or Connect
AVS › AVS Forum › Display Devices › Display Calibration › HCFR - Open source projector and display calibration software
New Posts  All Forums:Forum Nav:

HCFR - Open source projector and display calibration software - Page 63

post #1861 of 3436
Quote:
Originally Posted by Make73 View Post

Here´s that link about variables between tested units:
http://www.drycreekphoto.com/Learn/Calibration/MonitorCalibrationHardware.html

Many reported that there´s still some red tint with whitebalance after calibration with i1D3, if you can get rid of that with matrix I don´t see any reason why not to use.
Those are here that you can try them, you don´t have to use them. If spectral sample feels better then you should use it smile.gif

If the spectral file is not producing good results by eye, it is either your eyes or the spectral file.

The i1D3 Pro is meant to be a great colorimeter. HCFR interprets those signals to display the results in meaningful ways.

If the results look good in graphs, but not in real life (not personal preference), and it can be proven to be a bad calibration, then we are all wasting our time here.

With HCFR 2 and other sensors, these extra matrices and spectral files were not used. I recall great results (to me) with a LT2. I am embarking down the same route with a new sensor and HCFR 3, and I don't see the fundamental principles changing. But if my results do, I will be left looking for the root cause too.

Something is seriously amiss if red tints appear.

I would hope others with far more experience than I, will chime in and put this to bed.
post #1862 of 3436
Quote:
Originally Posted by zoyd View Post

no, don't use the D3 correction matrix for the D2, results would not be good. Assuming you have a plasma, I would use the plasma spectral sample (.ccss) correction option.

Thank you.

Yes, I have a Plasma. In HCFR v3.0.4.2 I can not select the type of plasma display. The only selectable items are CRT or LCD. I assume that I have to select CRT.
In the line "Reading Type" I select contact, while the line "Spectral Sample" is not selectable as it remains gray.
post #1863 of 3436
novice question 1

Anyhow I can get target values for primaries and secondaries colors based on my current White lum level?
Any table that calculates those and even better to recalculate upon White level re read?
I'm about target Y per color like in Gray Scale table for gamma Y. Also, would be nice to have gamma Y been recalculated upon any readings for 100% Gray (White), not just by running gray scale, but on demand.

Also, would be nice to see target x and y per colors. So far I use CIE Chart to get those x/y reference values.

Thanks.
post #1864 of 3436
post #1865 of 3436
Quote:
Originally Posted by realzven View Post

Mike look at this http://www.avsforum.com/attachments/66073

thanks
I thought that something like this is integrated into HCFR that I can't find.
post #1866 of 3436
Quote:
Originally Posted by wl1 View Post

If the spectral file is not producing good results by eye, it is either your eyes or the spectral file.
I meant without any correction, you can find such of comments also from this thread.
I calibrated whitebalance without matrix and with matrix and switching between two modes there you can notice that fine red tint at grayscale without any correction.
Quote:
Originally Posted by wl1 View Post

If the results look good in graphs, but not in real life (not personal preference), and it can be proven to be a bad calibration, then we are all wasting our time here
Who said so?
If you have reference meter and all other things are correct (like which patterns are best for your display, with plasma you can get different result with different patterns) then you should be fine.
I´m just rookie but already noticed in few years that calibration is not just like on/off switch, there´s so much things what affects to result.
You´re from UK, David from HDTVtest and Stephen from AVForums get different results with same set and same tools and those guys are professionals.
Quote:
Originally Posted by wl1 View Post

With HCFR 2 and other sensors, these extra matrices and spectral files were not used. I recall great results (to me) with a LT2. I am embarking down the same route with a new sensor and HCFR 3, and I don't see the fundamental principles changing. But if my results do, I will be left looking for the root cause too.
I remember that there was possibility to use matrix and choose plasma for display type (spectral sample).
At least I made correction file for Display 2 for using HCFR 2 when I sold that.
Quote:
Originally Posted by wl1 View Post

