Basic Guide to Color Calibration using a CMS (updated and enhanced) - Page 11

Anybody have a spreadsheet for Luv? I thought I saw one before but I don't remember where.

my LCH, LUV spreadsheet is in the color charts for dummies sticky thread

Thanks, Kras.

Perhaps a silly question, but in addition to the Gray Scale and Color adjustments outlined in this thread, what other adjustments should be made for a "complete" calibration. For example, I *presume* that sharpness would need to be calibrated. Is there anything else?

Tom - Your grayscale numbers should scale based upon an assumed gamma curve if folks want to use it for GS. Otherwise, this gives similar results to dEuv.

I would recommend that folks buy CalMAN, of course.

I am continuing to learn how to use HCFR with the i1 to calibrate my Epson 810, and each day I make more progress by reading many of the posts in this forum.

As with many of todays projectors, mine overshoots the HD color gamut by quite a bit, and the decoder is off a bit too in blue. The Epson has a pseudo CMS in that I can adjust xy but not Y.

Here is the before:



The fact that the red primary is off seems to have the most affect on the sunburned faces I have been seeing....bringing the Red Hue in line seems to have solved that. Now the faces look much more natural.

My first attempt I moved the primaries and secondaries to their expected locations:



As expected, Y was now way off, especially for green. So I did some more reading here on the forum, and the consensus seemed to be that it's better to have a large gamut with correct Y, than to have perfect xy. That is, adjust hue using the CMS and leave the saturation alone..or at very least don't use it to desaturate. So went back and did a third calibration:



This seemed to be a pretty decent compromise between proper accuracy and luminance. Y for blue is still pretty low, but I don't seem to be able to do much about it ... so I guess I can't worry about it too much.

However, you can see that cyan is not actually on the line between blue and green and lies outside the gamut defined by the primares...what are the side affects of this when the secondaries don't lie on the leg of the triangle?

In the case of cyan, I don't have any more ability to reduce the saturation in the CMS even if I wanted to...the only way to get it back onto the leg of the triangle is to move green back to where it was pre-calibration. Even if I did get cyan to lie on that line (in the last graph), its luminance would be too low. Luminance aside for a moment, I wonder if I could use the global color/saturation control to shrink my gamut to get cyan to line up, and then increase the saturation of the remaining colors to get them back to where they need to be.

Any thoughts are highly appreciated.

Patrick

Great spreadsheet. I think there might be an error though...assuming that if I use the same values for pre and post calibration that the dE values should be the same. Currently they are not equal when using the same xxY for pre and post.

The error appears to be in Column S starting at Row 22. The values for Sl are all shown as 1. When comparing these values to the post column AQ Row 22, which are 1,2,3,4,5,6.

PLincoln,
I suspect you may have altered the spreadsheet. When I downloaded it and just did a simple copy and paste so the before and after have the same x,y,Y values, I get the exact same dE's. Since you are talking about the columns that have been hidden in the original spreadsheet, it looks like you may have changed the original by mistake.

Sorry...I was referring to these spreadsheets. I see now there are multiple versions.

These two I get different values for pre/post
http://www.avsforum.com/avs-vb/attac...5&d=1190663045
http://www.avsforum.com/avs-vb/attac...4&d=1190676880

I enabled the hidden cells because the values didn't match. Not to alter the sheet...

I got very slow response reading from the Eye1 when placing the sensor facing the ceiling mount fron projector (about 6 inches from the screen) for measured the 30IRE. Please let me know where I should place the Eyeone sensor.

Many thanks.

Using your delta-eCIELAB_1994 spreadsheet from 10/4/07, I can't seem to make sense of Saturation dE and Hue dE. Could you maybe explain those values or give some clue of where to look for more information? I tried using google with the intent to check the math, but couldn't find anything. I just don't get how they would be useful the way they seem to work now. By playing with HSV colors in windows I think I might have a better grasp on how lightness, hue, and saturation relate to xyY, but I still don't get how the numbers I'm seeing in the Saturation dE and Hue dE would relate.

Okay, they add up to 100%. I get it now. That makes sense, thanks.

This is a good little nugget. Thanks, Tom.

Hi all,


Is it safe to assume that all the Sharp XV-Z series (wearing the 1080p dmd panel) have a proper CMS? I am particularly thinking the XV-Z21000. Does anyone know if this model has a proper CMS?

I was also thinking the BenQ W9000 ( I know this one has got it and it's quite cheaper than the sharp XV-Z21000-not by a huge amount-, but the sharp has a better gen panel DMD DC3 instead of the DC2 of the W9000 and it gives a big number in nits even in a 120" screeen).

Many thanks,

kopa13

Hi Tom,

Thanks for this. I quickly browsed through your first post abd did see the Sharp Models but I didn't see the Z21000. But I assume that this won't be the exception regarding having a proper CMS.

Another quick question (and please accept my apologies if the answer to this has been covered already), do you know if the VPL-VW60 has a CMS? I remember in another post from you in another thread that you weren't happy with the SXRD technology regarding pri/sec colours and the ability to correct them.

