The Official ChromaPure thread - Page 9

OK started over with `LCD Direct View' display type - Zeroed IREs and CMS - Calibrated from 100IRE down to 10IRE - Now I get the Red tinge like the ChromaPure. So Red Tinge with LCD Direct View and Yellow tinge with LCD (Led) Mode. Is there a Calibrating table that sits in the middle, between the two, that will give me the White Tinge? PS. Turned the Backlight, Contrast, Brightness and Colour down to 50 to prevent any possible over saturations. Konica Tech says that using the 2 Point versus the 10 Point, he didn't have the drifting setting values that he and I were getting with the 10 point calibration. Also the 2 Point cut down on the warm up problem.

Put back the BL to 69, Contrast to 84, Brightness to 52 and Colour to 48.
Setup to 100% white - the Only setting that affects this is the Gamma - all the other settings do not, excepting making the same screen brighter or darker. Pushing Gamma to Medium or High just enhances the tinge colour. If only the Gamma could have an Off setting. Completed another calibration. It appears that the ChromaPure is not setup for this type of LG yet. Possibly the new upgrade coming out? Calman's new upgrade? For now, do a complete calibration in Expert Mode, and watch the THX(User) mode.

After looking at all those East Coast snow scenes on FoxNewsHD via the TivoHD, there is no evidence of color tinging on my LG, the snow looks snow white. FoxHD PQ is very good BTW.

Somewhat OT, but FYI attached are the settings used by the (ISF trained) Calibrator when my LG was calibrated in October.
He used a Chroma 5 and the Sencore Color Pro 6000 software.

Although the post cal white balance shows IRE low and IRE high he actually used the 10 point calibration.

Note: The Tivo is connected to an input calibrated with ChromaPure, The ISF calibration is on an input connected to the Sony HDD250 HD DVR and are locked as ISF Day and ISF night. Keeping these for reference.

But this discussion really belongs to the xxLH90 calibration thread.

Yes, it does. A final non-specific comment: the level of difficulty reported points to either a faulty panel or operator error. It really sounds like the latter but if there are numerous credible reports of dramatic panel shift then it sounds like LG is making lemons.

In either case it certainly doesn't belong in a generic ChromaPure thread.

I was looking for an alternative for my problem when I came upon this ChromaPure calibration forum - put out the $200, only to find my Yellow Tinge had now turned into a Red tinge - Tom looked at my calibration results and stated they looked A-OK. I could see there being a problem with one LG but both? One a July build and the other a Nov build.

This is actually very simple. The display measures a nearly perfect white balance post-calibration. Yet, you claim to see a red tinge.

There are only three possible explanations for this:

1) Your subjective preferences and expectations are inconsistent with the standards by which the video material was mastered.
2) The meter you are using is defective and returning considerably inaccurate numbers.
3) Some aspect of the video performance other than white balance is causing this.

If it is (1), then there is nothing that can be done about this. If it is (2), the meter needs to be tested against a reference to determine if it is within spec. If it is (3), then you need to post xyY data for the color gamut so further recommendations can be made.

There is no fourth option--at least none I can think of.

Tinge problem only arises when White Balance is put on Warm. Going to change rooms, computer and 55LH90 tomorrow night while the wife's out of town to see if that changes things. I only have the one Chroma5 but it was sent back to Spectracal, who, after a weeks testing, said it was OK. Does the ChromaPure reference the tables in the Chroma5 Enhanced at all, or supply it's own reference tables? If it it's own tables, are any of the tables set up for the LG LED TVs and if so what Option settings get you the correct tables?

I'm just jumping in without looking back at past posts (for Display type/Display Model)... LOL


Uniformity.....?


I'm sure ChromaPure uses the tables/memories that X-Rite saved in your C5 (assuming you selected/CP is using the LCD table/memory)

That eliminates (2). I repeat, please post xyY data from a Color Gamut reading. All we have seen is grayscale data.

I'm confused (nothing abnormal about that though)....Nonetheless, when you say the tinge problem only occurs when white balance is set to warm....isn't that what one should expect? Warm is red......

My55LG0 has been calibrated with ChromaPure with white balance set to warm and does not show any red color tinge.

Well I understand you can calibrate it any which way you please...but it wasn't clear [to me] he calibrated using the warm setting.

From his post here :
http://www.avsforum.com/avs-vb/showp...&postcount=247


I use warm for White Balance but do not see a color tinge at all on my July build LG 55LH90

Well, without belaboring the point, I still didn't read where he calibrated the display in the warm mode. But if the problem only arises in that mode why use it then?

This is a long shot, but when calibrating the grayscale you should already have established the color temp setting you will be using and done at least a preliminary calibration of black level and white level. If you are calibrating grayscale in some cold temp setting (other than "warm") and then switching over to warm, I think your display would appear overly red.

Do not believe that is the case. When selecting the C5 in ChromaPure only Standard and CRT are listed whereas in CalMan Enhanced C5 is shown for a modified C5. But Tom can chime in on this.