Something is seriously amiss if red tints appear.
You can examine those matrixes which are posted here and especially which are delivered with i1 Diplay Pro meter, so customer paid for calibrating it against reference meter. You can notice that most of them corrects red reading of i1D3.
Also it´s good to remember that there´s different revisions of same meter, there could be and probably is differences. Using some others matrix may result to worse accuracy than without it.
That all is for DIY calibrations, if you take payment for calibration you should also have tools and knowledge like pro.
Edited by Make73 - 5/9/13 at 2:11am
post #1867 of 3436
Quote:
Originally Posted by Make73 View Post



I meant without any correction, you can find such of comments also from this thread.
I calibrated whitebalance without matrix and with matrix and switching between two modes there you can notice that fine red tint at grayscale without any correction.

Who said so?
If you have reference meter and all other things are correct (like which patterns are best for your display, with plasma you can get different result with different patterns) then you should be fine.
I´m just rookie but already noticed in few years that caibration is not just like on/off switch, there´s so much things what affects to result.
You´re from UK, David from HDTVtest and Stephen from AVForums get different results with same set and same tools and those guys are professionals.

I remember that there was possibility to use matrix and choose plasma for display type (spectral sample).
At least I made correction file for Display 2 for using HCFR 2 when I sold that.

You can examine those matrixes which are posted here and especially which are delivered with i1 Diplay Pro meter, so customer paid for calibrating it against reference meter. You can notice that most of them corrects red reading of i1D3.
Also it´s good to remember that there´s different revisions of same meter, there could be and probably is differences. Using some others matrix may result to worse accuracy than without it.
That all is for DIY calibrations, if you take payment for calibration you should also have tools and knowledge like pro.

I am a rookie too.  And keen to learn more. Diyer, not a Pro - payment wise :-)

 

My understanding is that the Spectral Files compensate the "already calibrated @xrite" i1D3Pro for the technology display it is being used on.  And that should be enough.

 

If you are judging errors by eye, eg, on tints, then the calibration methods employed did not get you to REC709.  Or your eyes are seeing things wrong.  Using offsets, and then judging by eye, to say the measurements are better, does not make sense to me.

 

These are my opinions by the way, so open to others greater experience to change my mind.

 

With regards using an i1Pro to profile other sensors (LT2 or i1D3 Pro), the data in the link you posted yesterday showed the i1Pro to be far from perfect.  I understand this device requires frequent calibration (annually) and during test runs.  

 

I am merely suggesting that with a capable measuring system, these variables of environment, calibrator, meter etc should not affect the end result (outside of the acceptable tolerances).

 

Ref:-  "David from HDTVtest and Stephen from AVForums get different results with same set and same tools and those guys are professionals."

I am not familiar with the exact tests you refer to - but when you say "same set and tools" - exactly the same units, or just same model type?  Done at similar times, same environment, same software tools and methods deployed?  And are the errors outside of the target (Delta 3). I suspect the results are well within this.

 

I have a Pioneer plasma, and it may not suffer the same way with window sizes as more modern units (with ABL). I am reading the threads on this forum that mention this to take the best advice, before seeing for myself.


Edited by wl1 - 5/9/13 at 1:35am
post #1868 of 3436
Quote:
Originally Posted by wl1 View Post

My understanding is that the Spectral Files compensate the "already calibrated @xrite" i1D3Pro for the technology display it is being used on.  And that should be enough.

Forget that matrix what I used, it´s from Argyll database and yes it is made against Pro1 with Panasonic plasma. I don´t know is that Pro1 calibrated or not.

When you buy i1DPro with matrix you can get different tables for different displays so there´s no need for spectral samples if I got it right, it´s included at matrix, you just choose right one for your display.
Those matrixes is what I´m talking about (kjgarrison´s, blaugarna´s and spongebob´s matrixes, all those guys paid for calibration(matrix) of i1 Display Pro)

What I meant for getting different results between HDTVtest and AVForums is like MLL with same meter (Klein K10-A) . There should not be differences like that between same TV model.

I end this discussion here smile.gif
Edited by Make73 - 5/9/13 at 2:33am
post #1869 of 3436
My calibration saga (or nobody here but me and my Munki)..

