (and use an XRite DTP94). I was part way through doing some trials with different settings. Now I'll have to download the new version and see how that goes.

This is where I had got to last night. All are at a D65 white point and 2.20 Gamma. I think the most indicative results must be the DeltaE of the White Point and the DeltaE for colour tracking. For the latter I had [Include Dark values] checked. I have also included Calibrated Black Level.

(1) Intensity: 120cd/m2, Contrast: Monitor Default, [Average low light measurements]: Off, Calibration steps: 32; Calibration priority: [Maximise Contrast Ratio]
DeltaE Results: White Point: 1.29; Colour: 1.07; Calibrated black level: 0.25.

(2) Intensity: 120cd/m2, Contrast: 200:1, [Average low light measurements]: Off, Calibration steps: 32; Calibration priority: [Maximise Contrast Ratio]
DeltaE Results: White Point: 1.08; Colour: 0.84; Calibrated black level: 0.74.

(3) Intensity: 120cd/m2, Contrast: 300:1, [Average low light measurements]: Off, Calibration steps: 32; Calibration priority: [Maximise Contrast Ratio]
DeltaE Results: White Point: 0.98; Colour: 0.83; Calibrated black level: 0.57.

(4) Intensity: 120cd/m2, Contrast: 300:1, [Average low light measurements]: Off, Calibration steps: 32; Calibration priority: [Best Greyscale Colour tracking]
DeltaE Results: White Point: 0.96; Colour: 0.72; Calibrated black level: 0.69.

(5) Intensity: 120cd/m2, Contrast: 300:1, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: [Best Greyscale Colour tracking]
DeltaE Results: White Point: 0.94; Colour: 0.72; Calibrated black level: 0.69.

Next I was going to increasing the number of steps and increasing the intensity.

Regards,
Murray
 
First calibration with the new software (at last settings)

(6) Intensity: 120cd/m2, Contrast: 300:1, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Not relevant
DeltaE Results: White Point: 0.95; Colour: 0.86 average, 1.12 maximum; Calibrated black level: 0.40.

Note, red and green colours in results compare to the previous calibration, rather than as an overall comment.

By the way, Ed, do you ever see anything in the DICOM conformance graphs? (Whether they are really useful may be another matter anyway).

I notice the interface says of [Best greyscale colour tracking] that "The option may provide better results if the display is used in a very dark environment and the target white point is not close to D65.". That seems to imply it's not normally required.

Regards,
Murray
 
Hi Murray - I have no idea what DICOM conformance is and no, I don't see anything there either.

I don't work in a VERY dark environment, and I don't know what "D65" means so I have no idea if my target white point is close to it!
 
OK, thank you. I found a little more on this in Wikipedia.

Also, as I should have replied in the other thread, thank you for pointing me to Spectraview. I am able to confirm that a European 2690 (purchased in this case in Australia) is compatible with US Spectraview II software (you just need to find someone in the US to buy it for you).

Regards,
Murray
 
Ed

Next I tried Contrast: Default (as it is now called) and best greyscale colour tracking.

(7) Intensity: 120cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 0.62; Colour: 0.48 (0.43 including dark values) average, 0.70 maximum; Calibrated black level: 0.34.

(A touch warm here at the moment - 38degC (100degF) - it has been around there for four or five days in a row and predicted to continue next week but at least that's not as bad as Melbourne and especially Adelaide where it has been as high as 46degC (114degF). Maybe global warming is kicking in.).

Regards,
Murray
 
Ed

Next I tried Contrast: Default (as it is now called) and best greyscale colour tracking.

(7) Intensity: 120cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 0.62; Colour: 0.48 (0.43 including dark values) average, 0.70 maximum; Calibrated black level: 0.34.

So... what does all that mean? I'll admit it, I was lazy and didn't go back and look up your other DeltaE measurements - how did they compare and which will you stick with and why?
 
Well it shouldn't be too hard for you to see how one compares against the other because the settings I changed from the last measurement are in bold blue, improved results are in bold green and poorer results are in bold red.

I still haven't finished the series and was going to leave comments mainly until the end but since you ask I'll make some up to this point.