Moreover, based on tests conducted by Tom Huffman, it is now my belief that the spectral content of CCFL vs LED is not so different as to materially affect grayscale calibration and especially the resulting PQ after calibration.
See here: http://www.chromapure.com/colorscience-myths.asp
Myth No 5.

This issue is still controversial and purveyors of calibration software and the modified C5 like Spectracal and now Sencore could welll see a commercial advantage here.

Bill Hergonson of CoastCalibration calibrated my LG in October using a C5 just re-certified and modified for the LED table by Sencore. Bill compared the calibration reports of the modified and previous (virgin) C5 and and concluded that the change would not show any discernable difference in PQ after calibration.

That said Tom Huffman may well decide to add the modified C5 in the next iteration of ChromaPure if that yields a commercial advantage.

I for now am very happy with ChromaPure and the standard C5 and have no intention of making a change.
Perhaps when my C5 needs to be re-certified I would elect to have the meter modified.

I think Tom is going out of his way to be helpful presumably because one of the issues raised was the possibility that ChromaPure might have some sort of defect relative to CalMAN and this is the ChromaPure thread.

However it's not the calibration 101 thread and it's not the help p5browne calibrate his panel thread. I always say why speculate when you can know and this is a rare case when it's all about measuring. So p5browne should post the relevant numbers (and avoid doing anything else) and people not familiar with the issue should read the rest of the p5browne calibration saga in the "Official" LG XXLH90 240hz LED User Setting/Calibration Forum thread.

I'll admit I have a vested interest here: I have a copy of ChromaPure and I'd like this thread to be useful to people interested in calibrating (not fiddling) displays using ChromaPure.

This is a very complex subject. We have looked at several LCDs, some LED backlit, some CCFL backlit. What we have found is that there is no consistent difference between the two that would allow us to create a single "LED" mode. For example, using the Chroma 5 and comparing against a reference instrument, Sony LEDs measure very differently than Samsung LEDs. In fact, Sony and Samsung CCFLs measure differently. We haven't looked at LGs yet. Plasmas and CRTs do not seem to suffer from this type of variability. Nor do measurements of front projectors, at least when taken from off-screen.

In our next release we are going to offer a Pro version of the Chroma 5 that wil have an individualized software-based calibration for a variety of different displays. On the basis of what we've seen, our position is that a single "LED" mode would not be useful. We may offer more specific modes tied to specific manufacturers, but we haven't formalized our plans yet.

However, whatever we decide upon will offer significantly enhanced accuracy for the Chroma 5 and Klein K-10. We may also offer something for the Display 2 as well. We hope to have this available sometime next month.

Yes, X-Rite only uses 2 of the 4 tables in the Chroma 5's so it makes sense you would only have Standard and CRT options in ChromaPure (and in CalMAN when a non enhanced Chroma 5 is used).

I'm assuming in ChromaPure you are using Standard for your Display..

I believe this is what Cliff is doing at PL. I think he has a Library available for his C5's (and customers)..

Wasn't Sencore doing this with the OTC-1000 as well?

Yes indeed.

But I certainly wish you the best and look forward to new announcements.
Your colorscience FAQ at ChromaPure.com is still my favorite source for learning.

Well, it is based on the assumption that there is some consistency among house implementations of LCD technology.

It would obviously be impractical and prohibitively expensive to create offsets for every model for every year, even from a single manufacturer, much less from two or three. But creating offsets from prominent examples of both CCFL and LED displays for 2 or 3 manufacturers would be, I think, doable and useful.

But this is why I said that this is a complex subject. Even if we restrict ourselves to LED-backlit displays only, there are edge-lit implementations, back-lit (local dimming) implementations, in both white and RGB versions. There is no reason to assume, indeed our experience has shown otherwise, that all of these implementations measure the same or even similarly.

Should I wait till next month purchase chromapure?

That's entirely up to you. Existing customers will have an upgrade option when this becomes available.

Finally got to calibrate my July build - totally different beast than the November build! How do you take into account for your Tables, that Manufacturer's maybe using the same Model # but over the course of that model's life, it may undergo a vast change in the electronics inside that the last will not even look like the first. Only the outside looks the same. The parameters as well may be totally different. Oh yes, about the tinge problem - didn't see it last night on the wife's TV - the difference - 10 feet back on her computer doing the calibration. When I do mine, my face is only 2 to3 feet from the screen.

I've done a quick and incomplete run using my VideoEq. The current tool involves a great deal of clicking so I ran out of energy before I ran out of work. All the changes were made in the VEq.

The attached file is a ChromaPure xml session (.calx).

 

2010-02-13.txt 17.3876953125k . file

Performance was pretty good to begin with, but there's not a huge change.

 

CalibrationBodosom.pdf 135.1005859375k . file

The issue is not whether the display looks different, but the precise profile of the colorimeter's readings of the display. We are working on the assumption that their will be some consistency within each manufacturer regarding the type of LCD and backlighting elements they use.

Well I was just messing around. I focused much of my attention on magenta. Rocky Horror and all.

I also noticed that although I selected fL I have cd/m2 for Y.