So two weeks ago I started to calibrate my 2013 Panasonic ST60 series plasma.
I had adjusted the greyscale by eye to match my old Pioneer plasma (first gen Kuro) with both sets side by side.
They are pretty close but the Panny has a lot of variations as you step thru 5% windows (from both Getgrey and AVS discs).
This is really evident when you put up reverse grey ramps but at 50% window they are nearly identical.
I used my DPT-94 & HCFR to check the old Pioneer and it was more than within 1% from a year or so when last checked with a very flat greyscale etc.
I switched over to the Panasonic and got what I expected and my "cal by eye" was pretty good except for the variations I mentioned above. Luckily this set has both 10% RGB and Gamma controls.
Before I had a chance to do anything the DTP-94 died forever.
Last weeek I got a ColorMunki display to replace the DTP-94.
I re-ran the Pioneer set (with spectral sample set to EDS plasma) and the results were within 1-2% Vs the DTP-94.
Switching over to the Panasonc set things changed dramatically with blue increasing 10-12% (same EDS plasma spectral setting) WTF ??
REsults were even worse at lower levels too.
I have been reading thru this entire thread and see the many examples of issues with plasmas and the i1 display/colormunki etc.
Can the spectrum of the new 2013 Panasonics be that different ??
Any ideas here ??
I have not yet tried the 3 or 4 cal factors for older panasonic set that others (Zoyd was one of them) but I wonder if any of those apply to this panel given the huge difference I see between readings (pioneer Vs pannny).
Any thoughts would be greatly appreciated...

Todd
post #1870 of 3436
Todd, its been reported by D-Nice and Chad B that the 2013 Panny Plasma behave like the 2011 series.

For reasons like you can have a perfect calibration one day and you go and measure a couple of days later and you can have different results in delta E...

Panasonics also have aggressive abl behavior and you have to find a patch size(window ire) to be in a good range for a decent calibration.

I have a Panny GT30 and depending on the patch size, I notice that the larger the patch the more blue I have to add.

So through test,trials and tribulation I have found the 4% patch size to worked best on my set because the rgb's and gamma are more linear and the blue is in about the right range.

So my advice to you is to keep testing different patch sizes off one the free calibration disc found here in this forum until your satisfied.
post #1871 of 3436
EODIS3 i1Display Pro

Does HCFR supports this meter? Is this the same meter referred to as the i1D3?
post #1872 of 3436
Quote:
Originally Posted by PeterLewis View Post


Todd, its been reported by D-Nice and Chad B that the 2013 Panny Plasma behave like the 2011 series.

Panasonics also have aggressive abl behavior and you have to find a patch size(window ire) to be in a good range for a decent calibration.

I have a Panny GT30 and depending on the patch size, I notice that the larger the patch the more blue I have to add.
.

From his review of the ST60 set, Chad B said these sets do not have the bad abl behavior that earlier sets do (or not as aggressive anyway).
I don't think it's safe to say the new 2013 ST panels are the same Vs even the 2012 sets since the black level is much better (equal to 2012 VT panels).
Panasonic changed something ??

I'm using small APL windows for all my tests and have tested the ABL with full field grey patterns and it does not kick in until about 75-80% level (almost the same Vs my Pioneer at a little higher level say 80%). It is a 50ST60 inch panel too and I think the 50 inch behaves better.
This is not the issue here anyway.

What I'm seeing is two radically different results measured on two different plasma panels with nearly identical grey levels (to the naked eye).
We are not talking small dE errors since those errors are off the charts with this meter and the new Panasonic panels.
This was not the case in the short time I had with my old DTP-94 and as I stated those measurements were quite close.

An example would be at 80% white (small window) where the picture shifts slightly red (clearly visible) but the meter is saying I'm 12% too high on blue ??
The same pattern with the same meter with the same settings measures nearly perfect with the pioneer...

I will try the Panasonic cal factors this weekend but somehow don't think they will compensate that much ?

Todd
post #1873 of 3436
Quote:
Originally Posted by AFryia View Post

EODIS3 i1Display Pro

Does HCFR supports this meter? Is this the same meter referred to as the i1D3?