The change in version halfway through complicates things because the new version might improve results more for some combination of settings than others. When I profiled with the new version using the settings from my most recent profile, there were differences. White point was about the same, Colour Tracking was worse and Black point was better.

We have read that best greyscale colour tracking makes no difference unless you choose Contrast = Monitor default (old version) or Default (new version). However when I changed to Best greyscale colour tracking at a contrast ratio of 300:1 there was some difference. White point was about the same, Colour Tracking was somewhat improved and black point was somewhat worse. Of course, to be rigorous it would be a good idea to run through the cases more than once to eliminate variation from extraneous factors and repeat all tests on the old software with tests on the new software (I doubt that I'll do that).

Setting Average low light measurements to on gave a slight improvement for white point and no difference for colour tracking or black point. This difference may not be significant.

I started off setting Contrast to Monitor default with calibration priority as Maximise contrast ratio. I found I got an improvement with setting contract ratio to 200:1 and then a further impovement setting it at 300:1. However, my last profile, with Contrast = Default and calibration priority = best greyscale colour tracking gave much better results than any of the previous setting combinations. So I think I'll stick with that.

Mind you, even the worst results were quite useable since the manual says a DeltaE of 1.0 is just perceptible to 50% of users. Presumably this means that noone could find distinctions at DeltaE values of say 0.5. Apart from initial accuracy, getting to around say 0.5 should mean the screen stays profiled longer.

I still have to try increasing the calibration steps and increasing the luminance. My preconception is that increasing the calibration steps will make little or no difference but increasing the luminance may make a significant difference. If this is so, then I won't necessarily go with the the luminance value that gives the best figures. One factor is that my other monitor is an NEC PF2141 CRT which last profiled at 118cd/m2. Profiling the 2690 at 120cd/m2 helps therefore for even display between the two monitors. However accuracy of printing is a more important criterion, especially for monochrome. I may end up printing some softproofed test images at different profile settings (using custom printer profiles) to see whether they come out darker or lighter than the screen. Should I happen to find a higher luminance is better for that (though I haven't particularly noticed any problems there) I suppose I might even have one profile for general viewing and another for soft proofing.

One other consideration is that although my tests might be useful for other people, they are not completely applicable. For example, you have a 2490 with 930 hours of use and sRGB gamut, whereas I have a 2690 with 3650 hours of use and aRGB gamut. Lighting, ambient temperature and other factors may also play a part.

Regards,
Murray
 
This was the last test:
(7) Intensity: 120cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 0.62; Colour: 0.48 (0.43 including dark values) average, 0.70 maximum; Calibrated black level: 0.34.
I still have to try increasing the calibration steps and increasing the luminance. My preconception is that increasing the calibration steps will make little or no difference but increasing the luminance may make a significant difference.
What I actually found was quite unexpected:

(8) Intensity: 120cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 52; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 1.46; Colour: 1.14 (0.96 including dark values) average, 1.58 maximum; Calibrated black level: 0.34.

Spectraview II Manual said:
A larger number of steps will generally result in a more accurate grayscale calibration, but will increase the calibration time. 32 steps is the recommended setting for most applications.
Here, increasing the steps from 32 to 52 has resulted in a significantly poorer profile. I found this so remarkable I reprofiled with 32 steps again, then 52 steps again, then 32 steps again:

(9) Intensity: 120cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 0.87; Colour: 0.76 (0.64 including dark values) average, 0.97 maximum; Calibrated black level: 0.35.

(10) Intensity: 120cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 52; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 1.00; Colour: 0.86 (0.72 including dark values) average, 1.15 maximum; Calibrated black level: 0.33.

(11) Intensity: 120cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 0.87; Colour: 0.74 (0.64 including dark values) average, 0.99 maximum; Calibrated black level: 0.34.

Tests 9 and 11 were both very similar but higher than Test 7 from the night before. The monitor was well warmed up in both cases so I guess it's something to do with the extremely warm temperatures we are having at the moment.

Test 10 had settled down somewhat from Test 8, perhaps also for reasons related to the record sustained high temperatures here, but profiling at 52 steps is showing worse results than testing at 32(!).

Next I tested at Intensity of 130cd/m2 and 140cd/m2 and once again the results were unexpected:

(12) Intensity: 130cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 0.97; Colour: 0.85 (0.72 including dark values) average, 1.14 maximum; Calibrated black level: 0.33.