Regards,

Kopa13

Many thanks for your help Tom.

I am looking for a FPJ that knowing it will not be perfect out of the box, regarding video, it will allow me as many corrections as possible. So please bare with me (and apologies for cloging up the thread) for one more question which will help me in great deal.

Which of the below stated HD models, from a Calibration friendly point of view for all major (video) aspects of it ie Grey Scaling tracking, Color Decoding and Color Gamut abilities, would anyone be better to live with:

Sharp XV-Z20000, BenQ W9000, Infocus IN82-(with the updated firmware), JVC RS1 (with the latest fw) or the Sony VPL-VW60 (this last one is listed just for the bargain price of ~$3100 I've found for it new)

Many thanks in advance,

Kopa13

One thing I still don't really understand is the HD vs SD colorspace issue. When I measure DVE HD I get one set of Y values for color, but if I measure GetGray I get a second set of Y values for red, green, and blue. The disks match for gray. Is that behavior is correct or not? If my HD DVD player was correctly upconverting, would the Y measurements for color bars from SD video match the color bars from HD video?

By Bear5k's suggestion that the luminance calculator can be more accurate, I'm inclined to use the calculator. So when I use the calculator, does it matter if I'm playing back HD or SD video to set my colors? I would have to think because I'm seeing different Y values for SD and HD color that only one or the other could be correctly set this way.

I know that the reference values for SD and HD show different Y values for the colors. Is there some reason why there aren't two different Y values in the calculator depending on if the video is HD or SD? The impression I get is that based on a set of primary locations that there's only one set of secondary locations and luminance values with D65 white. That would explain the SD standard, the HD standard, and the luminance calculator; the only odd-duck would be what the Y measurements for SD and HD video from my HD DVD player. If I'm using the luminance calculator, should I just stick to using the Y values from HD video so that there shouldn't be any colorspace issues possibly coming from the player? Or instead of using video at all, should I maybe just use my computer with RGB for setting the color decoder?

Hi there

First of all, thanks to Tom for the tutorial. I'll try to give this a whirl with my i1, ColourHCFR and my Pioneer LX608D when I get a free weekend. But first I have some initial questions:

1. I was intrigued by the discussion around CMS vs CRS. My new Pioneer Kuro includes the following adjustments with respect to colour:

Contrast
Brightness
Colour
Tint

Colour Temp (manual):
R High -/+
G High -/+
B High -/+
R Low -/+
B Low -/+
G Low -/+

Colour space (1=vivid, 2=standard)

Colour Management:
R - = closer to magenta + = closer to yellow
Y etc
G
C
B
M

I can with the help of ColorHCFR move primaries, secondaries and greyscale balance.

My assuption is that this is, to use the terms you've used above, Colour Management and not Colour Replacement. Is this correct?

(FYI I have a choice of three gamma settings. Apparently access to the service menu provides access to a 9-pt gamma function for the User mode but I'm not venturing anywhere near that menu for a long time.)

2. I'm perplexed by the SMPTE-C vs Rec 709 debate. Perhaps you can confirm if I have understood the debate correctly. SD material is encoded to SMPTE-C and ideally is displayed with SMPTE-C primaries etc. If displayed with a display calibrated for Rec709 it will look slightly oversaturated (OK not really oversaturated but darker). HD material should be encoded to Rec 709 but could well have been proofed on a SMPTE-C monitor in which case, again, when displayed on a Rec709 calibrated display will look darker than intended. Properly encoded HD material, which we currently question the availability of, will look lighter on a SMPTE-C calibrated display. If a DVD player upscales SD material who knows whether it is including a colour transform from SMPTE-C to Rec709.

Currently I have an old Theta DaVid which, of course, doesn't upscale anything. But I am interested in the new Pioneer BD player when it's available and so I am interested in whether I should calibrate to SMPTE-C or Rec709. I suspect the former, at least until my HD library grows considerably?

3. Lastly, should I come to this thread with questions about the results of my calibration efforts using Tom's tutorial or raise them in the ColorHCFR Calibration (Post your files here) thread:

http://www.avsforum.com/avs-vb/showthread.php?t=786966

in order to get your esteemed help?


Many thanks for all the effort you guys have put in to writing tutorials and helping me and others.

Regards

Steve

If the color encoding/decoding is done correctly the color bars should match (assuming they are both 75% or 100% color bars, with the same gray amplitude as color amplitude, same saturation, etc).

I don't disagree that SD vs. HD is much less an issue compared to the trend of displays with overblown color - but your CIE chart is very misleading as to color differences you can perceive. Plot the dL, dC, dH, dE metrics between SD vs. HD - you realize the differences are very perceivable. Easy to do in any of the spreadsheets floating around (including mine) - just make HD your target and SD your measure and see what differences you come up with. Obviously the difference is only worth worrying about on a display with good color to begin with.

dE(uv) from your spreadsheet between SMPTE-C and BT.709 RGBYCM is
12.6
6.7
4.3
0.9
5.7
9.8