Post 1 has a summary of the thread and this question is linked in Post 678.

 

"The i1 Display Pro and i1D3 are the same meter, the company that builds it calls it the Display Pro while other companies that sell the OEM version call it the Display 3. The i1Pro (not Display Pro) is a spectro and the Display Pro is a colorimeter."

post #1874 of 3436
Quote:
Originally Posted by todd95008 View Post

My calibration saga (or nobody here but me and my Munki)..

I re-ran the Pioneer set (with spectral sample set to EDS plasma) and the results were within 1-2% Vs the DTP-94.
Switching over to the Panasonc set things changed dramatically with blue increasing 10-12% (same EDS plasma spectral setting) WTF ??
REsults were even worse at lower levels too.

Can the spectrum of the new 2013 Panasonics be that different ??


Todd

Are you suggesting that the Spectral Sample file for Plasma may not be accurate for every Plasma?  As in, you feel it is OK for Pioneer, but Panasonic Plasma are not being measured correctly? 

 

If you are correct, it would explain a lot.  Certainly, this type of correction for different technologies, or even with the same technologies, wasn't apperent with other meters.  

 

Have you tried contacting Xrite?  Specifically asking about how their EDR Spectral file, how it was developed, what level of testing was done etc?  Obviously, they will not help with using their meter with HCFR 3 though.

post #1875 of 3436
How do I change BT.1886 default gamma value? HCFR has the BT.1886 reference gamma ~ 2.2 and can't be adjustable.
Graeme Gill said BT.1886 has the default value ~ 2.4.
post #1876 of 3436
Thread Starter 
The reference gamma number is greyed out because BT.1886 is hard coded to 2.4 as it should be.
post #1877 of 3436
zoyd, i want to buy a new laptop, there are no issue with windows 8 (32 or 64 bits) and HCFR 3.0.4.2 ?

thanks wink.gif
post #1878 of 3436
Another question : to profile a D3 with my rev D it's impossible in the 3.0.4.2., it's necessary to profile in the old v2 version ?

thanks again smile.gifwink.gif
post #1879 of 3436
Thread Starter 
No one has reported any errors in Windows 8 but I haven't tested myself. Yes, you can profile in 3.0.4.2 with either "Create using existing measures" or "Create using simultaneous measures"
post #1880 of 3436
the Rev D works with the 3.0.4.2 ? i thought the i1 PRO did not work with the new v3 colorHCFR version (3.0.4.2)...
post #1881 of 3436
Thread Starter 
Quote:
Originally Posted by realzven View Post

the Rev D works with the 3.0.4.2 ? i thought the i1 PRO did not work with the new v3 colorHCFR version (3.0.4.2)...

It didn't work in 3.0.4.0, works fine now as well as simultaneous meters
post #1882 of 3436
Quote:
Originally Posted by zoyd View Post

It didn't work in 3.0.4.0, works fine now as well as simultaneous meters

Thanks for restoring i1Pro compatibility, deeply appreciated!
post #1883 of 3436
Quote:
Originally Posted by realzven View Post

zoyd, i want to buy a new laptop, there are no issue with windows 8 (32 or 64 bits) and HCFR 3.0.4.2 ?

thanks wink.gif

I run HCFR 3.0.4.2 in Win8 64 bit - works Ok
post #1884 of 3436
Quote:
Originally Posted by zoyd View Post

It didn't work in 3.0.4.0, works fine now as well as simultaneous meters
thanks zoyd, you're da man smile.gifwink.gif
post #1885 of 3436
Quote:
Originally Posted by mikek753 View Post

I run HCFR 3.0.4.2 in Win8 64 bit - works Ok

thanks for the reply, i 'll buy a new laptop this weekend (2 pc in house , 2 broken mad.gif )
post #1886 of 3436
Quote:
Originally Posted by zoyd View Post

The reference gamma number is greyed out because BT.1886 is hard coded to 2.4 as it should be.
here is my display calibrated to BT.1886 gamma




Average gamma is 2.4 but the reference gamma ~ 2.2. Am I doing it wrong ?
post #1887 of 3436
Quote:
Originally Posted by wl1 View Post

Are you suggesting that the Spectral Sample file for Plasma may not be accurate for every Plasma?  As in, you feel it is OK for Pioneer, but Panasonic Plasma are not being measured correctly? 