(13) Intensity: 140cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 0.99; Colour: 0.86 (0.74 including dark values) average, 1.19 maximum; Calibrated black level: 0.36.

Profiling at 130cd/m2 and 140cd/m2 yields poorer results for me than profiling at 120cd/m2. This is presumably because my monitor is no longer new. Often it is difficult to reduce the brightness level too far on new monitors and still get goodresults.

So I've gone back to profile #7 and am using that. What I suspect all this shows is that everyone should do these sort of tests to find what works for them with their particular monitor and working environment. Also, I suspect that it may be a good idea to do some tests every 6 months or year (or when major upgrades of the software appear) to determine whether the optimal settings have changed for you. On the other hand, there's no point getting too concerned about small changes in the results that are below the level for human discernment.

Regards,
Murray
 
However accuracy of printing is a more important criterion, especially for monochrome.

It would be nice if there was a silver bullet type calibration that allowed you to edit your images and have them be perfectly ready for web and print, no fuss no muss. I guess the best you can aim for is a calibration that works well for your most common use (web/monitor for me) but also gets you in the ballpark for other uses (like printing). I'm going to have to think about developing a profile that I would switch to when working on a group of selected images for print. I would switch to that profile, make virtual copies and work on the images that way. This might allow me to zero in on the print settings for those images quicker?

I guess I'm still wondering how others do this. I recently sent a bunch of images off the printer without soft-proofing. About half turned out fine but the others required soft-proofing to get right. Maybe a print-oriented calibration profile would allow me to narrow the percentage that requires soft-proofing even further? Who prints drafts and then determines what to soft-proof? Or does everyone soft-proof everything before printing?
 
Hi Ed

I soft proof everything before printing, even to the extent of a round trip into Photoshop to print through Lightroom. This may be particularly necessary in my case because I have a lot of live music images with bright colours that can be out of gamut.

I also have an XRite Pulse so I can generate my own custom profiles. However, for the standard Epson papers, custom profiles are not really necessary. I got custom profiles some time ago from Dry Creek Photo and found only a very slight increase in gamut as the difference. With colour prints I sometimes print straight off the soft proof without further testing and it's usually OK. Monochrome I find usually requires test prints as well as soft proofing.

It will be interesting to see whether my new profiling settings appear to make any difference next time I come to print monochrome.

Do your tests on your monitor come out the same as my findings?

Regards,
Murray
 
Hi Murray,

My tests generally show a higher Delta E and seem to more consistent given a consistent change in parameters. I do not have extreme temperatures in my working area though either.

I am unable to find where you are getting the various DeltaE readings - I only see the one and it currently sits at 0.97 (Max I think - Average looks to be close to 0.54) ?
 
Hi Ed
(7) Intensity: 120cd/m2, Contrast: Default, [Average low light measurements]: On, Calibration steps: 32; Calibration priority: Best greyscale colour tracking
DeltaE Results: White Point: 0.62; Colour: 0.48 (0.43 including dark values) average, 0.70 maximum; Calibrated black level: 0.34.
All the results are in the Information window. The White point DeltaE is in the White Point box in the Summary tab. The colour tracking deltaE is in the Colour tracking window (average with and without dark values and maximum). The calibrated black level is not a DeltaE, it's just the black level, in the Display luminance box in the Summary tab.

I think all monitors must have an optimum age because it's difficult to calibrate them as low luminance values when new. So the DeltaEs probably improve for a while and then eventually decline. Yours is also a different monitor after all and no doubt has quite different characteristics.

Regards,
Murray
 
It seems obvious to me now, but I guess I wasn't paying close attention. I think I just assumed the Delta E value listed in the White Point box was a general calibration Delta E number. Now that I look closely, the Delta E on the Color Tracking tab is called a Grayscale Delta E. Thanks for prompting me to take a closer look! Sheesh.
 
I presume that the Colour tracking deltaE is more important.

The exercise has been useful to me as well, particularly trying Best greyscale colour tracking with default Contrast. I think they could probably improve the interface a little to sort out more clearly (1) Things you can change that may affect your monitor display; (2) other environmental settings and (3) measures of how good your profile is.
 
Top