If you are correct, it would explain a lot.  Certainly, this type of correction for different technologies, or even with the same technologies, wasn't apperent with other meters.  

Have you tried contacting Xrite?  Specifically asking about how their EDR Spectral file, how it was developed, what level of testing was done etc?  Obviously, they will not help with using their meter with HCFR 3 though.
Yes, that is what I'm suggesting.
It has already been established that 2011-2012 panasonic plasmas have a different spectral profile (2013 panels could be different even from that).
I don't think the spectral sample was from X-Rite.
I also don't think the ColorMunki software has any plasma profiles ?
I did not create or load this into HCFR as it was aways there with this meter.
This is not the generic sample on the meter page but the one on the next Argyll page (EDS plasma ....).
There are about 6 or 7 of them mostly LCD for different backlights and only this one for plasma.
I don't know where that specral sample comes from or what manufacture's panels.

Zoyd, any thoughts ??
post #1888 of 3436
Quote:
Originally Posted by todd95008 View Post

Yes, that is what I'm suggesting.
It has already been established that 2011-2012 panasonic plasmas have a different spectral profile (2013 panels could be different even from that).
I don't think the spectral sample was from X-Rite.
I also don't think the ColorMunki software has any plasma profiles ?
I did not create or load this into HCFR as it was aways there with this meter.
This is not the generic sample on the meter page but the one on the next Argyll page (EDS plasma ....).
There are about 6 or 7 of them mostly LCD for different backlights and only this one for plasma.
I don't know where that specral sample comes from or what manufacture's panels.

Zoyd, any thoughts ??

I believe the options you refer to are derived from the .edr files found within the iProfiler software that ships with the i1D3Pro. It's in the Calibration folder, IIRC. Post 1 has the links to several discussions on this thread about it (ccss, edr etc) if you want more info.

I contacted xrite to ask how the Plasma file was derived, and got an unsatisfactory response. If they come back with anything meaningful, i will post it.

It looks like xrite have developed a Plasma spectral file for use with their software - but it's crucial to know if that is good enough for the vast majority of displays (Samsung, Pioneer, Panasonic), or whether it just brings it closer than not using it at all.

Zoyd in one post, did say, its better to use a Spectral file (as you are doing) if you cannot profile your meter against a reference meter.

The options would appear to try the correction matrices, and compare which works best for you. But this takes the science out of it for me, so I could end up frustrated with this.
post #1889 of 3436
zoyd, if you continue developing this would it be possible to remove (or option to disble at parameters) saturation 0% measure or is there some special reason why that´s there?
Edited by Make73 - 5/11/13 at 1:28am
post #1890 of 3436
My official X-Rite i1pro Rev D (from "X-Rite EODUV Eye-One Design UV" bundle) continues to be non-functional in HCFR 3.0.4.2. Selecting "GretagMacbeth i1pro" causes HCFR 3.0.4.2 to hang unresponsive for a few seconds, then this dialog pops. It doesn't let me type anything for meter name, and any selections made are not saved after clicking okay. Clicking "Calibrate Meter" has no effect. Attempting to take any readings hangs or crashes HFCR 3.0.4.2. Alternately selecting "Eye One" as the meter results in "Unknown Error" everywhere.

Is there any one actively working on fixing i1pro support? Last version which functioned somewhat was the initial HCFR 3.0.0.0 version (Sensor: Argyll | GUI shows: I1Pro), but High-Res mode resulted in incorrect readings compared to Argyll CMS (standard mode was better) and it would still sporadically throw an "Incorrect driver" error between readings/calibrations requiring HCFR to be restarted.

For this reason I've been continuing to use original HCFR 2.1 which was stable with the official x-rite driver, but it'd be nice to see this HCFR 3.0 fork get proper i1pro support with the Argyll driver some day.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › HCFR - Open source projector and display calibration software